English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Massive access is a critical design challenge of Internet of Things (IoT)networks. In this paper, we consider the grant-free uplink transmission of anIoT network with a multiple-antenna base station (BS) and a large number ofsingle-antenna IoT devices. Taking into account the sporadic nature of IoTdevices, we formulate the joint activity detection and channel estimation(JADCE) problem as a group-sparse matrix estimation problem. This problem canbe solved by applying the existing compressed sensing techniques, which howevereither suffer from high computational complexities or lack of algorithmrobustness. To this end, we propose a novel algorithm unrolling framework basedon the deep neural network to simultaneously achieve low computationalcomplexity and high robustness for solving the JADCE problem. Specifically, wemap the original iterative shrinkage thresholding algorithm (ISTA) into anunrolled recurrent neural network (RNN), thereby improving the convergence rateand computational efficiency through end-to-end training. Moreover, theproposed algorithm unrolling approach inherits the structure and domainknowledge of the ISTA, thereby maintaining the algorithm robustness, which canhandle non-Gaussian preamble sequence matrix in massive access. With rigoroustheoretical analysis, we further simplify the unrolled network structure byreducing the redundant training parameters. Furthermore, we prove that thesimplified unrolled deep neural network structures enjoy a linear convergencerate. Extensive simulations based on various preamble signatures show that theproposed unrolled networks outperform the existing methods in terms of theconvergence rate, robustness and estimation accuracy.

Algorithm Unrolling for Massive Access via Deep Neural Network with Theoretical Guarantee