Deep Learning-Based Transmit Power Control for Device Activity Detection and Channel Estimation in Massive Access

Zhuo Sun*, Nan Yang, Chunhui Li, Jinhong Yuan, Tony Q.S. Quek

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    5 Citations (Scopus)

    Abstract

    We propose a transmit power control (TPC) scheme for grant-free multiple access, where each device is able to determine its transmit power based on a TPC function. For the proposed scheme, we design a novel deep learning framework to jointly design the TPC functions and the parametric Stein's unbiased risk estimate (SURE) approximate message passing (AMP) algorithm, which significantly improves the accuracy of active device detection and channel estimation, particularly for short pilot sequences. Simulations are conducted to demonstrate the advantages of our proposed deep learning framework on massive device activity detection and channel estimation compared to existing schemes.

    Original languageEnglish
    Pages (from-to)183-187
    Number of pages5
    JournalIEEE Wireless Communications Letters
    Volume11
    Issue number1
    DOIs
    Publication statusPublished - 1 Jan 2022

    Fingerprint

    Dive into the research topics of 'Deep Learning-Based Transmit Power Control for Device Activity Detection and Channel Estimation in Massive Access'. Together they form a unique fingerprint.

    Cite this