Geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method and system

Abstract

The present invention discloses a geographically-based, fully separated network-uplink transmission method and system for 6G high-speed rail mobile communications. The method includes the steps of: performing pure line-of-sight channel mapping based on the geographical locations of the train and each uplink base station, and using the mapping as channel estimation to perform joint receiver processing for secondary merging of uplink signals; designing a deep learning neural network to input the mapped pure line-of-sight channel and output multi-user joint precoding that satisfies constraints; offline training the network using historical channel data for each location; and inputting the pure line-of-sight channel corresponding to the train location during the actual deployment phase and outputting joint precoding at the corresponding location through the network, thereby realizing feedback-free transmission. Compared with traditional precoding design methods, the present invention effectively solves the problems of untimely channel feedback and large resource overhead caused by high-speed movement in high-speed rail scenarios, improves the spectral efficiency of wireless communications in high-speed rail scenarios, and provides an effective uplink transmission method for future high-speed rail mobile communications.

Description

[Technical Field]

[0001] The present invention belongs to the field of wireless mobile communications and relates to multi-user multi-input multi-output transmission in deep learning and high-speed rail scenarios, and more specifically to a geographically-based high-speed rail mobile communication 6G fully isolated network uplink transmission method and system. [Background technology]

[0002] With the rapid development of railway systems, high-speed train communication is attracting widespread attention. Fifth-generation mobile networks (5G) achieve significant performance improvements through large-scale multiple-input multiple-output (mMIMO) joint precoding designs, maximizing diversity and multipath gain. However, at high mobility, channels degrade in various aspects such as Doppler frequency offset (DFO), reduced coherence time, and phase noise, significantly reducing the performance of high-speed railway communication systems. Furthermore, because 5G uses higher frequency bands, handovers between base stations become more frequent in high-speed mobile scenarios.

[0003] Conventional solutions for Long-Term Evolutionary Technology (LTE-R) in rail networks plan to use more frequent pilot transmissions and channel status information (CSI) feedback. This leads to the consumption of large amounts of time-frequency resources, and furthermore, feedback delays become more severe in high-speed mobile scenarios, making it difficult to obtain accurate CSI even when frequent pilots are used for channel estimation.

[0004] Sixth-generation (6G) mobile communication networks are expected to offer more flexible resource allocation and improve the user experience. In "A Fully-Decoupled RAN Architecture for 6G Inspired by Neurotransmission," Academician Yu Quan and colleagues proposed a 6G fully decoupled radio access network (FD-RAN) architecture that separates conventional base stations into up-uplink base stations (UBS), down-uplink base stations, and control base stations. FD-RAN has significant advantages in high-speed rail communication systems. On the one hand, since all control plane signals are transmitted by the control base station, the amount of control information data is small, allowing the control base station to cover a wider area using lower frequency bands, which can significantly reduce handover overhead. On the other hand, FD-RAN employs a geographic location-based feedback-free transmission solution. This solution not only mitigates the negative effects of channel changes and resource overhead, but also avoids feedback delays through location prediction, and since the train's track is fixed, the train's position can be easily predicted in advance. However, determining appropriate precoding from location information alone remains difficult. Furthermore, considering the multiple mobile relays (MRs) installed on the train, a joint precoding design is required because a multi-user MIMO (MU-MIMO) scenario needs to be utilized. Additionally, to conserve more pilot resources, it is difficult to use the real-time channel estimated by the pilot for signal reception at the base station.

[0005] As described above, the problems with conventional technologies are as follows: (1) Conventional pilot-based channel estimation incurs significant resource overhead in high-speed rail scenarios, and feedback delays make it difficult for the transmitter to obtain real-time channel information. (2) In fully isolated 6G networks, large feedback delays caused by hardware isolation and isolation architectures significantly degrade the performance of transmission methods that rely on existing feedback mechanisms. [Overview of the Initiative] [Problems that the invention aims to solve]

[0006] Objective of the Invention: The objective of the present invention is to provide a geolocation-based high-speed rail mobile communication 6G fully isolated network uplink transmission method and system that, instead of the conventional channel estimation and channel feedback process, can directly select a precoding based on the geolocation of a train by fitting the mapping relationship between the geolocation of the train and the precoding using deep learning. [Means for solving the problem]

[0007] To achieve the above-mentioned objectives of the invention, the technical solutions employed in this invention are as follows: A geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method is Step 1 involves performing pure line-of-sight channel mapping based on the geographical location of the train and each uplink base station, and using this as channel estimation to perform joint reception processing for secondary merging of uplink signals. Step 2 involves designing a deep neural network that takes pure line-of-sight channel information mapped between a train and an uplink base station as input and outputs multi-user joint precoding information satisfying certain modulus constraints, designing the loss function of the deep neural network according to the system optimization goal, and performing data transmission using the precoding output by the deep neural network. Step 3 involves training a deep neural network offline using the historical channel data of the sample at each location, and updating the parameters until convergence occurs. The method includes step 4, which involves inputting a line-of-sight channel corresponding to the train position during the actual deployment phase, outputting joint precoding at the corresponding position using a deep neural network, thereby achieving feedback-free transmission.

[0008] Furthermore, in this scenario, we assume there are L base stations providing service to K train mobile relays, and the actual channel between the mobile relay on the kth train and the lth uplink base station is H kl Therefore, the pure line-of-sight channel mapping relationships are as follows:

[0009]

number

number

[0010] Furthermore, the pre-coding vector of the k-th relay on the train roof is

Number

Number

[0011] Furthermore, the step of performing joint reception processing for the secondary merge of the uplink signal includes the following.

[0012] The base station first performs primary dispersion processing on the received signal, and the merge vector of the k-th relay by the l-th base station is

Number

Number

Number

Number

[0014] Furthermore, it corresponds to the maximum signal-to-interference noise ratio.

number

number

number

[0015] Furthermore, by substituting the values ​​and performing calculations, the maximum spectral efficiency that the k-th relay can achieve is as follows:

number

[0016] Here, τ p and τ c |·| represents the pilot time and coherence time, respectively, and |·| represents the determinant of the matrix.

[0017] Furthermore, all user precoding vectors are P=[P1,···,P K If we represent this as ] and the system optimization goal is to maximize the sum of relay spectral efficiencies, the system optimization problem can be modeled as follows:

[0018]

number

number

[0019] Furthermore, the designed deep learning neural network takes all the channels between relays and base stations as inputs, and outputs the neural network's output α before outputting. k teeth,

number

[0020] In practical implementation, the deep learning neural network may be a fully connected deep learning neural network or a residual network.

[0021] The present invention also provides a computer system comprising memory, a processor, and a computer program stored in the memory and operable on the processor, wherein the computer program is loaded onto the processor to implement steps of a geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method. [Effects of the Invention]

[0022] The beneficial effects are as follows: Compared to conventional technology, the present invention has the following advantages: 1. The present invention provides a high-speed rail uplink communication system under an FD-RAN architecture that avoids problems related to feedback delay and resource overhead in high-speed rail scenarios. 2. The present invention enables multi-user multi-input multi-output precoding selection based on the geographic location of trains, saving pilot estimation and feedback overhead and further improving system spectral efficiency. 3. The present invention uses deep learning to adapt the mapping relationship between the geographic location information of trains and the precoding, and effectively mitigates the low-latitude problem of geographic location information by using pure line-of-sight channel mapping before input. [Brief explanation of the drawing]

[0023] [Figure 1] This is a schematic diagram of a scenario in which an embodiment of the present invention is applied. [Figure 2] This is a schematic diagram of the neural network parameter update process according to an embodiment of the present invention. [Figure 3] This is a schematic diagram of the training convergence process of a neural network according to an embodiment of the present invention. [Figure 4] This is a comparative curve diagram of spectral efficiency under various architectures according to embodiments of the present invention. [Modes for carrying out the invention]

[0024] To further clarify the object, technical solution, and advantages of the present invention, embodiments of the present invention will be described in detail below with reference to the drawings. These embodiments are carried out based on the technical solution of the present invention and provide detailed embodiments and specific operating procedures. It should be understood that the specific examples described herein are used solely for the purpose of illustrating the present invention, and the scope of protection of the present invention is not limited to the following embodiments.

[0025] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method disclosed in the embodiments of the present invention mainly comprises the following steps:

[0026] Step 1: Perform pure line-of-sight channel mapping based on the geographical location of the train and each uplink base station, and use this as channel estimation to perform joint receive processing for secondary merging of uplink signals.

[0027] Step 2: Design a deep neural network that takes pure line-of-sight channel information mapped between the train and the uplink base station as input and outputs multi-user joint precoding information that satisfies certain modulus constraints. Design the loss function of the deep neural network according to the system optimization goal, and perform data transmission using the precoding output by the deep neural network.

[0028] Step 3: Use the historical channel data of the samples at each location to train the deep neural network offline and update the parameters until convergence occurs.

[0029] Step 4: In the actual deployment phase, a pure line-of-sight channel corresponding to the train position is input, and a deep neural network outputs joint precoding at the corresponding position, thereby achieving feedback-free transmission.

[0030] The specific steps of an embodiment of the present invention will be described below with reference to the high-speed rail communication scenario shown in Figure 1. Assume that the uplink base station and the mobile relay are located on the same plane, and that there are L base stations providing service to K train mobile relays, and that the actual channel between the mobile relay on top of the k-th train and the l-th uplink base station is H kl Therefore, the pure line-of-sight channel mapping relationships are as follows:

[0031]

number

number

[0032] The specific process for secondary joint reception of signals between the base station and the edge cloud is shown below.

[0033] The precoding vector for the k-th relay is:

number

number

number

number

number

[0034] The base station first performs primary distributed processing on the received signal and then calculates the merge vector of the kth relay by the l-th base station.

number

number

number

number

number

number

number

[0035] The edge cloud centrally performs secondary merging of signals that have been primary processed at base stations, and sets the secondary merging coefficient to a kl In this representation, the signal of the k-th relay after merging is expressed as follows:

[0036]

number

number

number

number

number

number

[0037]

number

[0038] All user precoding vectors are P=[P1,···,P K Represented by ], and considering the maximization of the sum of relay spectral efficiencies, the optimization problem of maximizing the sum of system spectral efficiencies may also be modeled as follows:

[0039]

number

[0040] The following describes the deep learning neural network configuration used. In the embodiments of the present invention, the deep neural network may be a fully connected deep learning neural network to prevent the overhead of supervised learning labeling, or a residual network (ResNet) may be used to improve the convergence speed of training the deep neural network. All channels between relays and base stations are taken as inputs, and as an example, after being converted into vectors by a flattening layer, the number of neurons that pass through sequentially is 1024,512,256 and K*N. V Furthermore, batch normalization is performed for each layer to prevent overfitting. A Lambda processing layer is designed before output, and the output of the neural network is processed.

number

number

[0041] The network training process is shown in Figure 2. This process converges to minimize the loss function, finding the optimal solution to the optimization problem—that is, the precoding that maximizes the sum of the system's spectral efficiency—and thereby enabling the design of a channel-information feedback-free, geographically-based precoding. The convergence curve during network training is shown in Figure 3. This shows that the network can complete convergence within 20 iterations, indicating that it requires few computing resources during the training process.

[0042] To make this embodiment more intuitive and to compare the advantages and disadvantages of the FD-RAN architecture with other architectures, Figure 4 shows a schematic comparison of simulation results of the spectral efficiency of high-speed rail relays under different architectures. The results show that small cell performance is inferior to the FD-RAN structure because large inter-relay interference cannot be eliminated in individual signal reception. Compared to cellular networks, higher spectral efficiency can be achieved when the train is close to a cellular base station. However, spectral efficiency is lower at the cellular end. This demonstrates that the FD-RAN architecture in this embodiment can provide stable and high spectral efficiency in high-speed rail communications. A computer system disclosed in one embodiment of the present invention includes memory, a processor, and a computer program stored in memory and operable on the processor, wherein the computer program is loaded onto the processor to implement the steps of the high-speed rail mobile communication 6G fully isolated network uplink transmission method based on geographic location information relating to the steps described above.

[0043] The foregoing describes only preferred embodiments of the present invention and does not limit it. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are also covered by the present invention.

Claims

1. A geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method executed on a computer system, Step 1 involves performing pure line-of-sight channel mapping based on the geographical location of the train and each uplink base station, and using this as channel estimation to perform joint reception processing for secondary merging of uplink signals. Step 2 involves designing a deep neural network that takes pure line-of-sight channel information mapped between a train and an uplink base station as input and outputs multi-user joint precoding information satisfying certain modulus constraints, designing the loss function of the deep neural network according to the system optimization goal of maximizing the sum of relay spectral efficiencies, and performing data transmission using the precoding output by the deep neural network. Step 3 involves training a deep neural network offline using the historical channel data of the sample at each location, and updating the parameters until convergence occurs. A geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method, characterized by comprising step 4, inputting a pure line-of-sight channel corresponding to the train position during the actual deployment stage, outputting joint precoding at the corresponding position using a deep neural network, thereby achieving feedback-free transmission.

2. The pure line-of-sight channel mapping relationship between the mobile relay on top of the k-th train and the l-th uplink base station is as follows: [Number 41] Here, [Number 42] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 1, characterized in that θkl represents the phase offset between antennas, θkl represents the angle between the mobile relay on top of the k-th train and the l-th uplink base station, f represents the carrier frequency, λ represents the carrier wavelength, v represents the train's speed, c represents the speed of light, dB and dV represent the antenna spacing between the base station and the relay, respectively, NB and NV represent the number of antennas of the base station and the relay, respectively, and βkl is the fading coefficient.

3. The precoding vector for the k-th relay at the top of the train is: [Number 43] and power constraints [Number 44] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 1, wherein the following conditions are met, where NV represents the number of relay antennas and pmax is the maximum power limit.

4. The step of performing joint reception processing for secondary merging of uplink signals is: The base station first performs primary distributed processing on the received signal and then calculates the merge vector of the kth relay by the l-th base station. [Number 45] Let yl represent the received signal at the l-th base station, and estimate the signal of the k-th relay. [Number 46] Represented as, pure line-of-sight channel mapping. [Number 47] Based on this, by the local least mean squares error merging method [Number 48] The steps to calculate, The edge cloud centrally performs secondary merging of signals that have been primarily processed at the base station, and then processes the signal of the kth relay after merging. [Number 49] A geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 1, comprising the step of representing all quadratic merge coefficients as Ak[akl, ..., akL]T and the number of base stations as L.

5. Corresponding to the maximum signal-to-interference noise ratio [Number 50] Calculate, here, [Number 51] And Hnl represents the actual channel between the mobile relay on top of the nth train and the lth uplink base station. [Number 52] Here, Pn represents the precoding vector of the nth relay, Pk represents the precoding vector of the kth relay, Nk represents the interference term between relays, and σ² represents the noise power. [Number 53] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 4, wherein Hkl represents the actual channel between the mobile relay on top of the k-th train and the l-th uplink base station, and K represents the number of relays in the train.

6. The maximum spectral efficiency that the k-th relay can achieve is as follows: [Number 54] Here, τp and τc represent the pilot and coherence time lengths, respectively, and |・| represents the determinant of the matrix. [Number 55] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 5.

7. If all user precoding vectors are represented as P = [P1, ..., PK], and the system optimization goal is to maximize the sum of relay spectral efficiencies, then the system optimization problem can be modeled as follows: [Number 56] Here, SEk represents the maximum spectral efficiency that the k-th relay can achieve, NV represents the number of relay antennas, K represents the number of relays in the train, and the loss function of the deep neural network is defined as the negative number of the sum of the system spectral efficiencies. [Number 57] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 1, wherein N represents the number of samples.

8. The designed deep neural network takes a pure line-of-sight channel between all relays and base stations as input, and the output αk of the neural network before output is, [Number 58] The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 1, characterized in that it is converted into a pre-coded vector that satisfies certain modulus constraints.

9. The geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method according to claim 1, characterized in that the deep neural network employs a fully connected deep neural network or a residual network.

10. A computer system comprising memory, a processor, and computer programs stored in memory and operable on the processor, The computer system is characterized in that the computer program is loaded into a processor and controls network equipment comprising an uplink base station and an edge cloud to implement the steps of the geographic location-based high-speed rail mobile communication 6G fully isolated network uplink transmission method described in any one of claims 1 to 9.

Images