In order to address the issue of spectrum resource scarcity in 5G millimeter-wave Train-to-Train (T2T) communication systems for urban rail transit, the spectral efficiency of T2T communications is investigated. Firstly, a system model of Non-Orthogonal Multiple Access (NOMA) with downlink multi-user clusters is constructed. This model ensures successful power allocation among antennas within each cluster through Successive Interference Cancellation (SIC) while also achieving the design of total spectral efficiency in the NOMA system. Secondly, a power allocation scheme for antennas within clusters is proposed to achieve the maximum power allocation without setting rate constraints. Lastly, a Convolutional Neural Network and Long Short-Term Memory Neural Network-Assisted NOMA communication method (CNN-LSTM-NOMA) is employed to train the input data and obtain the optimal spectral efficiency. The results demonstrate that, under the same experimental dataset, the coefficient of determination for the proposed method reaches 0.995 71, which is higher than that of the CNN-NOMA and LSTM-NOMA methods assisted by a single neural network, and the determination coefficients of these two methods are 0.966 54 and 0.979 96, respectively. Moreover, the spectral efficiency of the CNN-LSTM-NOMA method is closer to the optimal unconstrained digital precoding. The research provides a theoretical basis for improving the spectral efficiency of T2T communications in future urban rail transit systems.
式中:为Tt和Tr间的NOMA信道矩阵;为与距离相关的NOMA信道矩阵;为与距离和载波频率相关的路径损耗常数;为载波频率,;为路径损耗常数,;A为额外的路径损耗常数,A=2 dB · km-1;为两车追踪距离,设初始追踪距离为;为离散传输信号向量;为高斯噪声,其取值服从均值为0,方差为的高斯分布,即,为单位矩阵。
采用传统NOMA求解目标函数时,将所有用户均固定在不同位置,并采用理想的信道状态信息(Channel State Information,CSI),使得信道增益较强的用户通过SIC消除信道增益较弱用户的干扰,再解码自身信号。因此,为了求解第个簇内最弱用户接收到的信号值,必须先求解前个用户的信号,这一过程需要进行次迭代。由于,为了求得所有簇内所有用户接收到的信号,则需要进行(Nc1-1) · (Nray-1)次计算,会产生极高的计算复杂度。此外,在列车实际追踪运行中,设每个用户固定位置且采用理想CSI并不符合实际情况,这必然会导致理想输出与实际输出之间的差异较大,从而降低频谱效率。然而,将深度学习方法引入无线通信领域,可以将信道估计、功率分配和信道编解码等传统SIC过程整合为单一模块,并能够确定特征与标签之间的关联规则。通过使用理想CSI建立模型,并利用非理想CSI进行迭代调优,尽可能减小理想输出与实际输出之间的差异,进而提高频谱效率。基于以上分析,提出1种CNN-LSTM-NOMA的频谱效率优化方案。
WANGZhuoran, JIAXuexiang. Development Direction of Urban Rail Transit Signal System in China [J]. TranspoWorld, 2019 (12): 158-159. in Chinese
[3]
FAROOQJ, SOLERJ. Radio Communication for Communications-Based Train Control (CBTC): a Tutorial and Survey [J]. IEEE Communications Surveys & Tutorials, 2017, 19 (3): 1377-1402.
[4]
3GPP. System Architecture for the 5G System: TS 23.501 V16.3.0 [S]. 2019: 18-90.
[5]
AIB, GUANK, RUPPM, et al. Future Railway Services-Oriented Mobile Communications Network [J]. IEEE Communications Magazine, 2015, 53 (10): 78-85.
[6]
艾渤,钟章队,沈超,等.轨道交通5G关键技术[M].北京:科学出版社, 2020.
[7]
AIBo, ZHONGZhangdui, SHENChao, et al. Key Technologies of Rail Transit 5G [M]. Beijing: Science Press, 2020. in Chinese
CHENYong, ZHANZhixian, ZHANGWei. Time Synchronization Method for Train-to-Train Communication Based on Model Predictive Control [J]. China Railway Science, 2022, 43 (5): 111-122. in Chinese
GAOYunbo, MARui, LICuiran, et al. Research on Train-to-Train Direct Communication Millimeter-Wave Horn Antenna Based on Improved Genetic Algorithm [J]. China Railway Science, 2023, 44 (3): 213-220. in Chinese
WANGFang, WUYu, LIXue, et al. Analysis of Engineering Applications of Different MlMO Base Stations in Railway 5G-R System [J]. Railway Signalling & Communication, 2024, 60 (7): 1-7. in Chinese
LINWei, WANGFang, LIHui. Research on the Transmission Power Parameter of 5G-R Mobile Terminal [J]. China Railway Science, 2023, 44 (6): 153-161. in Chinese
[18]
ZHAOJ H, LIUJ, NIS J, et al. Enhancing Transmission on Hybrid Precoding Based Train-to-Train Communication [J]. Mobile Networks and Applications, 2020, 25 (5): 2082-2091.
YANGZongye, WUWei, DONGYu. Research on Hybrid Precoding for T2T Communication Based on mmWave MIMO [J]. Journal of Railway Science and Engineering, 2023, 20 (12): 4484-4492. in Chinese
[21]
SONGH F, SCHNIEDERE. Availability and Performance Analysis of Train-to-Train Data Communication System [J]. IEEE Transactions on Intelligent Transportation Systems, 2019, 20 (7): 2786-2795.
WANGQiancheng, ZHAOHengkai, HANYi, et al. Resource Allocation of Rail Transit T2T Communication Network Based on Deep Reinforcement Learning [J]. Industrial Control Computer, 2023, 36 (9): 89-91, 94. in Chinese
LIQiang, CHENZijian. Research on Backup Mode of Train Control System Based on Train-to-Train Communication [J]. Journal of Railway Science and Engineering, 2022, 19 (11): 3407-3416. in Chinese
[26]
ZHAOJ H, ZHANGY, NIEY W, et al. Intelligent Resource Allocation for Train-to-Train Communication: a Multi-Agent Deep Reinforcement Learning Approach [J]. IEEE Access, 2020, 8: 8032-8040.
[27]
GEX H, WUQ, HANQ L, et al. Resilient Virtual Coupling Control of Automatic Train Convoys with Intermittent Communications [J]. IEEE Transactions on Vehicular Technology, 2024, 73 (5): 6183-6195.
ZHANGYinggui, ZHAOMinghui, ZHANGYunli. Simulation Research on Train Group Tracking Operation with Virtual Coupling for Urban Mass Rail Transit [J]. Journal of Transportation Systems Engineering and Information Technology, 2024, 24 (1): 199-209. in Chinese
[30]
LINC, CHANGQ, LIX X. A Deep Learning Approach for MIMO-NOMA Downlink Signal Detection [J]. Sensors, 2019, 19 (11): 2526.
[31]
DUBEYS R, CHAKRABORTYS. Average Biased ReLU Based CNN Descriptor for Improved Face Retrieval [J]. Multimedia Tools and Applications, 2021, 80 (15): 23181-23206.
[32]
DENGW, LIUH L, XUJ J, et al. An Improved Quantum-Inspired Differential Evolution Algorithm for Deep Belief Network [J]. IEEE Transactions on Instrumentation and Measurement, 2020, 69 (10): 7319-7327.