Cognitive Radio for Improved NBN Access in Regional Areas

Hardless, Benedict (2021) Cognitive Radio for Improved NBN Access in Regional Areas. [USQ Project]

Text (Project)
HARDLESS Benedict dissertation_redacted.pdf

Download (5MB) | Preview


Cognitive Radio (CR) has emerged as a potential solution to spectrum scarcity, whereby time division of spectrum use could allow Secondary Users (SU) to leverage bands when unoccupied by a Primary User (PU). This requires accurate detection of PU signal presence at low Signal to Noise Ratio (SNR).

Eigenvalue-based detection schemes have the potential to detect low SNR signals with periodicity. All schemes reviewed in the literature use Random Matrix Theory and combinations of eigenvalues to develop a threshold for Boolean detection. This project extends early work in Maximum-Minimum Eigenvalue (MME) detection by using eigenvalues as a proxy for SNR, examines the factors affecting it, and develops a framework for optimised detection. Novel Maximum-Minimum Percentage Difference (MMPD) detection is applied to simulated and real SDR signals to examine whether the SNR proxy hypothesis is valid under a range of conditions. Simulations were used to develop a mathematical model for the MMPD scheme. It is refined and tested against generated and real signals, and pre-processing requirements for received signals are defined. An alternative detection scheme called Eigenvector Augmented Fast Fourier Transform (EVA-FFT) is also developed based on Principal Component Analysis (PCA) and tested extensively.

This project defines the relationship between the eigenvalues of the sample covariance matrix and SNR, as well as how it changes with antenna count and listening time. Low SNR limits are proposed for eigenvalue based detection schemes based on antenna count and cycles captured. Neither MMPD or EVA-FFT employ energy detection as most other eigenvalue/eigenvector based schemes do.

Statistics for USQ ePrint 51815
Statistics for this ePrint Item
Item Type: USQ Project
Item Status: Live Archive
Faculty/School / Institute/Centre: Historic - Faculty of Health, Engineering and Sciences - School of Mechanical and Electrical Engineering (1 Jul 2013 - 31 Dec 2021)
Supervisors: Leis, John
Qualification: Bachelor of Engineering (Honours) (Electrical and Electronic)
Date Deposited: 03 Jan 2023 01:41
Last Modified: 26 Jun 2023 01:18
Uncontrolled Keywords: Cognitive radio, National Broadband Network, Regional, Secondary Users, Primary User, Signal to Noise Ratio, Principal Component Analysis, Random Matrix Theory, Eigenvalue

Actions (login required)

View Item Archive Repository Staff Only