Machine-learning-based detection and reconstruction from low-resolution samples

    公开(公告)号:US11139866B2

    公开(公告)日:2021-10-05

    申请号:US17135748

    申请日:2020-12-28

    Abstract: According to an aspect, there is provided an apparatus comprising a combiner for combining a received analog signal with an analog dithering signal to produce a combined analog signal, a one-bit analog-to-digital converter for converting the combined analog signal to a combined digital signal, means for performing joint downsampling and feature extraction for the combined digital signal, means for implementing a trained machine-learning algorithm for calculating one or more input parameters for waveform generation at least based on one or more features extracted from the combined digital signal and a parametric waveform generator for generating the analog dithering signal based on the one or more input parameters.

    MODIFYING A NUMBER OF UPLINK OR DOWNLINK INFORMATION STREAMS

    公开(公告)号:US20200344746A1

    公开(公告)日:2020-10-29

    申请号:US16757011

    申请日:2017-10-20

    Abstract: A method, in abase station of a wireless telecommunications system comprising multiple radiating elements, the method comprising modifying a number of uplink or downlink information streams, each information stream corresponding to an element in a space sampling basis for the system, wherein, in the downlink, the space sampling basis is configured so that the number of information streams is the same as the number of radiating elements of the system, and in the uplink, the space sampling basis is configured using a composite antenna derived 205f from multiple radiating elements of the system to form a synthesised beam.

Patent Agency Ranking