Performance comparison of Modulation and Coding set for IEEE 802.

16m system

Document Number:
IEEE C802.16m­09/0555r3
Date Submitted:
2009­03­02
Source:
Hua Zhou, Shengbao Liang, Yuanrong Lan, Jie Zhang, 
Yuantao Zhang, Jinyin Xue, Jian Wang, Jun Tian                   Voice: +86­10­59691519
Fujitsu R&D Center. E­mail:  zhouhua@cn.fujitsu.com
Venue:
Call for comment on 802.16m amendment working document
Comment for the text of “Channel coding and HARQ”
Base Contribution:
C802.16m­08_1425
Purpose:
To be discussed and adopted into the 802.16m AWD by TGm.
Notice:
This document does not represent the agreed views of the IEEE 802.16 Working Group or any of its subgroups. It represents only the views of the participants listed in 
the “Source(s)” field above. It is offered as a basis for discussion. It is not binding on the contributor(s), who reserve(s) the right to add, amend or withdraw material 
contained herein.
Release:
The contributor grants a free, irrevocable license to the IEEE to incorporate material contained in this contribution, and any modifications thereof, in the creation of an 
IEEE Standards publication; to copyright in the IEEE’s name any IEEE Standards publication even though it may include portions of this contribution; and at the 
IEEE’s sole discretion to permit others to reproduce in whole or in part the resulting IEEE Standards publication. The contributor also acknowledges and accepts that 
this contribution may be made public by IEEE 802.16.
Patent Policy:
The contributor is familiar with the IEEE­SA Patent Policy and Procedures:
<http://standards.ieee.org/guides/bylaws/sect6­7.html#6> and <http://standards.ieee.org/guides/opman/sect6.html#6.3>.
Further information is located at <http://standards.ieee.org/board/pat/pat­material.html> and <http://standards.ieee.org/board/pat >.

1
 Introduction
• FEC&HARQ DG work out the draft proposal for nominal
modulation and coding set. But this MCS table is lack of
system level simulation result support.
• Without simulation results, we can not say how it works.
We would like to investigate the existing MCS candidate
on system level simulation.
• In other hand, 4-bit table or 5-bit table should also be
evaluated, considering LTE has 5-bit MCS table. We
might not close the MCS table size discussion until we
have enough simulation results.

2
 Summary
• The simulation shows that two 4-bit MCS tables
have almost same performance. However,
Fujitsu’s proposal has a little gain over
contribution C802.16m-09_0510’s proposal.
• Going beyond that, we simulate LTE 5-bit MCS
table, which can bring 20+% throughput and
coverage gain over 16E 11-level MCS table. In
order to achieve higher performance, it’s
worthwhile having 5-bit MCS table as candidate.
• We propose not to conclude MCS table before
we have more simulation, for example, with
bigger size MCS table and with CL SU/MU-
MIMO scenario.
3
 Simulation results
• System level simulation
 MCS table
Sector Throughput (Mbps) User Coverage 5%(bps/Hz)

MIMO (2x2) MIMO(4x4) MIMO (2x2) MIMO(4x4)

16e MCS table (11­level) 14.1 24.1 0.027 0.049

Fujitsu’s MCS table (16­level) 15.7 (11.3%) 26.6 (10.3%) 0.032 (18.5%) 0.054 (10.2%)

C802.16m­09_0510’s MCS table  15.5 26.4 0.032 0.054

(16­level)

LTE MCS table (5­bit) 17.2 (22.0%) N/A 0.037 (37.0%) N/A

4
C802.16m-09_0510’s proposal
MCS index Modulation Code rate

‘0000’ QPSK 31/256

‘0001’ QPSK 48/256

‘0010’ QPSK 71/256

‘0011’ QPSK 101/256

‘0100’ QPSK 135/256

‘0101’ QPSK 171/256

‘0110’ 16QAM 102/256

‘0111’ 16QAM 128/256

‘1000’ 16QAM 155/256

‘1001’ 16QAM 184/256

‘1010’ 64QAM 135/256

‘1011’ 64QAM 157/256

‘1100’ 64QAM 181/256

‘1101’ 64QAM 205/256

‘1110’ 64QAM 225/256

‘1111’ 64QAM 237/256

5
 Fujitsu’s proposal

MCS Index Code Rate Mod Order Spectral Efficiency

0 0.1333 2 0.2667

1 0.2174 2 0.4348

2 0.3448 2 0.6897

3 0.4762 2 0.9524
4 0.6000 2 1.2000
5 0.3157 4 1.2632
6 0.4167 4 1.6667
7 0.4848 4 1.9394
8 0.5556 4 2.2222
9 0.6250 4 2.5000
10 0.7142 4 2.8571
11 0.5000 6 3.0000
12 0.6111 6 3.6667
13 0.7037 6 4.2222
14 0.8000 6 4.8000
15 0.9048 6 5.4286
6
Simulation parameter settings
Parameters Values
Bandwidth 10 MHz
FFT size 1024
Carrier frequency 2.5 GHz
Subframe structure 18x6 CRU, according to the latest revision of IEEE 80216m­08/003r6
Channel Model TU channel
Mobile speed 3 km/h
Antenna configuration 2x2, 4x4
Receiver algorithm MMSE
Channel coding & modulation 16e 11 MCS & new MCS & LTE 29­level MCS

Channel estimation Perfect channel estimation

Synchronous HARQ with Chase Combining, Retransmission delay = 8 TTI with maximum 3 
HARQ scheme
retransmissions and initial target PER of 10%

CQI feedback period and delay Every 8 TTI

Scheduling method PF

Cell to cell distance  500m

BTS power 46dbm

Shadow fading  8dB

UE noise figure 9dB

Link adaptation method RBIR for MMSE receiver and SISO 7
 Proposed MCS table text for IEEE
802.16m AWD
• 15.x.1.3

– Table 15.x.1-1 MCS table for downlink and uplink

data channel [TBD]