Implementation and performance evaluation of spectrum sensing using software defined radio

Huang, Yan (2017) Implementation and performance evaluation of spectrum sensing using software defined radio. MRes thesis, University of Nottingham.

[img] PDF (Thesis - as examined) - Repository staff only - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (3MB)

Abstract

As the spectrum becomes more and more congested, an accurate understanding of its usage is required. Software Defined Radio (SDR) systems provide a flexible prototyping platform to develop a wide range of transceivers operating different protocols which can be changed in real time conveniently. SDR is a promising technique with low requirement of hardware devices which made it popular in recent years.

In this project, a SDR test bed has been built as an analysis tool for mapping the use of the spectrum and identifying potential sources of interference. In particular, the project will make use of the Ettus SDR platform (USRP B210) and the GNU Radio development toolkit.

In order to map the use of the spectrum, a system needs to be set up to work as a tool to implement this function. There are some previous simulation results demonstrating that the spectrum sensing schemes have a good performance though these are implemented in software. Whereas, spectrum sensing implementation with SDR in hardware, and the verification of the performance in different system configuration especially for wide bandwidth spectrum sensing has not been explored well. Evaluating the practical performance of spectrum sensing is very important in practice, including the deep understanding of the different spectrum sensing algorithms and the understanding of the practical limitations of the SDR hardware.

Therefore, the aim of this project is to research the use of a SDR platform to implement spectrum sensing and identify its performance due to potential electromagnetic interference. The comparison of the performances among different spectrum sensing schemes is presented in this thesis. For the practical use, a wide range of spectrum sensing methods also are presented to overcome the bandwidth limitation of the SDR hardware. In order to understand what will happen if adding interference to the system, the Gigahertz Transverse Electromagnetic (GTEM) Cell is used in this project to study susceptibility to electromagnetic interference.

The results of this project include the successfully built test bed for spectrum sensing with SDR using the USRP B210 hardware board and GNU Radio, a detailed technical documentation on how to implement an Out-of-Tree spectrum sensing module in GNU Radio, a comparison of Receive Operation Characteristic (ROC) curves for different spectrum sensing schemes, analysing of the system reaction when adding interference to the GTEM Cell and comparison of the results with the system under simulated noise.

The successful development of the spectrum sensing system with SDR gives plenty of opportunities for further research in Dynamic Spectrum Access (DSA) including spectrum sensing technique, for example, improving the threshold of energy detection as a dynamical and adaptive one which will increase the accuracy of sensing results, and using a more powerful platform (USRP X310) to improve the performance and decrease the sensing time for energy saving. These directions will be subjects for future research work.

Item Type: Thesis (University of Nottingham only) (MRes)
Supervisors: Greedy, Steve
Nothorfer, Angela
Keywords: Software radio, Radio frequency allocation
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101 Telecommunication
Faculties/Schools: UK Campuses > Faculty of Engineering
Item ID: 41755
Depositing User: Huang, Yan
Date Deposited: 13 Jul 2017 04:40
Last Modified: 24 Nov 2017 13:01
URI: http://eprints.nottingham.ac.uk/id/eprint/41755

Actions (Archive Staff Only)

Edit View Edit View