Title:	Detection of Pilot-Hopping Sequences
DNr:	LiU-gpu-2020-3
Project Type:	LiU Compute
Principal Investigator:	Ema Becirovic <ema.becirovic@liu.se>
Affiliation:	Linköpings universitet
Duration:	2020-10-30 – 2021-11-01
Classification:	20203
Keywords:

Abstract

In this work, we study an active user detection problem for massive machine type communications (mMTC). The users transmit pilot-hopping sequences and detection of active users is performed based on the received energy. We use three different approaches for detecting the active users. Classical statistical methods, methods utilizing the channel hardening and favorable propagation properties of massive multiple-input multiple-output (MIMO) which simplifies the user detection, and a neural network which learns the complicated problem structure. By using the asymptotic approach, the problem becomes linear and can be solved with tools from compressed sensing. However, by linearizing the problem, we lose information that could be used to further improve the detection performance. We aim to bridge this with a neural network which learns the temporal properties of the problem.