Towards realisation of spectrum sharing of cognitive radio networks

Cognitive radio networks (CRN) have emerged as a promising solution to spectrum shortcoming, thanks to Professor Mitola who coined Cognitive Radios. To enable efficient communications, CRNs need to avoid interference to both Primary (licensee) Users (PUs), and among themselves (called self-coexistence). In this thesis, we focus on self-coexistence issues. Very briefly, the problems are categorised into intentional and unintentional interference. Firstly, unintentional interference includes: 1) CRNs administration; 2) Overcrowded CRNs Situation; 3) Missed spectrum detection; 4) Inter-cell Interference (ICI); and 5) Inability to model Secondary Users’ (SUs) activity. In intentional interference there is Primary User Emulation Attack (PUEA). To administer CRN operations (Prob. 1), in our first contribution, we proposed CogMnet, which aims to manage the spectrum sharing of centralised networks. CogMnet divides the country into locations. It then dedicates a real-time database for each location to record CRNs’ utilisations in real time, where each database includes three storage units: Networks locations storage unit; Real-time storage unit; and Historical storage unit. To tackle Prob. 2, our second contribution is CRNAC, a network admission control algorithm that aims to calculate the maximum number of CRNs allowed in any location. CRNAC has been tested and evaluated using MATLAB. To prevent research problems 3, 4, and to tackle research problem (5), our third contribution is RCNC, a new design for an infrastructure-based CRN core. The architecture of RCNC consists of two engines: Monitor and Coordinator Engine (MNCE) and Modified Cognitive Engine (MCE). Comprehensive simulation scenarios using ICS Designer (by ATDI) have validated some of RCNC’s components. In the last contribution, to deter PUEA (the intentional interference type), we developed a PUEA Deterrent (PUED) algorithm capable of detecting PUEAs commission details. PUED must be implemented by a PUEA Identifier Component in the MNCE in RCNC after every spectrum handing off. Therefore, PUED works like a CCTV system. According to criminology, robust CCTV systems have shown a significant prevention of clear visible theft, reducing crime rates by 80%. Therefore, we believe that our algorithm will do the same. Extensive simulations using a Vienna simulator showed the effectiveness of the PUED algorithm in terms of improving CRNs’ performance.

Level dc:type.qualificationlevel

Doctoral (Level 8)

Year dc:date.issued

2026

Author dc:creator

• Fakhrudeen, A

Language dc:language

en

Identifier

oai:salford-repository.worktribe.com:1389008

Author Identifier

0000-0001-8075-3341

OAI identifier oai:identifier

oai:salford-repository.worktribe.com:1389008