Electrical and Computer Engineering ETDs

Yang Li

Publication Date

2-13-2014

Abstract

The cognitive radios (CRs) have opened up new ways of better utilizing the scarce wireless spectrum resources. The CRs have been made feasible by recent advances in software-defined radios (SDRs), smart antennas, reconfigurable radio frequency (RF) front-ends, and full-duplex RF front-end architectures, to name a few. Generally, a CR is considered as a dynamically reconfigurable radio capable of adapting its operating parameters to the surrounding environment. Recent developments in spectrum policy and regulatory domains also allow more flexible and efficient utilization of wider RF spectrum range in the future. In line with the future directions of CRs, a new vision of a future autonomous CR device, called Radiobots, was previously proposed. The goals of the proposed Radiobot surpass the dynamic spectrum access (DSA) to achieve wideband operability and the main features of cognition. In order to ensure the practicality and robust operation of the Radiobot structure, the research focus of this dissertation includes the following aspects: 1) robust spectrum sensing and operability in a centralized CR network setup; 2) robust multivariate non-parametric quickest detection for dynamic spectrum usage tracking in an alien RF environment; 3) joint physical layer and medium access control layer (PHY/MAC) decision-making for wideband bandwidth aggregation (simultaneous operation over multiple modes/networks); and 4) autonomous spectrum sensing scheduling solutions in an alien ultra wideband RF environment. The major contribution of this dissertation is to investigate the feasibility of the autonomous CR operation in heterogeneous RF environments, and to provide novel solutions to the fundamental and crucial problems/challenges, including spectrum sensing, spectrum awareness, wideband operability, and autonomous PHY/MAC protocols, thus bringing the autonomous Radiobot one step closer to reality.

Keywords

cognitive radio, spectrum sensing, signal detection and estimation, quickest detection, machine learning, Markov model, cyclostationarity, wireless communication, hidden Markov model, wideband

Document Type

Dissertation

Language

English

Degree Name

Electrical Engineering

Level of Degree

Doctoral

Department Name

Electrical and Computer Engineering

First Committee Member (Chair)

Christodoulou, Christos

Second Committee Member

Shu, Wei

Third Committee Member

Zhang, Guoyi

Download

Share

COinS