information

Article
Physical Layer Security and Optimal Multi-Time-Slot
Power Allocation of SWIPT System Powered by
Hybrid Energy
Dandan Guo *, Baogang Li and Wei Zhao
Department of Electronic and Communication Engineering, North China Electric Power University,
Baoding 071003, China; baogangli@ncepu.edu.cn (B.L.); andyzhaoster@163.com (W.Z.)
* Correspondence: guodanstyle@163.com; Tel.: +86-0312-752-2472

Received: 29 June 2017; Accepted: 15 August 2017; Published: 16 August 2017

Abstract: In this paper, a new approach is proposed to solve the constrained optimization problem of
saving grid energy and increasing safety in a simultaneous wireless information and power transfer
(SWIPT) system. The traditional grid energy is combined with the renewable energy to form a hybrid
energy, which provides power for the system to achieve green wireless transmission. The transfer
process of SWIPT system is divided into multiple time slots. The renewable energy is harvested and
stored in battery at each time slot. A multi-time-slot artificial noise-assisted transmission strategy
is proposed to reduce the signal to noise ratio (SNR) of eavesdropping link. A power allocation
algorithm based on multi-time-slot golden section is given, which performs one-dimensional search
on the power ratio of artificial noise to determine the transmit power of source node. And then the
allocation algorithm is utilized to dynamically configure the harvested renewable energy for each
time slot. When the battery capacity is constant, the maximum renewable energy is being used to
reduce the grid power consumption. Finally, the performances of proposed schemes are evaluated by
simulations in terms of various tradeoffs.

Keywords: renewable energy; multi-slot artificial noise; SWIPT; time slots division; power allocation

1. Introduction
Exploiting new energy and transmission mode for wireless communication to reduce the grid
power consumption and improve the physical layer security (PLS) is the main issue faced by the
next generation (5G) of mobile communication [1,2], and is considered as a problem of wide concern.
Based on the above analysis, we consider a simultaneous wireless information and power transfer
(SWIPT) system powered by the hybrid energy to save grid resource, and utilize artificial noise (AN)
to interference eavesdropper.
The SWIPT technology received extensive attention with separation of destination nodes and
carrying energy transmission [3–5]. But the energy receiver (ER) is easy to eavesdrop the information
transmitted to the information receiver (IR) from source (S) in the SWIPT system [6,7]. And then
ER is seen as a potential eavesdropper that will affect physical layer security. Besides, according to
the lower energy conversion efficiency at ER, the high energy consumption is caused in the SWIPT
system. In terms of the security issue, the authors in [8] designed a secure beamforming scheme
in a SWIPT system, which can minimize the whole transmit power while maintaining the secrecy
rate and energy harvesting at ER. Meanwhile, one of the technologies for enhancing the secrecy
performance is to transmit artificial noise by jammers to interfere eavesdroppers [9]. The authors
in [10,11] recently considered the PLS with the AN by treating energy harvesting (EH) receivers as
potential eavesdroppers. A secure MISO SWIPT system was proposed in [12], where an artificial
noise is embedded to the information bearing signal to interfere the eavesdroppers and to harvest

Information 2017, 8, 100; doi:10.3390/info8030100 www.mdpi.com/journal/information

Information 2017, 8, 100 2 of 11

power by all receivers. However, these security schemes are only suitable for single time slot. Besides,
Information 2017, 8, 100 2 of 11
in order to save power consumption, the paper [13] proposed the use of green renewable energy to
mobile receivers in a multiple-antenna SWIPT system. The receivers obtain the renewable energy,
which by
power cannot be controlled
all receivers. and are
However, notsecurity
these always available
schemes are[14,15].
onlyThe authors
suitable in [16]time
for single proposed a using
slot. Besides,
in order to save power consumption, the paper [13] proposed the use of green renewable energyand
renewable energy scheme which divides time slots to allocate the dynamic renewable energy, to
gave a power allocation algorithm based on two-stage filling water. In
mobile receivers in a multiple-antenna SWIPT system. The receivers obtain the renewable energy, the literature [17–19], using
the directional
which cannot be water-filling
controlled and algorithm in different
are not always availablechannel
[14,15].models is considered
The authors to allocate
in [16] proposed the
a using
harvesting energy
renewable energy,scheme
including the Gaussian
which divides time fading
slotschannel [17],the
to allocate thedynamic
broadcast channel [18],
renewable andand
energy, the
multiple-access channel [19] to maximize the throughput in finite time horizon.
gave a power allocation algorithm based on two-stage filling water. In the literature [17–19], using the However, these
algorithms water-filling
directional cannot guarantee optimal
algorithm resourcechannel
in different allocation for multi-time-slot.
models is considered toAlthough
allocate themany studies
harvesting
have investigated
energy, including the theGaussian
performancefading inchannel
SWIPT,[17],
relatively less work
the broadcast has been
channel [18],done on multiple-access
and the multi-time-slot
wireless transmission.
channel [19] to maximize the throughput in finite time horizon. However, these algorithms cannot
In this
guarantee paper,resource
optimal a SWIPT modelfor
allocation based on the renewable
multi-time-slot. Althoughenergy and multi-slot
many studies artificial
have investigated
noise-assisted
the performancetransmission is proposed.
in SWIPT, relatively less work According
has been done to the dynamic arrival
on multi-time-slot characteristics
wireless transmission. of
renewable energy, the power splitting ratio and the transmitted power are determined
In this paper, a SWIPT model based on the renewable energy and multi-slot artificial noise-assisted at source
node. A part of
transmission is the transmitted
proposed. power istoused
According the to transmit
dynamic the signal
arrival to IR, and the
characteristics source served
of renewable as a
energy,
jammer utilizes the rest power to transmit AN to ER. Meanwhile, ER utilizes
the power splitting ratio and the transmitted power are determined at source node. A part of the the AN to harvest energy
to satisfy collection
transmitted power is demand.
used to Besides,
transmitwe thepropose
signal toa power
IR, andallocation
the sourcealgorithm
served aswhich can fully
a jammer utilize
utilizes the
renewable energy and guarantee secure transmission under the minimum grid
rest power to transmit AN to ER. Meanwhile, ER utilizes the AN to harvest energy to satisfy collection consumption.
demand. Besides, we propose a power allocation algorithm which can fully utilize renewable energy
2. The System Model and Mathematical Modeling
and guarantee secure transmission under the minimum grid consumption.

2.1.
2. TheSystem
The System Model
Model and Mathematical Modeling
We consider an energy cooperative network model as shown in Figure 1, which consists of a
2.1. The System Model
source, an information receiver, an energy receiver and a battery with a storage limit of max at the
E
We consider an energy cooperative network model as shown in Figure 1, which consists of a
source. The entire transmission process of the system is divided into N time slots and the size of each
source, an information receiver, an energy receiver and a battery with a storage limit of Emax at the
tf f i , i ∈ 1,..., N represents the i-th slot
slot is The
source. . All channels
entire are assumed
transmission to undergo
process blocking
of the system fading.into
is divided N time slots and the size of each
slot is t . All channels
channelf gain of the S-IR link andare assumed gto
i , i ∈ 1,...,
undergo N blocking f i ∈
represents the i-th slot1,channel
fading. i , . . . , N represents
gain of thethe i-thlink.
S-ER slot
channel gain of the S-IR link and gi , i ∈ 1, . . . , N represents the i-th slot channel gain of the S-ER link.
The transmit power splitting ratio is defined as α i , i ∈ 1,...1, N and the transmit power at S is
The transmit power splitting ratio is defined as αi , i ∈ 1, . . . 1, N and the transmit power at S is defined
p , i ∈ 1,..., N .
defined
as pi , i ∈as 1, . . .i , N.

Figure 1.
Figure 1. The system model.
Information 2017, 8, 100 3 of 11

2.2. The Mathematical Modeling

In this considered network, assuming that source is powered by hybrid energy consisted of the
traditional power grid and the renewable energy. Therefore, the mixed transmit signal at source
should be
√
q
xi = (1 − αi ) pi s0,i + αi pi s1,i , (1)

where s0,i , s1,i respectively represent the i-th information signal and AN signal at S, they are
independent circular symmetric complex Gaussian random variables with zero mean and unit variance.
The received signal expression of IR and ER are formulated as

y ER,i = gi xi + n ER,i , (2)

y IR,i = f i xi + n IR,i , (3)

where n ER,i and n IR,i respectively represent the additive white Gaussian noise (AWGN) at ER and IR
2 ) and n
with n ER,i ∼ CN (0, σ2,i 2
IR,i ∼ CN (0, σ1,i ). Then the SNR at IR and ER are respectively given by

(1 − α i ) f i p i
snr IRi = 2
, (4)
σ1,i

( 1 − α i ) gi p i
snr ERi = 2
, (5)
αi gi pi + σ2,i
According to Equations (4) and (5), the secrecy rate of i-th slot can be calculated as

Ri ( ai , pi ) = [log2 (1 + snr IRi ) − log2 (1 + snr ERi )]+ , (6)

where [ x ]+ = max(0, x ). In this scheme, we assume that the system is based on timeslot changing.
We introduce the following indicator function for the event of outage with respect to the minimum
secrecy rate r0 at each time slot:
(
1, R i ( α i , p i ) < r0
X (i ) = , (7)
0, otherwise.

where the outage probability can be expressed as

1 N
N i∑
E[ X (i )] = X ( i ),
=1

We consider a harvesting renewable energy plan at source as shown Figure 2, and assume that
source is equipped a battery with initial power E0 . When a time slot has just begun, the battery stores
the harvested renewable energy, i.e., Ei . During the wireless transmission process, the transmit power
consists of two parts, one part of which is provided by the battery, i.e., pri and the other obtains from
g
grid, i.e., pi .
 Information 2017, 8, 100 4 of 11
Information 2017, 8, 100 4 of 11

E0 E1 E2 EN −1 EN

...
...
1 2 N

Figure
Figure2.
2.The
Therenewable
renewableenergy
energyof
ofbattery.
battery.

After all time slots, the collected energy of ER should satisfy the average collection requirement,
After all time slots, the collected energy of ER should satisfy the average collection
i.e.,
requirement, i.e.,
11 N ξα p g ≥ p ,
N

N∑
 ξαii pii gii ≥ pave
ave , (8)
(8)
N i =1
i =1

where ξξ∈∈[[0,1]
where 0, 1] isisthethe
energy collection
energy efficiency
collection depending
efficiency on the rectification
depending and energy
on the rectification harvesting
and energy
circuitry, and p ave represents
harvesting circuitry,
p the need of average collection power. Besides, the collected renewable
and ave represents the need of average collection power. Besides, the collected
energy at each time slot must be stored in the battery before it is used, and the causal relationship is
renewable energy at each time slot must be stored in the battery before it is used, and the causal
shown as
relationship is shown as k k −1
∑k f i k∑
t p r
≤ −1 Ei , ∀k, (9)

i =1
f i 
i =1 t p r ≤ i =1E , ∀k ,
i =1
i (9)
Due to the limited battery capacity, the harvesting energy of source in each time slot cannot exceed
the maximum
Due to thelimit of the
limited battery,
battery i.e.,
capacity, the harvesting energy of source in each time slot cannot
exceed the maximum limit of the battery, i.e.,
k −1 k
∑ kE−i1 − ∑ kt f pri r≤ Emax , k < N, (10)
i =1  E i−=
i =1
1 t
i
i =1
f pi ≤ Emax , k < N , (10)

where Emax = MPr,ave t f with Pr,ave = 20 dBm. M represents the battery storage capacity coefficient and
E max = MPr , ave t f P
Pr,ave is the
where with r , ave
average harvesting energy
= 20ofdBm.
battery per second.the
M represents And we define
battery thecapacity
storage maximum average
coefficient
energyP consumption threshold of the grid in order to limit the actual grid consumption. Therefore, the
and r ,ave is the average harvesting energy of battery per second. And we define the maximum
constraint relation is expressed as
average energy consumption threshold ofNthe grid in order to limit the actual grid consumption.
1
N i∑
g
Therefore, the constraint relation is expressed pi − pri ) ≤ p ave ,
( as (11)
=1
1 N
g
where p ave represents the average grid consumption.  ( pi − pirBesides,
N i =1
g
) ≤ pave source is a power-constrained node(11)
and
the maximum power allowed for transmission is p peak . Therefore, , the transmit power of each slot must
p g following formula
satisfy the
where ave represents the average grid consumption. Besides, source is a power-constrained node
pk ≤ p peak , ∀k, (12)
p
and the maximum power allowed for transmission is peak . Therefore, the transmit power of each
We aim at minimizing the secrecy outage probability for the IR by jointly optimizing αi and pi .
slot must satisfy the following formula
Based on the definition in Equation (7), the optimization problem can be formulated as
pk ≤ ppeak , ∀k , (12)
min E[ X (i )]
αi ∈[0,1)
We aim at minimizing the secrecy α i and
s.t. outage
(8), (9)probability for the
, (10), (11) and (12)IR, by jointly optimizing (13)
pi . Based on the definition in Equation (7), the optimization
prk ≤ pi , ∀kproblem can be formulated as
prk ≥ 0, ∀k
Information 2017, 8, 100 5 of 11

3. The Optimal Power Splitting Ratio and Transmit Power

The above mathematical model involves both αi and pi that need to be optimized. We utilize
the two-stage optimization strategy in order to solve the above problem [20]. At the first stage,
one-dimensional search is performed for the power splitting ratio of each slot based on the golden
section principle, and then the minimum required power is obtained. Secondly, we apply the Lagrange
duality method to solve Equation (13) based on the ratio solution of the first stage.

3.1. The Power Splitting Ratio

Because of the transmission power of S is constrained, minimizing pi is also one of the goals [21].
Given any 0 ≤ αi ≤ 1, let p(αi ) denotes the minimum required power to maintain the secrecy rate r0
allowed for the system, i.e., Ri (αi , pi ) = r0 . It can be shown as

1 + snr IRi
= 2r 0 , (14)
1 + snr ERi

By substituting (4) and (5) into (14), and then pi and αi should satisfy the condition

2
(1 − αi ) f i αi gi p2i + (αi gi σ1,i 2
+ (1 − αi ) f i σ2,i − 2r0 gi σ1,i
2
) pi + (1 − 2r0 )σ1,i
2 2
σ2,i = 0, (15)

One can show from (15) that it is a quadratic function of pi . Therefore, the solution of pi can be
expressed as follows
 2 +(1− α ) f σ2 −2r0 g σ2 ]+ ∆
√
−[αi gi σ1,i
0 < αi < 1and ∆ > 0
i i 2,i i 1,i




p ( α i ) = 2 ( 1 − α i ) f i α i gi
 r 2 σ2
 (2 0 −1)σ1,i 2,i
p(0) < p peak , (16)
f 2 − 2r 0 g σ 2 2 > 2r 0 g σ 2


 p ( αi ) = σ
i 2,i i 1,i αi = 0 and f i σ2,i i 1,i
p peak p(0) ≥ p peak
 

where αi is the independent variable, p(αi ) is the dependent variable. αi∗ is defined as the optimal
power splitting ratio. ∆ is given by

2
2
∆ = (αi σ1,i 2
gi + σ2,i (1 − αi ) f i ) + 2r0 (2r0 σ1,i
4
gi 2 − 2αi σ1,i
4
gi 2 + (−4α2i + 6αi − 2)σ1i
2 2
σ2i hi gi ), (17)

One-dimensional search is performed for (15), and obtain αi which satisfies minimum p(αi ). When
2 = σ2 = 0.01
N is equal to 100, Figure 3 shows the optimal distribution of αi in i-th time slot with σ1,i 2,i
and r0 = 10 kbit/s. We see that the optimal distribution is in the range of 0.455 to 0.505, which changes
according to the channel state.
Information 2017,8,8,100
Information2017, 100 66of
of11
11

0.505

0.5

0.495

0.49

0.485

0.48
αi

0.475

0.47

0.465

0.46

0.455
0 10 20 30 40 50 60 70 80 90 100
i-th time slot

22
Figure 3. The optimal distribution of αi with σσ 22
α 1,i ==σσ
2,i2,i==0.01
0.01and r0 =r0 25
= bps/Hz.
25 bps/Hz .
Figure 3. The optimal distribution of i with 1,i and

3.2. The Optimal Transmit Power

3.2. The Optimal Transmit Power
In the first stage, we obtain the appropriate αi . In the second stage, the optimization problem is
In about
the first α i . In the second stage, the optimization problem
convex pi . stage, we obtain function
The Lagrangian the appropriate
of (13) is expressed as
p
is convex about i . The Lagrangian function of (13) is expressed as
N
L( R, λ, µi , ϕi , η ) = E[ X (i )] + λ( N1 ∑ ξ pNi gi − p ave )
1
E[ X (i)]k+ iλ=(1 kξ−p1 i gi − pave
L( R, λ , μi , ϕi ,η ) = N N −1
) k −1 k
− ∑ µk ( ∑ t f pNri − i =1
∑ Ei ) − ∑ ϕk ( ∑ Ei − ∑ t f pri − Emax )
k −1
Ni =1 k
k =N1−1 −11
ik= k i =1
−  μk ( t f pir − N  Ei ) −  ϕk ( Ei −  t f pi − Emax )
k =1 i =1 r , (18)
N
− ∑ β kk(=p1 rk − ip=1k ) + ∑ i =γ1 k prk k =1 i =1 i =1
k =1 N k =1 N , (18)
− η ( N1 −∑ (βpki (−p 
g γk
r r
pi ) −) p+ave
−
N
k r pk ) pk
k =1 k =1
i =1
1 N
− η
where λ, ui , ϕi and η are the dual variables( =1 ( pi − pir ) −with
N iassociated
g
pave ) average collected energy p at ER, the
the ave
harvesting renewable energy Ei at source, the battery capacity Emax and the average grid consumption
g
pwhere λ , ui , ϕi and
ave , respectively.
η the
Then arepartial Lagrange
the dual variables dual function with
associated of (18)the
is expressed as
average collected
p
energy ave at
E
i at source, the battery capacity E
max and the average grid
ER, the harvesting renewable
f (λ, µenergy
i , ϕi , η ) = min L( R, λ, µi , ϕi , η ), (19)
g {pk ≤ p peak },{αi ∈[0,1]}
p
consumption ave , respectively. Then the partial Lagrange dual function of (18) is expressed as
The dual problem of (13) fis(λgiven
, μi , ϕiby
,η ) = min L ( R , λ , μ i , ϕ i ,η ) ,
{p k ≤ p peak },{α i ∈[0,1]} (19)
max f (λ, µi , ϕi , η )
The dual problem of (13) is given by
s.t. pk ≤ p peak , (20)
maxλ ≥ f (λ0,, μ ϕ i ,0,
µii , ≥ η )ϕi ≥, η ≥ 0.
s.t . pk ≤ p peak , (20)
Substituting the optimal power splitting ratio from the first stage into Equation (18), we obtain
λ ≥ 0, μ i ≥ 0, ϕ i ≥,η ≥ 0.
the expression about the optimal transmit power of per time slot
Substituting the optimal power splitting ratio from the first stage into Equation (18), we obtain
3 2 + Cp + D = 0
the expression about theAP i + Bpitransmit
optimal i power of per time slot
A = cC1 (C22 + C2 )
B = c C22 + C2 + 2C1 C2 C3 + C1 C3 − C1 (C22 + C2 ) ,


(21)
C = 2C2 C3 c + cC3 + cC1 C32 − 2C1 C2 C3
D = cC32 − C1 C32 − C3
 D =cC32 − C1C32 − C3

ξλ N
c = (η − β i −
N
 α g ) / ln 2
i i
C = (1 − αi ) fi / σ1,2i C2 = α i / (1 − α i ) C = σ 2,i / ((1 − αi ) gi ) 2
where i =1 , 1 , and 3 .
Information 2017, 8, 100
Since Equation (21) is a cubic equation and is complex, we utilize MATLAB to solve the cubic 7 of 11
equation and obtain the optimal power.
N
2 , C = α / (1 − α ) and C = σ2 / ((1 − α ) g ).
where N ∑
− β i − ξλ
c = (ηAllocation
4. A Power αi gi )/ ln 2, C1 = (1 − αi ) f i /σ1,i
Algorithm 2 i i 3 2,i i i
i =1
SinceIn
Equation (21)we
this paper, is autilize
cubic the
equation
goldenand is complex,
section we to
principle utilize
searchMATLAB to solve
the power the ratio,
splitting cubic and
equation
then
and obtain the optimal power.
obtain the optimal transmit power at source based on the Lagrange duality method. Based on the
above analysis, a power allocation algorithm structure is proposed. The steps of the algorithm are as
4. A Power Allocation Algorithm
follows:
Step1:
In Initialize
this paper, wethe optimization
utilize the goldenproblem and algorithm
section principle parameters.
to search the power We want toratio,
splitting minimize
and thenthe
obtain the optimal transmit power at source α
based on the
outage probability which includes the variable i and i . The renewable energy power harvested p Lagrange duality method. Based on the
above analysis, a power allocation r algorithm structure is proposed. The steps of the algorithm are
in the
as battery of each slot is pi .
follows:
Step2: Perform
Step1: Initializeone-dimensional
the optimization searching problem and
α i and obtain
on algorithm the transmit
parameters. power
We want at source. the
to minimize
outage Step3: The power
probability which allocation
includesstarts from the
the variable αi i-th
and slot withrenewable
pi . The i = N, andenergy
the harvesting renewable
power harvested in
the battery of each slot is Emax
pri . pir Emax pir Emax
energy cannot exceed . If > , set = .
Step2: Perform one-dimensional searching on αi and obtain the transmit power at source.
Step4: In this step, we compare the renewable energy pir p in i-th slot.
power i renewable
Step3: The power allocation starts from the i-th slot withand i = the optimal
N, and the harvesting
energy
When cannot
pi is greater
exceedthen
r
Emax .piIf, anpri > Emax , setscheme
allocation pri = Emax .
is proposed referring to the model (a) in Figure 4.
Step4:inInthe
As shown thismodel
step, we
(a), compare
all of thethe renewable
renewable energy
energy pri and
of the i-th the optimal
time slot is power pi in
used, the i-th
rest ofslot.
the
When pi is greater then pri , an allocation scheme
g r is proposed referring to the model (a) in Figure 4.
transmit power is filled by grid, i.e., pi = pi + pi . And then set i = i − 1. In contrast, utilize the model
As shown in the model (a), all of the renewable energy of the i-th time slot is used, the rest of the
(b) to allocate the hybrid power. In the model g (b), find the j-th slot with j < i satisfying with
transmit i
power is filled byi grid, i.e., pi = pi + pri . And then set i = i − 1. In contrast, utilize the
pir − (b)
model ≤ 0allocatepthe
pm to r
 pm >power.
i − hybrid 0 In the model (b), find the j-th slot with j < i satisfying with
m =i j m = j +i1
and , and then set i = j. During this process, no grid power is used.
pri − ∑ pm ≤ 0 and pri − ∑ pm > 0, and then set i = j. During this process, no grid power is used.
Step5:
m= j If i ≠ 1 , repeatm= stepj+1 3 and step 4 until i = 1.
Step5: If i 6= 1, repeat step 3 and step 4 until i = 1. p < p peak
Step6: If there is still remaining renewable energy at rest of the time slot and i , the
Step6: If there is still remaining renewable energy at rest of the time slot and pi < p peak , the
remaining energy will be directly supplied to the source node of the corresponding time slot to
remaining energy will be directly supplied to the source node of the corresponding time slot to transmit
transmit information for IR.
information for IR.
In this algorithm, we build an inverse sequence for the time slot to guarantee all the renewable
In this algorithm, we build an inverse sequence for the time slot to guarantee all the renewable
energy be utilized.
energy be utilized.

i
pir −  pm > 0
pi
m=i−1
pi −1
pir −pi >0
pi +1 i pi
g pir − pm ≤ 0
p pj
pig i +1 m= j

pir pir+1

Figure 4. Two
Figure 4. Twotypes
typesofofpower
powerallocation model.
allocation (a)(a)
model. The lessless
The renewable energy
renewable allocation
energy model;
allocation (b) The
model; (b)
enough renewable energy allocation model.
The enough renewable energy allocation model.

5. Simulation Results
In this section, we provide simulation results to validate our theoretical analysis and evaluate the
performance of the proposed scheme. We run N = 100 frames in each simulation and set the maximum
transmit power as 30 dBm, i.e., p peak = 30 dBm. The harvested energy of ER follows non-negative
uniform distribution with mean p ave = 20 dBm. The energy conversion efficiency is set as ξ = 0.8 [22].
Besides, in order to make fully use of the renewable energy, we set a little average grid consumption
g
power as p ave = 20 dBm.
Figure 5 shows that the relationship between the power ratio, the outage probability and the
transmit power. We define the power splitting ratio vector α = {α1 , α2 , . . . , α N }. Three vector schemes
non-negative uniform distribution with mean pave = 20 dBm . The energy conversion efficiency is set
as ξ = 0.8 [22]. Besides, in order to make fully use of the renewable energy, we set a little average
g
grid consumption power as pave = 20 dBm .
Figure 5 shows that the relationship between the power ratio, the outage probability and the
transmit power. We define the power splitting ratio vector α = {α1 , α 2 ,..., α N } . Three vector schemes
Information 2017, 8, 100 8 of 11

are given: all the elements of vector are 0.1, all the elements of vector are 0.3 and all the elements of
are given:
vector are all
thethe elements
optimal of vector
search values.are
We 0.1,
canallsee
thethat
elements of vector
the transmit are 0.3
power and all the
gradually elements
decreases of
and
vector are the optimal search values. We can see that the transmit power gradually decreases
tends to be stable with the increasing outage probability. Besides, when the outage probability is and tends
to be stable
fixed, with theαincreasing
the optimal correspondsoutage probability.
to the Besides,power
whole transmit when is theminimum.
outage probability
When theisdifference
fixed, the
optimal α
between corresponds
the to the
power splitting whole
ratio andtransmit
the optimalpower α is larger,
minimum. When theofdifference
the difference between
transmit power the
value
power splitting ratio and the optimal α is larger, the difference of transmit power value will
will be larger. Therefore, in the case of satisfying a certain outage probability, utilizing the golden be larger.
Therefore, in the the
section to search caseoptimal α can
of satisfying a certain
effectivelyoutage probability,
reduce utilizing
the transmit powertheatgolden
source.section to search
the optimal α can effectively reduce the transmit power at source.

120
the total transmitted power of N time slots

α =0.1
α =0.3
100 the optimal power spliting ratio vector α

80

60

40

20

0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
the outage probability
2 = σ2 = 0.01 and g
Figure 5. The average transmit power with different αi , M = 3, p ave = 20 dBm, σ1,i
α g σ 2 2,i
= σ 2,2 i = 0.01
Figure 5. The
p ave = 20 dBm.average transmit power with different i , M = 3, pave = 20 dBm , 1,i
and pave = 20 dBm .
From Figure 6, we can see that the grid power consumption is gradually decreasing when
From Figure
the battery capacity6, we can see that
increases. the gridallocation
The power power consumption
algorithm whichis gradually decreasing
we proposed whenthan
is better the
battery capacity
the directly increases.
filling The powerand
water algorithm allocation algorithm
the two-stage which
filling we proposed
water algorithm isonbetter than the
reducing
directly fillingof water
consumption algorithmfully
grid. Therefore, and utilizing
the two-stage
renewable filling
energywater
can algorithm
improve the oncapacity,
reducing the
which
consumption of grid.power
reduces unnecessary Therefore, fully utilizing
consumption of the renewable
grid as much energy can improve the capacity, which
as possible.
reduces unnecessary
Figure 7 shows power consumption
that the of theof
curve changing grid
theassecrecy
much as possible.
rate with the influence of M and αi .
According
Figureto7 the three-dimensional
shows figure, weofobtain
that the curve changing that therate
the secrecy secrecy
withrate
the gradually increases
influence of M and and αi .
tends to be to
According stable with changing of M,figure,
the three-dimensional and the
wecurve
obtainof that
the secrecy rate has
the secrecy rateagradually
peak at around αi =and
increases 0.5.
Therefore, choosing the appropriate battery capacity and the power splitting ratio can effectively
tends to be stable with changing of M, and the curve of the secrecy rate has a peak at around i
α = 0.5
improve the secrecy performance.
. Therefore, choosing the appropriate battery capacity and the power splitting ratio can effectively
improve the secrecy performance.
Information 2017, 8, 100 9 of 11

Information 2017, 8,
Information 2017, 8, 100
100 99 of
of 11
11
18
the directly filling water algorithm

(mW)(mW)
16
18 the two-stage filling water algorithm
the
thepower allocation
directly algorithm
filling water proposed in this paper
algorithm
14
16 the two-stage filling water algorithm

consumption
the power allocation algorithm proposed in this paper
12
14

consumption
10
12

8
10
grid energy

68
the energy

46
the grid

24

02
0 1 2 3 4 5 6 7 8 9 10

0 the value of M
0 1 2 3 4 5 6 7 8 9 10
the value of M
Figure 6. The grid power consumption with different power allocation algorithm. i i , α = α*
∗
σ 1,i = σ 2,i = 0.01
Figure 6. The grid 2 power2 consumption with different power allocation algorithm.
α
α =
i = αα *i ,
gg = 206.dBm
pFigure ,
The grid
2 power
2 and p =
consumption =with 20 dBm .
different power allocation algorithm. i i ,
ave = 20 dBm, σ1,i = σ2,i = 0.01 and p ave ave
pave 20 dBm.
σ 1,2i = σ 2,2 i = 0.01
g
pave = 20 dBm , and pave = 20 dBm .

25
rate (kbit/s)

20
25
rate (kbit/s)

15
20
the secrecy

10
15
the secrecy

5
10

05
10
0 1
10 0.8
5
M 0.4
0.6
1
0.2 α 0.8
5
M 0 0
0.4
0.6

0.2 α
2 = σ 0 M g
= 20 dBm,σσ21,i 2 2 = 0.01 and
0
1, i = σ 2, i = 0.01
Figure 7. The secrecy rate with the influence and αi . of p ave
Figure
α i g
. pave = 20 dBm ,
2,i
and
p ave = 7.
20The secrecy rate with the influence of M and
dBm.
pave = 20 dBm .
Figure 7. The secrecy rate with the influence of M and
α i . p g = 20 dBm , σ 1,2i = σ 2,2 i = 0.01 and
ave
6. Conclusions and Future Works
pave = 20 dBm .
6. Conclusions and Future Works
This paper considers the problems of safety and energy in a SWIPT system, in presence of
This eavesdropper,
potential paper and
6. Conclusions considers i.e.,the
Future ER.problems
WorksThe paperofshows
safetythat
andsecure
energy in a SWIPT
transmission cansystem, in presence
be achieved of
by adding
potential eavesdropper, i.e., ER. The paper shows that secure transmission can be achieved
AN based on multi-slot. And we have optimized the power splitting ratio and the transmit power at by adding
This paper considers the problems of safety and energy in a SWIPT system, in presence of
AN based
source on multi-slot.
based on the time And we have
domain. optimized
Also, we proposethe power
a powersplitting ratioalgorithm
allocation and the transmit power at
for the system to
potential eavesdropper, i.e., ER. The paper shows that secure transmission can be achieved by adding
source
allocatebased on the time
the renewable domain.
energy Also,
at each weslot
time propose a powerthe
to guarantee allocation
minimum algorithm for the
grid power system to
consumption.
AN based on multi-slot. And we have optimized the power splitting ratio and the transmit power at
allocate the renewable
The simulation examples energy
have at each time
validated theslot to guarantee
proposed methods theand
minimum
confirmedgridthe
power
effectconsumption.
on the secrecy
source based on the time domain. Also, we propose a power allocation algorithm for the system to
The simulation
performance andexamples have A
grid energy. validated the proposed
system with methods
a full duplex node orandIRconfirmed
served as athe effect may
jammer on the
be
allocate the renewable energy at each time slot to guarantee the minimum grid power consumption.
secrecy performance
interesting future topic.and grid energy. A system with a full duplex node or IR served as a jammer
The simulation examples have validated the proposed methods and confirmed the effect on the
may be interesting future topic.
secrecy performanceThe
Acknowledgments: andworkgrid energy. A system
in this paper was with a full duplex
supported nodeNatural
by Beijing or IR served
ScienceasFoundation
a jammer
(Grants
may beNo. 4164101),future
interesting National Natural Science Foundation of China (Grants No. 61501185), Hebei Province
topic.
Information 2017, 8, 100 10 of 11

Natural Science Foundation (Grants No. F2016502062), and the Fundamental Research Funds for the Central
Universities (2016MS97).
Author Contributions: Dandan Guo proposed the idea and wrote this paper; Dandan Guo and Wei Zhao
conceived and carried out the simulation; Dandan Guo and Baogang Li analyzed the data; all authors have read
and approved the final manuscript.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Bockelmann, C.; Prata, N.; Nilopour, H.; Au, K.; Svensson, T.; Stefanovic, C.; Popovski, P.; Dekorsy, A.
Massive machine-type communications in 5G: Physical and MAC-layer solu tions. IEEE Commun. Mag. 2016,
54, 59–65. [CrossRef]
2. Zhang, X.; Cheng, W.C.; Zhang, H.L. Full-duplex transmission in PHY and MAC layers for 5G mobile
wireless networks. IEEE Wirel. Commun. 2015, 22, 112–121. [CrossRef]
3. Zhu, Z.Y.; Lee, K.J.; Wang, Z.Y.; Lee, I. Robust beamforming and power splitting design in distributed
antenna system with SWIPT under bounded channel uncertainty. In Proceedings of the IEEE 81st Vehicular
Technology Conference, Glasgow, UK, 11–14 May 2015.
4. Zhu, Z.Y.; Chu, Z.; Wang, Z.Y.; Lee, I. Joint optimization of AN-aided beamforming and power splitting
designs for MISO secrecy channel with SWIPT. In Proceedings of the 2016 IEEE International Conference on
Communications (ICC), Kuala Lumpur, Malaysia, 22–27 May 2016.
5. Zhang, H.; Du, J.; Cheng, J.; Leung, V.C.M. Resource allocation in SWIPT enable heterogeneous cloud small
cell networks with incomplete CSI. In Proceedings of the 2016 IEEE Global Communications Conference
(GLOBECOM), Washington, DC, USA, 4–8 December 2016.
6. Wu, Y.N.; Chen, X.M. Robust beamforming and power splitting for secrecy wireless information and power
transfer in cognitive relay networks. IEEE Commun. Lett. 2016, 20, 1152–1155. [CrossRef]
7. Chen, X.M.; Ng, D.W.K.; Chen, H.H. Secrecy wireless information and power transfer: Challenges and
opportunities. IEEE Wirel. Commun. 2016, 23, 54–61. [CrossRef]
8. Zhang, H.Y.; Li, C.G.; Huang, Y.M.; Yang, L.X. Secure beamforming for SWIPT in multiuser MISO broadcast
channels with confidential messages. IEEE Commun. Lett. 2015, 19, 1347–1350. [CrossRef]
9. Romero-Zurita, N.; Mclernon, D.; Ghogho, M.; Swami, A. PHY layer security based on protected zone and
artificial noise. IEEE Signal Process. Lett. 2013, 20, 487–490. [CrossRef]
10. Zhang, H.; Hong, X.; Cheng, J.; Nallanathan, A.; Leung, V. Secure resource allocation for OFDMA two-way
relay wireless sensor networks without and with cooperative jamming. IEEE Trans. Ind. Inf. 2016, 12,
1714–1725. [CrossRef]
11. Zhang, L.Y.; Zhang, H.X.; Wu, D.L.; Yuan, D.F. Improving physical layer security for MISO systems via
using artificial noise. In Proceedings of the 2015 IEEE Global Communications Conference (GLOBECOM),
San Diego, CA, USA, 6–10 December 2015.
12. Chu, Z.; Zhu, Z.Y.; Hussein, J. Robust optimization for AN-Aided transmission and power splitting for
secure MISO SWIPT system. IEEE Commun. Lett. 2016, 20, 1571–1574. [CrossRef]
13. Ng, D.W.K.; Schober, R. Secure and green SWIPT in distributed Antenna Networks with limited backhaul
capacity. IEEE Trans. Wirel. Commun. 2015, 14, 5082–5097. [CrossRef]
14. Yuan, B.; Yang, X.; Li, D.H. Analysis and design of a high efficiency current fed multi-resonant converter
for high step-up power conversion in renewable energy harvesting. In Proceedings of the 2010 2nd IEEE
International Symposium on Power Electronics for Distributed Generation Systems (PEDG), Hefei, China,
16–18 September 2010; pp. 138–143.
15. Che, L.; Zhang, X.P.; Shahidehpour, M.; Alabdulwahab, A.; Abusorrah, A. Optimal interconnection planning
of community microgrids with renewable energy sources. IEEE Trans. Smart Grid 2017, 8, 1054–1063.
[CrossRef]
16. Gong, J.; Zhou, S.; Niu, Z.S. Optimal power allocation for energy harvesting and power grid coexisting
wireless communication systems. IEEE Trans. Commun. 2013, 61, 3040–3049. [CrossRef]
17. Ozel, O.; Tutuncuogl, K.; Yang, J.; Ulukus, S.; Yener, A. Transmission with energy harvesting nodes in fading
wireless channels: Optimal policies. IEEE J. Sel. Areas Commun. 2011, 29, 1732–1743. [CrossRef]
Information 2017, 8, 100 11 of 11

18. Yang, J.; Ozel, O.; Ulukus, S. Broadcasting with an energy harvesting rechargeable transmitter. IEEE Trans.
Wirel. Commun. 2012, 29, 571–583. [CrossRef]
19. Yang, J.; Ulukus, S. Optimal packet scheduling in a multiple access channel with rechargeable nodes.
In Proceedings of the IEEE ICC, Kyoto, Japan, 5–9 June 2011; pp. 1–5.
20. Yang, G.J.; Xie, L.P.; Zeng, J.C. A hybrid vector artificial physics optimization with one-dimensional
search. In Proceedings of the 2010 International Conference on Computational Aspects of Social Networks
(CASoN 2010), Taiyuan, China, 26–28 September 2010; pp. 19–22.
21. Hong, X.; Liu, L.; Zhang, R. Secrecy wireless information and power transfer in fading wiretap channel.
IEEE Trans. Veh. Technol. 2016, 65, 180–190.
22. Jiang, X.; Zhang, C.; Zhang, Z.; Karagiannidis, G.K. Power beacon assisted wiretap channels with jamming.
IEEE Trans. Wirel. Commun. 2016, 15, 8353–8367. [CrossRef]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).