CN101977096B - Amplify-and-forward (AF) cooperative communication method for performing network coding on MPSK signal - Google Patents

Abstract

The invention discloses an amplify-and-forward (AF) cooperative communication method for performing network coding on an MPSK signal, belonging to the technical field of wireless communication. The method is implemented as follows: based on a system model for communication between two wireless users which relay with each other and a base station, respectively modulating data for sent MPSK at a user side at time slot 1 and at time slot 2; performing AF on received signals and local signals after network coding by a relay user at time slot 3; and performing maximum likelihood detection on the received three time slots by the base station to recover user data. Simultaneously, the technical scheme of the invention also provides a method similar to the maximum likelihood detection aiming to the BPSK, which can simplify the detection operation complexity of a receiving end. The method can be suitable for occasions such as a sensor network and the like which have low requirements on relays and can save one transmission time slot under the condition of having the same error rate as a common AF cooperation communication method, thus saving frequency spectrum resources (improving the frequency spectrum resource utilization factor) and total system transmission power.

Description

A Coordinated Communication Method for Amplifying and Fronthauling MPSK Signals with Network Coding

technical field

The invention belongs to the technical field of wireless communication, and relates to network coding technology and collaborative communication technology.

Background technique

The network coding technology was first proposed by Ahlswede et al. Its purpose is mainly to improve the communication capacity of the wired network. Yingda Chen et al first introduced network coding technology into collaborative communication. Assuming that the information transmission between users is reliable, they studied the decoding-forward (DF) cooperative communication method based on network coding (NC), and proved that the introduction of network coding into cooperative communication can improve the overall performance of the system (improve network transmission capacity and system throughput, enhance network robustness, balance network load, etc.). In addition, some literatures proposed a high spectral efficiency DF cooperative communication method based on network coding. In this method, the cooperative user acting as a relay decodes the data of the cooperative partner and XORs it with its own data, so as to realize the network Coding saves transmission time slots and improves spectral efficiency. However, the biggest disadvantage of this method is its high complexity, because: the relay user first needs to decode the data of the cooperative partner and perform XOR with its own data, and then encode the XOR result again; while the decoding process Usually adopts the loop iterative decoding process, its computational complexity and hardware overhead are very high, and it has high requirements on the relay users, so it is difficult to realize.

In some application scenarios (such as sensor networks), cooperative users cannot meet such high-complexity requirements. In such scenarios, amplified fronthaul (AF) cooperative communication is the first choice due to its simple implementation and full diversity order. collaborative communication method. Therefore, how to apply network coding technology to AF cooperative communication to obtain system performance gain is a problem worth studying.

After network coding, the user and the relay user transmit the same code word bits in the time slot for forwarding the coded information. Network Coding-based AF Cooperative Method (AF-NC) utilizes this property to improve spectral efficiency. In the optimal receiving algorithm, the base station will inevitably realize the maximum ratio combination of the symbol through a certain method, and what simple and effective power allocation method should be adopted to make the signal-to-noise ratio of the combined signal the highest is an urgent problem to be solved. one of the problems.

At the base station, maximum likelihood detection is performed on the received signal. However, due to the complexity of the Maximum Likelihood (ML) receiver for the AF-NC method, is it possible to propose a simplified receiving method and how to implement a simplified detection method under the premise of maintaining a good bit error rate performance? It is also a question that needs to be studied.

Contents of the invention

Based on the system model of communication between two wireless users relaying each other and a base station, the present invention proposes an amplified pretransmission cooperative communication method for performing network coding on MPSK signals. This method is based on the AF cooperative communication method of simple network coding (AF-NC), and adopts the maximum likelihood detection method, so that when MPSK modulation is used, AF-NC can obtain the same information as ordinary AF while improving spectrum utilization and saving transmission power. Comparable performance of cooperative communication methods.

Technical scheme of the present invention is:

An amplified fronthaul cooperative communication method for performing network coding on MPSK signals, based on a system model of communication between two users relaying each other and a base station (as shown in Figure 1), using a time-division multiplexing transmission method, and assuming The channel is a Rayleigh channel, as shown in Figure 2, including the following steps:

Step 1: In time slot 1, user 1 broadcasts the MPSK-modulated signal of user 1's own coded data with conventional transmission power, and user 2 and the base station detect and judge the received signal of user 1;

Step 2: In time slot 2, user 2 broadcasts the MPSK-modulated signal of user 2's coded data by using conventional transmission power, and user 1 and the base station detect and judge the received signal of user 2;

Step 3: The base station feeds back the pros and cons of the SNRs of the two uplink channels through the downlink control signaling channel;

Step 4: In time slot 3, the user with a higher channel SNR transmits network coded data with a power exceeding the normal transmission power, and the base station performs joint processing on the signals received in the three time slots, and finally obtains user 1 and User 2's encoded data.

In the above scheme: 1) the MPSK modulation method described in step 1 and step 2 is a kind of MPSK modulation method based on the constellation point mapping method of the circular domain; 2) the network coding process described in the step 4 adopts user 1 and user 2 The coded data is implemented by performing the product operation in the analog domain to replace the XOR operation of the digital signal in the usual network coding; 3) The joint processing described in step 4 is that the base station performs the maximum likelihood on the signals received in the 3 time slots After detection and demodulation, detection values of coded data of user 1 and user 2 are obtained.

In the three transmission modes shown in Figure 2, the diversity of the two user data received by the base station is guaranteed to be 2 (it has been proved that the AF-NC method can transmit within 3 time slots and obtain order 2 diversity). Compared with the non-cooperative transmission method, the AF cooperative communication transmission method avoids the possibility that the deep fading of the same channel interferes with the signal twice in a row, and the maximum ratio combination (weighted sum) of the same data sent by the two channels by the base station Then the transmission signal is detected. The AF-NC scheme proposed by the present invention adds a network coding part on this basis. Compared with the AF cooperative communication transmission method, a transmission time slot can be saved under the condition of achieving a considerable bit error rate, thereby saving the frequency spectrum. resource (improved spectrum resource utilization ratio), improved the data transmission rate of system (in unit time, the transmission capacity of common AF method is 1/2 bit, the transmission capacity of AF-NC method provided by the present invention can reach 2/ 3 bits); under the same total amount of transmit power, the saved transmit power of the 4th time slot can be increased to the 3rd time slot, so that, under the same channel conditions, the transmit signal can be further improved Signal-to-noise ratio, resulting in a lower bit error rate. At the data sending end, the present invention adopts the MPSK modulation mode to map M consecutive bits into a symbol, and the modulation result is a complex signal for transmission on the channel. At the receiving end of the base station, the maximum likelihood detection is performed on the received signal three times, and then demodulated, and finally the detected value of the encoded data is obtained, which greatly reduces the computational complexity and greatly reduces the difficulty of hardware implementation (as shown in Figure 3) .

To sum up, the present invention proposes an amplified fronthaul cooperative communication method for performing network coding on MPSK signals based on a system model of communication between two wireless users relaying each other and a base station. The method is based on the AF cooperative communication method of simple network coding (AF-NC), and adopts the maximum likelihood detection method in the signal detection at the receiving end, so that when MPSK modulation is used, a transmission can be saved under the condition of achieving a considerable bit error rate. time slots, thereby saving spectrum resources (improving the utilization rate of spectrum resources) and the total transmission power of the system. When the BPSK signal modulation mode is adopted, the present invention performs approximate processing on the maximum likelihood detection method in the aspect of signal detection at the receiving end, thereby greatly reducing the computational complexity and greatly reducing the difficulty of hardware implementation. In addition, compared with the AF coordinated communication transmission method, the present invention saves one transmission time slot, and under the same total amount of transmission power, the saved transmission power of the fourth time slot can be increased to the third time slot In this way, under the same channel conditions, the signal-to-noise ratio of the transmitted signal can be further improved, thereby achieving a lower bit error rate.

Description of drawings

Figure 1 is a schematic diagram of a system application scenario. User 1 and user 2 transmit data to the base station. In order to avoid interference, the two users use time division multiplexing for transmission.

Fig. 2 shows non-coordinated, AF-coordinated, and AF-NC-coordinated transmission methods. Figure 2(a) and (b) show the non-cooperative and common AF cooperative transmission methods respectively. In the non-cooperative transmission method, user 1 transmits the same data in time slot 1 and time slot 3; user 2 transmits the same data in time slot 2 and time slot 4. In the common AF transmission method, user 1 transmits data in time slot 1, and user 2 and the base station receive it; user 2 retransmits user 1's data in time slot 4, and user 1 and the base station receive it. Fig. 2(c) shows the AF-NC cooperative transmission method provided by the present invention.

Fig. 3 is a schematic diagram of AF-NC approximate maximum likelihood detection.

Figure 4 is the theoretical analysis and simulation results of AF-NC bit error rate (the average uplink signal-to-noise ratio of the two users is equal).

Figure 5 is the theoretical analysis and simulation results of AF-NC bit error rate (uplink average SNR of user 2 is 20dB).

Figure 6 is the simulation result of bit error rate of AF-NC MPSK modulation (the average SNR of uplink of two users is equal).

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

An amplified fronthaul cooperative communication method for performing network coding on MPSK signals, based on a system model of communication between two users relaying each other and a base station (as shown in Figure 1), using a time-division multiplexing transmission method, and assuming The channel is a Rayleigh channel, as shown in Figure 2, including the following steps:

Step 1: In time slot 1, user 1 broadcasts the MPSK-modulated signal of user 1's own coded data with conventional transmission power, and user 2 and the base station detect and judge the received signal of user 1;

Step 2: In time slot 2, user 2 broadcasts the MPSK-modulated signal of user 2's coded data by using conventional transmission power, and user 1 and the base station detect and judge the received signal of user 2;

Step 3: The base station feeds back the pros and cons of the SNRs of the two uplink channels through the downlink control signaling channel;

Step 4: In time slot 3, the user with a higher channel SNR transmits network coded data with a power exceeding the normal transmission power, and the base station performs joint processing on the signals received in the three time slots, and finally obtains user 1 and User 2's encoded data.

In the above scheme: 1) the MPSK modulation method described in step 1 and step 2 is a kind of MPSK modulation method based on the constellation point mapping method of the circular domain; 2) the network coding process described in the step 4 adopts user 1 and user 2 The coded data is implemented by performing the product operation in the analog domain to replace the XOR operation of the digital signal in the usual network coding; 3) The joint processing described in step 4 is that the base station performs the maximum likelihood on the signals received in the 3 time slots After detection and demodulation, detection values of coded data of user 1 and user 2 are obtained.

The present invention proposes an amplified pretransmission cooperative communication method for performing network coding on MPSK signals. If the MPSK signal is a BPSK signal, the maximum likelihood detection process is an approximate maximum likelihood detection process. The specific method is:

表示基站在时隙2接收信号的后验概率对数似然比；

表示基站在时隙3接收信号的后验概率对数似然比；Pr(x[n]＝+1)表示发送方发送编码、调制后的数据“+1”的先验概率；Pr(x[n]＝-1)表示发送编码、调制后的数据“-1”的先验概率；First define as follows: x ₁ [n] means the data sent by user 1; x ₂ [n] means the data sent by user 2; y ₁ [n] means the signal received by the base station in time slot 1; y ₂ [n] Indicates the signal received by the base station in time slot 2; y ₃ [n] indicates the signal received by the base station in time slot 3; Indicates the log-likelihood ratio of the posterior probability of the base station receiving the signal in time slot 1;

Indicates the logarithmic likelihood ratio of the posterior probability of the base station receiving the signal in time slot 2;

Represents the logarithmic likelihood ratio of the posterior probability of the base station receiving the signal in time slot 3; Pr(x[n]=+1) represents the priori probability of the sender sending encoded and modulated data "+1"; Pr(x[n]=+1) [n]=-1) represents the prior probability of sending coded and modulated data "-1";

对x₁[n]的最大似然检测值

(即最大后验概率检测值)为：

对x₂[n]的最大似然检测值

(即最大后验概率检测值)为：

从中可以看出，采用传统最大似然检测方法计算x₁[n]、x₂[n]的检测值时需要进行复杂的指数运算。本发明提出了一种复杂度较低的近似最大似然检测(Approximate Maximum Likelihood，AML)方法。考虑到Turbo编码中Max-Log-MAP解码算法用到的

可以达到一种很好的复杂度与性能的折衷。可以得到近似最大似然检测方法：对x₁[n]的最大似然检测值 ${\hat{x}}_1\left[n\right]\≈\mathrm{sign}\left(r_1\right[n]+\max \left(r_2\right[n]+r_{\mathrm{xor}}[n],0)-\max \left(r_2\right[n],r_{\mathrm{xor}}[n\left]\right)),$ 对x₂[n]的最大似然检测值 ${\hat{x}}_2{\hat{x}}_2\left[n\right]\≈\mathrm{sign}\left(r_2\right[n]+\max \left(r_1\right[n]+r_{\mathrm{xor}}[n],0)-\max \left(r_1\right[n],r_{\mathrm{xor}}[n\left]\right)).$ Due to the statistical prior probability

Maximum likelihood detection value for x ₁ [n]

(that is, the maximum posterior probability detection value) is:

Maximum likelihood detection value for _x2 [n]

(that is, the maximum posterior probability detection value) is:

It can be seen from the figure that complex exponent operations are required when calculating the detection values of x ₁ [n] and x ₂ [n] using the traditional maximum likelihood detection method. The present invention proposes an approximate maximum likelihood detection (Approximate Maximum Likelihood, AML) method with low complexity. Considering the Max-Log-MAP decoding algorithm used in Turbo encoding

A good compromise between complexity and performance can be achieved. The approximate maximum likelihood detection method can be obtained: the maximum likelihood detection value of x ₁ [n] ${\hat{x}}_1\left[\mathrm{no}\right]\≈\mathrm{sign}\left(r_1\right[\mathrm{no}]+\max \left(r_2\right[\mathrm{no}]+r_{\mathrm{xor}}[\mathrm{no}],0)-\max \left(r_2\right[\mathrm{no}],r_{\mathrm{xor}}[\mathrm{no}\left]\right)),$ Maximum likelihood detection value for _x2 [n] ${\hat{x}}_2{\hat{x}}_2\left[\mathrm{no}\right]\≈\mathrm{sign}\left(r_2\right[\mathrm{no}]+\max \left(r_1\right[\mathrm{no}]+r_{\mathrm{xor}}[\mathrm{no}],0)-\max \left(r_1\right[\mathrm{no}],r_{\mathrm{xor}}[\mathrm{no}\left]\right)).$

在图5中横坐标值为

仿真中数据帧长度为256，采用BPSK调制。图6为采用4PSK和8PSK调制时AF-NC的误码率性能仿真结果，图中横坐标值为在仿真中，传输资源分配及仿真条件与BPSK方法时完全一致。A flat Rayleigh fading channel is used in the simulation. Figure 4 and Figure 5 show the bit error rate simulation and numerical results of AF-NC and non-cooperative bit error rate. In Figure 4, the average SNR of each link is equal, and the abscissa value is

In Figure 5, the abscissa value is

The length of the data frame in the simulation is 256, using BPSK modulation. Figure 6 shows the simulation results of bit error rate performance of AF-NC when using 4PSK and 8PSK modulation, the abscissa in the figure is In the simulation, the transmission resource allocation and simulation conditions are completely consistent with the BPSK method.

Claims (4)

1. amplification forward pass collaborative communication method that mpsk signal is carried out network code, based on the system model that communicates between two users of mutual relaying and base station, adopt time-multiplexed transmission means, and suppose that channel is Rayleigh channel, comprise the following steps:

Step 1: at time slot 1, adopted the signal of coded data after the MPSK modulation of conventional transmitted power broadcasting user 1 self by user 1, differentiation is detected to the user's 1 that receives signal in user 2 and base station;

Step 2: at time slot 2, adopted the signal of coded data after the MPSK modulation of conventional transmitted power broadcasting user 2 self by user 2, differentiation is detected to the user's 2 that receives signal in user 1 and base station;

Step 3: by the good and bad result of base station by descending control signaling channel feedback two up channel signal to noise ratios;

Step 4: at time slot 3, adopt the power that surpasses conventional transmitted power to send the data of network code by the higher user of channel signal to noise ratio, Combined Treatment is carried out to 3 received signals of time slot in the base station, finally obtains user 1 and user's 2 coded data; Wherein said network code is user 1 and user's 2 coded data to be carried out the product calculation of analog domain.

2. the amplification forward pass collaborative communication method that mpsk signal is carried out network code according to claim 1, is characterized in that, the described MPSK modulation system of step 1 and step 2 is a kind of MPSK modulation system of the constellation point mapping mode based on justifying the territory.

3. the amplification forward pass collaborative communication method that mpsk signal is carried out network code according to claim 1, it is characterized in that, Combined Treatment described in step 4 is that the base station obtains the detected value of user 1 and user's 2 coded data to 33 received signals of time slot after carrying out Maximum Likelihood Detection, demodulation.

4. the amplification forward pass collaborative communication method that mpsk signal is carried out network code according to claim 3, is characterized in that, when mpsk signal was bpsk signal, described Maximum Likelihood Detection process was a kind of approximate Maximum Likelihood Detection process, and concrete grammar is:

At first be defined as follows: x ₁The data that [n] expression user 1 sends; x ₂The data that [n] expression user 2 sends; y ₁The signal that [n] expression base station receives at time slot 1; y ₂The signal that [n] expression base station receives at time slot 2; y ₃The signal that [n] expression base station receives at time slot 3;

The expression base station receives the posterior probability log-likelihood ratio of signal at time slot 1;

The expression base station receives the posterior probability log-likelihood ratio of signal at time slot 2; The expression base station receives the posterior probability log-likelihood ratio of signal at time slot 3; Pr (x[n]=+ 1) represent that transmit leg sends the prior probability of the data "+1 " after encoding, modulating; Pr (x[n]=-1) expression sends the prior probability of the data " 1 " after coding, modulation;

To x ₁The Maximum Likelihood Detection value of [n]

To x ₂The Maximum Likelihood Detection value of [n]

Images