Dissertation/Thesis Abstract

Low Complexity Sub-Nyquist Spectrum Sensing for Wideband Cognitive Radios
by Shaban, Mohamed, Ph.D., University of Louisiana at Lafayette, 2016, 116; 10003781
Abstract (Summary)

Cognitive Radios (CRs) are expected to enhance the spectrum efficiency by enabling the opportunistic access to unused licensed frequency bands. To fully maximize capacity gains, CRs must be able to detect spectral opportunities across a wide spectral band. A wideband spectrum sensing architecture requires analog-to-digital converters (ADC) with extremely high sampling rates. Currently, however, it is infeasible to achieve such rates using the state of the art ADC technologies. On the other hand, the compressed sensing theory has been introduced as a technique for efficiently acquiring and reconstructing a signal using sub-Nyquist sampling rates. The recently proposed wideband spectrum sensing techniques based on the compressed sensing theory acquire wideband signals at low sampling rates while requiring prior knowledge of the signal sparsity. In this dissertation, we proposed a wideband spectrum sensing architecture that employs a recently-proposed compressed sensing technique, “channel multiplexer”, with a single low switching rate analog to digital converter and a low chipping rate pseudorandom generation circuit.

Our proposed architecture offers a simple low-power design with enhanced performance over the-state-of-the-art wideband spectrum sensing architectures. Moreover, we proposed a sub-Nyquist wideband spectrum sensing method for sensing wide frequency band with unknown occupation by measuring a sufficient statistic (average) of the frequency spectra per each licensed channel. The proposed method provides a low complexity approach compared with the-state-of-the-art sub-Nyquist wideband spectrum sensing methods. Finally, we introduced high-speed architectures of the recently proposed compressive sensing detection method. The proposed architecture is fully pipelined, where 14 clock cycles are required to detect a 1024-length signal occupying 8 channels from 16 measurements. The design was implemented in 45 nm CMOS operating at 165 MHz. Since, the sensing time for 1024-length signal was roughly 84.8 ns, the proposed design offered a high speed signal detection compared with the-state-of-the-art orthogonal matching pursuit architectures.

Indexing (document details)
Advisor: Bayoumi, Magdy
Commitee: Bayoumi, Magdy, Fekih, Afef, Perkins, Dmitri, Wu, Hongyi
School: University of Louisiana at Lafayette
Department: Computer Engineering
School Location: United States -- Louisiana
Source: DAI-B 77/06(E), Dissertation Abstracts International
Subjects: Computer Engineering
Keywords: Cognitive, Complexity, Radios, Spectrum, Sub-nyquist, Wideband
Publication Number: 10003781
ISBN: 978-1-339-42889-5
Copyright © 2020 ProQuest LLC. All rights reserved. Terms and Conditions Privacy Policy Cookie Policy