Massive Connectivity in MIMO-OFDM Systems With Frequency Selectivity Compensation

Abstract

This paper considers the joint design of device activity detection and channel estimation in multiple-input multiple-output (MIMO) orthogonal frequency-division multiplexing (OFDM) based grant-free non-orthogonal multiple access (NOMA) systems. In specific, we leverage the correlation of the channel frequency responses in typical narrow-band massive machine-type communication (mMTC) to establish a blockwise linear channel model. In the proposed channel model, the continuous OFDM subcarriers are divided into several subblocks. A linear function with only two variables (mean and slope) is used to approximate the frequency-selective channel in each sub-block. This significantly reduces the number of variables to be determined in channel estimation. We then formulate the joint active device detection and channel estimation as a Bayesian inference problem. By exploiting the block-sparsity of the channel matrix, an efficient turbo message passing (Turbo- MP) algorithm is developed to resolve the Bayesian inference problem with near- linear complexity. We show that Turbo-MP achieves superior performance over the state-of-the-art algorithms.