Design and allocation of low correlated signatures for sequence block-based compressive sensing multiuser detection in massive machine-type communication
- Authors
- Bakhtawar, Malika; Alam, Mehmood; Wakeel, Abdul; Nisar, Shibli; Syed, Ikram; Choi, Jaehyuk
- Issue Date
- Mar-2023
- Publisher
- ELSEVIER GMBH
- Keywords
- Code domain NOMA; Massive MTC; 5G; CSMUD
- Citation
- AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS, v.161
- Journal Title
- AEU-INTERNATIONAL JOURNAL OF ELECTRONICS AND COMMUNICATIONS
- Volume
- 161
- URI
- https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/88205
- DOI
- 10.1016/j.aeue.2023.154537
- ISSN
- 1434-8411
- Abstract
- Compressed sensing-based multiuser detection (CSMUD) is a promising technique to enable non-orthogonal multiple access (NOMA) for massive machine-type communication (mMTC) in beyond 5G wireless commu-nication systems. CSMUD facilitates grant-free multiple access by jointly detecting the activity and data at the receiver. The performance of CSMUD is predominantly determined by the activity detection, governed by the maximum correlation between the signatures. Sequence block-based CSMUD (SB-CSMUD), an enhanced CSMUD, exploits temporal and spreading diversity to improve the activity detection. In this paper, we propose a three-step approach to design and allocate low correlated signatures and improve activity detection in SB-CSMUD. In the first step, a low correlated random sensing matrix is designed, followed by the designing of low correlated signatures. The designed signatures are then allocated to the nodes according to the traffic pattern of the mMTC. The proposed signatures reduces the maximum correlation by 4.48%, which is further reduced by the proposed signature allocation scheme. The simulation results validate the analysis and show that our proposed scheme significantly reduces the detection error rate (DER) without increasing computational complexity.
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - ETC > 1. Journal Articles
![qrcode](https://api.qrserver.com/v1/create-qr-code/?size=55x55&data=https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/88205)
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.