Joint Activity Detection and Channel Estimation for mmW/THz Wideband Massive Access
Xiaodan Shao, Xiaoming Chen, Caijun Zhong, Zhaoyang Zhang
Abstract
Millimeter-wave/Terahertz (mmW/THz) communications have shown great potential for wideband massive access in next-generation cellular internet of things (IoT) networks. To decrease the length of pilot sequences and the computational complexity in wideband massive access, this paper proposes a novel joint activity detection and channel estimation (JADCE) algorithm. Specifically, after formulating JADCE as a problem of recovering a simultaneously sparse-group and low rank matrix according to the characteristics of mmW/THz channel, we prove that jointly imposing l <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">0</sub> norm and low rank on such a matrix can achieve a robust recovery under sufficient conditions, and verify that the number of measurements derived for the mmW/THz wideband massive access system is significantly smaller than currently known measurements bound derived for the conventional simultaneously sparse and low-rank recovery. Furthermore, we propose a multi-rank aware method by exploiting the quotient geometry of product of complex rank-L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">max</sub> matrices with the maximum number of scattering clusters L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">max</sub> . Theoretical analysis and simulation results confirm the superiority of the proposed algorithm in terms of computational complexity, detection error rate, and channel estimation accuracy.