CN114884550A - Relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network - Google Patents

Abstract

The invention discloses a relay cooperative transmission method for a bidirectional transmission FD multi-relay cooperative SWIPT network, and belongs to the technical field of calculation, calculation or counting. The invention is based on the SWIPT receiver adopting the dynamic PS strategy, the single link interruption probability is minimized as an optimization target, and the power distribution coefficients of signals in different frequency bands are further optimized while the relay PS coefficient is optimized. Meanwhile, the performance and the applicable scenes of various RS algorithms are analyzed by taking the traversal capacity and the maximization of the bidirectional link as optimization targets. The relay node of the SWIPT receiver structure is selected to participate in bidirectional communication of the FD multi-relay cooperation SWIPT network oriented to bidirectional transmission, effective power distribution is achieved, and system transmission efficiency is improved.

Description

Relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network

technical field

The invention relates to wireless information and energy simultaneous transmission technology, and specifically discloses a relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network, which is suitable for IoT nodes, terminal nodes and other equipment that are inconvenient or difficult to replace batteries It belongs to the technical field of computing, reckoning or counting.

Background technique

Energy Harvesting (EH, Energy Harvesting) technology has become a solution for extending the life cycle of energy-constrained wireless communication networks. Among various energy sources, the energy carried by the radio frequency (RF, Radio Frequency) signal is more stable and can be adaptively adjusted according to system requirements. As the carrier of the signal, RF signal has been widely used in wireless signal transmission (WIT, Wireless Information Transfer), and the RF signal can carry energy, which makes the wireless information and energy simultaneous transmission (SWIPT, Simultaneous Wireless Information and Power Transfer) technology possible. .

However, the existing state-of-the-art RF energy receivers cannot directly process the information carried by the signal, so researchers propose two practical strategies, Time Switching (TS, Power Splitting), to coordinate operation. Information processing module (IP, Information Processing) and EH module. In the TS strategy, the running time of the IP module and the EH module are staggered, the circuit design and implementation are relatively simple, but the information transmission is not continuous, and in order to adapt to the discontinuity of information transmission, the signal sender needs to make changes in signal coding. Adjusted to avoid sending useful signals while the EH module is running. The PS strategy divides the received signal of the RF energy receiver into two parts according to the power, one part is used for the IP module, and the other part is used for the EH module. The PS strategy solves the discontinuity of information transmission and makes better use of the time resources of the communication system, but it also increases the difficulty of circuit implementation of the RF energy receiver.

Since RF signals will encounter problems such as path loss and multipath attenuation, in order to expand the coverage of the signal and improve the quality of communication services, relay cooperative communication is increasingly used in conjunction with the SWIPT technology. In a relay cooperative communication network, Amplify and Forward (AF, Amplify and Forward) protocol and Decode and Forward (DF, Decode and Forward) protocol are two basic forwarding protocols. The DF protocol first decodes the received signal at the relay and then forwards it to the signal receiving end. The decoding circuit filters the antenna noise contained in the received signal but also increases the power loss at the relay. The AF forwarding protocol only performs simple linear amplification processing on the received signal, and amplifies the power of the noise while amplifying the power of the useful signal. However, the AF forwarding protocol is simple to implement and does not require a high-energy decoding circuit. In the relay cooperative SWIPT network, the circuit operation power of the relay is completely provided by the RF signal received by the relay. In order to reduce the energy consumption of the relay, the AF forwarding protocol is more used in practical applications instead of the more energy-consuming DF. forwarding protocol.

In the relay cooperative SWIPT network, EH relays can be divided into half-duplex (HD, Half-Duplex) EH relays and full-duplex (FD, Full-Duplex) EH relays according to the working mode. When the EH relay works in HD mode, the transmission cycle of each communication is divided into two stages. In the first stage, the EH relay receives the signal sent by the signal transmitter, and in the second stage, the EH relay forwards the signal received in the first stage to the signal receiver. When the EH relay works in the FD mode, the EH relay forwards the received signal to the signal receiving end while receiving the signal sent by the signal transmitter. Considering the influence of Self-Interference (SI, Self-Interference) at the relay, most of the early researches on relay cooperative networks focused on HD relay communication. In recent years, with the breakthrough of Self-Interference Cancellation (SIC, Self-Interference Cancellation) technology, FD relay networks have gained widespread attention due to their significant advantages in spectral efficiency compared to HD relay networks. Because the FD relay scheme based on the TS strategy can only send or receive information in a part of the transmission period, the time resource utilization of the system is low. On the other hand, benefiting from the breakthrough of SIC technology in recent years, the EH relay adopts the PS strategy to coordinate the EH module and the IP module well, sending and receiving information at the same time without affecting the communication quality of the system. Therefore, adopting the PS strategy at the EH relay can make full use of the FD characteristics of the system.

In a multi-relay cooperative communication network, relay selection (RS, Relay Selection) can enhance the diversity gain of the system. Using appropriate criteria to select the optimal relay can achieve the optimal system performance. Different from the traditional relay cooperative network, in the relay cooperative SWIPT network, the signal transmission power at the relay is not limited by the battery capacity of the relay itself, and the energy collected from the RF signal can support the relay in different Work non-stop with or without battery replacement. Therefore, choosing more relays will not consume extra battery power, but can collect more energy for signal forwarding. However, the greater the number of selected relays, the less the transmission bandwidth allocated to each relay, and how to select an appropriate set of relays to optimize the system capacity has become a research hotspot.

In practical engineering applications, due to the needs of business scenarios, the use of two-way relay networks is unavoidable. At present, the research on bidirectional transmission FD multi-relay cooperative SWIPT network mainly focuses on co-frequency full-duplex transmission. In fact, in the bidirectional transmission FD multi-relay cooperative network, due to the different distances between the two signal transmission sources and each relay and the different instantaneous channel states, the signals transmitted by the two signal transmission sources will be different when they reach the relay. The experienced channel attenuation is inconsistent, which naturally brings about the power allocation problem of the mixed signal. Under the same-frequency full-duplex relay transmission mechanism, since the signals of the two signal transmission sources share the same frequency band, the separation operation of the mixed signal cannot be performed at the relay, and the power distribution cannot be completed.

In the prior art document, "Full-Duplex Wireless-Powered Relay in Two Way Cooperative Networks" This paper considers the scenario of bidirectional signal transmission in a multi-relay cooperative SWIPT network, but mainly focuses on the different effects of the SWIPT receiver strategy on the system throughput and interruption probability, and takes the same-frequency full-duplex transmission as the research background. There is no separation of mixed signals at the repeater, and it does not have the ability to distinguish and separate mixed signals by frequency band, so power distribution cannot be achieved; Switched in-band full-duplex wireless power supply bidirectional relay, "2016 URSI Asia-Pacific Radio Science Conference, pp.438-441, 2016") considers the scenario of bidirectional signal transmission in a single relay cooperative SWIPT network, and proposes a This relay forwarding protocol can deal with co-frequency self-interference, but it adopts the TS strategy with low time slot utilization, and still takes co-frequency full-duplex transmission as the research background, and cannot separate mixed signals at the relay. operate.

SUMMARY OF THE INVENTION

The purpose of the present invention is to address the above-mentioned deficiencies of the background technology, by redesigning the structure of the SWIPT receiver at the relay to support Frequency Division Duplex (FDD) communication, to solve the problem of hybrid transmission in a two-way transmission FD relay cooperative SWIPT network. The technical problems of signal separation and power redistribution, a relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network is proposed. The information waveform signal is separated by a band-pass filter, and the signals corresponding to the two signal source frequency bands are extracted respectively, and the relay will redistribute the power stored in the energy waveform signal output by the PS strategy to a single chain. The optimization goal is to minimize the probability of channel interruption, and the optimal PS coefficient and the optimal power distribution coefficient are obtained. The redistributed two parts of power are respectively used for power amplification of the frequency band signals of the two signal sources, so as to realize the separation of signals in different frequency bands in the mixed signal and Inventive object of optimal power distribution.

The present invention adopts following technical scheme for realizing above-mentioned purpose of invention:

The relay cooperative transmission method for the bidirectional transmission FD multi-relay cooperative SWIPT network, by designing a SWIPT receiver that can support FDD communication and applying it to the bidirectional transmission FD multi-relay cooperative SWIPT network, the signals of different frequency bands can be transmitted in the Separate and redistribute power.

Further, in the relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network, the SWIPT receiver supporting FDD communication includes:

部分的混合信号至IP模块，输出

部分的混合信号至EH模块，ρ_j为第j个中继的PS系数，ρ_j∈[0,1]；The power division module receives the mixed signal from two signal sources, divides the mixed signal into two parts according to the PS coefficient of the current relay, and outputs

Part of the mixed signal to IP block, output

Part of the mixed signal is sent to the EH module, ρ _j is the PS coefficient of the jth relay, ρ _j ∈ [0,1];

部分的混合信号进行解调，从所述

部分的混合信号中分离抽取出来自两个信号源的信号，对分离抽取出的来自两个信号源的信号进行放大处理，输出来自两个信号源的频段信号；IP module, for the received

part of the mixed signal is demodulated from the

Separate and extract signals from two signal sources from part of the mixed signal, amplify the separated and extracted signals from the two signal sources, and output the frequency band signals from the two signal sources;

部分的混合信号，将

部分的混合信号转换为直流电能后储存至电池模块；及，EH module, receive

part of the mixed signal, will

Part of the mixed signal is converted to DC power and stored in the battery module; and,

The power distribution module distributes the power of the DC power stored in the battery module according to the power distribution coefficient of the current relay, and transmits the frequency band signals from the two signal sources to the corresponding signal sources according to the distributed power.

进一步采用带通滤波器进行分离，有

为经带通滤波器i(i＝1,2)分离抽取出的来自信号源S_i(i＝1,2)的信号，ρ_j为中继R_j处的动态PS系数，h_ij为信号源S_i与中继R_j之间的信道响应，

代表信号源S_i的发射信号，

为中继R_j处的残余自干扰信道响应，

为中继R_j处发射信号中对应于信号源S_i(i＝1,2)信号频段的部分。Further, in the relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network, the information waveform signal separated by the PS strategy at the selected relay R _j

A bandpass filter is further used for separation, there are

is the signal from the signal source S _i (i=1, 2) separated and extracted by the band-pass filter i (i=1, 2), ρ _j is the dynamic PS coefficient at the relay R _j , h _ij is the signal the channel _{response between source Si and relay R j ,}

represents the transmitted signal of the signal source _Si ,

is the residual self-interference channel response at relay R _j ,

is the part of the signal frequency band corresponding to the signal source Si ( _i =1, 2) in the transmitted signal at the relay R _j .

在进入EH模块后被转化为直流电能储存至电池模块，将中继R_j的电池模块能提供的总发射功率记为

功率分配模块将对

进行重新分配，设定系统调整功率分配系数至

由于

有：

为中继R_j处总发射功率

中用于信号源S₁频段信号的功率部分，

为中继R_j处总发射功率

中用于信号源S₂频段信号的功率部分，s_j为中继R_j的功率分配系数。Further, in the relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network, the energy waveform signal separated by the PS strategy at the selected relay R _j

After entering the EH module, it is converted into DC power and stored in the battery module, and the total transmit power that can be provided by the battery module of the relay R _j is recorded as

The power distribution module will

To redistribute, set the system to adjust the power distribution coefficient to

because

Have:

is the total transmit power at the relay R _j

is used for the power part of the S ₁ -band signal of the signal source,

is the total transmit power at the relay R _j

is used for the power part of the signal source S ₂ frequency band signal, s _j is the power distribution coefficient of the relay R _j .

或

其中，

为中继R_j对信号源

频段信号的放大倍数，

为中继R_j处带通滤波器i(i＝1,2)所引入的加性高斯白噪声的方差，

表示信号源S_i(i＝1,2)接收天线端的加性高斯白噪声的方差。Still further, in the relay cooperative transmission method for the bidirectional transmission FD multi-relay cooperative SWIPT network, the signal-to-noise signal that has finally been forwarded by the selected relay R _j and received and processed by the signal source S _i (i=1, 2) is calculated. than

or

in,

For the relay R _j pair of signal sources

The amplification factor of the frequency band signal,

is the variance of the additive white Gaussian noise introduced by the bandpass filter i (i=1,2) at the relay R _j ,

Represents the variance of the additive white Gaussian noise at the receiving antenna of the signal source Si ( _i =1, 2).

为目标函数求解中继R_j处的最优PS系数

和最优功率分配系数

有：

其中，

为当前已选中继集，S_R为原有已选中继集。Further, in the relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network, the

Solve the optimal PS coefficients at relay R _j for the objective function

and the optimal power distribution coefficient

Have:

in,

is the currently selected relay set, and _SR is the original selected relay set.

The present invention adopts the above-mentioned technical scheme, and has the following beneficial effects with respect to the prior art:

(1) The present invention designs a SWIPT receiver structure that supports FDD communication at the relay. By realizing the signal separation operation of signals in different signal source frequency bands and completing the power redistribution operation after signal separation, the SWIPT network can be adaptive according to different channel conditions. The power allocation coefficient is adjusted to make an adaptive response to the dynamic channel weakening of the mixed signal at the relay in the multi-relay cooperative network.

(2) On the basis of the relay node of the SWIPT receiver structure, the present invention determines the optimal PS system and power distribution coefficient of the relay with the minimum probability of one-way link interruption as the goal, and then uses the sum of the bidirectional link traversal capacity The maximum target is to select the relays that participate in the two-way communication. Since each relay has the ability to separate mixed signals through frequency bands, the relays selected to participate in the two-way communication can forward signals under their respective optimal power distribution coefficients, improving the orientation of the two-way communication. Transmission efficiency of the FD multi-relay cooperative SWIPT system.

(3) The relay of the SWIPT receiver structure adopted in the present invention can separate the self-interference signal, and the separated self-interference signal and the signal transmitted by the signal source are both amplified and processed, and the power is distributed, so the relay node of the present invention can be used It is flexibly applied to the relay network where the signal transceiver terminal can support FDD communication, and can effectively suppress the self-interference of each node in the relay network.

Description of drawings

Fig. 1 is a topology diagram of a bidirectional transmission FD AF multi-relay cooperative SWIPT network model based on a PS strategy.

FIG. 2 is a schematic diagram of a SWIPT receiver inside a relay node.

Figure 3 shows the traversal capacity and comparison of different RS algorithms.

FIG. 4 is a comparison diagram of outage probability of different RS algorithms.

FIG. 5 is a schematic diagram of selecting the number of relays in the GRS algorithm.

Detailed ways

The relay cooperative transmission method for bidirectional transmission FD multi-relay cooperative SWIPT network proposed by the present invention designs a relay cooperative transmission scheme based on the FDD mechanism, and redesigns the SWIPT receiver structure at the relay to support FDD communication. After the separation operation of signals in different frequency bands is completed, the power is redistributed, and the optimal PS coefficient and the optimal power allocation coefficient are obtained with the minimum probability of single link interruption as the optimization goal. Based on this, the performance of various RS algorithms is analyzed. The theoretical derivation of the technical solution of the invention will be described in detail below with reference to FIGS. 1 to 3 .

代表被选择中继所组成的集合，网络可用总带宽为B，|·|运算符表示一个集合的基，则每个被选中继所分得的信道带宽为

假设该多中继协作网络中所有链路的信道均为瑞利衰弱信道，并将信号源S_i(i＝1,2)与中继R_j(j＝1,2,...,N)之间的信道响应记作h_ij。Consider the bidirectional transmission double-hop FD AF multi-relay cooperative SWIPT network model based on the PS strategy, as shown in Figure 1. It is assumed that there are N operational relays R ₁ , R ₂ ,..., R _N and two signal sources S ₁ , S ₂ with information encoding and decoding functions in the network, and each relay has the function of energy harvesting SWIPT receiver. It is assumed that each relay adopts a passive design, that is, all the power consumed by the relay during operation comes from the energy collected by itself from the RF signal. All nodes including two signal sources work in FD mode. At the same time, in order to avoid signal interference between relays in FD mode, equal frequency bands are used to divide the total bandwidth of the system so that each selected relay works in an incoherent mode. within the frequency range. It is assumed that the signal source S ₁ and the signal source S ₂ are not reachable, so the direct link is not considered. Assume

represents the set composed of selected relays, the total available bandwidth of the network is B, and the |·| operator represents the basis of a set, then the channel bandwidth allocated by each selected relay is

It is assumed that the channels of all links in the multi-relay cooperative network are Rayleigh fading channels, and the signal source S _i (i=1,2) and the relay R _j (j=1,2,...,N ) between the channel responses are denoted as h _ij .

在本发明所设定的双向通信场景下，两信号源S₁和S₂均工作在FD模式，因此信号源S₁和S₂处的自干扰也需要考虑，将信号源S_i处的RSI信道响应记作

假设所有信道的信道状态信息(CSI)对信号源完全已知，系统在应用相应的RS算法时，将根据CSI的瞬时性和随机性，动态选择“最优”中继集以最大化系统容量。为了简化分析，忽略在中继内部电路和在CSI获取阶段的信号处理所消耗的电能。FIG. 2 shows the structure model of the relay node SWIPT receiver designed by the present invention and capable of supporting FDD communication. The residual self-interference (RSI) channel response at relay R _j is denoted as

In the bidirectional communication scenario set by the present invention, the two signal sources S ₁ and S ₂ both work in FD mode, so the self- _interference at the signal sources S ₁ and S ₂ also needs to be considered. channel response

Assuming that the channel state information (CSI) of all channels is completely known to the signal source, when the system applies the corresponding RS algorithm, it will dynamically select the "optimal" relay set to maximize the system capacity according to the temporality and randomness of the CSI. . To simplify the analysis, the power consumed by the internal circuits of the relay and the signal processing in the CSI acquisition stage is ignored.

During the operation of the relay network, when the relay R _j is selected, the relay R _j will receive the mixed signals of different frequency bands from the signal source S ₁ and the signal source S ₂ , and at the same time, the processed mixed signals will be The broadcast form is forwarded to the signal sources S ₁ and S ₂ respectively. After each signal source receives the mixed signal, it separates and extracts the signal from another signal source, and uses the maximum ratio combining (MRC) diversity reception method to process the forwarded signals from multiple selected relays. The selected relay R _j divides the received mixed signal into two parts according to the PS coefficient ρ _j : (1-ρ _j ), wherein the ρ _j part enters the IP module, and the 1-ρ _j part enters the EH module. The part entering the IP module will be filtered by band-pass filter 1 and band-pass filter 2 respectively after being demodulated to filter out the signals from signal sources S ₁ and S ₂ belonging to different frequency bands. The signal is amplified using the AF protocol. The signal entering the EH module is converted into DC energy in the form of energy waveform by the internal circuit and temporarily stored in the battery module, and then the relay will redistribute the power of the energy stored in the battery to the signal source S ₁ Power amplification of band signal and source S ₂ band signal. The following is a detailed derivation of the analytic expressions of the system traversal capacity sum and outage probability under the FDD-based relay cooperative transmission scheme proposed by the present invention, and the RS algorithm is explained with the greedy relay selection (GRS, Greedy Relay Selection) algorithm as a typical example. ideas.

(1) FDD-based relay cooperative transmission scheme

During network operation, when relay R _j is selected, the mixed signal received at R _j can be expressed as:

为中继R_j处的接收信号，h_ij为信号源S_i与中继R_j之间的信道响应，

代表信号源S_i的发射信号，

为中继R_j处的RSI信道响应，

为中继R_j处的发射信号，

为中继R_j处的天线噪声，

服从均值为0，方差为

的复高斯分布

其中，

N_A代表中继R_j接收天线处的噪声功率谱密度，[n]表示时域。in,

is the received signal at the relay R _j , _{h ij} is the channel response between the signal source Si and the relay R _j ,

represents the transmitted signal of the signal source _Si ,

is the RSI channel response at relay R _j ,

is the transmitted signal at the relay R _j ,

is the antenna noise at the relay R _j ,

The mean is 0, and the variance is

The complex Gaussian distribution of

in,

N _A represents the noise power spectral density at the receiving antenna of the relay R _j , and [n] represents the time domain.

After the relay R _j receives the mixed signal from the signal sources S ₁ and S ₂ , it will be divided into two parts according to the PS coefficient: ρ _j : (1-ρ _j ), ρ _j ∈ [0,1], respectively Delivered to the IP module and EH module as follows:

首先进入IP模块完成信号解调，随后带通滤波器1和带通滤波器2按照不同频段将混合信号进行分离，所采用频段分别对应信号源S₁和S₂的频率范围，即带通滤波器i(i＝1,2)分离抽取出的信号为来自信号源S_i(i＝1,2)的信号，将中继R_j处带通滤波器i(i＝1,2)分离抽取出的信号记作

有：

First, enter the IP module to complete the signal demodulation, and then the band-pass filter ₁ and the band-pass filter ₂ separate the mixed signal according to different frequency bands. The signal separated and extracted by the device _i (i=1,2) is the signal from the signal source Si (i=1,2), and the bandpass filter i (i=1,2) at the relay R _j is separated and extracted The outgoing signal is recorded as

Have:

为中继R_j所发射信号

中对应于信号源S_i(i＝1,2)信号频段的部分，考虑在实际电路系统中，信号

中热噪声相比有用信号占比很小，即：Among them, the signal

Signal transmitted for relay R _j

The part corresponding to the signal frequency band of the signal source S _i (i=1, 2), considering that in the actual circuit system, the signal

Compared with the useful signal, the medium thermal noise has a small proportion, namely:

采用AF协议进行放大处理，记中继R_j对来自信号源S_i(i＝1,2)频段信号的放大倍数为

有：The system then separates the extracted signal with a bandpass filter i (i=1,2)

The AF protocol is used for amplification processing, and the amplification factor of the signal from the signal source Si ( _i =1,2) frequency band by the relay R _j is recorded as

Have:

为系统对来自信号源S_i(i＝1,2)频段信号进行放大处理后的输出信号，

代表中继R_j处带通滤波器i(i＝1,2)所引入的加性高斯白噪声(AWGN)，服从均值为0，方差

为的复高斯分布

其中，

代表中继R_j处内部电路的噪声功率谱密度。In formula (6), the signal

is the output signal after the system amplifies the frequency band signal from the signal source Si ( _i =1,2),

Represents the additive white Gaussian noise (AWGN) introduced by the bandpass filter i (i=1,2) at the relay R _j , obeying the mean of 0 and the variance

The complex Gaussian distribution of

in,

represents the noise power spectral density of the internal circuit at the relay R _j .

中噪声占比很小，有

式(6)又可改写为：due to the signal

The proportion of noise in the medium is very small, and there are

Equation (6) can be rewritten as:

Then the transmit power corresponding to the frequency band signal from the signal source Si ( _i =1,2) that the relay R _j needs to provide is:

将式(8)视为关于变量

的一元方程，解得：To simplify the analysis, it is _assumed that the signal transmission powers of the _two signal sources S1 and S2 are both P _S , namely:

Equation (8) is considered to be about the variable

The one-variable equation of , solves:

其完全由电池模块所储存的电能所驱动，分别用于信号源S₁和S₂频段信号的功率放大，有：The total transmit power required to be provided by the relay R _j is recorded as

It is completely driven by the electric energy stored in the battery module, and is used for power amplification of the signal source S ₁ and S ₂ frequency bands respectively, including:

进入EH模块后被转化为直流电能储存至电池模块，记EH模块的能量收集效率为ξ，则中继R_j能够提供的总发射功率

的值可表示为：The signal in formula (3)

After entering the EH module, it is converted into DC electric energy and stored in the battery module. Denote the energy collection efficiency of the EH module as ξ, then the total transmit power that the relay R _j can provide

The value of can be expressed as:

的一元方程，解得：Equation (11) is considered to be about the variable

The one-variable equation of , solves:

进入EH模块后被转化为直流电能储存至电池模块，随后功率分配模块将对电池模块中的电能进行功率上的重新分配，设定系统调整功率分配系数至

联立式(10)，有方程组：Signal

After entering the EH module, it is converted into DC electric energy and stored in the battery module, and then the power distribution module will redistribute the electric energy in the battery module, and set the system to adjust the power distribution coefficient to

Simultaneous equation (10), there is a system of equations:

Solving this system of equations, we get:

和式(12)分别代入式(14)中

和

分别求解关于变量

和变量

的一元方程，得：Put the formula (9) in

and formula (12) are substituted into formula (14) respectively

and

Solve separately about the variables

and variable

The one-variable equation of , we get:

The signal forwarded by the relay R _j and received by the signal source Si ( _i =1,2) can be expressed as:

为信号源S_i(i＝1,2)处的RSI信道响应，而w_ij[n](i＝1,2)代表在中继R_j工作频率范围内作用于信号源S_i(i＝1,2)接收天线端的AWGN，其服从均值为0，方差为

的复高斯分布

其中，

N_S表示信号源S_i(i＝1,2)接收天线端的噪声功率谱密度。in,

is the RSI channel response at the signal source Si ( _i =1,2), and w _ij [n] (i=1,2) represents the signal source Si ( _i =1, 2) acting on the signal source Si ( _i = 1,2) The AWGN at the receiving antenna end, which obeys the mean value of 0 and the variance is

The complex Gaussian distribution of

in,

N _S represents the noise power spectral density at the receiving antenna end of the signal source Si ( _i =1, 2).

Then the signal power forwarded by the relay R _j and received by the signal source Si ( _i =1, 2) can be calculated as:

代入式(18)，

可表示为：The formula (8) is given by

Substitute into equation (18),

can be expressed as:

信号源S_i(i＝1,2)作为支持FDD通信的收发机，接收信号时滤除了与其发送信号同频段的信号，即式(19)中第一项、第二项、第三项、第七项信号分量将被滤除，而对应另一端信号源

发送信号频段的有用信号为第四项、第五项信号分量，第六项和第八项信号分量则为噪声分量，其中，

分量在经过信号源S_i(i＝1,2)接收机过滤处理后将变为

因此得到经中继R_j转发并由信号源S_i(i＝1,2)接收处理后的信号信噪比为：Among them, when i=1,2, there are

The signal source S _i (i=1,2) is used as a transceiver supporting FDD communication. When receiving the signal, it filters out the signal in the same frequency band as the transmitted signal, that is, the first item, the second item, the third item, The seventh signal component will be filtered out and correspond to the signal source at the other end

The useful signals of the transmitted signal frequency band are the fourth and fifth signal components, and the sixth and eighth signal components are noise components, where,

The components will _become

Therefore, the signal-to-noise ratio of the signal forwarded by the relay R _j and received and processed by the signal source Si ( _i =1, 2) is:

或

or

的值已由式(15)或式(16)给出，而

的值可由式(14)和式(12)给出。In formula (20),

The value of is given by Eq. (15) or Eq. (16), and

The value of can be given by equation (14) and equation (12).

链路的信道容量可以表示为：Finally, the signal source Si ( _i =1, 2) uses the MRC diversity reception method to process the signal samples from multiple forwarding relays,

The channel capacity of a link can be expressed as:

(2) Modeling the relay selection problem

代表系统最终选择用于协作通信的中继集合，本发明在评估不同中继选择算法性能时，主要考虑了S₁-S₂链路和S₂-S₁链路遍历容量之和与中断概率两个性能指标。其中，S₁-S₂链路和S₂-S₁链路遍历容量之和可被表示为：In formula (21),

The representative system finally selects the relay set for cooperative communication. When evaluating the performance of different relay selection algorithms, the present invention mainly considers the sum of the traversal capacity of the S ₁ -S ₂ link and the S ₂ -S ₁ link and the probability of interruption Two performance indicators. where the sum of the S ₁ -S ₂ link and the S ₂ -S ₁ link traversal capacity can be expressed as:

The outage probability can be defined as:

低于此值则认为通信中断。Among them, C _th is the interruption threshold of channel capacity, if

Below this value, communication is considered to be interrupted.

链路的信道容量可以记为

其中，

为中继R_j加入原有中继集S_R后的已选中继集，，k表示已选中继集

中第k个已选中继，则中继R_j参与协作通信中断概率可定义为：When comparing the relay and the relay which is better, different evaluation criteria will select different relays. The present invention is based on the two-way transmission relay cooperative network based on the FDD mechanism, and considers the power distribution of the two signal source frequency band signals at the relay. The optimal relay will cause the optimal power distribution coefficient at the relay to be seriously unbalanced, that is, the signal sent by one of the two signal sources is not effectively forwarded, which is contrary to the scenario setting of the bidirectional transmission relay network of the present invention. In order to maintain the communication fairness between the S ₁ -S ₂ link and the S ₂ -S ₁ link in the bidirectional transmission scenario, the present invention still uses the channel capacity of the relay participating in the cooperative communication as the criterion for judging the quality of the relay. When solving the optimal PS coefficient and the optimal power distribution coefficient, the outage probability of the relay participating in the cooperative communication is taken as the objective function. Based on equation (21), the relay R _j participates in the cooperative communication

The channel capacity of the link can be written as

in,

The selected relay set after adding the original relay set SR to the relay _{R j} , k represents the selected relay set

The k-th selected relay in , then the interruption probability of relay R _j participating in cooperative communication can be defined as:

最小为优化目标，求得最优PS系数和最优功率分配系数。从式(24)、式(21)容易分析出，最小化中断概率

等价于最大化

也就等价于最大化

记中继R_j参与协作通信所要优化的目标函数为

其中

为中继R_j加入原有已选中继集S_R后的已选中继集，则中继R_j处的最优PS系数

和最优功率分配系数

为：First, with the probability of interruption

The minimum is the optimization objective, and the optimal PS coefficient and the optimal power distribution coefficient are obtained. From equations (24) and (21), it is easy to analyze that the probability of interruption is minimized

is equivalent to maximizing

is equivalent to maximizing

Note that the objective function to be optimized by the relay R _j to participate in cooperative communication is

in

For the selected relay set after adding the original selected relay set _SR to the relay R _j , the optimal PS coefficient at the relay R _j

and the optimal power distribution coefficient

for:

作为判断中继优劣的标准，则在中继数量为N的中继集合中选出的单个最优中继的下标为：Then, the sum of the channel capacities of the bidirectional links participating in the cooperative communication with the relay R _j

As a criterion for judging the quality of relays, the subscript of the single optimal relay selected from the relay set with the number of relays N is:

和

由式(25)计算得出。in

and

It is calculated by formula (25).

Several common RS algorithms are single relay selection (SRS, Single Relay Selection) algorithm, all selection (AP, All Participate) algorithm, greedy relay selection (GRS, Greedy Relay Selection) algorithm and exhaustive search (ES, Exhaustive Search) algorithm )algorithm. Among them, the AP algorithm selects all relays for cooperative communication; while the SRS algorithm is essentially a special case where the GRS algorithm only selects one "optimal" relay in the first step; although the ES algorithm can achieve the optimal performance in the true sense , but when the number of optional relays is large, the algorithm complexity is huge and cannot be applied in practice. The GRS algorithm can achieve almost the same performance as the optimal ES algorithm in the average capacity of the system, and its algorithm complexity is low, which can meet the real-time processing requirements, so it is widely used in practice. Therefore, the relay cooperative transmission method will be described below using the GRS algorithm as an implementation algorithm.

(3) Greedy relay selection algorithm

第n步结束后的已选中继集

可表示为：In a relay network with N number of relays, the greedy relay selection (GRS) algorithm is divided into N steps, and the initial state of the relay set is

The selected relay set after the end of step n

can be expressed as:

为第n步在剩余可选中继集中选择的用于加入中继集

的最优中继的下标，有：in

Selected in the remaining optional relay set for step n to join the relay set

The subscripts of the optimal relays are:

和

可由式(25)在

下计算得出。in

and

It can be obtained from equation (25) in

Calculated below.

中选出最优中继集

有：exist

select the optimal relay set

Have:

3 to 5, the performance performance and applicable scenarios of the three RS algorithms of the SRS algorithm, the AP algorithm and the GRS algorithm under the relay cooperative transmission method of the present invention are described respectively. Obviously, the described embodiments are only some, but not all, embodiments of the present invention. All other embodiments of the same inventive concept obtained by those of ordinary skill in the art based on the embodiments of the present invention fall into the protection scope of the present invention. All wireless link channels are modeled by Rayleigh fading channels. Table 1 lists the main parameters used in the simulation process.

Table 1 List of simulation parameters

system bandwidth 10MHz noise power spectral density -174dBm/Hz Energy Harvesting (EH) Efficiency 90% Residual Self-Interference (RSI) Channel Gain -80dB The number of relays in the candidate relay set 3 or 8 Spectral efficiency cutoff threshold 0.5bps/Hz

FIG. 3 compares the bidirectional link traversal capacity sum of different RS algorithms under the relay cooperative transmission scheme based on the FDD mechanism proposed by the present invention, and discusses two cases where the number of relays in the network is 3 and 8. Under the framework of the bidirectional relay network of the present invention, the GRS algorithm still achieves almost completely similar performance to the ES algorithm, so the ES algorithm is not presented and analyzed here. The performance of the ES algorithm represents the optimal performance of the system, but because the complexity of the algorithm is too high, it is inconvenient for practical application. Here, the GRS algorithm with lower algorithm complexity represents the optimal performance of the system. It is also known from Figure 3 that the sum of traversal capacity of the GRS algorithm on all source transmit power coordinates is higher than that of the AP algorithm and the SRS algorithm.

Looking at Figure 3, when the source transmit power is low, since the AP algorithm selects all the relays to collect more energy, more signal copies are generated, and finally the received signal-to-noise ratio at the two signal sources is strengthened. Therefore, the AP algorithm The ergodic capacity sum is similar to that of the GRS algorithm, which represents the best performance, and is significantly ahead of the ergodic capacity sum of the SRS algorithm. It can also be known from Figure 5 that the GRS algorithm will select a relatively large number of relays when the source transmit power is low. In view of the extremely low algorithm complexity of AP, the AP algorithm is generally enabled in the system under low source transmit power. However, when the transmission power of the source increases gradually, the performance of the AP algorithm is more and more behind the performance of the GRS algorithm and the SRS algorithm. This is because when the transmission power of the source is high, a single optimal relay can already collect considerable energy for signal transmission. At this time, only selecting a single optimal relay can fully and effectively utilize the system bandwidth. Therefore, GRS The algorithm also tends to select a single optimal relay to achieve the optimal performance of the system when the source transmit power is high. In view of the extremely low linear complexity of the SRS algorithm, the system generally enables the SRS algorithm when the source transmit power is high. Similarly, when the number of relays increases from 3 to 8, the traversal capacity sum of the AP algorithm will increase under low source transmit power, while the traversal capacity sum will decrease at high source transmit power. In summary, the SRS algorithm and the AP algorithm achieve near-optimal sum of ergodic capacity under high and low source transmit power, respectively, and have extremely low algorithm complexity.

In addition, when the number of relays increases from 3 to 8, the performance improvement of the SRS algorithm is more obvious at low source transmit power, while the performance improvement is very limited at high source transmit power. This is because at low source transmit power, the increase in the number of relays provides more channel state possibilities, and the probability of being able to select a better relay increases, while high source transmit power makes up for the defect of channel state fluctuations. The GRS algorithm more or less benefits from the increase in the number of relays in all source transmit power ranges. Similar to the SRS algorithm, the GRS algorithm has less ergodic capacity and improvement under high source transmit power. In summary, the AP algorithm, the SRS algorithm and the GRS algorithm can obviously benefit from the increase in the number of relays at low source transmit power, while only the GRS algorithm can obtain a small performance gain from the increase in the number of relays at high source transmit power. The SRS algorithm is almost unaffected, and the AP algorithm even achieves a lower traversal capacity sum.

的概率分布差异，在仿真参数所设定的频谱效率中断阈值为0.5bps/Hz的前提下，SRS算法下

的概率大于AP算法下

当中继数量从3增加至8时，无论是SRS算法、AP算法还是GRS算法都获得中断性能的明显提升，由于AP算法和GRS算法在较低源发射功率下选择的中继数量较多，能够从中继数量的增加中获益，因此AP算法和GRS算法在较低源发射功率下的性能提升就已经较为明显。所以，当所设计中继网络对中断性能要求较高时，可通过增加中继数量并且优先使用AP算法和GRS算法来实现，如果AP算法与GRS算法性能相近，可考虑使用算法复杂度极低的AP算法。此外，在图3中，中继数量为8的SRS算法相较中继数量为3的GRS算法在所有源发射功率区间内都获得了更高的遍历容量和，并且在图4中，当源发射功率超过25dB时，中继数量为8的SRS算法在中断性能上也超过了中继数量为3的GRS算法，这说明中继数量对系统性能有着很大影响，如果所设计中继网络预算充足，应该尽量布置更多的中继。FIG. 4 compares the outage probability of different RS algorithms under the relay cooperative transmission scheme based on the FDD mechanism proposed by the present invention, and also considers two situations in which the number of relays in the network is 3 and 8. Different from the traversal capacity of the SRS algorithm and the leading AP algorithm under the high source transmit power in Figure 3, the SRS algorithm is weaker than the AP algorithm in the performance of the outage probability. This stems from the relationship between the SRS algorithm and the AP algorithm.

On the premise that the spectral efficiency interruption threshold set by the simulation parameters is 0.5bps/Hz, under the SRS algorithm

The probability is greater than that under the AP algorithm

When the number of relays is increased from 3 to 8, the interruption performance of the SRS algorithm, the AP algorithm and the GRS algorithm is significantly improved. Since the AP algorithm and the GRS algorithm select a large number of relays at a lower source transmit power, they can Benefiting from the increase in the number of relays, the performance improvement of the AP algorithm and the GRS algorithm at a lower source transmit power is already obvious. Therefore, when the designed relay network requires high interruption performance, it can be achieved by increasing the number of relays and giving priority to using the AP algorithm and the GRS algorithm. If the performance of the AP algorithm is similar to the GRS algorithm, it can be considered to use a very low algorithm complexity. AP algorithm. In addition, in Figure 3, the SRS algorithm with 8 relays achieves higher ergodic capacity sums in all source transmit power intervals than the GRS algorithm with 3 relays, and in Figure 4, when the source When the transmit power exceeds 25dB, the SRS algorithm with 8 relays also exceeds the interrupt performance of the GRS algorithm with 3 relays, which shows that the number of relays has a great impact on the system performance. If the designed relay network budget Sufficient, should try to arrange more relays.

Figure 5 depicts the variation of the number of relays selected by the GRS algorithm representing the optimal performance of the system with the source transmit power when the number of relays is 3 and 8 respectively. Consistent with the previous analysis, the GRS algorithm tends to select a larger number of relays when the source transmit power is low, and selects a number of relays that tends to 1 when the source transmit power is high. When the source transmit power is 0dB, the GRS algorithm with 3 relays selects the number of relays close to 3, and the utilization of the relays is close to saturation, while the GRS algorithm with 8 relays only selects the relays close to 6 The utilization rate of the relays is only 75%, which shows that when the number of relays increases to a certain extent, the channel diversity between different relays is sufficient, and it is not advisable to blindly increase the number of relays to improve the system performance. In addition, Figure 5 also shows that when the source transmit power starts to exceed about 17dB, the number of relays required to be selected by the GRS algorithm with the number of relays becomes less than that of the GRS algorithm with the number of relays 3, which means that when the source transmits When the power reaches a certain level, the number of relays selected by both is less. At this time, the former only needs to select fewer relays to achieve the optimal performance of the system by virtue of the richer channel diversity.

Claims (5)

1. The relay cooperative transmission method facing the bidirectional transmission FD multi-relay cooperative SWIPT network is characterized in that each relay in the bidirectional transmission FD multi-relay cooperative SWIPT network is an SWIPT receiver supporting FDD communication, for a link between two signal sources working in different frequency bands in the bidirectional transmission FD multi-relay cooperative SWIPT network, the optimal PS coefficient and the optimal power distribution coefficient of each relay are determined by taking the minimum unidirectional link interruption probability as a target, the maximum sum of the traversal capacity of the bidirectional links is taken as a target, and the optimal relay participating in the bidirectional link communication is selected according to the optimal PS coefficient and the optimal power distribution coefficient of each relay.

2. The relay cooperative transmission method oriented to the FD multi-relay cooperative SWIPT network of claim 1, wherein the SWIPT receiver supporting FDD communication comprises:

a power division module for receiving the mixed signal from the two signal sources, dividing the mixed signal into two parts according to the PS coefficient of the current relay, and outputting

Part of the mixed signal is output to the IP module

Part of the mixed signal to the EH module, p _j PS coefficient for the jth relay, p _j ∈[0,1]；

IP module to received

Demodulating part of the mixed signal from said

Separating and extracting signals from the two signal sources from part of mixed signals, amplifying the separated and extracted signals from the two signal sources, and outputting frequency band signals from the two signal sources;

EH module, receive

Part of the mixed signal is

Part of the mixed signals are converted into direct current electric energy and then stored in the battery module; and a process for the preparation of a coating,

and the power distribution module is used for carrying out power distribution on the direct current electric energy stored in the battery module according to the power distribution coefficient of the current relay and transmitting the frequency band signals from the two signal sources to the corresponding signal sources according to the distributed power.

3. The relay cooperative transmission method oriented to the FD multi-relay cooperative SWIPT network of claim 2, wherein the method is characterized in that

Part of the mixed signal being

The above-mentioned

Part of the mixed signal being

Wherein,

are respectively time domain [ n]The IP module and the EH module in the jth relay receive the mixed signal,

is the time domain [ n ]]The mixed signal h from two signal sources received by the power distribution module in the jth relay _ij For the ith signal source S _i With the jth relay R _j In the channel response between the first and second channels,

is the time domain [ n ]]Internal ith signal source S _i Is transmitted in the same manner as in the previous embodiment,

for the jth relay R _j The response of the RSI channel at (a),

is the time domain [ n]Inner jth relay R _j The transmission signal of (a) is transmitted,

is the time domain [ n ]]Inner jth relay R _j The antenna noise of (a) is detected,

obedience mean of 0 and variance of

Complex gaussian distribution of

N _A For the jth relay R _j A noise power spectral density at a receiving antenna, B is a total bandwidth of the bidirectional transmission FD oriented multi-relay cooperative SWIPT network,

for the jth relay R _j Join original selected relay set S _R The subsequent selected relay set is selected to be,

4. the relay cooperative transmission method oriented to the FD multi-relay cooperative SWIPT network of claim 3, wherein the expression for determining the optimal PS coefficient and the optimal power distribution coefficient at each relay with the minimum probability of interruption of the unidirectional link as a target is as follows:

wherein,

optimal PS coefficient and optimal power distribution coefficient, s, for the jth relay _j The power allocation coefficient for the jth relay is,

the probability of a unidirectional link outage for the jth relay to engage in bidirectional communication,

in order to obtain the signal-to-noise ratio of the signal which is forwarded by the jth relay and received and processed by the 1 st signal source,

in order to obtain the signal-to-noise ratio of the signal which is forwarded by the jth relay and received and processed by the 2 nd signal source,

in order to obtain the signal-to-noise ratio of the signal which is forwarded by the jth relay and received and processed by the ith signal source,

reference numerals for the other of the two signal sources relative to the ith signal source,

for the jth relay R _j For signals from signal sources

Amplification of frequency band signals, P _S Is the transmission power of the signal source,

for the jth relay R _j Provided corresponding to the signal from the signal source

The transmission power of the frequency band signal is,

for the jth relay R _j The variance of the complex gaussian distribution obeyed by the additive white gaussian noise introduced by the band-pass filter used for separating the radio frequency signal corresponding to the signal source frequency band signal,

for the jth relay R _j The noise power spectral density of the internal circuitry,

to relay R at jth _j The variance of the complex gaussian distribution obeyed by the additive white gaussian noise acting at the receive antenna end of the signal source within the operating range,

N _S the noise power spectral density at the antenna end is received for the signal source.

5. The relay cooperative transmission method for the FD multi-relay cooperative SWIPT network as claimed in claim 4, wherein the table is used to select the optimal relay participating in the bidirectional link communication according to the optimal PS coefficient and the optimal power distribution coefficient at each relay with the maximum sum of the bidirectional link traversal capacities as the targetThe expression is as follows:

wherein j is ^* For optimal relay index, N is the set of relays { R } _j The number of relays in the (j) is,

the sum of the traversal capacities of the bidirectional links participating in bidirectional communication with the optimal PS coefficient and the optimal power distribution coefficient is taken as the jth relay,

the sum of the bi-directional link traversal capacities for the jth relay participating in the bi-directional communication,

k represents the selected relay set

The kth selected relay.

Images