Channel performance estimation for massive multiple input multiple output

Press, Randall Allister

Channel performance estimation for massive multiple input multiple output

Author(s)

Press, Randall Allister

Date Issued

2023

Type

Thesis

Publisher

Cape Peninsula University of Technology

Abstract

Global demand for connectivity has risen exponentially in the 4th industrial revolution. The 4th generation (4G) networks experienced difficulties meeting high demands of media centric communication. Due to spectrum limitation, the orthogonal resources were reaching their limits to satisfy user requirements. Demand has driven the need for 5th generation (5G) communication to provide spectral efficient, low latency, dense network communication. Investigation into non-orthogonal multiple access (NOMA) has been proposed over orthogonal multiple access (OMA) schemes. These include power domain NOMA and code domain NOMA. These schemes are capable of overloading resource elements with multiple users. These techniques show promise in addressing spectral efficiency, low latency, and dense network demands. Thus, a review of literature discusses the need for multiple user access and highlights the earlier techniques. Analysis shows that the SCMA codebook and decoders are effective at providing improved spectral efficiency while delivering a good bit error rate performance. This research aims to design an optimal sparse code multiple access (SCMA) codebook to improve bit error and spectral efficiency of multiple access uplink communication system. An SCMA codebook design is implemented on the uplink of communication system, where a maximum likelihood decoder is implemented at the receiver. The 16-APSK and 16-QAM constellations are used to design the SCMA codebooks, while optimization of the minimum Euclidean distance between neighbouring points is implemented to assist the decoder.

Additional information

Thesis (MEng (Electrical Engineering))--Cape Peninsula University of Technology, 2023

File(s)

Name

Press_Randall_205059392.pdf

Size

2.59 MB

Format

Adobe PDF

Checksum

(MD5):d05c74228d73857051caee3fcaef96e6