Abstract
For highly dynamic communication scenarios, channel tracking can reduce the pilot overhead for channel re-estimations in massive multiple-input multiple-output (MIMO) systems. In this paper, we propose an angle domain channel tracking method based on deep learning (DL) by combining the temporal correlation features with inter-frame correlation features of millimeter wave (mmWave) systems in Vehicle-to-Everything (V2X) communications. To characterize the changes in fast time-varying mmWave channel more closely, our angle variation model concerns the influence between multiple frames and the random disturbance referring to the action recognition of each frame in video. Based on this, we construct a novel framework for channel tracking called fused channel prediction network (FCPN) based on the data stream of azimuth angle of arrival (AAoAs) and the differences between adjacent frames, which is composed of two recurrent neural network (RNN)-based predictors with the same input size. Furthermore, we design a fusion strategy for the results of the FCPN, where the scale factors are proportional to the performance of the predictors during the previous frames. Numerical results are presented to verify our studies.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
