An end-to-end communication system with variable rate mechanism

Abstract

The present invention belongs to the field of communication technology, and specifically relates to an end-to-end communication system with a variable rate mechanism. The present invention expands the transmission rate modes supported by the existing E2E system and can realize the effective utilization of channel resources by the E2E system: first, through reasonable network structure division and expansion, the E2E system can operate in multiple rate modes; second, by adopting a variable rate network training mechanism, the E2E system can achieve near-optimal performance in various rate modes through less parameter training; finally, the trained network is combined with the signal-to-noise ratio estimation technology, which can enable the E2E system to evaluate the surrounding environmental conditions and adjust the rate, thereby realizing the rate adaptation of the network to the environmental conditions, thereby maintaining a certain BER performance while communicating with the highest possible efficiency.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to an end-to-end communication system with a variable rate mechanism. Background Art

[0002] In traditional communication system design, the entire communication process is broken down into multiple independent functional modules, each performing a specific function, such as source / channel coding, modulation, channel estimation, and equalization. While this approach can produce a communication system that meets application requirements, it cannot guarantee that the individually optimized processing blocks, when combined, will achieve optimal link performance. In contrast, the recently emerged end-to-end (E2E) communication technology uses neural networks to build autoencoders and employs deep learning techniques to jointly optimize the transmitter and receiver. This approach can theoretically achieve globally optimal communication waveforms (T. O'Shea and J. Hoydis, “An introduction to deep learning for the physical layer,” IEEE Trans. Cognit. Commun. Netw., vol. 3, no. 4, pp. 563–575, December 2017).

[0003] End-to-end (E2E) communication is a data-driven communication model based on deep learning technology. Each component of an E2E system is composed of neural networks. Training these neural networks under the supervision of E2E losses allows the system to approach global optimality, thereby resolving the problem of traditional communication systems struggling to achieve optimal performance in scenarios where channel modeling is difficult. In real-world environments, channel conditions (including noise and fading) are constantly changing. When channel conditions are poor, the probability of bit errors is high, the channel capacity is low, and error-free transmission is limited to a relatively low rate. When channel conditions are favorable, the channel capacity is high, allowing error-free transmission of information at a higher rate. Summary of the Invention

[0004] To address the problem that existing E2E systems cannot fully utilize limited channel resources, the present invention proposes an E2E communication system with a variable rate mechanism. To fully utilize limited channel resources, the E2E system needs to support multiple rate modes and be able to adaptively adjust the information transmission rate based on channel conditions. Therefore, based on the structure of the E2E system, the present invention invents a novel E2E communication waveform system that can support multiple transmission rates and achieve near-optimal performance in all rate modes with minimal parameter training. Ultimately, the trained E2E system can fully utilize channel resources in a changing channel.

[0005] The technical solution of the present invention is:

[0006] An end-to-end communication system with a variable rate mechanism, wherein the transmitter and receiver of the system are constructed by neural networks, the transmitter is divided into a relationship extraction part and a transmission rate control part, and the receiver is divided into a receiving rate control part and a relationship interpretation part, respectively. and To represent the mapping they represent; Among them, the transmission rate control part and the receiving rate control part expand the transmission rate mode supported by the network by changing the modulation order. The communication process is a joint action of the four-part network. The transmitter converts the transmission data into a transmission signal suitable for transmission in the channel according to the channel characteristics, and the receiver recovers the data, thereby realizing communication based on the variable rate mechanism; Correspondingly, the network constituting the transmitter and receiver adopts a variable rate training mechanism for training, specifically:

[0007] Initialization settings have a total of T sets of rate transmission schemes, the training group size is b, the number of training rounds is E, the Adam optimizer parameters are α, β1, β2, and the initial weights of the relationship extraction part and the relationship interpretation part are The initial weights of the transmission rate control part and the reception rate control part are

[0008] First, the network is trained based on the first set of rate transmission schemes. For each training round, the network parameters are optimized by the Adam optimizer using the binary cross entropy loss function. To update:

[0009]

[0010] After the first set of schemes is trained, for the remaining rate transmission schemes, each training round is fixed with the rate obtained by the first set of schemes. Part of the weight, using the binary cross entropy loss function, To update:

[0011]

[0012] After training is completed, the bit error rate performance corresponding to different rate transmission schemes under different signal-to-noise ratios is obtained through testing. Then, based on the signal-to-noise ratio estimation technology, the communication rate is switched according to the communication environment of the current communication system to achieve the optimal communication rate selection while meeting the bit error performance.

[0013] The present invention has the following beneficial effects: it expands the transmission rate modes supported by existing E2E systems and enables the E2E system to effectively utilize channel resources: first, through reasonable network structure division and expansion, the E2E system can operate in multiple rate modes; second, by adopting a variable-rate network training mechanism, the E2E system can achieve near-optimal performance in various rate modes with less parameter training; finally, the trained network, combined with signal-to-noise ratio estimation technology, enables the E2E system to assess the surrounding environmental conditions and adjust the rate, achieving rate adaptation of the network to environmental conditions, thereby maintaining a certain BER performance while communicating with the highest possible efficiency. DETAILED DESCRIPTION

[0014] Figure 1 FIG2 is a diagram comparing the E2E network architecture provided by an embodiment of the present invention and a traditional communication system.

[0015] Figure 2 This is a schematic diagram of the principle of implementing a rate-varying mechanism in an E2E network according to an embodiment of the present invention.

[0016] Figure 3 2 is a schematic diagram of BER performance of different waveform schemes for an E2E system that completes variable rate network training according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention is described in detail below with reference to the accompanying drawings and embodiments.

[0019] The end-to-end communication system architecture based on the autoencoder adopted by the present invention and its comparison with the traditional communication system are shown in Figure 1 The transmitter and receiver of the system of the present invention are both constructed by neural networks, using Θ t and Θ r Denote their parameters. First, the launch network f t (·) Transmit data s=[s1,…,s L ],s i ∈{0,1},i∈{1,2,…,L} is converted into a transmission signal suitable for transmission in the channel according to the channel characteristics Where L represents the input data length, N represents the transmission signal length, and L / N≤1 can be regarded as the code rate of the system. The output of the transmitter passes through the channel to obtain the received signal Send it to the demodulation network f r (·) Perform data recovery and finally obtain the recovered data

[0020] Since Convolutional Neural Network (CNN) has excellent feature extraction capabilities, requires few parameters, and can easily change the dimension of input data, the present invention uses convolutional neural networks to design transmitters and receivers. In order to ensure that the physical constraints on the transmitted signal x are met, the transmitter hardware will limit x. Therefore, a normalization layer is added at the end of the transmitter to impose specific constraints on x, limiting its average power to The last layer of the demodulation network is activated with a Sigmoid layer, and the binary cross entropy loss function is used as the loss function of the E2E system. Based on this, stochastic gradient descent and back propagation are performed to optimize the entire network.

[0021] The present invention Figure 1 Based on the E2E architecture shown in the figure, the transmitting network and the demodulation network are divided into four parts: the symbol relationship extraction part (transmitting network), the rate control part (transmitting network), the rate control part (demodulation network) and the symbol relationship interpretation part (demodulation network), respectively. and To represent the mapping they represent, this partitioning relationship is shown in Figure 2 Among them, the rate control part mainly expands the transmission rate mode supported by the network by changing the modulation order (MO). In the convolutional neural network, this is achieved by changing the length of the output signal by setting different step sizes.

[0022] Figure 2 The upper middle part represents the transmitter, and the lower middle part represents the receiver, and shows three cases where MO is 2, 4, and 16. Based on this segmentation relationship, the present invention further proposes a training mechanism for a variable rate network. As shown in the figure, during training, the system first performs integrated training in the first waveform scheme, and obtains four trained parts. In the subsequent training of the waveform scheme, the trained inter-symbol relationship extraction and interpretation parts are fixed, and only the network of the rate control part is updated. Therefore, under different MOs, that is, different transmission rates, the inter-symbol relationship extraction part of the transmitter and the inter-symbol relationship interpretation part of the receiver are fixed, and their respective rate control parts are equivalent to a replaceable "control head". There are as many sets of rate control parts as there are rate modes that can be replaced when changing the system rate.

[0023] The process of the variable rate network training mechanism is as follows:

[0024] 1. Initialize the T-set rate transmission scheme, training group size b, number of training rounds E, Adam optimizer parameters α, β1, β2, and initial weights of the inter-symbol relationship extraction part (transmitter) and the inter-symbol relationship interpretation part (receiver) Initial weight of the rate control part of the T-suite scheme

[0025] 2. For the first set of rate transmission schemes, for each training round, the network parameters are optimized by Adam optimizer using binary cross entropy loss function. To update:

[0026]

[0027] 3. After the first set of schemes are trained, for the remaining rate transmission schemes, each training round is fixed with the rate obtained by the first set of schemes. Part of the weight, using the binary cross entropy loss function, To update:

[0028]

[0029] After the network is trained according to this variable rate training mechanism, the bit error rate performance corresponding to different rate transmission schemes under different signal-to-noise ratios can be tested. Based on the bit error rate performance obtained by the network, combined with the existing signal-to-noise ratio estimation technology, it is possible to evaluate the quality of the current environmental conditions, thereby switching to network parameters suitable for the current environment and achieving rate adaptation of the E2E system. Specifically, if Figure 2 Taking the three waveform systems in Figure 1 as an example, labeled Waveforms 1, 2, and 3 from left to right, and assuming that the entire link has an allowable bit error rate (BER), the entire rate adaptation process can proceed as follows: the system begins operating with Waveform 2. When the estimated signal-to-noise ratio indicates that the current network performance is better than the allowable BER, the system can adjust to Waveform 3 and continue operating. When the estimated signal-to-noise ratio indicates that the network performance has deteriorated below the allowable BER, it is necessary to downgrade to Waveform 2. If Waveform 2 still cannot meet the performance requirements, it can be further downgraded to Waveform 1. If Waveform 1 also fails to meet the requirements, the entire system needs to be retrained based on the current channel environment.

[0030] In the present invention, the communication model of the E2E system is:

[0031]

[0032] Where h is the channel coefficient vector, n is the channel noise vector, represents the Hadamard product, i.e., the element-by-element multiplication between vectors. In theory, the training mechanism proposed in this invention is equivalent to solving the following optimization problem in advance, i.e., the integrated training under the waveform scheme:

[0033]

[0034] The constraints in the first row represent the information transmission and recovery process, and the constraints in the second row represent the channel and noise distribution preset during training. And the signal power P0. Get the optimal solution Then solve the following optimization problems in sequence, that is, separate training of the rate control part under other waveform schemes:

[0035]

[0036] Finally, all the optimal solutions are obtained After the training is completed, the receiver can control the transmission rate of the system according to the channel conditions. If the error performance is the top priority and the transmission rate is increased as much as possible while ensuring the system performance, the system can first apply the Tth set of solutions; when the error performance cannot meet the requirements, it will start to downgrade to the T-1th set; if it still does not meet the requirements, it will continue to downgrade. If the system is in a completely good working state, you can try to upshift to increase the transmission rate. In theory, when the transmitter and receiver can be strictly divided into four parts, that is, when the rate control part does not have the function of extracting the relationship between symbols, It is the optimal solution, which is not completely true in practice. However, the present invention has shown through experiments that The system performance obtained when is the network weight is close to global training:

[0037]

[0038] The performance obtained, the simulation results are shown in Figure 3 , the experiment is carried out under AWGN channel, and the E b / N0 are both 10dB. As can be seen from the figure, the results of non-global training are almost the same as those of global training. It is at least a suboptimal solution or even close to the optimal solution, that is, the four-part division of the E2E system is feasible. In addition, it can be found that the system trained non-globally performs even better than the global training at low signal-to-noise ratio (i.e., non-training area, that is, distribution outside the distribution included in the training set). This means that the adaptation result of the E2E system for tasks outside the training distribution during global training may not be as good as the result obtained by partial initialization using a pre-trained network. This result proves the feasibility of the E2E communication system with a variable rate mechanism proposed in the present invention and the advantages of the variable rate network training mechanism adopted.

[0039] Example

[0040] The parameters of the transmitting network and the demodulating network in this example are shown in Table 1. The rate control part consists of the last convolutional layer of the transmitting network and the first convolutional layer of the demodulating network. Table 1 shows the parameters when MO is 2.

[0041] Table 1 Transmission and demodulation network parameters of the embodiment

[0042]

[0043] The specific structure of the rate control part under different modulation orders is shown in Table 2.

[0044] Table 2 Neural network structures of three rate control parts

[0045]

[0046] The specific implementation steps of this example are as follows:

[0047] Step 1: Build an E2E communication system according to Table 1 and Table 2.

[0048] Step 2: For the E2E system, perform variable rate network training to enable the network to work in multiple rate modes. The training method process is as follows:

[0049] 1. Initialize the T-set rate transmission scheme, training group size b, number of training rounds E, Adam optimizer parameters α, β1, β2, and initial weights of the inter-symbol relationship extraction part (transmitter network) and the inter-symbol relationship interpretation part (demodulation network) Initial weight of the rate control part of the T-suite scheme

[0050] 2. For the first set of rate transmission schemes, for each training round, a set of 0-1 information bits s∈{0,1} is randomly generated b×1×L As the transmission data, the transmission data s is transmitted through the transmission network to obtain the transmission signal The two parts of the second dimension represent the real part and imaginary part of the baseband signal respectively. Let the transmitted signal x pass through the channel to obtain the received signal y, and then let the received signal y pass through the demodulation network to obtain the recovered information data Use binary cross entropy loss function to calculate s and The E2E loss between the two networks is calculated and the network parameters are optimized by Adam optimizer. To update:

[0051]

[0052] 3. After the first set of schemes are trained, for the remaining rate transmission schemes, each training round is fixed with the rate obtained by the first set of schemes. Repeat step 2 for the weight of part, and use the binary cross entropy loss function to To update:

[0053] Step 3: Use the trained network to test and record the bit error rate performance of each rate scheme under different signal-to-noise ratios. This will serve as the basis for rate adjustment in actual network applications.

[0054] Step 4: Deploy the extensively trained transmit and demodulation networks to real-world scenarios. Combine the networks with existing signal-to-noise ratio estimation techniques to achieve rate adaptation based on environmental conditions. The specific steps are as follows:

[0055] 1. First, apply the T-th set of rate transmission schemes and estimate the signal-to-noise ratio in the current environment. Based on the estimation results and the results of network testing, evaluate the bit error rate performance of the current system.

[0056] 2. If the bit error rate performance at this time cannot meet the system requirements, start downgrading to the T-1 solution and repeat step 1. If it still does not meet the requirements, continue downgrading.

[0057] 3. If the bit error rate performance at this time is evaluated to be better than the allowed bit error rate, it is considered that the system is in a good working state, and an attempt is made to upgrade to increase the transmission rate.

[0058] The data to be transmitted is divided into frames according to a predetermined frame length and handed over to the transmitting network. The transmitted signal passes through the channel and is received by the demodulation network. The signal-to-noise ratio estimation technology is used to monitor the BER performance of the system. Based on the estimated performance, the system can change the waveform scheme in time to enable the system to communicate with the highest possible efficiency while maintaining the BER level, thereby achieving effective utilization of channel resources.

[0059] In this embodiment, the BER performance of the system is observed under an AWGN channel to verify the performance of the proposed E2E network with a variable rate mechanism. The length L of the input data is 128. Figure 3 The BER performance test comparison under different transmission rate schemes is described after the E2E system in this embodiment completes training using the proposed variable rate network training mechanism. b / N0 are both 10dB. As can be seen from the figure, the results of dividing the network into four parts and performing non-global training are almost the same as the results of global training of the entire network (baseline). At least it is a suboptimal solution or even close to the optimal solution. The four-part segmentation of the E2E system by the present invention is feasible. In addition, the performance of the system trained by non-global training is not good at low signal-to-noise ratio (i.e., non-training E b / N0 region) even outperforms the performance of global training. This means that the adaptation results of the E2E system for tasks outside the training distribution during global training may not be as good as those achieved by partial initialization using a pre-trained network. This further demonstrates the advantages of the variable-rate E2E communication system proposed in this invention.

Claims

1. An end-to-end communication system with a variable rate mechanism, characterized in that: The transmitter and receiver of the system are constructed by neural networks. The transmitter is divided into a relationship extraction part and a transmission rate control part, and the receiver is divided into a receiving rate control part and a relationship interpretation part. and To represent the mapping they represent; Among them, the transmission rate control part and the receiving rate control part expand the transmission rate mode supported by the network by changing the modulation order. The communication process is a joint action of the four-part network. The transmitter converts the transmission data into a transmission signal suitable for transmission in the channel according to the channel characteristics, and the receiver recovers the data, thereby realizing communication based on the variable rate mechanism; Correspondingly, the network constituting the transmitter and receiver adopts a variable rate training mechanism for training, specifically: Initialization settings have a total of T sets of rate transmission schemes, the training group size is b, the number of training rounds is E, the Adam optimizer parameters are α, β1, β2, and the initial weights of the relationship extraction part and the relationship interpretation part are The initial weights of the transmission rate control part and the reception rate control part are First, the network is trained based on the first set of rate transmission schemes. For each training round, the network parameters are optimized by the Adam optimizer using the binary cross entropy loss function. To update: After the first set of schemes is trained, for the remaining rate transmission schemes, each training round is fixed with the rate obtained by the first set of schemes. Part of the weight, using the binary cross entropy loss function, To update: After training is completed, the bit error rate performance corresponding to different rate transmission schemes under different signal-to-noise ratios is obtained through testing. Then, based on the signal-to-noise ratio estimation technology, the communication rate is switched according to the communication environment of the current communication system to achieve the optimal communication rate selection while meeting the bit error performance.

Images