642 436 Cvoice

Cisco Voice Over IP (CVOICE)
(642-‐436 CVOICE 6.0)

CiscoVoiceGuru.com -‐ Guru Guide

Prepared by Matthew Berry, CCIE #26721
For more Cisco Voice study resources visit www.ciscovoiceguru.com
Prepared by Matthew Berry, CCIE #26721 1 For more Cisco Voice study resources visit www.ciscovoiceguru.com

1: Introducing Voice over IP Networks • SCCP:

o Recent versions of IOS gateways also support SCCP controlled FXS ports

VOIP FUNDAMENTALS

Components of VoIP:
• IP Phones: Provide IP endpoints for voice communication.
• Gatekeeper: Provides Call Admission Control (CAC), bandwidth control and

management, and address translation.
• Gateway: Translation between VoIP and non-‐VoIP networks (PSTN) and physical access VoIP Service Considerations
for local analog and digital voice devices.
• Multipoint Control Unit (MCU): Provides real-‐time connectivity for participants in
multiple locations to attend the same videoconference or meeting.
• Call agent: A call agent provides call control for IP phones, CAC, bandwidth control and
management, and address translation. Unlike a gatekeeper. Typically runs on a server
platform (CUCM)
• Application servers: Provide services such as voice mail, unified messaging, and Cisco
Communications Manager Attendant Console.
• Videoconference station: A videoconference station provides access for end-‐user
participation in videoconferencing.

VoIP Functions:
• Signaling: Capability to generate and exchange control information that will be used to
establish, monitor, and release connections between two endpoints.
• Database Services: Query a database to determine whether the call can be placed or
information can be made available.
• Bearer Control: Supervises appropriate call connect and call disconnect signaling be
passed between end devices. Carried by SS7 in the PSTN network. SIP, H.323, H.248, or
MGCP within the IP network.
• Codecs: Provide the coding and decoding translation between analog and digital
facilities

VoIP Signaling Protocols: H.323, MGCP, SCCP, SIP

H.323 Umbrella – most widely deployed standards-‐based voice/video standard
• H.225 Call Signaling: establish a connection between two H.323 endpoints, exchanging
H.225 protocol messages on call-‐signaling channel.
• H.225 Registration, Admission, and Status (RAS): used to perform registration, Real-‐Time Transport Protocol
admission control, bandwidth changes, status, and disengage procedures between
endpoints and gatekeepers. RAS channel is used to exchange RAS messages. • Typically runs on top of UDP, utilizing its multiplexing and checksum services
• H.245 Control Signaling: used to exchange end-‐to-‐end control messages governing the • Only standard RTP has is the use of even ports, next higher odd port for RTCP
operation of an H.323 endpoint. • Dynamic port range: 16384-‐32767 >> makes it difficult to traverse firewalls
o Capabilities exchange • Often used for unicast sessions, but initially designed for multicast use
o Open/Close of logical channels used to carry media streams o Roles: sender, receive, translator (multicast), mixer (multicast)
o Flow-‐control messages • Enables destination device to reorder and retime the voice packets before playback
o General commands and indications
• Audio Codecs: All H.323 terminals must have at least one audio codec supported RTP Control Protocol
• Video Codecs: Encodes/decodes video for transmission. Any H.323 terminal providing • Provides out-‐of-‐band control information for an RTP flow
voice communications must support video encoding/decoding as specified in the ITU-‐T • Does not transport any multimedia data itself >> used for QOS reporting
H.261 recommendation. • Types of RTCP packets:
o Sender report
Session Initiation Protocol (SIP) o Receiver report
• Standard based on the logic of the World wide Web >> very easy to implement o Source description
• CUCM does not control SIP devices >> doe not register with call agent o Goodbye
o Application-‐specific
Note: Since H.323/SIP are peer-‐to-‐peer protocols, H.323/SIP gateways are not registers with CUCM. An • Mechanism for hosts involved in RTP session to exchange information about monitoring
IP addressed is configured to confirmation that communication is possible. and controlling the session.
• Feedback for NTP synchronization
MGCP • Separate flow from RTP >> 4 ports per two-‐way conversion, 2 ports per direction
• Plain-‐text protocol used by call-‐control devices to manage IP telephony gateways
• CUCM controls individual voice ports on an MGCP gateway Compressed RTP (cRTP)
• UDP port 2427 • Header portion of RTP is significantly larger than the data portion
• cRTP compresses the IP/UDP/RTP headers from 40 bytes to approximately 2-‐4 bytes
PRI/BRI Backhaul o IP (20 bytes) + UDP (8 bytes) + RTP (12 bytes) = 40 bytes
• Internal interface between call agent (CUCM) and Cisco gateways o RFC 2508 – Compressing IP/UDP/RTP headers for low-‐speed serial links
• Separate channel for backhauling signaling information o RFC 2509 – IP Header Compressions over PPP
• PRI backhaul forwards PRI Layer 3 (Q.931) signaling via TCP connections o RFC 3545 – Enhanced Compressed RTP (ECRTP) for Links with High Delay,
Packet Loss and Reordering
Skinny Client Control Protocol
• Cisco IP phones that use SCCP can coexist in an H.323 environment Why/When to use cRTP
• Suppresses the sending of redundant information and caches it in the gateways at each
Comparing VoIP Signaling Protocols end of a link, sending reduced headers, then reassembling the full header upon arrival.
• H.323: suite of protocols >> complex configuration • Reduction in overhead results in a delay reduction
o ISDN protocols (Q.921, Q.931) only used on ISDN link to PSTN • Not recommended to use cRTP on any high-‐speed interfaces (i.e. over T1 speed)
because the CPU utilization offsets any bandwidth savings.

Secure RTP (sRTP) – FLOW ENCRYPT // AUTH INT // REPLAY PROT
• Partnered with sister protocol called sRTCP
• Can be used in conjunction with cRTP
• SIP: responsible for all signaling between CUCM cluster and a gateway • FLOW ENCRYPTION
o ISDN protocols (Q.921, Q.931) only used on ISDN link to PSTN o Standardizes encryption with use of Advanced Encryption Standard (AES)
o Mode 1: Segmented Integer Counter Mode – default encryption
algorithm, def. key length 128 bits, def. session salt key length 112 bits.
o Mode 2: f8-‐mode – variation of output feedback mode
o sRTP gives the user ability to disable encryption outright, using the so-‐
called NULL cipher. This is useful when confidentiality is not required,
• MGCP: but other sRTP features (authentication, message integrity) are needed.
o ISDN Layer 3 (Q.931) information is backhauled to CUCM • AUTHENTICATION and INTEGRITY
o ISDN Layer 2 (Q.921) is terminated on the gateway o HMAC-‐SHA1 algorithm authenticates a message and protects its integrity.
o Produces 160-‐bit result, truncated to 80 bits, becomes authentication
tag, which is appended to a packet.
• REPLAY PROTECTION
o Receiver maintains indices of previously received messages, comparing

Direct signaling model = no gatekeeper them with the index of newly received messages and admitting new
message only if it has not been played before


INTRODUCTING VOIP GATEWAYS

Multisite WAN with Centralized Call-‐Processing Deployment
Single call-‐processing agent cluster provides services for multiple remote sites, IP WAN connection
Analog Gateways
options of leaded lines, frame relay, ATM, ATM and Frame Relay Server Inter-‐Working (SIW),
• Station gateways connect an IP telephony network to POTS, provide FXS ports
Multiprotocol Label Switching (MPLS) VPN, IPsec VPN
• Trunk gateways connect an IP telephony network to the PSTN CO or PBX, provide FXO
Design Characteristics
ports and E&M ports for analog trunk connection to a legacy PBX
• Single CUCM cluster
Digital Gateways
• Maximum 30,000 SCCP/SIP IP phones per cluster
• Connect an IP telephony network to the PSTN or to a PBX via digital trunks, such as PRI
• Maximum 1,000 locations per UCM cluster
or BRI CCS and T1/E1 CAS
• Maximum 1,100 H.323 devices (gateways, MCUs, trunks, clients) or MGCP gateways
IP Telephony Gateway Core Feature Requirements • PSTN for all external calls
• Gateway protocol support • DSP resources for conferencing, transcoding, and MTP
• Advanced gateway functionality – DTMF relay, supplementary services such as hold, • Voicemail, Unified Messaging, Cisco Unified Presence, A/V components
transfer, and conferencing • Integration with legacy PBX and VM systems
• Work with redundant Cisco UCM • H.323 clients, MCUs, and H.323/H.320 gateways that require gatekeeper to place calls
• Call survivability in Cisco UCM – preserves the bearer streams must register with a Cisco IOS gatekeeper. CUCM then uses H.323 trunk to integrate
• Q Signaling (QSIG) support – standard for PBX interoperability, Cisco voice packets with a gatekeeper and provide call routing and bandwidth management services
appears to PBXs as a distributed transit PBX that can establish calls to any PBX or other • MCU resources required for multipoint video conferencing (SCCP/H.323)
telephony endpoint served by a Cisco gateway, including non-‐QSIG endpoints. • H.323/H.320 video gateways needed to communicate with H.320 videoconferencing
• Fax and modem support devices on public ISDN network
• High-‐bandwidth audio between devices in same site (G.711, G.722, Cisco Wideband)
• High-‐bandwidth video between devices within the site (CUAVA Wideband Codec)
• Minimum 768 kbps or greater WAN link speeds
• UCM locations provide CAC and automated alternate routing (AAR) is also supported for
video calls, which allows calls to flow over PSTN if a call across the WAN is rejected by
the locations feature
• SRST version 4.0 and later support video
• Cisco Unified CME 4.0 and later might be used for remote site survivability instead of an
SRST router
• Unified CME can be integrated with the Cisco Unity server in the branch office in case of
a UCM outage
Design Guidelines
• Minimize delay between CUCM and remote locations to reduce voice cut-‐through
delays (clipping) – 150 ms maximum one way
• Use HSRP for network resiliency
• Use locations mechanism in CUCM to provide CAC into and out of remote branches
• Centralized/Distributed call processing depends on a number of factors:
o IP WAN bandwidth or delay limitations
o Criticality of the voice network
o Feature set needs
o Scalability
o Ease of management
o Cost
[1] [2] [3]
• For remote sites, use following features to ensure call processing survivability in the
Conferencing and transcoding only FXS only DSP farm
event of a WAN failure

o SCCP phones, use SRST on Cisco IOS gateway or CME running SRST mode
o SIP phones, use SIP SRST
o MGCP phones, use MGCP gateway fallback

Multisite WAN with Distributed Call-‐Processing Deployment
IP WAN interconnects all distributed call processing sites. PSTN serves as a backup connection between
sites in case the IP WAN connection fails or no more bandwidth. Allows each site to be self-‐contained.
Design Characteristics (as they differ from a multisite, centralized call-‐processing deployment)
• Cisco recommends that each UCM cluster have its own set of endpoint gatekeepers and
that a separate set of gatekeepers be used to manage intercluster calls. It is possible in
some circumstances to use the same set of gatekeepers for both functions, depending
on the size of the network and complexity of the dial plan.
• Low-‐bandwidth audio (for example, G.729 or G.728) between devices in different sites.
• High-‐bandwidth video (for example, 384 kbps or greater) between devices in the same
site, but low-‐bandwidth video (for example, 128 kbps) between devices at different
sites. The Cisco Unified Video Advantage Wideband Codec, operating at 7 Mbps, is
recommended only for calls between devices at the same site. Note that the Cisco VT
Camera Wideband Video Codec is not supported over intercluster trunks.
• Call admission control is provided by UCM locations for calls between sites controlled by
the same UCM cluster and by the Cisco IOS Gatekeeper for calls between UCM clusters
(that is, intercluster trunks). AAR is also supported for both intracluster and intercluster
video calls.

Benefits
Single-‐site Deployment Model
• Cost savings when you use the IP WAN for calls between sites
Call-‐processing agent cluster located at single site, no telephony services provided over IP WAN
Design Characteristics • Use of the IP WAN to bypass toll charges by routing calls through remote site gateways,
closer to the number dialed (tail-‐end-‐hop-‐off or TEHO)
• Single CUCM cluster
• Maximum utilization of available bandwidth by allowing voice traffic to share an IP WAN
• Maximum 30,000 SCCP/SIP IP phones per cluster
with other types of traffic
• Maximum 1,100 H.323 devices (gateways, MCUs, trunks, clients) or MGCP gateways
• No loss of functionality during an IP WAN failure
• PSTN for all calls outside the site
• Scalability to hundreds of sites
• DSP resources for conferencing, transcoding, and MTP
Design Guidelines (similar to those for a multisite, centralized call-‐processing deployment)
• Voicemail, Unified Messaging, Cisco Unified Presence, A/V components
Best Practices
• Integration with legacy PBX and VM systems
• Use a Cisco IOS gatekeeper to provide call admission control into and out of each site
• H.323 clients, MCUs, and H.323/H.320 gateways that require gatekeeper to place calls
• High availability of the gatekeeper, use HSRP gatekeeper pairs, gatekeeper clustering,
must register with a Cisco IOS gatekeeper. CUCM then uses H.323 trunk to integrate
and/or alternate gatekeeper support. Use multiple gatekeepers for redundancy.
with a gatekeeper and provide call routing and bandwidth management services
• Size platforms appropriately to ensure performance and capacity requirements are met.
• MCU resources required for multipoint video conferencing (SCCP/H.323)
• Use only one type of codec on the WAN because the H.323 specification does not allow
• H.323/H.320 video gateways needed to communicate with H.320 videoconferencing
for Layer 2, IP, UDP, or RTP header overhead in the bandwidth request.
devices on public ISDN network
• Using one type of codec on the WAN simplifies capacity planning by eliminating the
• High-‐bandwidth audio within the site (G.711, G.722, Cisco Wideband Audio)
need to over-‐provision the IP WAN to allow for a worst-‐case scenario.
• High-‐bandwidth video between devices within the site (Cisco Unified Video Advantage
• Gatekeeper networks can scale to hundreds of sites, and the design is limited only by
Wideband Codec, operating at 7 Mbps, supported)
the WAN topology.
Design Guidelines -‐ Subset of distributed/centralized call-‐processing model.
• For SIP, provide adequate redundancy for the SIP proxies and ensure that SIP proxies
Benefits -‐ Ease of deployment, common infrastructure, simplified dial plan, no transcoding resources
have the capacity for the call rate and number of calls required in the network.
required because of a single high-‐bandwidth codec

Best Practices
• Provide high-‐availability, fault-‐tolerant infrastructure
• Know calling patterns for your enterprise
• Use G.711 codecs for all endpoints, eliminates consumption of DSP resources
• Use SIP, SRST, MGCP gateways for the PSTN, simplifies dial-‐plan configuration
• Implement recommended network infrastructure

Acceptable Delay

Clustering over the IP WAN Deployment
• Local Failover Deployment Model: Local failover requires that you place UCM subscriber
and backup servers at the same site, with no WAN between them. This deployment
model is ideal for two to four sites with UCM.
• Remote Failover Deployment Model: Remote failover allows you to deploy the backup
servers over the WAN. Using this deployment model, you might have up to eight sites
with UCM subscribers being backed up by UCM subscribers at another site.
Note 1: This recommendation is for connections with echo that are adequately controlled, implying
Note: You can also use a combination of the two deployment models to satisfy specific site that echo cancellers are used. Echo cancellers are required when one-‐way delay exceeds 25 ms (G.131).
requirements. For example, two main sites might each have primary and backup subscribers, with Note 2: For private networks, a 200 ms delay is a reasonable goal and a 250 ms delay is a limit per Cisco
another two sites containing only a primary server each and utilizing either shared backups or
dedicated backups at the two main sites. Calculating Delay Budget

Benefits of the Clustering over the IP WAN Deployment
• Single point of administration for users for all sites within a cluster
• Feature transparency
• Shared line appearances
• Extension mobility within the cluster
• Unified dial plan
Note: These features make this solution ideal as a disaster recovery plan

WAN Considerations

Intra-‐Cluster Communication Signaling (ICCS) between UCM servers consists of many traffic types. The
Packet Loss
ICCS traffic types are classified as either priority or best effort. Priority ICCS traffic is marked with IP
• Cisco DSPs correct for 20 ms to 50 ms of lost voice through the use of Packet Loss
Precedence 3 (DSCP 24 or PHB CS3). Best-‐effort ICCS traffic is marked with IP Precedence 0 (DSCP 0 or
Concealment (PLC) algorithms, which generates a reasonable replacement packet to
PHB BE).
improve the voice quality.
• Delay – Maximum one-‐way delay between any UCM servers for all priority ICCS traffic
• Effective codec algorithms require that only a single packet can be lost at a time.
should not exceed 20 ms, or 40 ms round-‐trip time (RTT)

o Propagation delay between two sites >> 8 microseconds per kilometer
Audio Quality Measurement
without any other network delays. Theoretical maximum distance of
• MOS -‐ Subjective tests, scale 1-‐5, based on the experience of the listeners
approximately 3000 km for 20 ms delay or approximately 1860 miles.
• PSQM – Automated method of measuring speech quality “in service,” or as the speech
• Jitter – Variance in delay >> processing, queue, buffer, congestion, path variation delay.
happens. Resides with the IP call management systems, which sometimes integrated
o Jitter for IP Precedence 3 ICCS traffic must be minimized using QOS
into SNMP systems. Scale 0-‐6.5, where 6.5 is the worst. PSQM does not include
• Packet loss and errors – Network engineered to provide sufficient prioritized bandwidth
jitter/delay problems.
for all ICCS traffic, especially the priority ICCS traffic.
• PESQ – P.862, considered the current standard for voice-‐quality measurement. PESQ
o Standard QOS mechanisms must be implemented to avoid congestion
can take into account codec errors, filtering errors, jitter problems, and delay problems.
and packet loss.
Scale 1-‐4.5, where 4.5 is the best and 3.8 considered toll-‐quality.
o Lost packets must be retransmitted because of TCP error recovery >>

delayed call setup, disconnect, or other supplementary services during
Voice-‐Quality Measurement Comparison
the call.
• Bandwidth – Provision correct amount of bandwidth between each server for the
expected call volume, type of devices, and number of devices
o Bandwidth is in addition to any other bandwidth for other applications
sharing the network, including voice/video traffic between the sites.

o QOS must be enabled to provide prioritization/scheduling for the
different classes of traffic (overprovision and undersubscribe) VoIP and QoS
• QoS -‐ QoS-‐enabled bandwidth must be engineered into the network infrastructure. • Header Compression: Used with RTP and TCP, compresses the extensive RTP or TCP
header >> resulting in decreased consumption of available bandwidth for voice traffic >>
Chapter 2: Considering VoIP Design Elements reduction in delay is realized.
• Frame Relay Traffic Shaping (FRTS): Delays excess traffic using a buffer or queuing
mechanism to hold packets and shape the flow when the data rate of the source is
VOIP FUNDAMENTALS
higher than expected.
• FRF.12 (and Higher): Ensures predictability for voice traffic, aiming to provide better
Factors that Affect Audio Clarity throughput on low-‐speed Frame Relay links by interleaving delay-‐sensitive voice traffic
• Fidelity -‐ Degree to which a system, or a portion of a system, accurately reproduces at on one virtual circuit (VC) with fragments of a long frame on another VC utilizing the
its output the essential characteristics of the signal impressed upon its input, or the same interface.
result of a prescribed operation on the signal impressed upon its input. • Public Switched Telephone Network (PSTN) Fallback: Provides a mechanism to monitor
• Echo – Result of electrical impedance mismatches in the transmission path. Always congestion in the IP network and either redirect calls to the PSTN or reject calls based
present, but usually not detectable. Use suppressors or cancellers to minimize. on the network congestion.
• Jitter – Variation in the arrival of coded speech packets at the far end of a VoIP network • IP RTP Priority and Frame Relay IP RTP Priority: Provides a strict priority queuing
induced by variation in the routes of individual packets, contention, or congestion. Use scheme that allows delay-‐sensitive data, such as voice, to be dequeued and sent before
dejitter buffers to resolve variable delay. packets when other queues are dequeued. These features are especially useful on slow-‐
• Delay – Time between the spoken voice and the arrival of the electronically delivered speed WAN links, including Frame Relay, Multilink PPP [MLP], and T1 ATM links. It works
voice at the far end. Caused by distance (propagation delay), coding, compression, with weighted fair queuing (WFQ) and Class-‐Based WFQ (CBWFQ).
serialization, and buffers. • IP to ATM Class of Service (CoS): Includes a feature suite that maps QoS characteristics
• Packet loss – Caused by unstable network, congestion, or too much variable delay. Lost between IP and ATM. Offers differential service classes across the entire WAN, not just
voice packets not recoverable resulting in gaps, voice clipping, and skips in conversation. the routed portion. Gives mission-‐critical applications exceptional service during periods
• Side tone – Purposeful design that allows the speakers to hear their spoken audio in the of high network usage and congestion.
earpiece. Otherwise, they think the phone is dead. • Low Latency Queuing (LLQ): Provides strict priority queuing on ATM VCs and serial
• Background noise – Low-‐volume audio heard from the far-‐end connection. Effect of interfaces. This feature enables you to configure the priority status for a class within
VAD is often that speakers think the connection is broken because they hear nothing CBWFQ and is not limited to User Datagram Protocol (UDP) port numbers, as is IP RTP
from the other end. VAD is often combined with comfort noise generation (CNG) to Priority.
prevent illusion that the call has been disconnected. • MLP: Allows large packets to be multilink encapsulated and fragmented so they are
small enough to satisfy the delay requirements of real-‐time traffic. MLP also provides a
Delay special transmit queue for smaller, delay-‐sensitive packets, enabling them to be sent
• Fixed delay (Carl Purchases Some Pizzas) – Predictable, add directly to overall delay earlier than other flows.
o Coding – Time to translate audio signal >> digital signal • Resource Reservation Protocol (RSVP): Supports the reservation of resources across an
o Packetization – Time to put digital voice information in/out of packets IP network, allowing end systems to request QoS guarantees from the network. For
o Serialization – Insertion of bits onto a link networks supporting VoIP, RSVP (in conjunction with features that provide queuing,
o Propagation – Time it takes a packet to traverse a link traffic shaping, and voice call signaling) can provide call admission control (CAC) for
• Variable delay – Queuing delays in the egress trunk buffers located on the serial port voice traffic. Cisco also provides RSVP support for LLQ and Frame Relay.
connected to the WAN. These buffers create variable delays (jitter) across the network.

Objectives of QoS Method 1b: Modem Pass-‐Through (over TDM networks)
• Support guaranteed bandwidth • Transports modem signals through a packet network using PCM-‐encoded packets
• Improve loss characteristics: Designing the Frame Relay network, for example, so • Does not support switch from G.Clear to G.711
discard eligibility is not a factor for frames containing voice, keeping voice below the • VAD and echo cancellation need to be disabled
committed information rate (CIR) • Functions:
• Avoid and manage network congestion o Represses processing functions like compression, echo cancellation, high-‐
• Shape network traffic pass filter, and VAD
• Set traffic priorities across the network o Issues redundant packets to protect against random packet drops
o Provides static jitter buffers of 200 ms to protect against clock skew
Using QoS to Improve Voice Quality o Discriminates modem signals from voice and fax signals, indicating the
LLQ provides strict priority queuing (PQ) in conjunction with CBWFQ. LLQ configures the priority status detection of the modem signal across the connection, and placing the
for a class within CBWFQ, in which voice packets receive priority over all other traffic. connection in a state that transports the signal across the network with
the least amount of distortion
Transporting Modulated Data over IP Networks o Reliably maintains a modem connection across the packet network for a
• Data is assembled by a packet assembler/disassemble (PAD) into individual packets of long duration under normal network conditions
data, involving a process of segmentation or subdivision of larger sets of data as
specified by the native protocol of the sending device. Method 2b: Modem Relay (over TDM networks)
• Each packet has a unique identifier that makes it independent and has its own • On detection of modem answer tone, gateways switch into modem pass-‐through mode
destination address. Because the packet is unique and independent, it can traverse the and then, if the call menu (CM) signal is detected, the two gateways switch into modem
network in a stream of packets and use different routes. relay mode.
• Modem relay significantly reduces the effects that dropped packets, latency, and jitter
Understanding Fax/Modem Pass-‐Through, Relay, and Store and Forward have on the modem session. Compared to modem pass-‐through, reduces the amount
• Fax transmissions are designed to operate across a 64 kbps pulse code modulation of bandwidth used.
(PCM) encoded voice circuit, but in packet networks, the 64 kbps stream is often • Option 1: Modem Pass-‐through – Modem traffic carried between gateways in RTP
compressed into a much smaller data rate by passing it through a DSP. packets, uncompressed voice codec (G.711 mu-‐law or a-‐law)
• Faxes/modems are rarely used in a VoIP network without some kind of relay or pass-‐ • Option 2: Modem Relay – Modem signals demodulated at one gateway, converted to
through mechanism in place since DSPs not designed to compress fax/modem tones. digital form, and carried in the Simple Packet Relay Transport (SPRT) protocol which
runs over UDP packets to the other gateway.
Method 1: Fax Pass-‐through (over IP networks) (voice band data) – Modulated fax information from • Features (covered in following sections):
the PSTN is passed in-‐band end-‐to-‐end over a voice speech path in an IP network. o Modem tone detection and signaling
• Review: Simplest method, Not default, Not most desirable method for t(x) o Relay switchover
• Supported by: H.323, SIP, MGCP o Controlled redundancy
• Gateways do not distinguish fax call from voice call o Packet size
• T(x) occurs in-‐band over a voice call o Clock slip buffer management
• Fax traffic carried between two gateways in RTP packets using an uncompressed format
resembling the G.711 codec. Takes constant 64 kbps (payload) stream plus IP overhead Modem Tone Detection and Signaling
• Supports V.34 modulation and V.42 error correction and link layer protocol
• Maximum transfer rates of up to 33.6 kbps, forces higher speeds to slow down
• Supported by: SIP, MGCP, and H.323
o For MGCP/SIP/H.323, during call setup, gateways negotiate these items:
§ To use or not use modem relay mode
§ To use or not use gateway exchange identification (XID)
§ Value of payload type for Named Signaling Event (NSE)
packets (unique to MGCP/SIP)
Relay Switchover
When the gateways detect data modem, both gateways switch to modem pass-‐through mode by
performing the following actions. They revert to previous configuration at the end of the call.
• Switching to G.711 codec
• Disabling the high-‐pass filter
• Disabling VAD
• Technique 1: Configured voice codec is used for the fax transmission. • Using special jitter buffer management algorithms
o Only works when configured codec is G.711 with no VAD and no echo
• Disabling the echo canceller upon detection of a modem phase reversal tone
cancellation (EC) or when the configured codec is a clear-‐channel codec

or G.726/32.
Payload Redundancy
o Low bit-‐rate codecs cannot be used for fax transmissions.
• When only a single gateway is configured, the other gateway receives packets correctly,
• Technique 2: “Codec up speed” or “fax pass-‐through with up speed”
but does not produce redundant packets
o Gateway dynamically changes the codec from the codec configured for
• Redundancy enabled: 10 ms sample-‐sized packets sent
voice to G.711 with no VAD and no EC for the duration of the fax session.
• Redundancy disabled (default): 20 ms sample-‐sized packets sent

Method 2: Method: Fax Relay (over IP networks) – The T.30 fax from the PSTN is demodulated at the
Dynamic and Static Jitter Buffers
sending gateway. The demodulated fax content is enveloped into packets, sent over the network, and
When gateways detect data modem, both gateways switch from dynamic buffers to static jitter buffers
remodulated into T.30 fax at the receiving end.
of 200 ms depth to compensate for PSTN clocking differences. They revert back at the end of the call.
• Review: Oldest method, Default method, Uses RTP as method of transport

• Supported by: H.323, SIP, MGCP
Gateway-‐Control Modem Relay
• Cisco IOS supports: (1) T.38 fax relay and (2) Cisco Fax Relay
Beginning with IOS 12.4(4)T, Cisco supports gateway-‐controlled negotiation parameters for modem
• Gateways terminate T.30 fax signaling by spoofing a virtual fax machine to the locally relay. It is a non-‐negotiated, bearer-‐switched mode that does not involve call-‐agent-‐assisted
attached fax machine. negotiation during call setup.
• Unlike fax pass-‐through, demodulates fax bits at local gateway, sends information • Gateways use NSEs to switchover from voice >> voice band data >> modem relay
across voice network using fax relay protocol, and then remodulates bits back into tones • Upon detecting a 2100 Hz tone…
at far gateway. The fax machines on either end are sending and receiving tones and are o Terminating gateway sends NSE 192 to originating gateway and switches
not aware that a demodulation/modulation fax relay process is occurring.
over to modem pass-‐through.
• Technique 1: Cisco Fax Relay – Cisco-‐proprietary method, default on most platforms o The terminating gateway also sends NSE 199 to indicate modem relay.
• Technique 2: T.38 Fax Relay – Based on ITU.T T.38 standard, real-‐time fax transmission, o If this event is recognized by the originating gateway, the call occurs as
requires a few additional commands on gateway dial peers modem relay. If the event is not recognized, the call occurs as modem
pass-‐through.

Method 3: Store-‐and-‐forward fax (on/off-‐ramp gateway) – Breaks the fax process into distinct sending
and receiving processes and allows fax messages to be stored between those processes.
• Based on the ITU-‐T T.37 standard
• Enables fax t(x) to be received/delivered to computers rather than faxes
• Basic functionality facilitated by SMTP, with additional functionality that provides
confirmation of deliver using existing SMTP mechanisms >> Extended SMTP (ESMTP)
• On-‐ramp faxing – voice gateway that handles incoming calls from a standard fax
machine or the PSTN converts traditional G3 fax to an email message with a TIFF
attachment, which are handled by an email server
• Off-‐ramp faxing – voice gateway that handles outgoing calls from the network to a fax
machine or the PSTN converts a fax e-‐mail with a TIFF attachment into a traditional fax
format that can be delivered to a standard fax machine or the PSTN

The ANSam or CED tone causes a switch to modem pass-‐through, if enabled, to allow
the tone to pass cleanly to the remote fax.
3. A normal fax machine, after generating a CED or hearing a CNG (CalliNG) tone, sends a
DIS (digital identification signal) message with the capabilities of the fax machine. The
DSP in the Cisco IOS gateway attached to the fax machine that generated the DIS
message (normally the TGW) detects the High-‐Level Data Link Control (HDLC) flag
sequence at the start of the DIS message and initiates fax relay switchover. The DSP also
triggers an internal event to notify the call control stack that fax switchover is required.
The call control stack then instructs the DSP to change the RTP payload type to 96 and
to send this payload type to the OGW.
4. When the DSP on the OGW receives an RTP packet with the payload type set to 96, it
triggers an event to inform its own call control stack that a fax changeover has been
requested by the remote gateway. The OGW then sends an RTP packet to the TGW with
Gateway Signaling Protocols and Fax Pass-‐through and Relay payload type 97 to indicate that the OGW has started the fax changeover. When the
Fax Pass-‐Through Operation (NSE) TGW receives the payload type 97 packet, the packet serves as an acknowledgement.
The TGW starts the fax codec download and is ready for fax relay.
5. After the OGW has completed the codec download, it sends RTP packets with payload
type 96 to the TGW. The TGW responds with an RTP packet with payload type 97, and
fax relay can begin between the two gateways. As part of the fax codec download, other
parameters such as VAD, jitter buffers, and echo cancellation are changed to suit the
different characteristics of a fax call.
#3 -‐ H.323 T.38 Fax Relay
1. Same as Cisco Fax Relay
4. The detecting TGW sends a ModeRequest message to the OGW, and the OGW responds
with a ModeRequestAck.
5. The OGW sends a closeLogicalChannel message to close its VoIP UDP port, and the TGW
Cisco Fax Relay (Codec Downloaded) responds with a closeLogicalChannelAck message while it closes the VoIP port.
6. The OGW sends an openLogicalChannel message that indicates to which port to send
the T.38 UDP information on the OGW, and the TGW responds with an
openLogicalChannelAck message.
7. The TGW sends a closeLogicalChannel message to close its VoIP UDP port, and the OGW
responds with a closeLogicalChannelAck message.
8. The TGW sends an openLogicalChannel message that indicates to which port to send the
T.38 UDP stream, and the OGW responds with an openLogicalChannelAck message.
9. T.38-‐encoded UDP packets flow back and forth. At the end of the fax transmission,
either gateway can initiate another ModeRequest message to return to VoIP mode.
#4 -‐ SIP T.38 Fax Relay
4. The TGW detects a fax V.21 flag sequence and sends an INVITE message with T.38
details in the SDP field to the OGW or to the SIP proxy server, depending on the network
H.323 T.38 Fax Relay (Node Request/Ack) topology.
5. The OGW receives the INVITE message and sends back a 200 OK message.
6. The TGW acknowledges the 200 OK message and sends an ACK message directly to the
OGW.
7. The OGW starts sending T.38 UDP packets instead of VoIP UDP packets across the same
ports.
8. At the end of the fax transmission, another INVITE message can be sent to return to
VoIP mode.

MGCP T.38 Fax Relay
• Mode 1: Gateway-‐controlled mode – gateways negotiate fax relay t(x) by exchanging
capability information in SDP messages, which is transparent to the call agent. Allows
the use of a MGCP-‐based T.38 fax without necessity of upgrading call agent software.

SIP T.38 Fax Relay (INVITE) • Mode 2: Call agent-‐controlled mode – call agents use MGCP messaging to instruct
gateways to process fax traffic. For MGCP T.38 fax relay, call agents can tell gateways to
revert to gateway-‐controlled mode if call agent cannot handle the fax control messaging
traffic (overloaded/congested networks)
Call flow for an MGCP-‐based T.38 fax relay:
1. A call is initially established as a voice call.
2. The gateways advertise capabilities in an SDP exchange during connection
establishment.
3. If both gateways do not support T.38 fax relay, fax pass-‐through is used for fax
transmission. If both gateways support T.38, they attempt to switch to T.38 upon fax
tone detection. The existing audio channel is used for T.38 fax relay, and the existing
connection port is reused to minimize delay. If failure occurs at some point during the
switch to T.38, the call reverts to the original settings it had as a voice call. If this failure
occurs, a fallback to fax pass-‐through is not supported.

4. Reverts to a voice call using the previously designated codec, unless the call agent
#1 -‐ Fax Pass-‐Through Operation
instructs the gateways to do otherwise.
If pass-‐through is supported, these events occur:

1. For the duration of the call, the DSP listens for the 2100-‐Hz CED tone to detect a fax or
Gateway-‐Controlled MGCP T.38 Fax Relay
modem on the line.
• Call agent uses the fx: extension of the local connection option (LCO) to instruct a
2. If the CED tone is heard, an internal event is generated to alert the call control stack that
gateway how to process a call.
a fax or modem changeover is required.
• In gateway-‐controlled mode, gateways exchange NSEs with these steps:
3. The call control stack on the OGW instructs the DSP to send an NSE to the TGW,
o Instruct the peer gateway to switch to T.38 for a fax transmission.
informing the TGW of the request to carry out a codec change.
o Either acknowledge the switch and the readiness of the gateway to
4. If the TGW supports NSEs, it responds to the OGW instruction and loads the new codec.
accept T.38 packets or indicate that the gateway cannot accept T.38
The fax machines are able to communicate on an end-‐to-‐end basis with no further
packets.
intervention by the voice gateways.

CA-‐Controlled MGCP T.38 Fax Relay
Note: NSEs are a Cisco-‐proprietary version of IETF-‐standard named telephony events (NTEs), which are
specially marked data packets used to digitally convey telephony signaling tones and events. • Capability to accept the MGCP FXR package, to receive the fxr prefix in commands from
• NSEs = RTP payload type 100 the call agent, and to send the fxr prefix in notifications to the call agent.
• Capability to accept a new port when switching from voice to fax transmission during a
• NTEs = RTP payload type 101
call. This new capability allows successful T.38 call-‐agent-‐controlled fax communications
• NSE/NTEs provide a more reliable way to communicate tones and events using a single
between H.323 and MGCP gateways in those situations in which the H.323 gateway
packet rather than a series of in-‐band packets that can be corrupted or partially lost.
assigns a new port when changing a call from voice to fax. New ports are assigned in

H.323 gateways using images from Cisco IOS Release 12.2(2)T through Cisco IOS Release
#2 -‐ Cisco Fax Relay
12.2(7.5)T. MGCP gateways in MGCP-‐to-‐MGCP fax calls reuse the same port, but call-‐

agent-‐controlled T.38 fax relay enables MGCP gateways to handle both situations, either
When DSP put into voice mode at beginning of a VoIP call, DSP informed by call control stack whether
switching to a new port or reusing the same port, as directed by the call agent.
fax relay is supported. If so, finds out if it is Cisco Fax Relay or T.38 fax relay. If Cisco Fax Relay, the

following occurs:
1. VoIP call established as if it were a normal speech call. Call control procedures are
followed >> DSP put into voice mode, which expects human speech
2. At anytime during the life of the call, if a fax answer or calling tone (ANSam [modified
ANSwer tone] or CED) is heard, the DSP does not interfere with the speech processing.

DTMF Support Calculating the Total Bandwidth for a VoIP Call

When using low-‐bit-‐rate voice codec, DTMF tones get distorted. DTMF relay solves the problem of To perform the calculations, you must consider these contributing factors as part of the equation:
DTMF distortion by transporting DTMF tones “out of band,” separate from the RTP stream. • More bandwidth required for the codec requires more total bandwidth.
#1 -‐ H.323 DTMF Support • More overhead associated with the data link requires more total bandwidth.
Four methods supported on Cisco gateways • Larger sample size requires less total bandwidth.
1. Cisco-‐proprietary: DTMF tones sent in same RTP channel as voice data, but they are • RTP header compression requires significantly less total bandwidth.
encoded differently and identified as payload type 121. Requires use of Cisco gateways
at both ends of H.323 call. Total Bandwidth Requirements (in kbps)
2. H.245 Alphanumeric: Separates DTMF tones from voice stream, sends through H.245
signaling channel, transported in H.245 user input indication messages. H.245 channel [(AudioBytes+L2+L3+Xtra) / SampleSizems] * 8
is reliable, resulting in guaranteed delivery. Not send tone length information.
3. H.245 Signal: Passes along tone length information, optional on H.323 gateways. Calculation for G.729 codec, 20-‐bye sample size, using Frame Relay without cRTP
4. NTEs: Transports DTMF tones in RTP packets, using an RTP payload type that prevents • Total_BW = ([6 + 40 + 20] / 20) * 8000
tones from being compressed via the codec being used to encode the voice traffic. • Total_Bandwidth = 26,400 bps
#2 -‐ MGCP DTMF Support
Four current implementations Effects of Voice Activity Detection on Bandwidth
• Cisco proprietary: DSPs on gateways send/receive DTMF digits in-‐band in voice RTP • VAD can suppress packets containing silence
stream, but code them differently so they can be identified by receiver. • 35 percent bandwidth savings is average. Does not consider loud background sounds,
• NSE: Uses NTEs in RTP packets differences in languages, and other factors.
• NTE: Two modes of implementation • VAD shouldn’t be considered on links carry fewer than 24 voice calls simultaneously.
o Gateway-‐controlled mode: gateways negotiate DTMF transmission by • VAD enabled by default on all VoIP calls.
exchanging capability information in SDP messages.
• Also provides CNG.
o Call agent-‐controlled mode: Call agents use MGCP to instruct gateways

to process DTMF traffic.
DSP (TRANS VOICE MEDIA CONF)
#3 -‐ SIP DTMF Support
Four major functions of DSPs in a voice gateway:
• Can use Cisco proprietary NOTIFY-‐based out-‐of-‐band DTMF relay. It can also be used by 1. Transcoding – Direct digital-‐to-‐digital conversion from one codec to another.
analog phones attached to analog voice ports on the router. Compresses/decompresses voice streams to match endpoint-‐devices. Processed by
• Messages sent bi-‐directionally between gateways for a DTMF event during a call. DSPs on the DSP farm. Sessions initiated/managed by CUCM, who refers to transcoders
• Takes precedence over other DTMF relay mechanisms configured on a SIP dial peer. as hardware MTPs
2. Voice termination – applies to call that has two legs, one on a TDM interface and the
PROCESSING VOICE PACKETS WITH CODECS AND DSPs other on a VoIP connection. TDM leg must be terminated by hardware that performs
coding/decoding and Packetization of the streams. DSP also provides each cancellation,
Codecs VAD, and jitter management at the same time it performs termination.
• G.711 – Encodes audio on 64 kbps channel. PCM scheme operating at 8 kHz sample 3. Media Termination Point (MTP) – entity that accepts two full-‐duplex voice streams
rate, 8 bits per sample. Used for voice delivery by PSTN or through PBXs. using the same codec. Bridges media streams and allows them to be set up and torn
o Mu-‐law: North America/Japan down independently. Can be used to transcode a-‐law to mu-‐law and provide further
o A-‐law: Europe and everywhere else call-‐processing.
• G.726 – ITU-‐T Adaptive Differential Pulse Code Modulation (ADPCM) coding at 4. Audio Conferencing – joins multiple participants into a single call. Mixes streams
40/32/24/16 kbps. ADPCM-‐encoded voice can be interchanged between packet voice, together and creates a unique output stream for each connected party. Call also mixed
PSTN, and PBX networks (if configured). Often referred to by the bit size of a sample, multiple-‐codec calls, supporting any call-‐type scenario in multisite environments.
which are 2/3/4/5-‐bits, respectively. 5. Repacketization: transcode a-‐law to mu-‐law and vice versa
• G.728 – 16 kbps Low-‐Delay Code Excited Linear Prediction (LDCELP) variation of CELP 6. H.323 Supplementary Services – can be used to extend supplementary services to
voice compression. Must be translated to PSTN format for delivery. H.323 endpoints that do not support the H.323v2 OpenLogicalChannel and
• G.729 – 8 kbps, Conjugate Structure Algebraic Code Excited Linear Prediction (CS-‐ACELP) CloseLogicalChannel request features of the Empty Capabilities Set (ECS) – Infrequent.
compression. G.729a requires less computation, but quality lower. G.729b adds VAD When MTP is required on an H.323 call and none is available, the call will process but will not be able to
and CNG for smarter bandwidth usage, but more computation. G.729ab combines both. invoke the supplementary services.
• G.723 – Dual-‐rate speech coder
o r63 – 6.3 kbps, 24-‐byte frames, MPC-‐MLQ (Multipulse LPC with Maximum MTP Types
Likelihood Quanitzation) algorithm, somewhat higher sound quality • Software MTP
o r53 – 5.3 kbps, 20-‐byte frames. ACELP algorithm, additional flexibility o Can be implemented by installing the Cisco IP Voice Media Streaming
• GSM Full Rate Codec (GSMFR) – 13 kbps, 20 ms frame size. Regular Pulse Excited – Application on a CUCM server or by using a Cisco IOS gateway without
Linear Predictive (RPE-‐LTP) coder. Required for VoiceXML scripts to function. using DSP resources.
• Internet Low Bit Rate Codec (iLBC) – 13.33 kbps (30 ms) and 15.20 kbps (20 ms). o Supports G.711 to G.711 and G.729 to G.729 streams
Algorithm version of Block-‐Independent Linear Predictive Coding. Graceful speech o Can be implemented on a Cisco IOS router by configuring a software-‐only
degradation in case of lost frames, caused by lost/delayed IP packets. MTP under a DSP farm. This DSP farm can be used only as a pure MTP
and does not require any hardware DSPs on the router.
Impact of Voice Samples and Packet Size on Bandwidth o Cisco IP Voice Media Streaming Application can be installed on a server
• Voice sample – digital output from codec’s DSP encapsulated into PDU and configured to be used by CUCM.
o Cisco IOS Based allows configuration of G.711 mu-‐law and a-‐law, G.729a,
• 20 ms PDUs (protocol data unity) by default
G.729, G.729ab, G.729b, GSM, and pass-‐through, but only one at a time.
o More samples per PDU >> larger PDUs >> reduced bandwidth, but can
• Hardware MTP
cause variable delay and sever gaps if PDUs are dropped
o A resource that uses gateway-‐based DSPs to interconnect two G.711
streams, without using the gateway CPU.
o Uses a DSP resource for endpoints using the same codec but a different
packetization time, which cannot be done by software only.

Codec Complexity
Refers to the amount of processing required to perform voice compression.
• Affects call density, that is, the number of calls reconciled on the DSPs
• Difference with complexities is the amount of CPU needed to process codec algorithm.

Configuring Codec Complexity
C549 DSPs: (config)# voice-‐card 1
(config-‐voicecard)# codec complexity {high | medium | <cr>}
Audio Bandwidth (in bytes) C5510 DSPs: (config)# voice-‐card 1
(SampleSizesec * CodecBW)/8
Verify:
(Config)# codec complexity {flex | high | medium | secure}
(config)# show voice dsp
Where codec bandwidth is in bps and sample size is in sec (ex. 64000 * 0.03)
Flex allows 6-‐16 calls to be completed per DSP depending on codec used for the call.
Secure supports SRTP, providing authentication and encryption services to RTP.
Data Link Overhead (18 bytes LAN / 6 bytes WAN)
• Ethernet II – 18 bytes overhead DSP Requirements for Media Resources
• MLP – 6 bytes overhead Requirements based on (1) DSP type and (2) codec used
• Frame Relay Forum Standard 12 (FRF.12) – 6 bytes overhead • Each DSP configured/functions independently
• Single DSP can support only one function at a time (e.g. conferencing, transcoding, etc.)
Security and Tunneling Overhead • Conferencing resources either be G/711-‐only or mixed mode (i.e. one party with G.729)
• IPsec – 50-‐57 bytes overhead
• L2TP/GRE – 24 byes overhead
• MLP – 6 bytes overhead
• MPLS – 4 bytes overhead


Configuring Conferencing and Transcoding on Voice Gateways

1. Determine DSP Resource Requirements
Chapter 3: Routing Calls over Analog Voice Ports
a. DSPs directly on a voice network module (such as the NM-‐HD-‐2VE)
b. PVDM2s installed in a voice network module (such as the NM-‐HDV2) INTRODUCING ANALOG VOICE APPLICATIONS ON CISCO IOS ROUTERS
c. PVDM2s that are installed directly onto the motherboard (such as on the
Cisco 2800 and 3800 Series voice gateway routers). Local Calls – between two telephones connected to one Cisco voice-‐enabled router
2. Enable SCCP: IOS router containing DSP resources talks with CUCM using SCCP On-‐net Calls – between two telephones on the same data network, attach via FXS/PBX
(config)#sccp local FastEthernet 0/1 Off-‐net Calls – call outside data network, routed over the PSTN
(config)#sccp ccm 10.1.1.201 identifier 1 priority 1 version 4.1 PLAR Calls
(config)#sccp • User does not receive dial-‐tone
(config)#sccp ccm group 1 • Works between any type of signaling: E&M, FXO, FXS, analog/digital interfaces
(config-‐sccp-‐ccm)#bind interface FastEthernet0/1 PBX-‐to-‐PBX Calls
(config-‐sccp-‐ccm)#associate ccm 1 priority 1 • PBXs connected with private tie trunks, digital T1/E1 with CAS/PRI signaling, or analog
(config-‐sccp-‐ccm)#associate profile 1 • Another form of toll-‐bypass
(config-‐sccp-‐ccm)#register {DeviceNameForCUCM} Intercluster Trunk Calls – between two IP phones at different remote offices over IP WAN using CUCM
Where 10.1.1.201 is the primary CUCM server running CCM 4.1 On-‐net to Off-‐net Calls
3. Configure Enhanced Conferencing and Transcoding • Between two IP phones, routed thru PSTN because IP WAN congested/down
a. Enable DSP farm services. • Might occur due to dial plan configuration or be redirected by CAC
(config)# voice-‐card 0
(config)# dsp services dspfarm
INTRODUCING ANALOG VOICE PORTS ON CISCO IOS ROUTERS
b. Configure a DSP farm profile to allocate DSP-‐farm resources
(config)# dspfarm profile 1 {transcoding | conference | mtp}
(config-‐dspfarm-‐profile)# codec g711ulaw Voice Ports
(config-‐dspfarm-‐profile)# codec g711alaw • Signaling interfaces generate on-‐hook status, ringing, line seizure
(config-‐dspfarm-‐profile)# codec g729ar8 • Important to know which signaling method the telephone side of connection is using
(config-‐dspfarm-‐profile)# codec g729abr8 and match router configuration and voice interface hardware to that signaling method.
(config-‐dspfarm-‐profile)# codec g729r8
(config-‐dspfarm-‐profile)# maximum sessions {#} Analog Voice Ports
(config-‐dspfarm-‐profile)# associate application SCCP • Two-‐wire circuits connect to analog phone/fax, four-‐wire circuits connect to PBXs
(config-‐dspfarm-‐profile)# no shutdown • FXO, FXS (trunk/tie line or PBX w/o E&M support), E&M (common analog trunk circuit)
c. Associate a DSP Farm Profile to a CUCM Group. • E&M – uses special signaling paths separate from the trunk audio path
Conference Bridge o Signaling paths known as the E-‐lead and M-‐lead
• Service >> Media Resource >> Conference Bridge o Preferred because provide better answer/disconnect supervision
• Select Conference Bridge Type o DCE (CO router) and CTE (customer PBX)
• FXO/FXS – indicate on-‐hook status, off-‐hook status, and line seizure with a 48V circuit
using loop-‐start or ground-‐start signaling
o Loop Start Signaling
§ Downside: No way to prevent CO and subscriber from
seizing the same line simultaneously (glare)
§ Downside: Doesn’t provide switch-‐side disconnect

Transcoding Resource supervision for FXO calls
• Service >> Media Resource >> Transcoder
• Select Transcoder Type

d. Verify DSP Farm configuration.

Cisco IOS Configuration Commands for Enhanced Media Resources
• voice-‐card slot
• dsp services dspfarm

• dspfarm profile profile-‐identifier {conference | mtp | transcode| o Ground Start Signaling
• codec {codec-‐type | pass-‐through} § Idle state: both tip/ring disconnected from ground.
o Pass-‐through option is available only for MTPs and is typically used for Battery (-‐48V DC) is still connected to the ring line
CUCM 5.0 controlled RSVP-‐based CAC § PBX/FXO grounds the ring line to tell CO/FXS that there is
• maximum sessions number an incoming call. CO/FXS senses the ring ground and then
• associate profile sccp grounds the tip lead to let the PBX/FXO know that it is
• sccp ccm {ip-‐address | dns} identifier identifier-‐number [priority priority] [port port-‐ ready to receive the incoming call.
number] [version version_number] § PBX/FXO senses the tip ground and closes the loop
• sccp local interface-‐type interface-‐number [port port-‐number] • DTMF Frequencies
• sccp
• sccp ccm group group-‐number
• associate ccm identifier-‐number priority priority
• associate profile profile-‐identifier register device-‐name
• bind interface interface-‐type interface-‐number
• Informational Signaling
Verifying Media Resources
• Verify configuration of a DSP farm profile – show dspfarm profile

• E&M Signaling
o Type I – Most common. Two signaling wires:
E-‐lead, M-‐lead, remaining two pairs of wires serve as audio path
o Type II – Used in sensitive environment. Four signaling wires:
E-‐lead, M-‐lead, Signal Ground (SG), Signal Battery (SB)
• Check DSP status used for DSP farm profiles – show dspfarm dsp all o Type III – Not common. Four wires used for signaling.
o Type IV – Not supported by Cisco. Four signaling wires.
o Type V – Most common outside North America. Two signaling wires.
o SSDC5 – Common in England. Similar to Type 5.

CAMA Trunk Configuration

Example: PSAP expects NPD signaling, with the area code 312 being represented by the digit 1.
(config)# voice-‐port 1/1/1
(config-‐voiceport)# ani mapping 1 312
(config-‐voiceport)# signal cama kp-‐npd-‐nxx-‐xxxx-‐st

(config)# dial-‐peer voice 911 pots
(config-‐dialpeer)# destination-‐pattern 911

• E&M Physical Interface – RJ-‐48 connector, two-‐wire or four-‐wire (config-‐dialpeer)# prefix 911
o Actually has 6-‐8 wires, but still referred to as 2/4 wire (config-‐dialpeer)# port 1/1/1
o Two-‐wire (half-‐duplex), Four-‐wire (full-‐duplex)
• E&M Address Signaling (config)# dial-‐peer voice 9911 pots
o Immediate-‐start: Simplest method. E&M tie trunk interfaces. (config-‐dialpeer)# destination-‐pattern 9911
Calling side seizes line by going off-‐hook on its E-‐lead, waits minimum 150 (config-‐dialpeer)# prefix 911
ms and then sends DTMF digits or pulses (config-‐dialpeer)# port 1/1/1
o Wink-‐start: Most common method. Default for E&M voice ports.
Developed to minimize glare. (config)# dial-‐peer voice 910 pots
Calling side seizes line by going off-‐hook on its E-‐lead, waits for short (config-‐dialpeer)# destination-‐pattern 9[2-‐8].........
temporary off-‐hook pulses (wink) from other end on its M-‐lead before (config-‐dialpeer)# port 0/0/0:23
sending DTMF digits or pulses
o Delay-‐start: E&M tie trunk interfaces Direct Inward Dial
Calling side seizes line by going off-‐hook on its E-‐lead, after timed interval, (config)# voice-‐port 0/0/0
calling side looks at status of called side. If called side is on-‐hook, calling (config-‐voiceport)# signal did wink-‐start
side starts sending DTMF digits.
(config)# voice-‐port 0/1/0
Configuring Analog Voice Ports (config-‐voiceport)# signal groundstart

FXS Voice Port Configuration
(config)# voice-‐port slot/port (config)# dial-‐peer voice 1 pots
(config-‐dialpeer)# incoming called-‐number .
(config-‐voiceport)# signal {loopstart | groundstart}
(config-‐voiceport)# cptone locale (config-‐dialpeer)# direct-‐inward-‐dial
(config-‐voiceport)# ring cadence {pattern-‐number | define pulse interval} (config-‐dialpeer)# port 0/0/0
(config-‐voiceport)# no shutdown
Note: signal change requires shut/no shut (config-‐dialpeer)# destination-‐pattern 9[2-‐8].........
Note: patternXX provides preset ring-‐cadence patterns (config-‐dialpeer)# port 0/1/0
Note: define allows creation of custom ring cadence
Timers and Timing
FXO Voice Port Configuration • timeouts initial seconds – initial digit timeout value in seconds, time dial tone is
Utilizes PLAR feature to ring 4001 phone presented before first digit is expected
(config)# voice-‐port slot/port • timeouts interdigits seconds – time between last CED and sending the call
(config-‐voiceport)# signal {loopstart | groundstart} • timeouts ringing {seconds | infinity} – time call can ring when there is no answer, use
(config-‐voiceport)# connection plar opx 4001 this to reduce 180 seconds default and not tie up voice port on gateway
(config-‐voiceport)# no shutdown • timing digit milliseconds – DTMF signal duration for voice port, use for a device having
(config)# dial-‐peer voice 90 pots trouble recognizing dialed digits
(config-‐dialpeer)# destination-‐pattern 9T • timing interdigit milliseconds –DTMF interdigit duration for voice port
(config-‐dialpeer)# port 0/0/0 • timing hookflash-‐input milliseconds – maximum duration of a hookflash indication.
o Hookflash is an indication by a caller that wants to do something specific
Using the opx option, the location voice port provides a local response before the remote with the call, such as transfer the call or place call on hold.
voice port receives an answer. On FXO interfances, the voice port does not answer until • timing-‐hookflash-‐output milliseconds
the remote side has answered.
Verifying Voice Ports
E&M Voice Port Configuration 1. Check for dial tone.
Connects a voice gateway to a legacy PBX, configuring DID connectivity between systems a. Is the plug firmly seated?
(config)# voice-‐port slot/port b. Is the voice port enabled?
(config-‐voiceport)# signal {immediate-‐start | wink-‐start | delay-‐start } c. Is the voice port recognized by the Cisco IOS
(config-‐voiceport)# operation {2-‐wire | 4-‐wire} d. Is the router running the correct version of Cisco IOS to recognize port
(config-‐voiceport)# type 1 e. Is a dial peer configured for that port?
(config-‐voiceport)# no shutdown 2. If dial tone >> Check for DTMF voice band tones
(config)# dial-‐peer voice 10 pots 3. Use show voice port to verify data configured correctly
(config-‐dialpeer)# destination-‐pattern 1... 4. If trouble connecting a call >> use show voice port to make sure port is enabled
(config-‐dialpeer)# direct-‐inward-‐dial a. If port is in a hung state, use shut/no shut
(config-‐dialpeer)# forward-‐digits all 5. If configured E&M interfaces, make sure values associated with your specific PBX setup
(config-‐dialpeer)# port 1/1/1 are correct. Check for two-‐wire/four-‐wire wink start, immediate-‐start, or delay-‐start
signaling types, and the E&M interface type. Must match those set on PBX
Trunks 6. Confirm that the VNM (module containing voice ports) is correctly installed.
Single physical/logical interface containing several physical interfaces, connecting to single destination
• FXO port can forward the FXS-‐configured station-‐ID name and station-‐ID number to the
CO switch. CO switch discards information and replaces them with the information it
had configured for this connection.
• Inbound – Caller ID feature is supported on FXO port in gateway. If gateway is
configured for H.323, caller ID is displayed on IP phones and on analog phones
• MGCP Gateway – CUCM doesn’t support caller ID feature on MGCP gateways

Centralized Automated Message Accounting
Special analog trunk originally developed for LD billing, but mainly used for emergency call services.
Can use CAMA ports to connect to a PSAP for emergency. Can only send ANI information. test voice port port_or_DS0-‐group_identifier detector {m-‐lead | battery-‐reversal | ring | tip-‐
ground | ring-‐ground | ring-‐trip} {on | off | disable}
• Calling number used to reference Automatic Location Identification (ALI) database to
Forces a detector into specific states for testing.
find exact location of the caller and any extra information.
test voice port port_or_DS0-‐group_identifier inject-‐tone {local | network} {1000hz | 2000hz |
• Calling number used as a callback number in case the call is disconnected.
200hz | 3000hz | 300hz | 3200hz | 3400hz | 500hz | quiet | disable}
• During E911 call setup, calling number is transmitted to each switching point before
Injects a test tone into a voice port. A call must be established on the voice port under
audio channel is connected (selective router)
test. When you are finished testing, be sure to use the disable option to end test tone.
• Supported: VIC2-‐2FXO, VIC2-‐4FXO, 2800 ISRs, 3800 ISRs
test voice port port_or_DS0-‐group_identifier loopback {local | network | disable}
• Five CAMA signaling options exist: Performs loopback testing on a voice port. A call must be established on the voice port
o KP-‐0-‐NXX-‐XXXX-‐ST 7-‐digit ANI transmission. The Numbering Plan Area under test. When you finish the loopback testing, be sure to use the disable option to
(NPA), or area code, is implied by the trunk group and is not transmitted. end the forced loopback.
o KP-‐0-‐NPA-‐NXX-‐XXXX-‐ST test voice port port_or_DS0-‐group_identifier relay {e-‐lead | loop | ring-‐ground | battery-‐
10-‐digit transmission. The E.164 number is fully transmitted. reversal | power-‐denial | ring | tip-‐ground} {on | off | disable}
o KP-‐0-‐NPA-‐NXX-‐XXXX-‐ST-‐KP-‐YYY-‐YYY-‐YYYY-‐ST Tests relay-‐related functions on a voice port.
Supports CAMA signaling with ANI/Pseudo ANI (PANI). test voice port port_or_DS0-‐group_identifier switch {fax | disable}
o KP-‐2-‐ST Default transmission when the CAMA trunk cannot get a Forces a voice port into fax or voice mode for testing. If voice port does not detect fax
corresponding Numbering Plan Digit (NPD) in the look-‐up table or when data, voice port remains in fax mode for 30 seconds and then reverts automatically to
the calling number is fewer than 10 digits. (NPA digits are not available.) voice mode. After you enter test voice port switch fax command, you can use show
o KP-‐NPD-‐NXX-‐XXXX-‐ST: 8-‐digit ANI transmission, where the NPD is a voice call command to check whether the voice port is able to operate in fax mode.
single MF digit that is expanded into the NPA. The NPD table is csim start dial-‐string
preprogrammed in the sending and receiving equipment (on each end of Simulates a call to the specified dial string. This command is most useful when testing
the MF trunk). For example: 0=415, 1=510, 2=650, 3=916 dial plans.
05551234 = (415) 555-‐1234 // 15551234 = (510) 555-‐1234

INTRODUCING DIAL PEERS Chapter 4: Performing Call Signaling over Digital Voice Ports

Understanding Call Legs
INTRODUCING DIGITAL VOICE PORTS
• Call legs are logical connections between any two telephony devices, such as gateways,
routers, Cisco Unified Communication Managers, or telephony endpoint devices.
• Call legs are router-‐centric. Digital Trunks
• T1: TDM to transmit digital data over 24 voice channels using CAS
• E1: TDM to transmit digital data over 30 voice channels using either CAS/CCS
• ISDN: A circuit-‐switched telephone network system using CCS. Variations of Integrated
Services Digital Network (ISDN) circuits include the following:

Understanding Dial Peers
• A dial peer is an addressable call endpoint. The address is called a destination pattern
and is configured in every dial peer.
• POTS dial peers: connect to traditional telephony network, such as PSTN, PBX, or
telephony edge device such as a telephone or fax machine.
o Provide an address for edge network or device
o Point to specific voice port that connects the edge network or device
• VoIP dial peers: connect over an IP network
o Provide destination address for edge device located across network
o Associate destination address with next-‐hop router or destination router

Configuring POTS Dial Peers
(config-‐dialpeer)# port 1/0/0

Configuring VoIP Dial Peers
(config)# dial-‐peer voice 2 voip
(config-‐dialpeer)# session target ipv4:10.18.0.1 • The ds0-‐group command creates a logical voice port (a DS0 group) from some or all of
the DS0 channels, which allows you to address those channels easily, as a group.
Configuring Destination Pattern Options
Period (.) Matches any dialed digit from 0-‐9 or the * key (20.. = 2011, 2022, 2045) T1 CAS
Plus (+) Matches one or more instances of preceding digit (5+23 = 5523) • DS0 = 64 kbps = 8000 samples/sec x 8 bits/sample = 64,000 bits/sec
Brackets ([ ]) Matches a range of digits ([1-‐3]22 = 122, 222, 322) • Uses the same signaling types available for analog trunks: loop start, ground start, and
Caret (^) Does NOT match (^[1-‐3]22 = 022, 422, 522, 622) E&M variants such as wink-‐start, delay-‐start, and immediate-‐start.
T Matches any number of dialed digits (from 0-‐32 digits) o E&M FG-‐B: Inbound/Outbound DNIS, inbound ANI (only on Cisco AS5x00)
Comma (,) Inserts a one-‐second pause between dialed digits o E&M FG-‐D: Inbound/Outbound DNIS, inbound ANI
o E&M FG-‐D EANA: Inbound/Outbound DNIS, outbound ANI
Matching Inbound Dial Peers • SF – 12 T1 frames, bit-‐robbing in frames 6 and 12, least significant bit of each channel is
• Inbound POTS dial peers are associated with incoming POTS call legs of originating robbed, leaving 7 bits for voice data
router/gateway. • ESF – Of the total 8000 F bits used in T1 >> 2000 framing, 2000 CRC, 4000 intelligent
• Inbound VoIP dial peers are associated with incoming VoIP call legs of terminating supervisory channel to control functions end-‐to-‐end (e.g. loopback and error reporting)
router/ gateway.
• Information send in call setup message E1 R2 CAS
o Called number dialed number identification service (DNIS) • Widely used in Europe, Asia, Central/South America
DNIS = DESTINATION • Bundles 32 time slots instead of 24 >> 2.048 Mbps bandwidth
Call-‐destination dial string, derived from the ISDN setup message or • E1 Multiframe Format
channel associated signaling dialed number identification service. o 16 consecutive 256-‐bit frames, each frame carrying 32 time slots
o Calling number automatic number identification (ANI/CLID) o Time slot 1 >> frame synchronization
ANI/CLID = ANDY CALLED o Time slot 17 >> signaling
Represents the origin, and it is derived from the ISDN setup message or o Time slot 2-‐16, 18-‐32 >> voice traffic
channel associated signaling (CAS) automatic number identification.
o Voice port – POTS physical voice port. ISDN
• Router/Gateway matches call setup element parameters in the following order • Circuit-‐switched telephone network system designed to allow digital transmission of
IAN AND DAVE PUNCH DOGS voice/data over ordinary telephone copper wires
1. Match dialed number (DNIS) using incoming called-‐number • ISDN provides additional supplementary services such as Call Waiting, Do Not Disturb,
2. Matchcaller ID information (ANI) using answer-‐address etc. CAS/R2 only provide DNIS
Note: To match calls based on originating calling number • ISDN-‐PRI Nonfacility Associated Signaling (NFAS) – Single D channel controls multiple
3. Match caller ID information (ANI) using destination-‐pattern ISDN PRIs on a chassis. Primary channel with option of another D channel for backup.
4. Match an incoming POTS dial peer by using port Frees up additional channel to carry information.
5. If multiple dial peers match, pick the first dial peer added to config • Benefits using ISDN for voice traffic
6. If no match, use dial peer 0. o Perfect for G.711 PCM because each B channel is full 64 kbps
o Built-‐in call control protocol >> ITU-‐T Q.931
Characteristics of the Default Dial Peer o Can convey standards-‐based voice features
Only used for inbound matches. Dial peer 0 cannot be used to match outbound calls. o Supports standards-‐based dial-‐up capabilities like G4 fax and audio
• Characteristics of Dial Peer 0 • Drop and Insert – allows for dynamic multiplexing of B channels between different
o Any codec interfaces. Only possible if all interfaces share common clock source (w/ ISRs)
o IP precedence 0
o VAD enabled ISDN Signaling (debug isdn q931)
o No RSVP support • Q.921 – Layer 2 ISDN signaling protocol is Link Access Procedure, D channel (LAPD)
o fax-‐rate service Similar to HDLC and LAPB. Used across D channel to supervise control/signaling
• For inbound POTS peers, dial peer 0 is configured with the no ivr application command • Q.931 – Layer 3 ISDN signaling protocol. Support user-‐to-‐user, circuit-‐switched (B
• The user will get a dial tone and then need to proceed with dialed digits channels) and packet-‐switched (D-‐channels) connections.
• Cannot be viewed with the show command o D channel directs the CO switch to send incoming calls to particular time
slots on the Cisco access server or router.
Matching Outbound Dial Peers
• Uses the destination-‐pattern command
o POTS >> port command
o VoIP >> session target command
• show dialplan number string command to determine which dial peer is matched to a
specific dialed strong. Displays all matching dial peers in the order that they are used.


• Components of the ISDN frame

o Protocol discriminator – protocol used to encode remainder of layer USING QSIG FOR DIGITAL SIGNALING
o Length of call reference value – define length of next field (1-‐2 byes long)
o Flag – set to zero (0) for messages sent by the part that allocated the call QSIG Overview
reference value. Otherwise, it is set to one (1).
• Variant of ISDN Q.921/Q.931 D-‐channel signaling for PBXs or key systems that are called
o Call reference value – value allocated for duration of session, identifies Private Integrated-‐services Network-‐eXchange (PINX).
call between device maintaining call and the ISDN switch.
• Router can route incoming calls from a PINX across WAN to peer router, which can then
o Message type – identifies message type that determines what additional
transfer call to another PINX
information is required and allowed. Sent between endpoints on D
• Two sub layers:
channel, single byte (octet)
o QSIG Basic Call – signaling procedures/protocol for circuit-‐switch Call
o ISDN Information Element (IE) – Additional information needed for call
Control. Based on Q.931.
processing, such as calling party number, called party number, and CID.
o QSIG Generic Function – signaling protocol for control of Supplementary
§ Cause IE – diagnosing network or customer premises
Services and Additional Network Features (ANFs) at the Q reference
equipment (CPE) problems. When a call is terminated, a
point. Enables capabilities beyond basic call capability.
cause ID indicates reason for the termination.
• QSIG Features
§ Facility IE – activate supplementary services, widely used
o Basic Call – dynamic establishment of voice connections, must use digital
by PBXs and in the PSTN.
T1/E1 PRI trinkets to support QSIG protocol
§ Progress IE -‐ ring back, busy tones, and announcements,
o Call Completion – rely on facility selection and reservation feature and
such as “Number trying to reach is no longer in service.”
provide Cisco Call Back functionality
Used in ISDN and H.323 networks.
§ Completion of Calls to Busy Subscribers (CCBS)
§ Display IE – sends text to provide output for an LCD
When caller gets busy tone, can request that call
display
completes when line is no longer busy.
o Coding Standard
§ Completion of Calls on No Reply (CCNR)
o Location
When caller gets no answer, can request that call
• Non-‐Facility Associate Signaling (NFAS)
completes after activity occurs on the line.
o Supported only with channelized T1 controlled, must be ISDN PRI-‐capable
o Call Diversion – PINX receives request to divert call to another user, new
o Disable specified channel or PRI interface using isdn service interface
call place, additional CDR record generated

§ Call Forward All (CFA)
Configuring a T1 CAS Trunk
§ Call Forward Busy (CFB)
Network Clock Timing
§ Call Forward No Answer (CFNA)
• Clock Slip – repetition/deletion of a block of bits in a synchronous bit stream because of
o Call Transfer
a discrepancy in the read/write rates at a buffer. Caused by inability of an equipment
o Identification Services
buffer store to accommodate differences between the phases/frequencies of
o Message Waiting Indication Service
incoming/outgoing signals in cases where the time of the outgoing signal is not derived
§ MWI Activate
from that of the incoming signal.
§ MWI De-‐activate
• Recommended Cisco Configuration
o Do Not Disturb (DND) and Do Not Disturb Override (DNDO)
network-‐clock-‐participate [slot slot-‐number | wic wic-‐slot | aim aim-‐slot-‐number]
• Path Placement
network-‐clock-‐select priority {bri | t1 | e1} slot/port
show network clocks

DS0 Groups
• ds0-‐group ds0-‐group-‐no timeslots timeslot-‐list type signal-‐type
o Automatically creates logical voice port number as slot/port:ds0-‐group-‐no
o Defines T1/E1 channels for compressed voice calls
o Creates a logical voice port
o Defines emulated analog signaling method used to connect to PSTN/PBX

T1 CAS Controller Configuration Example
(config)# controller {t1 | e1} slot/port
(config-‐controller)# framing {sf | esf}
(config-‐controller)# clock source {line [primary | bits] | internal | free-‐running}
(config-‐controller)# linecode {ami | b8zs}

(config-‐controller)# ds0-‐group ds0-‐group-‐number timeslots timeslot-‐list [service service-‐type] Configuring QSIG Support
type {e&m-‐fgb | e&m-‐fgd | e&m-‐immediate-‐start | fgd-‐eana | fgd-‐os | fxs-‐ground-‐start | fxs-‐
loop-‐start | none | r1-‐itu | r1-‐modified | r1-‐turkey}
(config-‐controller)# no shutdown

T1 CAS Digital Voice Parameters
(config)# voice-‐port slot/port:ds0-‐group-‐number Configure global QSIG support for BRI or PRI
(config-‐voiceport)# cptone locale BRI (config)# isdn switch-‐type basic-‐qsig
(config-‐voiceport)# compand-‐type {u-‐law | a-‐law}
(config-‐voiceport)# no shutdown PRI (config)# isdn switch-‐type primary-‐qsig
(config)# card type t1 0
Configuring an E1 R2 Trunk
(config)# controller E1 0/0/0 Configure T1/E1 controller for QSIG over PRI
(config-‐controller)# ds0-‐group 0 timeslots 1-‐31 type r2-‐digital (config)# controller t1 0/1
r2-‐compelled ani (config-‐controller)# pri-‐group timeslots 1-‐24
(config-‐controller)# cas-‐custom 0 (config)# interface serial 0/1:23
(config-‐ctrl-‐cas)# country china use-‐defaults (config-‐if)# isdn switch-‐type primary-‐qsig
(config-‐if)# isdn protocol-‐emulate user
(config)# dial-‐peer voice 90 pots User side only – configure layer 2/3 port mode emulation and clock status
(config-‐dialpeer)# destination-‐pattern 9T for user, terminal equipment (TE), default
(config-‐dialpeer)# direct-‐inward-‐dial (config-‐if)# isdn protocol-‐emulate network
(config-‐dialpeer)# port 0/0/0:0 Network side only – configure layer 2/3 port mode emulation and clock
status for network, network termination (NT)
Configuring an ISDN Trunk
After clock source, framing, and line code are configured, ISDN voice functionality requires commands: Configure QSIG for BRI
(config)# isdn switch-‐type primary-‐qsig (config)# interface bri 1/1
(config)# controller t1 0/0 (config-‐if)# isdn layer1-‐emulate user
(config-‐controller)# pri-‐group timeslots 1-‐24 Configure Layer 1 port mode emulation and clock status for the user—that is, the TE.
(config-‐controller)# interface serial 0/0:23
(config-‐if)# isdn incoming-‐voice voice (config)# interface bri 1/1
(config-‐if)# no shutdown (config-‐if)# isdn layer1-‐emulate network
(config-‐if)# isdn incoming-‐voice voice
Verifying Digital Voice Ports (config-‐if)# isdn protocol-‐emulate user
1. Check for dial tone. Current version of ISO require a POTS dial peer be configured for Network side only – configure layer 2/3 port mode emulation and clock status for
the voice port being tested before a dial tone will be heard. network, network termination (NT)
2. If dial tone, check for DTMF detection. If dial tone stops when you dial a digit, the voice
port is probably configured correctly. (config)# interface bri 1/1
3. show voice port summary >> identify port numbers of voice interfaces in router (config-‐if)# isdn protocol-‐emulate network
4. show voice port >> verify voice-‐port parameter settings Network side only – configure layer 2/3 port mode emulation and clock status for
5. show controller >> verify digital T1/E1 controller is up and no alarms have been network, network termination (NT)
reported, and to display information about clock sources and other settings
6. show voice dsp >> display voice-‐channel configuration information for all DSP channels
7. show voice call summary >> verify the call status for all ports
8. show call active voice >> display contents of the active call table, which shows all calls
currently connected through the router/concentrator
9. show call history voice >> display contents of the call history table

• Translations can be defined per gateway – allows translations to be made on gateway
Verifying QSIG Trunks and then send transformed number to CUCM for processing, useful for system that
spans multiple countries
• More specific call-‐router than CUCM – allows matching by called and calling part
numbers whereas CUCM only handles called number
• No need for extra SRST-‐related call-‐routing configuration
• No dependency on CUCM version – due to peer-‐to-‐peer nature of H.323
• More voice interface types are supported
• Support for ISDN Nonfacility Associate Signaling (NFAS) -‐ MGCP does not
• Fax support is improved – better on H.323 than MGCP gateways because H.323
supports T.37 and T.38 and can route faxes directly to FXS
• Call preservation is enhanced

H.323 Network Components
• H.323 Terminals – endpoint that provides real-‐time voice communication with another
endpoint, such as an H.323 terminal or multipoint control unit (MCU). Must be capable

of t(x)/r(x) G.711 (a-‐law and mu-‐law) 64 kbps PCM
• H.324 Terminals – uses 33.6 kbps modem for transmission, H.263 codec for video
encoding, and G.723 for audio
• H.323 Gateways – endpoint that provides real-‐time voice communications between
H.323 terminals on LAN and other ITU terminals on a WAN or to other H.323 gateways
• Cisco Unified Border Elements (UBE or session border controllers) – optional H.323
component, joins two VoIP call legs, functions like a PSTN-‐to-‐IP gateway, can pass codec
preferences to the terminating leg of a VoIP call, configured to use transparent codec
• H.323 Gatekeepers – provides call management, including admission control,
bandwidth management, and routing services for calls in the network
o Zone – scope of endpoints over which gatekeeper exercises authority
o Functions of Gatekeeper
§ Address translation – coverts alias address to IP address
§ Admission control – limits access to network resources
Remember: if network emulation is not correctly set, Layer 2 will not come up!! based on bandwidth restrictions
§ Bandwidth control – responds to bandwidth
debug isdn q921 request/modifications
• Limited to commands/responses during peer-‐to-‐peer communication over D channel § Zone management – services for registered endpoints
• Does not include B channels that are part of the same ISDN interface § Call control signaling – performs call signaling on behalf of
• Does not display data link layer access procedures taking place on ISDN network side the endpoint (gatekeeper-‐routed call signaling)
§ Call authorization – rejects calls based on auth. Failure
debug isdn q931 § Bandwidth management – limits number of concurrent
• Watch Q.931 signaling messages go back/forth while router negotiates ISDN connection accesses to IP internetwork resources (CAC)
• Displays call setup/teardown of ISDN network connections § Call management – maintains record of ongoing calls
• show dialer retrieve information about status/configuration of ISDN interface on router • Multipoint Control Units (MCUs) – allows three or more endpoints to participate in a
multipoint conference

Chapter 5: VoIP Gateways & Gateway Control Protocols H.323 Call Establishment and Maintenance
• Endpoint to endpoint – endpoints locate other endpoints through nonstandard
Configuring H.323 mechanisms and initiate direct communication between endpoints
• Endpoint to gatekeeper – endpoints interoperate with gatekeeper using RAS channel
H.323 Gateway Overview • Gatekeeper to gatekeeper – with multiple gatekeepers, they communication with each
• Gateway – point where circuit-‐switched call is encoded and repackaged into IP packets other using RAS channel
o Provides admission control, address lookup/translation, accounting
H.323 Call Flows
H.323 Basic Call Setup

• H.323 and IP
o H.225.0 – Call setup
§ Call-‐signaling function – uses call-‐signaling channel that
allows endpoint to communicate with another endpoint
o H.225.0 – Registration, admission, status (RAS) control for call routing
§ Uses separate signaling channel (RAS channel) to perform
registration, admissions, bandwidth changes, status, and
disengage procedures between endpoints and a
gatekeeper.

o H.245 – Capabilities exchange
§ H.245 control function – uses control-‐channel to transport
control messages between endpoints, separate from call
signaling channel
§ Logical channel signaling – opens/closes a channel that
carries a media stream
§ Capabilities exchange – negotiates audio, video, and
coder-‐decoder (codec) capability between endpoints
§ Master or responder determination – determines which
endpoint is the master and which the responder, used to
resolve conflicts during the call
§ Mode request – requests a change in mode or capability
of the media stream
§ Timer and counter values – establishes values for timers
and counts and agreement of those values by endpoints

Why H.323
• Dial plans can be configured directly on the gateway – possible to handle call
setup/teardown with endpoints or route calls to another site without use CUCM cluster

H.323 Fast Connect Call Setup • Fax Relay

Reduces number of round-‐trip exchanges and achieve capability exchange and logical (config)# voice service {pots | voatm | vofr | voip}
channel assignments in one round trip (conf-‐voi-‐serv)# fax protocol t38 [nse [force]] [ls-‐redundancy value
[hs-‐redundancy value]] [fallback {none | pass-‐through {g711ulaw | g711alaw}}]
(conf-‐voi-‐serv)# exit
(config)# dial-‐peer voice id voip
(config-‐dial-‐peer)# fax-‐relay ecm disable
(config-‐dial-‐peer)# fax-‐relay sg3-‐to-‐g3

Note 1: Dial-‐peer configs takes precedence over voice-‐service configs
Note 2: fax-‐relay ecm disable disables error correction mode (ECM) by modifying
Digital information Signal (DIS) T.30 message in both directions, even if only one
gateway is configured with ECM disabled.
Note 3: fax-‐relay sg3-‐to-‐g3 enables fax stream between two SG3 machines to
negotiate down to G3 speeds

Configuring H.323 DTMF Relay
For SIP calls, the most appropriate method to transport DTMF tones is RTP-‐NTE or SIP-‐NOTIFY.
(config-‐dial-‐peer)# dtmf-‐relay {[cisco-‐rtp] [h245-‐alphanumeric] [h245-‐signal]
[rtp-‐nte [digit-‐drop]] [sip-‐notify]}

Verifying H.323 Gateways

H.323 Multipoint Conferences
All types of multipoint conferences rely on a single MC to coordinate membership of a conference.
Each endpoint has an H.245 control channel connection to the MC. Either MC or endpoint initiates the Implementing MGCP Gateways
control channel setup.
1. Centralized multipoint conference – endpoints must have their A/V/D channels MGCP Overview
connected to multipoint processor (MP) • Centralized device control protocol with simple endpoints
2. Distributed multipoint conference – endpoints do not have connection to an MP, • Stimulus protocol – endpoints and gateways cannot function alone
endpoints multicast A/V/D streams to call participants in conference, any mixing of • Plaintext commands sent to gateways on UDP 2427, send back to call agent UDP 2727
streams is a function of the endpoint, all endpoints must use same comm. Parameters • Gateways interacted with call agent, also called Media Gateway Controller (MGC)
3. Ad hoc multipoint conference – if no collocated MC, they use services of a gatekeeper. • Uses “hairpinning” to return a call to the PSTN when the packet network is not available

Configuring H.323 Gateways Why MGCP
Options for setting up H.323 gateway include the following: • Alternate dial tone for VoIP environments – enables VoIP system to control call
• Enable H.323 VoIP call services globally (required) setup/teardown and Custom Local Area Subscriber Services (CLASS) features
(config)# voice service {pots | voatm | vofr | voip} • Simplified configuration for static VoIP network dial peers – call agent provides
(conf-‐voi-‐serv)# h323
functions similar to VoIP dial peers, but still requires POTS dial peers
(conf-‐voi-‐serv)# no shutdown
• Migration paths – systems using earlier version of protocol can migrate easily

• Centralized dial plan configured on CUCM
• Configure interface as an H.323 gateway interface (required)
• Centralized gateway configuration on CUCM
(config)# interface interface
• Simple Cisco IOS gateway configuration – far few commands needed
(config-‐if)# ip address ip-‐address network-‐mask
• Supports QSIG supplementary services with CUCM – with QSIG, you can used MGCP to
(config-‐if)# h323-‐gateway voip interface
connection CUCM with a traditional PBX
(config-‐if)# h323-‐gateway voip h323-‐id name

(config-‐if)# h323-‐gateway voip bind srcaddr ip-‐address
MGCP Architecture

• Endpoints – the point of interconnection between the packet network and the
• Configure codecs (optional)
(config)# voice class codec tag traditional telephone network
(config-‐class)# codec preference value codec-‐type [bytes payload-‐size] • Gateways – handle translation of audio between a switched-‐circuit and packet network
(config)# dial-‐peer voice tag voip o Connections can be point-‐to-‐point or multipoint
(config-‐dial-‐peer)# voice-‐class codec tag o For point-‐to-‐point, endpoints could be in separate/same gateway
o Creating call connection involves a series of signals and events that
• Adjust H.225 Timers describe the connection process
(config)# voice class h323 tag • Call Agents – exercises control over operation of gateway
(config-‐class)# h225 timeout tcp establish seconds >> establish timeout o What events should be reported back to call agent
(config-‐class)# h225 timeout setup value >> SETUP response timeout o How endpoints should be connected
(config)# dial-‐peer voice tag voip o What signals should be implemented on endpoints
(config-‐dial-‐peer)# voice-‐class h323 tag o Uses directory of endpoints and the relationship each endpoint has with
(config)# voice service {pots | voatm | vofr | voip} dial plan to determine appropriate call routing
(conf-‐voi-‐serv)# h323 o Call agents initiate all VoIP call legs
(conf-‐serv-‐h323)# h225 timeout tcp call-‐idle {value value | never} >> Idle timer
Timer value in minutes (0-‐1440) MGCP Gateways
0 = disables timer, TCP connection closed immediately after call clears • Trunking gateway (TGW) – provides interface between PSTN and VoIP networks, can be
Never = connection maintained forever or until other endpoint closes it DS0/T1/E1, access servers or routers
• Residential gateway (RGW) – provide interface between analog (RJ-‐11) calls from a
Configuring H.323 Fax Pass-‐Through and Relay telephone and a VoIP network, cable-‐modem/2600-‐series-‐routers
Tones used by fax machines can be degraded by a codec to the point they’re unintelligible. Two
approaches exist to preserve fax tones across an IP WAN. Basic MGCP Concepts
• Fax Pass-‐Through • MGCP calls and connections – call agent instructs gateway to make a connection with a
(config)# dial-‐peer voice id voip specific endpoint, gateway returns session parameters to call agent, which then sends
(config-‐dial-‐peer)# destination-‐pattern pattern to other gateway, all connections for same call with share common Call ID and same
(config-‐dial-‐peer)# session target ipv4:ip-‐address media stream
(config-‐dial-‐peer)# fax protocol {cisco | none | system | pass-‐through {g711ulaw • MGCP control commands (8 command verbs)
|g711alaw}} o Used by call agent to query state of media gateway
(config-‐dial-‐peer)# fax rate {2400 | … | 14400} {disable | voice} [bytes rate] § AuditEndpoint (AUEP) – status of endpoint
§ AuditConnection (AUCX) – status of connection
fax protocol none = disables all faxing o Used by call agent to manage RTP connection on media gateway
no fax protocol = sets fax protocol for dial peer to default, which is system § CreateConnection (CRCX) – tells gateway to establish
connection with the endpoint
Note 1: If the fax protocol (voice-‐service) command is used to set fax relay options § DLCX – tells gateway to delete a connection
for all dial peers and the fax protocol (dial-‐peer) command is used on a specific § ModifyConnection (MDCX) tells gateway to update its
dial-‐peer, the dial-‐peer configuration takes precedence. connection parameters for a previously established
Note 2: If fax rate transmission is higher than the codec rate in the same dial peer, connection
the data send over the network for fax is above bandwidth reserved for RSVP o Used by call agent to request notification of events and apply signals
§ NotificationRequest (RQNT) – tells gateway to watch for
events on an endpoint and specifies action to take

o Used by media gateway to indicate it has received an event that call Implementing SIP Gateways
agent had requested notification for.
§ RestartInProgress (RSIP) – notifies call agent that the
SIP Overview
gateway and its endpoints are removed from service or
• ASCII text-‐based application layer control protocol
are being placed back in service
• Enables Internet endpoints (user agents) to discover one another and agree on the
• Package types
parameters of the session they want to share
o MGCP groups events/signals into packages
• Supports five facets of establishing/terminating multimedia communications:
o Supported by default on certain types of gateways
o Determines location of target endpoint – supports address resolution,

name mapping, and call redirection
o Trunk: mgcp package-‐capability trunk-‐package
o Determines media capabilities of the target endpoint – finds lowest
o Line: mgcp package-‐capability line-‐package
level of common services between endpoints through SDP, conferences
o DTMF: mgcp package-‐capability dtmf-‐package
established only use level of capability that all participants share
o Generic media: mgcp package-‐capability gm-‐package
o Determines availability of target endpoint – if target endpoint not
o RTP: mgcp package-‐capability rtp-‐package
available, SIP returns a message indicating why target unavailable
o Announce server: mgcp package-‐capability as-‐package
o Establishes a session between originating and target endpoints – also
o Script: mgcp package-‐capability script-‐package
supports mid-‐call changes, i.e. adding another endpoint in conference

o Handles transfer/termination of calls
MGCP Call Flows

How SIP Works
• User identified by unique SIP address, similar to an email address in the format of
sip:userID@gateway.com
• Users register with a registrar server using their assigned SIP address

Why SIP
• Dial plan configuration directly on the gateway – allows special calls to be routed
without the assistance of CUCM
• Translations defined per gateway – allows calling part transformations or special
number formats, allows translation of incoming calls directly on gateway to meet
internally used number formats
• Advanced support of third-‐party telephony systems
• Interoperability with third-‐party voice gateways – most feasible way to connect a Cisco
IOS voice gateway to a third-‐party voice gateway

SIP Architecture
• Session Peers / User Agents (UA)
o User agent client (UAC) – client application that initiates SIP request
o User agent server (UAS) – server application that contacts user when SIP
invitation received and returns a response on behalf of the user to the
invitation originator
• Clients (endpoints)
o Phone – act as UAS/UAC on session-‐by-‐session basis
§ Initiate/Respond to SIP requests
Configuring MGCP Gateways § ephones are IP phone not configured on gateway
MGCP Residential Gateway o Gateway – acts as UAS/UAC and provides call control support, translation
(config)# ccm-‐manager mgcp >> only required if call agent is CUCM function between SIP conferencing endpoints, performs call-‐setup and
(config)# mgcp call-‐clearing on both IP side and SCN side
(config-‐mgcp)# mgcp call-‐agent 172.20.5.20 service-‐type mgcp >> at least one required • Servers
(config)# dial-‐peer voice 1 pots o Proxy server – receives SIP requests from client and forwards requests on
(config-‐dialpeer)# application mgcpapp >> MGCP application to run on port behalf of client to next SIP server in network, perform authentication,
(config-‐dialpeer)# port 1/0/0 >> specify voice port to bind with MGCP authorization, network access control, routing, reliable request
(config)# dial-‐peer voice 2 pots transmission and security
(config-‐dialpeer)# application mgcpapp o Redirect server – provides client with information about next hop(s) that
(config-‐dialpeer)# port 1/0/1 a message should take, and then client contacts next hop server directly,
(config-‐dialpeer)# exit can be another network server or a UA
(config)# mgcp package-‐capability dtmf-‐package o Registrar server – receives requests from UACs for registration
(config)# mgcp package-‐capability gm-‐package o Location server – address resolution services to SIP proxy/redirect
(config)# mgcp package-‐capability line-‐package servers, resolves addresses
(config)# mgcp package-‐capability rtp-‐package
(config)# mgcp default-‐package line-‐package SIP Call Flow
Direct Call Setup
MGCP Trunk Gateway
(config)# ccm-‐manager mgcp >> only required if call agent is CUCM
(config)# mgcp 4000
(config)# mgcp call-‐agent 10.1.1.201 4000 >> at least one required
(config)# controller t1 0/1/1
(config-‐controller)# framing esf
(config-‐controller)# clock source internal
(config-‐controller)# ds0-‐group 1 timeslots 1-‐24 type none service mgcp >> only for trunk
(config)# mgcp package-‐capability dtmf-‐package
(config)# mgcp package-‐capability gm-‐package
(config)# mgcp package-‐capability line-‐package
(config)# mgcp default-‐package line-‐package
Note: Specify event extra packages supported
on trunking gateway, default is trunk-‐package

Configuring Fax Pass-‐Through and Relay with MGCP Gateway
(config)# ccm-‐manager mgcp >> Configure call agent
(config)# no ccm-‐manager fax protocol cisco >> Disable Cisco Fax Relay
(config)# mgcp >> Enable MGCP (add port if not default) Call Setup Using a Proxy Server
(config)# mgcp call-‐agent 10.1.1.10 service-‐type mgcp version 0.1
(config)# mgcp package-‐capability fxr-‐package >> specify additional MGCP package cap.
(config)# mgcp fax rate 14400 >> specify maximum fax rate
(config)# mgcp timer 300 >> adjust NSE timers for network conditions
(config)# mgcp fax-‐relay sg3-‐to-‐g3 >> configure fax machines to negotiate down to G3 speeds

Verifying MGCP
• show mgcp
• show ccm-‐manager
• show mgcp endpoint
• show mgcp statistics
• debug voice ccapi inout – shows every interaction with call control API on telephone
interface and VoIP side, allows you to follow progress of a call
• debug mgcp [all | errors | events | packets | parser] – reports all command activity


Call Setup Using a Redirect Server debug ccsip transport: Enables tracing of the SIP transport handler and the TCP or UDP process.

Chapter 6: Identifying Dial Plan Characteristics
Introducing Dial Plans

Dial Plan Overview
• A dial plan is a numbering plan for the voice-‐enabled network
• Endpoint addressing (Numbering plan) – assigning numbers to endpoints
• Call routing and path selection – option to select different paths to reach the same
destination, using a secondary path if the primary path goes down
• Digit manipulation – can occur prior to or after a routing decision has been made
• Calling privileges – assigning different groups of devices to different classes of service
• Call coverage – special groups of devices to handle incoming calls for a certain service
according to different rules (top-‐down, circular-‐hunt, longest-‐idle, broadcast)

SIP Addressing
• sip: or ssip: (for secure SIP connections) as URL type
• When two UAs communicate with each other, the current destination and final
destination are the same. Different with use of proxy/redirect server.

SIP DTMF Considerations
• SIP DTMF relay method is required in following situations:
o Connecting Cisco SRST system to remote SIP-‐based IVR or vm app
o Connecting Cisco SRST system to remote SIP PSTN voice gateway that
goes through PSTN to a voice-‐mail or IVR application
• SIP usually sends DTMF in-‐band digits, whereas SCCP supports only out-‐of-‐band digits.
The software-‐based Media Termination Point (MTP) device receives the DTMF out-‐of-‐
band tones and generates DTMF in-‐band tones for the SIP client. Digit Manipulation
• POTS dial peers support wider range of commands for simple digit stripping/prefixing
• VoIP dial peers are primarily dependent on voice translation profiles.

Calling Privileges – implemented on Cisco IOS gateways using COR and COR lists

Call Coverage – goal is to lose as few calls as possible
• Call coverage for individual users – tries to forward call to other uses or to voice mail in
case this user does not answer a call

• NOTIFY-‐based out-‐of-‐band DTMF relay sends messages bidirectionally between • Pilot numbers with associated user groups – used to distribute incoming calls
originating and terminating gateways for a DTMF event during a call.
• If multiple DTMF relay mechanisms are enabled on a SIP dial peer and are negotiated Scalable Dial Plans
successfully, NOTIFY-‐based out-‐of-‐band DTMF relay takes precedence. • Dial-‐plan logic distribution – effective distribution of dial-‐plan logic among various
components, try to localize processing as much as possible
Configuring SIP • Hierarchical numbering plan -‐ scale number devices without introducing interdigit
Enable the SIP voice service within Cisco IOS timeout or routing issues due to overlapping ranges, need good route summarization
(config)# voice service {pots | voatm | vofr | voip} • Simplicity in provisioning – keep it simple and symmetrical, made consistent using
(conf-‐voi-‐serv)# sip translation rules to manipulate local digit dialing patterns, standardize before entering
(conf-‐voi-‐serv)# no shutdown VoIP core
• Post dial delay -‐ time between when the last digit is dialed and the phone rings at the
Specify the parameters for the SIP service receiving location, excessive translations/digit-‐manipulations/lookups >> lag time
(conf-‐serv-‐sip)# session transport { tcp | udp } • Availability and fault tolerance – SRST is a good option here
(conf-‐serv-‐sip)# bind {control | media | all} source-‐interface interface-‐id • Conformance to public standards
(conf-‐serv-‐sip)# exit
PSTN Dial Plan Requirements
Configure the SIP UA • Inbound call routing
(config)# sip-‐ua • Outbound call routing
(config-‐sip-‐ua)# authentication username name password password realm string • Correct PSTN Automatic Number Identification (ANI) presentation – should include the
(config-‐sip-‐ua)# registrar name expires secs PSTN access code and any other identifiers required by the PSTN to successfully place a
(config-‐sip-‐ua)# sip-‐server {dns:[hostname] | ipv4:ip_addr:[port-‐num]} call using that ANI
(config-‐sip-‐ua)# retry {invite number | response number | bye number | cancel number}
ISDN Dial Plan Requirements
Configure SIP-‐based VoIP dial peers to connect and route calls to the service provider's SIP network • Correct PSTN inbound ANI presentation depending on TON – number displayed must
(config)# dial-‐peer voice 2000 pots be a number that can be dialed back
(config-‐dial-‐peer)# destination-‐pattern 2... • Correct PSTN outbound ANI presentation depending on TON
(config-‐dial-‐peer)# session protocol sipv2
(config-‐dial-‐peer)# session target sip-‐server >> sip-‐server was entered in SIP UA config Configuring PSTN Dial Plans
(config-‐dial-‐peer)# dtmf-‐relay rtp-‐nte >> how DTMF signaling will be sent Configure digit manipulation for PSTN calls
(config)# dial-‐peer voice 2001 voip Inbound
(config-‐dial-‐peer)# destination-‐pattern 2… (config)# voice translation-‐rule 1 >> Create voice translation rule
(config-‐dial-‐peer)# session protocol sipv2 (cfg-‐translation-‐rule)# rule 1 /^4085552/ /2/ >> Create sub-‐rule to the rule
(config-‐dial-‐peer)# session target ipv4:10.1.1.15 >> Distinguishes one session from another (cfg-‐translation-‐rule)# exit
(config-‐dial-‐peer)# dtmf-‐relay sip-‐notify (config)# voice translation-‐profile pstn-‐in >> Create voice translation profile
(cfg-‐translation-‐profile)# translate called 1 >> Assign v.t. rule to v.t. profile
Verifying SIP Gateways (cfg-‐translation-‐profile)# exit
• show sip service – displays if service is up/down (config)# voice-‐port 0/0/0:23
• show sip-‐ua status – displays status for SIP user agent (config-‐voiceport)# translation-‐profile incoming pstn-‐in >> Assign v.t. profile to port
• show sip-‐ua timers – displays current SIP UA timers
• show sip-‐ua register status – display status of E.164 numbers that a SIP gateway has Outbound
registered with an external primary SIP registrar (config)# voice translation-‐rule 2
• show sip-‐ua calls – detailed information about current SIP calls (cfg-‐translation-‐rule)# rule 1 /^2/ /4085552/
(cfg-‐translation-‐rule)# exit
debug asnl events: Use this command to verify that the SIP subscription server is up (config)# voice translation-‐profile pstn-‐out
debug voip ccapi inout: Follow progress of call from inbound intf. or VoIP peer to outbound side of call (cfg-‐translation-‐profile)# translate calling 2
debug voip ccapi protoheaders: Displays messages sent between originating/terminating gateways (cfg-‐translation-‐profile)# exit
debug ccsip all: Enables all ccsip-‐type debugging. (config)# voice-‐port 0/0/0:23
debug ccsip calls: Displays all SIP call details as they are updated in SIP call control block (config-‐voiceport)# translation-‐profile outgoing pstn-‐out
debug ccsip errors: Traces all errors encountered by the SIP subsystem.
debug ccsip events: Traces events, such as call setups, connections, and disconnections Configure digit manipulation for intersite calls
debug ccsip info: Enables tracing of general SIP Service Provider Interface (SPI) information (config)# num-‐exp 3... 915125553...
debug ccsip media: Enables tracing of SIP media streams.
debug ccsip messages: Shows the headers of SIP messages exchanged between a client and a server.
debug ccsip preauth: Enables diagnostic reporting of authentication, authorization, and accounting
debug ccsip states: Displays SIP states and state changes for sessions within SIP subsystem.

Configure dial peer matching 911 Services

Outbound
(config-‐dial-‐peer)# destination-‐pattern 9[2-‐9]..[2-‐9]......
(config-‐dial-‐peer)# direct-‐inward-‐dial
(config-‐dial-‐peer)# port 0/0/0:23

Note: direct-‐inward-‐dial command is configured to enable call routing using the received called-‐
number information.

Inbound
(config-‐dial-‐peer)# destination-‐pattern 9[2-‐9]..[2-‐9]......
(config-‐dial-‐peer)# direct-‐inward-‐dial
(config-‐dial-‐peer)# incoming called-‐number 2...
• Automatic Number Identification (ANI) – calling party number
• Automatic Location Identification (ALI) – map phone number >> geographical location

• Public Safety Answering Point (PSAP) – where emergency calls are terminated
Note: Although call routing might work without any additional configuration, always configure
appropriate inbound dial-‐peer matching using incoming called-‐number or answer-‐address to • Emergency Response Location (ERL) – used in mobility environments by associating
ensure that a call is actually routed as desired. ports and devices to ERL groups, which define exact location
• Master Street Address Guide (MSAG) – database maintained by government agency,
Verifying PSTN Dial Plans lists geographic location of PSAPs
• show dial-‐peer voice number – Displays information for a specific voice dial peer • Selective router – dedicated 911 switch in service provider network that routes 911 calls
• show dial-‐peer voice summary – Displays a short summary of each voice dial peer to appropriate PSAP based on the calling number
• show dialplan number dial-‐string [carrier identifier] [fax | huntstop | voice] [timeout] • Centralized Automated Message Accounting (CAMA) – type of analog trunk that
Displays which outgoing dial peer reached when a particular telephone number is dialed connects a customer switch directly to selective router, only carries 911 calls

• debug isdn q931 – Debugs ISDN Layer 3 information, including DNIS/ANI information Chapter 7: Configuring Advanced Dial Plans
• debug voip dialpeer – Debugs dial-‐peer matching
• debug voice translation – Debugs voice-‐translation-‐rule operation Configuring Digit Manipulation

Number Plan Fundamentals Digit Manipulation
Adding/subtracting from original dialed number, accommodates user dialing habits.
Numbering Plan Categories Because call agent performs digit manipulation in MGCP network, only might need to be performed on
• Private numbering plans – make plan coincide with publicly assigned number ranges H.323 and SIP gateways.
• Public/PSTN number plans
o International Numbering Plan (E.164) – 1-‐3 digit country code (CC)
followed by a national destination code (NDC) following by a subscriber
number (DN), up to 15 digits in an E.164 number
o National number plans – specific country or group of countries, example
is NANP which defines a 10-‐digit numbering plan

Private Numbering Plan Design Considerations
• Number of addressable devices – use fewest possible digits
• Number of locations – use fewest possible digits
• Inbound call routing – match numbers one-‐to-‐one or use single number for company
• PSTN access codes – cannot use PSTN access code digit to begin an extension since 9 is a
PSTN access code, same for operator digits (example, 0)
• Basic digit manipulation
PSTN Number Plan Common Elements o num-‐exp – global command, performs match-‐and-‐replace operation,
• Emergency services occurs prior to matching a dial peer
• Information or directory services o digit-‐strip – dial-‐peer command that strips off the matched digits, occurs
• Local calls – usually free and within a local geographic area after outbound dial-‐peer is matched and before any digits sent out,
• Mobile calls – most countries charge for mobile phone calls enabled by default on POTS dial peers
• Long distance or toll – usually require an access code o prefix – adds prefix to number, only applies to POTS dial peers
• Toll free o forward-‐digits {num-‐digits | all | extra} – specifies digits to forward
• Premium – per-‐minute or per-‐call basis o clid – used to modify calling line ID
• International • Voice translation rules and profiles – most powerful tool

Scalable Numbering Plans Digit Stripping
• Simple provisioning – capability to easily add new groups and modify existing groups • By default, POTS dial peer strips any explicitly defined digits in a destination pattern
• Simplified routing – keep local calls local and uses specialized number key, such an an • No need to use no digit-‐strip for VoIP dial-‐peers
area code, for long-‐distance calls
• Summarization – groups numbers in specific geographical area or functional group Digit Forwarding – forward-‐digits {value | all}
• Scalability – provides additional high-‐level number groups • Forwards right-‐justified-‐number of digits
• Management – controls number groups from a single point in the overall network
Digit Prefixing – prefix {value}
GOAL:
• Minimal impact on existing systems Number Expansion – num-‐exp {match} {change}
• Minimal impact on users of the system • show num-‐exp – view configured number expansion table
• Minimal translation configuration • show dialplan number string – confirm presence of a valid dial peer to match string

• Consideration of anticipated growth
Caller ID Number Manipulation
Overlapping Numbering Plans • clid network-‐number number – sets network-‐provided number in the Information
Occur when DIDs assigned by service provider have the same extension ranges at two or more Element (IE) message, sets presentation bit to allow calling part number to be presented
locations. This can be resolved in several ways: • clid second-‐number strip – removes user-‐provided number, or second number, from
• Redesign directory number ranges to ensure non-‐overlapping directory numbers this IE message.
• Use an intersite access code and a site code that will be prepended to directory • clid restrict – sets presentation bit to prevent display of CLID information, but does not
numbers to create unique dialable numbers. remove from the IE message
o Completely remove numbers using clid strip
• Do not assign DIDs. Instead, publish a single number and use a receptionist or AA
• station-‐id {[name string] | [number string]}
Private and Public Numbering Plan Integration • show dialplan number string – determine what CLID string will be sent out
Challenges faced with number plan integration:
• Varying number length – might need digit manipulation to accommodate varying Voice Translation Rules and Profiles
number lengths for local, long-‐distance, key system and Centrex dialing • Configuring voice-‐ translation rule is a two-‐step process
• Specialized services – services such as Centrex and other equivalents have four-‐digit or o Translation rule defined at the global level
five-‐digit dialing. Dialing from PSTN into private VoIP network and then out to a Centrex o Translation rule applied at the dial-‐peer level as inbound/outbound
extension call also require extensive digit manipulation. translation on either the called/calling number
• Voice mail – sometimes voice-‐mail system can require a different number plan than the • Maximum of 15 sub-‐rules that include Stream Editor-‐like expressions
endpoint telephones, thus requiring translation • Maximum of 128 translation rules are supported
• Necessity of prefixes or area codes – example, rerouting calls from IP network to the
PSTN for failure recovery can require extra digits
• International dialing consideration
• End user training and “ease of use”


• CME configured to use voice-‐translation profile instead of dialplan-‐pattern command

• Verify using the test voice translation-‐rule

Regular expressions for voice translation rules
^ Match the expression at the start of a line.
$ Match the expression at the end of the line.
/ Delimiter that marks the start and end of both the matching and replacement strings. • Verify using the show voice translation-‐rule and show voice translation-‐profile
\ Escape the special meaning of the next character.
– Indicates a range when not in the first/last position. Used with the "[" and "]"
characters.
[list] Match a single character in a list.
[^list] Do not match a single character specified in the list.
. Match any single character.
* Repeat the previous regular expression (regex) zero or more times.
+ Repeat the previous regular expression one or more times.
? Repeat the previous regular expression zero or one time (use CTRL-‐V to enter in Cisco
IOS, because Cisco IOS interprets a "?" character as a request for context-‐sensitive help).
() Groups regular expressions.

Example:
rule 1 /$9$$[^10].*$/ /\11408\2/

Configuring Path Selection

Call Routing and Path Selection
• Each call passing through a Cisco IOS router is considered to have two call legs, one
entering the router and one exiting the router.
• Two main types of call legs:
Note: The first set of parenthesis is referenced as \1 and the second set as \2 o Traditional TDM call legs that connect router to PSTN, analog phone/fax
o IP call legs that connect router to other gateways/keepers or CUCM
Voice Translation Profiles • Two main types of dial peers
o POTS dial peers
• Types of call numbers in a translation profile o VoIP dial peers
o called – defines for called number (DNIS) • Router selects a dial peer by matching the information elements in the setup message
o calling – defines for calling number (ANI)
(called number/DNIS and calling number/ANI) with four configurable dial-‐peer
o redirect-‐called –defines for redirect-‐called number
attributes. Inbound dial-‐peer matching is prioritized as follows:
• Reference translation profile by: o Dialed number (DNIS) using incoming called-‐number
o Dial peer – two different profiles can be defined: incoming/outgoing o ID information (ANI) using answer-‐address
o Trunk group – two different profiles can be defined: incoming/outgoing o Caller ID information (ANI) using destination-‐pattern
o Voice port – two different profiles can be defined: incoming/outgoing, if o Incoming POTS dial peer by using port
voice port also trunk group member >> voice port overrides trunk group o No match >> dial peer 0
o NFAS interface – defined through the translation-‐profile command,
• Outbound dial-‐peer matching is prioritized as follows by default:
higher precedence than translation profile of a voice port or trunk group o Searches through all dial peers and tries to match the called number (the
o Source IP group – can be defined in source IP group for VoIP calls
DNIS) with the destination-‐pattern >> closest match is selected.
o VoIP incoming – defined globally for all incoming VoIP (H.323/SIP) calls,
o If multiple equal matches >> lowest preference configuration wins
lower precedence than source IP group
o If equal preferences are found, a random dial peer is selected.

• ANI and DNIS Matching on Dial Peers
Voice Profile Example
o (*) and (#) As they appear on standard touch-‐tone dial pads
(config)# voice translation-‐rule 1
o (,) Inserts pause between digits
(config-‐translation-‐rule)# rule 1 /^4085552/ /2/
o (.) Wildcard
(config-‐translation-‐rule)# exit
o (%) Preceding digit occurred zero or more times
o (+) Preceding digit occurred one or more times
(config-‐translation-‐rule)# rule 1 /^.*/ /9&/ type subscriber subscriber
o (^) Match to the beginning of the string
(config-‐translation-‐rule)# rule 2 /^.*/ /91&/ type national national
o ($) Match null string at end of input string
(config-‐translation-‐rule)# rule 3 /^.*/ /9011&/ type international international
o (\) Followed by single character and matches that character
o (?) Preceding digit occurred zero or one times
(config)# voice translation-‐profile Pstn-‐In
o ( [ ] ) Range, only numbers 0-‐9
(cfg-‐translation-‐profile)# translate called 1
o ( ( ) ) Indicate pattern and are same as regular expression rule
(cfg-‐translation-‐profile)# translate calling 2

• direct-‐inward-‐dial – Enable DID call treatment for incoming called number
Call Blocking Example
• preference value – Indicate the preferred order
• no dial-‐peer outbound status-‐check pots – Check status of outbound POTS dial peers
(config-‐translation-‐rule)# rule 1 reject /408555*/
during call setup and to disallow, for that call, any dial-‐peers who status is down
(config)# voice translation profile BlockCall
Matching Dial Peers in a Hunt Group
(cfg-‐translation-‐profile)# translate calling 1
(cfg-‐translation-‐profile)# exit • Longest match in phone number – looks for match with greatest number dialed digits
(config)# dial-‐peer voice 111 pots • Explicit preference – lower the preference number, higher the priority
(config-‐dial-‐peer)# call-‐block translation-‐profile incoming BlockCall • Random selection – all weighted equally
(config-‐dial-‐peer)# call-‐block disconnect-‐cause incoming invalid-‐number
Optional values call-‐reject, invalid-‐number, unassigned-‐number, user-‐busy Can be changed using the dial-‐peer hunt global configuration command (0 is default)
• 0: Longest match in phone number, explicit preference, random selection.
Voice Translation Profiles Versus the dialplan-‐pattern Command • 1: Longest match in phone number, explicit preference, least recent use.
• SRST and CME use dialplan-‐pattern command to map ephone-‐dns to DID numbers, but • 2: Explicit preference, longest match in phone number, random selection.
they also increase the number of dial peers, which makes troubleshooting hard. • 3: Explicit preference, longest match in phone number, least recent use.
• 4: Least recent use, longest match in phone number, explicit preference.
• 5: Least recent use, explicit preference, longest match in phone number.
• 6: Random selection.
• 7: Least recent use.

• Problem: no support for FXS and voice-‐mail pilots >> Solution: voice translation-‐profiles

Inbound Site-‐Code Dialing Example

H.323 Dial-‐Peer Configuration Best Practices

Note: Notice the default dial peer with incoming called-‐number . – It should be the only dial
peer that contains a “.” For the destination pattern and direct inward dial. Should not contain a
port number.

Path Selection Strategies
• Site-‐code dialing – allows users to place intersite call by dialing a site code
• Toll-‐bypass – uses WAN link for call routing to avoid PSTN charges

• TEHO – similar to toll-‐bypass but extends WAN usage for PSTN calls as well Configuring TEHO
Site-‐Code Dialing and Toll-‐Bypass
Idea: routing intersite calls over an IP WAN link

Tail-‐End Hop-‐Off (TEHO) -‐ Routing PSTN calls over IP WAN to PSTN demark close to dialed destination

Configuring Site-‐Code Dialing and Toll-‐Bypass

1. Define the VoIP outbound digit manipulation.

2. Define the outbound VoIP dial peer.

1. Configure voice translation rules and voice translation profiles for inbound/outbound
VoIP intersite routing 3. Define the outbound POTS dial peer.

Implementing Call Privileges on Cisco IOS Gateways

Calling Privileges
Capability to deny certain call attempts based on incoming/outgoing CORs provisioned on the dial peers
2. Define dial peers for VoIP intersite routing that route the call using the WAN link

Understanding COR on Cisco IOS Gateways
Define incoming/outgoing COR lists. Each list includes members, which are tags defined within IOS
• If COR applied on an incoming dial-‐peer is a super set or equal to the COR applied to the

3. Configure voice translation rules and voice translation profiles for inbound/outbound outgoing dial-‐peer, the call goes through. Otherwise, call is rejected.
PSTN intersite routing • If not corlist statements are applied to some dial peers, the following apply:
o No incoming corlist setup on dial-‐peer >> default corlist -‐highest prior.
o No outgoing corlist setup on dial-‐peer >> use default corlist -‐lowest prior.
• Two components of COR
o COR – building block of calling privileges
o corlist – contains multiple CORs and is bound to dial peers
4. Define dial peers for PSTN intersite routing that route the call using the PSTN link in case
• Call Routing with Corlists
the WAN link is not available

Outbound Site-‐Code Dialing Example


Four CORs : Local, LD, INTL, 911

Chapter 8: Configuring H.323 Gatekeepers
Used to create three incoming
corlists: Local, LD, INTL H.323 Gatekeeper Fundamentals

Gatekeeper Overview

Understanding COR for SRST and CME
COR with SRST and CME >> corlist cannot be simply bound to all dial peers, because one call leg will be
represented by dynamic dial peers derived from ephones.
CME: ephone-‐dn 1
corlist incoming INTL
Configured on ephone
SRST: call-‐manager failback
cor incoming INTL 1 2000-‐2100
Configured for range of numbers • Address resolution – must be able to resolve the alias or the E.164 telephone number
Up to 20 tags can be configured into the network address for the destination terminal, performed using a translation
table that is updated with registration messages
• Admission control – can control the admission of the endpoints into the H.323 network,
uses the following RAS messages: Admission Request (ARQ), Admission Confirmation
Configuring COR for Cisco Unified Communications Manager Express (ACF), and Admission Reject (ARJ). Might also be a null function that admits all requests.
1. Define individual “tags” (CORS) to be used as corlist members • Bandwidth control – manages endpoint bandwidth requirements, endpoint specifies
(config)# dial-‐peer cor custom preferred codec during gatekeeper registration, following RAS messages used to control
(config-‐dp-‐cor)# name 911 negotiation: Bandwidth Request (BRQ), Bandwidth Confirmation (BCF), and Bandwidth
(config-‐dp-‐cor)# name local Reject (BRJ)
(config-‐dp-‐cor)# name ld • Zone management – required to provide address translation, admission control, and
(config-‐dp-‐cor)# name intl bandwidth control for terminals/gateways/MCUs located within its zone of control
2. Define corlists that will be assigned “outgoing” to the PSTN dial peers • Call authorization (opt.) – can restrict access to certain endpoints/gateways based on
(config)# dial-‐peer cor list 911cal policies, such as time of day
(config-‐dp-‐corlist)# member 911 • Call management (opt.) –maintains active call information and uses it to indicate busy
(config)# dial-‐peer cor list localcall endpoints or redirect calls
(config-‐dp-‐corlist)# member local • Bandwidth management (opt.) –reject admission when req. bandwidth not available
(config)# dial-‐peer cor list ldcall
(config-‐dp-‐corlist)# member ld Zones – set of H.323 nodes controlled by a single local gatekeeper,
(config)# dial-‐peer cor list intlcall • Only one active gatekeeper per zone
(config-‐dp-‐corlist)# member intl • Endpoints discover gateway and zone membership using RAS messages
3. Define corlists that will be assigned “incoming” to the PSTN dial peers • If message sent via multicast, endpoints registers non-‐deterministically with first
… gatekeeper that responds to the message
(config)# dial-‐peer cor list intl • zone subnet command defines the subnets that constitute gatekeeper zones
(config-‐dp-‐corlist)# member 911 Zone Prefixes – determines to which zone calls are sent
(config-‐dp-‐corlist)# member local • Zone controlled by gatekeeper >> zone prefixes help route call to right endpoint
(config-‐dp-‐corlist)# member ld • Typically area codes
(config-‐dp-‐corlist)# member intl
• Serve same purpose as the domain names in the H.323-‐ID address
4. Associate corlists with existing VoIP or POTS PSTN dial-‐peers
• zone prefix GK1 952....... >> Gatekeeper knows to send those calls to GK1
…

Technology Prefixes – denote different types or classes of gateways
(config-‐dial-‐peer)# destination-‐pattern 9011T
• Voice gateways could register with technology prefix 1#, H.320 gateways 2#, etc.
(config-‐dial-‐peer)# prefix 011
(config-‐dial-‐peer)# corlist outgoing intlcall • If there is more than one possible gateway, the registration is done randomly
Technology Prefix with Hop-‐off
5. Assign corlists for user privileges to corresponding ephone-‐dns • When endpoint/gateway makes call admission request to its gatekeeper,
(config)# dial-‐peer voice 1004 pots gatekeeper resolves the dest. address by looking for the technology prefix
(config-‐dial-‐peer)# destination-‐pattern 1004$ • When matches, remaining string is compared against known zone prefixes
(config-‐dial-‐peer)# port 1/0/1 • If resolves to remote zone, entire address including both technology/zones
(config-‐dial-‐peer)# corlist incoming 911 prefixes is sent to the remote gatekeeper in a Location Request (LRQ)
(config-‐dial-‐peer)# corlist incoming local • Remote gatekeeper uses technology prefix to device which gateways to hop off
(config-‐dial-‐peer)# corlist incoming ld o Zone prefix >> determines routing to a zone
(config-‐dial-‐peer)# corlist incoming intl o Technology prefix >> determines gateway in that zone
(config)# ephone-‐dn 1 • This can be overridden by associating a forced hop-‐off zone with a particular
(config-‐ephone-‐dn)# corlist incoming intl technology prefix, which forces the call to the specified zone, regardless of zone
prefix in the address.
Configuring COR for SRST
(config)# call-‐manager-‐fallback Gatekeeper Signaling – H.323 RAS / GUP / GKTMP
(config-‐cm-‐fallback)# cor incoming INTL 1 2000 – 2100 {<cr> | default} • H.323 RAS– subset of H.225 signaling protocol (call
Note 1: accomplishes COR for range of extensions 2000-‐2100 control/setup, signaling), uses UDP,
Note 2: default command instructs the router to use an existing default corlist o Sets up connections between H.323 endpoints
o No gatekeeper >> H.225 messages exchanged
Verifying COR directly between endpoints
show dial-‐peer cor command to display corlists and members o RTP ports negotiated
o H.245 TCP port negotiated during H.225
signaling for std. H.323 connection
o RTP UDP port range 16384-‐32767
• RAS Messages
o Gatekeeper discovery messages (GRQ/GCF/GRJ) – endpoint unicasts/multicasts
§ Initial signaling done through H.225 RAS
§ Unicast discovery: UDP 1718
§ Multicast: UDP 224.0.1.41
o Terminal and Gateway registration messages (RRQ/RCF/RRJ)
§ Lightweight registration – endpoints specify TTL value in it RRQ
message, when close to expiration gateway sends Keepalive=True
which refreshes existing registration
o Terminal and Gateway unregistration messages (URQ/UCF/URJ)
o Call Admission messages (ARQ/ACF/ARJ)


o Location messages (LRQ/LCF/LRJ) – used between interzone gatekeepers Gatekeeper-‐Routed Call Signaling with call-‐signaling channel handled by gatekeeper
§ lrq reject-‐resource-‐low >> gatekeeper rejects LRQ requests if no 1. The GK responds to ARQ and advises
terminating endpoints are available endpoint to perform call setup
§ LRQ Sequential >> provides redundancy and load-‐sharing features procedure with GK, not with
>> default forwarding mode, orig. gatekeeper forwards LRQ to terminating endpoint.
first gatekeeper in matching list, then waits for response before 2. Endpoint initiates setup request w/ GK
sending LRQ to next gatekeeper, uses fixed timer 3. GK sends its own request to terminating
• GKA(config)# gatekeeper endpoint and incorporates some details
• GKA(config-‐gk)# zone local GKA cisco.com acquired from originating request.
• GKA(config-‐gk)# zone remote GKB cisco.com 4. When connect message received from
• GKA(config-‐gk)# zone remote GKC cisco.com terminating endpoint, GK sends connect
• GKA(config-‐gk)# zone remote GKD cisco.com message to originating endpoint.
• GKA(config-‐gk)# zone prefix GKB 1408555.... seq 5. Two endpoints establish H.245 control channel between them. Call procedure continues normal
• GKA(config-‐gk)# zone prefix GKC 1408555.... seq
• GKA(config-‐gk)# zone prefix GKD 1408555.... seq Call flow with multiple gatekeepers
Note: Speed up timer using lrq lrj immediate-‐advance
§ LRQ Blast >> simultaneously send LRQs to all gatekeepers that
match the zone prefix
• GKA(config)# gatekeeper
• GKA(config-‐gk)# zone local GKA cisco.com
• GKA(config-‐gk)# zone remote GKB cisco.com
• GKA(config-‐gk)# zone remote GKC cisco.com
• GKA(config-‐gk)# zone remote GKD cisco.com
• GKA(config-‐gk)# zone prefix GKB 1408555.... blast
• GKA(config-‐gk)# zone prefix GKC 1408555.... blast
• GKA(config-‐gk)# zone prefix GKD 1408555.... blast
Note: if all three reply with positive conformation (LCF),
gatekeeper chooses one. Tailor the choice using cost and
priority keywords at end of zone remote statement.
Lower = Better
zone remote GKB cisco.com cost {#} priority {#}
o Status messages (IRQ/ICF/IRR/IACK/INAK)

o Bandwidth messages (BRQ/BCF/BRJ)
o Resource availability messages (RAI/RAC/RIP) – indicate/confirm/busy-‐wait Zone Prefixes
o Disengage messages (DRQ/DCF/DRJ)

Intrazone Call Setup
1. PA dials phone number 408 5555-‐2001 to PB
2. GA sends ARQ to GK, asking permission to call PB.
3. GK does a lookup, finds PB registered to GB, returns ACF
with IP address of GB. Technology Prefixes
4. GA sends an H.225 call setup message to GB with phone
number of PB.
5. GB sends H.255 call proceeding message to GA.
6. GB sends ARQ to GK, asking permission to answer GA's
call.
7. GK returns an ACF with IP add. of GA.
8. GB and GA initiate an H.245 capability exchange and
open logical channels.
9. GB sets up a plain old telephone service (POTS) call to PB at 408 555-‐2001.
10. When PB answers, GB sends an H.245 call connect message to GA. • Gatekeepers use technology prefixes to route calls when no E.164 addresses registered
11. Dual RTP streams flow between gateways. match the called umber. Without E.164 addresses registered, GK relies on two options:
o With technology prefix matches option – GK uses technology prefix
Interzone Call Setup appended in the called number to select destination gateway or zone
Difference between Intrasite and Interzone, is steps 3 and 4 o With default technology prefixes option – GK assigns default gateway(s)
3. GK1 does a lookup and does not find PB registered. for routing unresolved call addresses.
GK1 does a prefix lookup and finds a match with GK2. • If you use mostly voice gateways in network, you can configure gatekeeper to use 1#*
GK1 sends an LRQ to GK2 and a RIP to GA. (1# followed by zero or more characters) gateway as the default
4. GK2 does a lookup, finds PB registered, returns LCF to o (config-‐gk)#gw-‐type-‐prefix 1#* default-‐technology
GK1 with the IP address of GB.
Gatekeeper Call Routing

Call Disconnect
1. PB hangs up.
2. GWB sends a DRQ to GK2, disconnecting the call
between PA and PB. A DCF is received later
3. GWB sends Q.931 release complete mess. to GWA.
4. GWA sends a DRQ to GK1, disconnecting the call
between PA and PB. A DCF is received later.
5. GWA signals a call disconnect to the voice network.

Call Flows with a Gatekeeper
Typical call flow with call-‐signaling channel created from endpoint to endpoint

Gatekeeper Call Routing: Zone Prefixes and Default Technology Prefixes


Gatekeeper Call Routing: Zone Prefixes and Technology Prefixes Directory Gatekeeper Signaling

Gatekeeper Call Routing: Zone Prefixes and Registered Numbers Directory Gatekeeper
Create single local zone on DGK with local address
(config)# gatekeeper
(config-‐gk)# zone local DGK cisco.com 10.4.1.1
Create remote zones for each GK controlled by DGK
(config-‐gk)# zone remote SJCGK cisco.com 10.1.1.1 1719
(config-‐gk)# zone remote AUSGK cisco.com 10.2.1.1 1719
(config-‐gk)# zone remote NYCGK cisco.com 10.3.1.1 1719
Specify zone prefixes for remote gatekeepers
(config-‐gk)# zone prefix SJCGK 408*
(config-‐gk)# zone prefix DFWGK 972*
(config-‐gk)# zone prefix NYCGK 212*
Enable forwarding of location requests
(config-‐gk)# lrq forward-‐queries
(config-‐gk)# lrq lrj immediate-‐advance

Individual Gatekeeper

Create single local zone on GK with local address
Gatekeeper Call Routing: Remote Zone (config)# gatekeeper
(config-‐gk)# zone local SJCGK cisco.com 10.1.1.1
Create remote zone for DGK
(config-‐gk)# zone remote DGK cisco.com 10.4.1.1 1719
Specify zone prefix to register with DGK
(config-‐gk)# zone prefix SJCGK 408* gw-‐priority 10 SJCGW
Specify zone prefix of DGK
(config-‐gk)# zone prefix DGK *

Gatekeeper Transaction Message Protocol
• Can extend call control intelligence of a gatekeeper by converting incoming RAS
messages to text messages and sends them to an external server.
• Overrides default gatekeeper behavior

Verifying Gatekeepers

Gatekeeper Call Routing: Hop-‐Off Technology Prefix

Directory Gatekeepers
• LRQ forwarding allows gatekeeper to be appointed as a directory/super gatekeeper
• Then it is only necessary to configure each gatekeeper with its own local zones and zone
prefixes and a single match-‐all wildcard prefix for the zone of the directory gatekeeper
• Only directory gatekeeper has to be configured with the full set of all zones and zone
prefixes within the network.
• LRQ from a non-‐Cisco gatekeeper cannot be forwarded
• Directory Gatekeeper Characteristics Configuring H.323 Gatekeepers
o No longer need a full-‐mesh configuration between interzone gatekeepers
o LRQ messages are RAS messages triggered by an ARQ message from Gatekeeper Configuration Steps
endpoints that are forwarded from gatekeeper to gatekeeper 1. Configure local and remote zones on the gatekeeper.
o Limit five hops for an LRQ message >> allows up to a 4-‐tier GK hierarchy 2. Configure zone prefixes for all zones where calls should be routed.
o Centralize dial plan and server a potential interface to other centralized 3. Configure technology prefixes to provide more flexibility in call routing.
applications. 4. Configure gateways to use H.323 gatekeepers.
o In large-‐scale VoIP network, a centralized interface point is required, 5. Configure dial peers.
which can interact with Advanced Intelligent Network (AIN) in SS7,
GKTMP router servers, and AAA. Gateway Selection Process
• GK maintains separate gateway list, ordered by priority, for each zone prefix
• If GW does not have assigned priority for a zone prefix >> defaults to priority 5
• Priority 0 disables the use of a GW in a particular zone
• H.323 version 2 provides gateway resource reporting to notify gatekeeper when
H.323 resources are getting low. GK uses this info to determine which gateway
is used to complete a call.

Configuration Considerations Step 3: Configuring Technology Prefixes

• Multiple zones can be defined –supports intrazone/interzone communication
• Only one RAS IP argument can be defined for all local zones -‐ you cannot
configure each zone to use a different RAS IP address; if you define this IP in the
first zone section you can omit it for all subsequent zones
• Cannot remove a local zone if there are endpoints/gateways registered to it -‐
first shut down the gatekeeper, which forces the endpoints, gateways, and local
zones to unregister
• Multiple logical gatekeepers control the multiple zones on the same platform
• Maximum number of local zones defined by gatekeeper should not exceed 100

Step 1: Configuring Gatekeeper Zones

(config)# gatekeeper
(config-‐gk)# zone local SanJose cisco.com 10.1.1.10
(config-‐gk)# zone local Houston cisco.com
(config)# zone prefix SanJose 2... gw-‐priority 10 SanJose
(config)# zone prefix Houston 3... gw-‐priority 10 SanJose
(config-‐gk)# gw-‐type-‐prefix 99#* gw ipaddr 192.168.1.1 1720
(config-‐gk)# gw-‐type-‐prefix 1#* default-‐technology
(config-‐gk)# no shutdown
Note: By using the gw-‐type command above, users no longer have to prepend a technology
prefix on the address. Any address that does not contain an explicit technology prefix will be
routed to one of the voice gateways that registered with 1#

gw-‐type-‐prefix type-‐prefix [[hopoff gkid1] [hopoff gkid2] [hopoff gkidn] [seq | blast]]
[default-‐technology] [[gw ipaddr ipaddr [port]]]

(config)# gatekeeper • type-‐prefix – technology prefix recognized and stripped before checking
(config-‐gk)# zone local SanJose cisco.com 10.1.1.10 for zone profile
(config-‐gk)# zone local Houston cisco.com enable-‐intrazone • hopoff gkid (optional) – specify gatekeeper where call is to hop off,
(config-‐gk)# zone remote Austin cisco.com 10.1.1.12 regardless of zone prefix in destination address
• default-‐technology (optional) – gateways registering with this prefix
(config-‐gk)# no shutdown
option are used as the default for routing any addresses that are
Note: Setting the IP address for one local zone makes it the address used for all local zones
otherwise unresolved

• gw ipaddr ipaddr [port] (optional) – use this to indicate gateway is
zone local gatekeeper-‐name domain-‐name [ras-‐ip-‐address] [invia inbound_gatekeeper |
incapable of registering technology prefixes; when it registers, it adds the
outvia outbound_gatekeeper [enable-‐intrazone]]
gateway to the group for this type of prefix, just as if it had sent the
• gatekeeper-‐name – specifies the zone name, usually fully qualified
technology prefix in its registration
domain name of GK

• domain-‐name – specifies domain name served by this GK
Step 4: Configuring Gateways to Use H.323 Gatekeepers
• ras-‐ip-‐address (optional) – specifies IP address of one of the interfaces on
Enable the gateway process on the router.
the GK. When the GK respons to GK discovery messages, it signals the
(config)# gateway
endpoint or GW to use this address in future communications

• invia inbound_gatekeeper – GK used for calls entering this zone
Enter interface configuration mode for the interface you intend to use for communication with
• outvia outbound_gatekeeper – GK used for calls leaving this zone the H.323 gatekeeper, give an IP address, and configure it as an H.323 interface
• enable-‐intrazone – forces all intrazone calls to use the via gatekeeper (config)# interface loopback 0
(config-‐if)# ip-‐address 192.168.1.3 255.255.255.0
zone remote other-‐gatekeeper-‐name other-‐domain-‐name other-‐gatekeeper-‐ip-‐address (config-‐if)# h323-‐gateway voip interface
[port-‐number] cost cost-‐value [priority priority-‐value]] [foreign-‐domain] [invia
inbound_gatekeeper] | [outvia outbound_gatekeeper] Define the IP address on the gateway to be used for H.323 communication.
• other-‐gatekeeper-‐name (config-‐if)# h323-‐gateway voip bind srcaddr 192.168.1.3
• other-‐domain-‐name
• other-‐gatekeeper-‐ip-‐address Define the name and location of the gatekeeper.
• port-‐number (optional) – RAS signaling port for remote zone (1-‐65535, (config-‐if)# h323-‐gateway voip id Houston ipaddr 192.168.1.15 1719 priority 1
default 1719)
• cost cost-‐value (optional) – cost of zone (1-‐100, default 50) Specify the H.323 gateway name to identify it to its associated gatekeeper.
• priority priority-‐value (optional) – priority of zone (1-‐100, default 50) (config-‐if)# h323-‐gateway voip h323-‐id Houston
• foreign-‐domain (optional) – cluster is in a different administrative domain
Specify the technology prefix the gateway registers with the gatekeeper.
Step 2: Configuring Zone Prefixes (config-‐if)# h323-‐gateway voip tech-‐prefix 1#

Step 5: Dial-‐Peer Configuration
(config)# gateway
(config-‐dial-‐peer)# destination pattern 2...
(config-‐dial-‐peer)# tech-‐prefix 1#
(config-‐dial-‐peer)# session target ras
Note 1: Because of the GK setup, we don’t actually have to setup a technology prefix
Note 2: session target ras specifies the RAS protocol to determine IP address of the session
target, meaning a gatekeeper translates the E.164 address to an IP address

IMPORTANT: When dealing with service numbers, such as 911, make sure to include the no
e.164 register command
(config-‐dial-‐peer)# destination pattern 911
(config-‐dial-‐peer)# prefix 911
(config)# gatekeeper (config-‐dial-‐peer)# no e.164 register
(config)# zone local SanJose cisco.com 10.1.1.10 (config-‐dial-‐peer)# session target ras
(config)# zone local Houston cisco.com
(config)# zone prefix SanJose1 2... gw-‐priority 5 SanJose Verifying Gatekeeper Functionality
(config)# zone prefix SanJose2 2... gw-‐priority 10 SanJose • show gatekeeper gw-‐type-‐prefix: Displays the technology prefix of a gateway
(config)# no shutdown • show gatekeeper status: Displays the overall gatekeeper status, including zone status
• show gatekeeper zone prefix: Displays the zone prefixes known to a gatekeeper
zone prefix gatekeeper-‐name e164-‐prefix [blast | seq] [gw-‐priority priority gw-‐alias • show gatekeeper calls: Displays current calls known to a gatekeeper
[gw-‐alias, ...]] • show gatekeeper endpoints: Displays endpoints currently registered with a gatekeeper
• e164-‐prefix – E.164 prefix in standard form followed by dots (.) • show gatekeeper zone status: Displays the status of zones registered with a gatekeeper
• gw-‐priority priority gw-‐alias (optional) – defines how the gatekeeper • debug h225 {asn1 | events}: Displays H.225 activity in real-‐time
selects gateways in its local zone for calls to numbers beginning with the • debug h245 {asn1 | events}: Displays H.245 activity in real-‐time
specified e164-‐prefix (1-‐10, where 0 disables and 10 highest priority) • debug ras: Displays RAS messages, in real-‐time, to and from a gatekeeper


Providing Call Admission Control with H.323

Gatekeeper Zone Bandwidth Operation
Call Admission Control – After provisioned bandwidth has been fully utilized, the Cisco Unified IP
Communications system must reject subsequent calls to avoid oversubscription of priority queue on IP

WAN link, which would cause quality degradation for all voice calls.
• Limits the number of calls that can be active on a particular link at the same time
• Not guarantee a level of audio quality, does allow regulation of bandwidth consumption
• Gatekeeper is responsible for CAC between CUCM/CME and H.323 Gateways
• Bandwidth configuration for all zones is done on the gatekeeper

Zone Bandwidth Calculation
• (Number of Calls) * (Codec Payload Bandwidth) * 2 = Zone Bandwidth

• Example for three simultaneous G.711 calls in GK network: 3 * 64 kbps * 2 = 384 kbps
• Gatekeepers and CUCM servers have different bandwidth values for the same codecs

Chapter 9: Establishing Connection with ITSP

Introducing Cisco Unified Border Element Gateway

Cisco Unified Border Element Overview
• Standard GW without Cisco UBE functionality will not allow VoIP-‐to-‐VoIP connections
Configuring CAC on Gatekeeper • Provides network-‐to-‐network interface point for the following:
(config)# gatekeeper o Signaling interworking (H.323, SIP)
(config-‐gk)# bandwidth interzone default 128 o Media interworking (DTMF, fax, modem, codec transcoding)
(config-‐gk)# bandwidth total default 5000 o Address and port translations (privacy and topology hiding)
(config-‐gk)# bandwidth session default 384 o Billing and call detail record (CDR) normalization
(config-‐gk)# bandwidth session zone denver 256 o Quality-‐of-‐service (QoS) and bandwidth management

bandwidth {interzone | total | session | remote | check-‐destination} {default | zone zone-‐ Cisco IOS Image Support for Cisco UBE Gateways – IOS 12.2(13)T and later
name} bandwidth-‐size
• Interzone – total bandwidth for H.323 traffic from a zone to any other zone Cisco UBE Gateways in Enterprise Environments
• Total -‐ total bandwidth for H.323 traffic allowed in a zone • Serve two main purposes:
• Session – maximum bandwidth allows for a session in a zone o External connections – used as a demarcation point within a unified
• Remote – total bandwidth for H.323 traffic between this GK and any other GK communications network and provides interconnectivity with external
• check-‐destination – enables the GK to verify available bandwidth resources at a networks
destination endpoint o Internal connections – when used in a VoIP network, a Cisco UBE can be
• zone zone-‐name – specifies particular zone used to increase flexibility and interoperability between different devices
• bandwidth-‐size – maximum bandwidth in seconds, in kbps • Key features offered by Cisco UBE
o Protocol interworking
Between one zone and any other zone >> interzone default o Address hiding – can hide or replace endpoint IP addresses
Within one zone or for traffic between that zone and another zone >> total default o Security – can be placed in a DMZ and provide outside connectivity to
Single session within a specific zone >> session zone external networks
Single session within any zone >> session default o Video integration – supports H.323 video connections
o Call Admission Control (CAC) – can use CAC mechanisms, including RSVP

RAI in Gatekeeper Networks

Protocol Internetworking on Cisco UBE Gateways
• Interworking between the same signaling protocol
Gateways can be configured to report on DS0 and DSP resource status to its gatekeeper
• Interworking between different signaling protocols – can interconnect dial peers that
• Accomplished with the use of RAI RAS messages use different signaling protocols >> greater flexibility
• resource threshold [all] [high percentage-‐value] [low percentage-‐value]
o Default for high and low values is 90 Media Flows on Cisco UBE Gateways
o (Opt.) report-‐policy >> specify how resource utilization calculated • Media flow-‐through – UBE replaces source IP address used for media connections with
§ Idle-‐only >> includes free and in-‐use channels (default) its own IP address >> replaces potential duplicate IPs with one address and hides
§ Addressable >> includes free, in-‐use, and disables channels original endpoint IP address from the remote endpoints
o Higher CPU load on the router >> lowers number concurrent call flows
o Might result in suboptimal traffic flows because direct endpoint-‐to-‐
endpoint communication is prohibited


• Media flow-‐around – no duplicate IP address ranges exists, IP address hiding is not Cisco UCM to Cisco UCM
required, use when not concerned with hiding your network addresses

Codec Filtering on Cisco UBEs SIP Carrier Interworking

Gatekeeper and SIP Carrier Interworking

RSVP-‐Based CAC on Cisco UBEs

Cisco UBE and Via-‐Zone Gatekeeper

Cisco UBE Gateways and Gatekeeper Internetworking

Configuring Cisco Unified Border Elements

Protocol Interworking Command
(config)# voice service voip
(config-‐voice-‐service)# allow-‐connections h323 to 323
(config-‐voice-‐service)# allow-‐connections sip to sip
(config-‐voice-‐service)# allow-‐connections h323 to sip
(config-‐voice-‐service)# allow-‐connections sip to h323

• UBE can register with the gatekeeper
• Gatekeeper can use a register UBE router with via zones – when routing a call between Configuring H.323-‐toH.323 Interworking
two zones, a gatekeeper can be configured to route the call via a zone containing a UBE Step 1: Enabling H.323-‐to-‐H.323 Interworking
router. This enables interzone networking using a central UBE router without the need (config)# voice service voip
to deploy a Cisco UBE router at every site or redesign an already-‐deployed H.323 (config-‐voice-‐service)# allow-‐connections h323 to h323
network.
Step 2: Configuring H.323 Dial Peers
• When UBE router is used as an outbound voice gateway >> same concepts that apply
(config)# dial-‐peer voice 2001
when using traditional voice gateways with gatekeepers apply to UBE deployments
(config-‐dial-‐peer)# description To Cisco Unified Communications Manager
• When routing calls between zones that require UBE functionality >> via-‐zones should
(config-‐dial-‐peer)# destination-‐pattern 2...
be used, existing GK deployments can easily be modified to include UBE
(config-‐dial-‐peer)# session-‐target ipv4:192.168.1.1

(config-‐dial-‐peer)# exit
Cisco UBE Gateway Call Flows
(config)# dial-‐peer voice 3000
Cisco UCM to Cisco UCME
(config-‐dial-‐peer)# description To Cisco Unified Communications Manager Express
(config-‐dial-‐peer)# session-‐target ipv4:192.168.2.254


Configuring H.323-‐to-‐SIP Interworking Configure Cisco UBE

Step 1: Enabling H.323-‐to-‐SIP Interworking 1. Enable H.323 interworking.
(config)# voice service voip 2. Create a loopback interface to use as the source interface for Cisco UBE to register with
(config-‐voice-‐service)# allow-‐connections h323 to sip the gatekeeper. The Loopback1 interface is used as the H.323 gateway interface. The
(config-‐voice-‐service)# allow-‐connections sip to h323 Cisco UBE router will register in zone VIA with the H.323 ID CUBE and the technology
prefix 1#.
Step 2: Configuring Dial Peers 3. Create two dial peers—one pointing to San Jose and the other to Chicago.
For a SIP (rtp-‐nte)-‐to-‐H.323 (h245-‐alphanumeric) call via a Cisco UBE router, if any RTP named 4. Enable the gateway process.
telephony event (NTE) packets are sent before the H.323 endpoint answers the call, the DTMF
signal is not heard on a terminating gateway.
(config-‐dial-‐peer)# description To Cisco Unified Communications Manager
(config-‐dial-‐peer)# session target ipv4:192.168.1.1
(config-‐dial-‐peer)# dtmf-‐relay h245-‐alphanumeric
(config-‐dial-‐peer)# exit
(config-‐dial-‐peer)# description To International SIP Carrier
(config-‐dial-‐peer)# session protocol sipv2
(config-‐dial-‐peer)# destination-‐pattern 9011T
(config-‐dial-‐peer)# session target ipv4:192.168.10.254
(config-‐dial-‐peer)# dtmf-‐relay rtp-‐nte digit-‐drop h245-‐alphanumeric

Media Flow and Transparent Codec Commands
• media [flow-‐around | flow-‐through] –entered in dial-‐peer or globally under the voice service

• codec transparent – to configure transparent codec pass-‐through
Verifying Cisco UBEs and Via-‐Zone Gatekeepers

When you use the show gatekeeper endpoints command on the gatekeeper, a Cisco UBE router will be
Configuring Transparent Codec Pass-‐Through and Media-‐Flow-‐Around
displayed as an H323-‐GW type.
Codec transparency enables a Cisco UBE router to pass codec capabilities between endpoints. If you
configure transparency, a Cisco UBE router uses the codec that was specified by the endpoints for
setting up a call.

Show active calls on the gatekeeper

Configuring Cisco UBEs and Via-‐Zone Gatekeepers

Configure Gatekeeper
1. Create a loopback interface to use for the gatekeeper.
2. Create local, remote, and VIA zones.
3. Specify zone and technology prefixes.


642 436 Cvoice

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

642 436 Cvoice

Uploaded by

Copyright:

Available Formats

1: Introducing Voice over IP Networks • SCCP:

INTRODUCTING VOIP GATEWAYS

DTMF Support Calculating the Total Bandwidth for a VoIP Call

Configuring Conferencing and Transcoding on Voice Gateways

CAMA Trunk Configuration

• Components of the ISDN frame

H.323 Fast Connect Call Setup • Fax Relay

Configure dial peer matching 911 Services

Inbound Site-‐Code Dialing Example

Four CORs : Local, LD, INTL, 911

Configuration Considerations Step 3: Configuring Technology Prefixes

Providing Call Admission Control with H.323

Configuring H.323-‐to-‐SIP Interworking Configure Cisco UBE

You might also like

642 436 Cvoice

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

642 436 Cvoice

Uploaded by

Copyright:

Available Formats

1: Introducing Voice over IP Networks • SCCP:

INTRODUCTING VOIP GATEWAYS

DTMF Support Calculating the Total Bandwidth for a VoIP Call

Configuring Conferencing and Transcoding on Voice Gateways

CAMA Trunk Configuration

• Components of the ISDN frame

H.323 Fast Connect Call Setup • Fax Relay

Configure dial peer matching 911 Services

Inbound Site-­‐Code Dialing Example

Four CORs : Local, LD, INTL, 911

Configuration Considerations Step 3: Configuring Technology Prefixes

Providing Call Admission Control with H.323

Configuring H.323-­‐to-­‐SIP Interworking Configure Cisco UBE

You might also like

Inbound Site-‐Code Dialing Example

Configuring H.323-‐to-‐SIP Interworking Configure Cisco UBE