A Fast Beam Alignment Scheme for Dual-Band HSR Wireless Networks

To satisfy the ever increasing capacity requirements resulting from the rapid developments of the railway industry, future 5G based high-speed railway (HSR) wireless networks begin to consider using millimeter wave (mmWave) bands to expand the frequency spectrum together with the beamforming technology to guarantee the radiation range. Nevertheless, due to the high time consumption of beam alignments and the high mobility of trains, significant angle offsets will be induced during the initial access (IA) processes, which may completely invalidate the final determined beam pair. To solve this problem, in this paper we take advantage of the periodicity and regularity of trains' trajectory and propose a fast IA scheme, in which the beam training set is reduced by learning from historical beam training results and further adjusted during the IA process to compensate the angle offsets due to movements of trains. To guarantee the whole network robustness, in this study a dual-band HSR network by integrating both legacy sub-6 GHz bands and mmWave bands is considered, under which the standby omnidirectional sub-6 GHz link can help facilitate the IA process of the mmWave communication part. Theoretical and numerical results have shown that the proposed scheme can enhance the IA success probability and reduce the IA time consumption.