A novel method for reducing the effect of doppler frequency shift in high-speed railway mobile communication using machine learning

Currently, high-speed railway systems are rapidly expanding worldwide, necessitating reliable and efficient mobile communication solutions. However, the Doppler frequency shift caused by the high speeds of trains presents significant challenges to communication systems, particularly those using OFDM in LTE. This paper presents a novel for Doppler frequency compensation in high-speed railway communication based on the results of estimating the train's velocity using machine learning algorithms. By leveraging advanced algorithm such as neural networks, our method dynamically predicts and compensates for Doppler shifts in real-time. In this proposed novel, the Doppler frequency value is calculated based on the actual train’s velocity and the scenario of a high-speed railway, then the Doppler frequency is used to compensate the carrier frequency offset (CFO) directly at the Access Point (AP) device on the train. We conduct simulations to evaluate the effectiveness of our proposed solution in a high-speed railway scenario. The results demonstrate a marked improvement in communication reliability and data integrity, highlighting the potential of machine learning to enhance the performance of mobile communication systems in high-speed railways. The results of the proposed model are evaluated based on the system’s bit error rate BER after the CFO compensation decreases and the ratio of average energy per bit (Eb) to noise power density (N0) (Eb/N0) increases. This shows that the proposed solution works effectively and reduces the system’s bit errors while improving the communication performance. This study opens new avenues for integrating intelligent systems in transportation networks, ensuring seamless connectivity, and improving passenger experience