Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.716175
Title: Application of computational intelligence in cognitive radio network for efficient spectrum utilization, and speech therapy
Author: Iliya, Sunday
Awarding Body: De Montfort University
Current Institution: De Montfort University
Date of Award: 2016
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
communication systems utilize all the available frequency bands as efficiently as possible in time, frequency and spatial domains. Society requires more high capacity and broadband wireless connectivity, demanding greater access to spectrum. Most of the licensed spectrums are grossly underutilized while some spectrum (licensed and unlicensed) are overcrowded. The problem of spectrum scarcity and underutilization can be minimized by adopting a new paradigm of wireless communication scheme. Advanced Cognitive Radio (CR) network or Dynamic Adaptive Spectrum Sharing is one of the ways to optimize our wireless communications technologies for high data rates while maintaining users’ desired quality of service (QoS) requirements. Scanning a wideband spectrum to find spectrum holes to deliver to users an acceptable quality of service using algorithmic methods requires a lot of time and energy. Computational Intelligence (CI) techniques can be applied to these scenarios to predict the available spectrum holes, and the expected RF power in the channels. This will enable the CR to predictively avoid noisy channels among the idle channels, thus delivering optimum QoS at less radio resources. In this study, spectrum holes search using artificial neural network (ANN) and traditional search methods were simulated. The RF power traffic of some selected channels ranging from 50MHz to 2.5GHz were modelled using optimized ANN and support vector machine (SVM) regression models for prediction of real world RF power. The prediction accuracy and generalization was improved by combining different prediction models with a weighted output to form one model. The meta-parameters of the prediction models were evolved using population based differential evolution and swarm intelligence optimization algorithms. The success of CR network is largely dependent on the overall world knowledge of spectrum utilization in both time, frequency and spatial domains. To identify underutilized bands that can serve as potential candidate bands to be exploited by CRs, spectrum occupancy survey based on long time RF measurement using energy detector was conducted. Results show that the average spectrum utilization of the bands considered within the studied location is less than 30%. Though this research is focused on the application of CI with CR as the main target, the skills and knowledge acquired from the PhD research in CI was applied in ome neighbourhood areas related to the medical field. This includes the use of ANN and SVM for impaired speech segmentation which is the first phase of a research project that aims at developing an artificial speech therapist for speech impaired patients.
Supervisor: Not available Sponsor: Petroleum Technology Development Fund (PTDF), Nigeria
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.716175  DOI: Not available
Share: