METHOD AND DEVICE FOR LEARNING-BASED CHANNEL ESTIMATION WITH DATA PRE-PROCESSING

    公开(公告)号:US20240414030A1

    公开(公告)日:2024-12-12

    申请号:US18465517

    申请日:2023-09-12

    Abstract: Methods and apparatuses are provided in which frequency domain (FD) interpolation is performed on wideband reference signal (RS) resource elements (REs) to generate channel estimates on corresponding RS symbols. Denoising may be performed on the channel estimates in a time domain (TD) or a delay-Doppler domain (DD). Learning-based channel estimation (LCE) is applied on the channel estimates using a neural network (NN) to generate refined channel estimates on the corresponding RS symbols. Time domain (TD) interpolation is performed on the refined channel estimates to produce a channel estimate across symbols.

    SYSTEMS AND METHODS FOR DECODING IN WIRELESS COMMUNICATIONS

    公开(公告)号:US20250150308A1

    公开(公告)日:2025-05-08

    申请号:US18907137

    申请日:2024-10-04

    Abstract: A system and a method are disclosed for decoding in wireless communications. In some embodiments, a method includes receiving a reference signal; generating a channel estimate, based on the reference signal; determining a channel estimation error metric for the channel estimate; receiving a transmission; calculating a log likelihood ratio, based on the channel estimation error metric, for each of a plurality of bit positions of the transmission; and decoding the transmission based on the log likelihood ratio. The calculating of the log likelihood ratio may include calculating a corrected log likelihood ratio based at least on an uncorrected log likelihood ratio.

Patent Agency Ranking