Blind localization of radio emitters in wireless communications

Magowe, K 2017, Blind localization of radio emitters in wireless communications, Doctor of Philosophy (PhD), Engineering, RMIT University.

Document type: Thesis
Collection: Theses

Attached Files
Name Description MIMEType Size
Magowe.pdf Thesis Click to show the corresponding preview/stream application/pdf; Bytes
Title Blind localization of radio emitters in wireless communications
Author(s) Magowe, K
Year 2017
Abstract The proliferation of wireless services is expected to increase the demand for radio spectrum in the foreseeable future. Given the limitations of the radio spectrum, it is evident that the current fixed frequency assignment policy fails to accommodate this increasing demand. Thus, the need for innovative technologies that can scale to accommodate future demands both in terms of spectrum efficiency and high reliable communication. Cognitive radio (CR) is one of the emerging technologies that offers a more flexible use of frequency bands allowing unlicensed users to exploit and use portions of the spectrum that are temporarily unused without causing any potential harmful interference to the incumbents. The most important functionality of a CR system is to observe the radio environment through various spectrum awareness techniques e.g., spectrum sensing or detection of spectral users in the spatio-temporal domain.

In this research, we mainly focus on one of the key cognitive radio enabling techniques called localization, which provides crucial geo-location of the unknown radio transmitter in the surrounding environment. Knowledge of the user’s location can be very useful in enhancing the functionality of CRs and allows for better spectrum resource allocations in the spatial domain. For instance, the location-awareness feature can be harnessed to accomplish CR tasks such as spectrum sensing, dynamic channel allocation and interference management to enable cognitive radio operation and hence to maximize the spectral utilization. Additionally, geo-location can significantly expand the capabilities of many wireless communication applications ranging from physical layer security, geo-routing, energy efficiency, and a large set of emerging wireless sensor network and social networking applications.

We devote the first part of this research to explore a broad range of existing cooperative localization techniques and through Monte-Carlo simulations analyze the performance of such techniques. We also propose two novel techniques that offer better localization performance with respect to the existing ones. The second and third parts of this research put forth a new analytical framework to characterize the performance of a particular low-complexity localization technique called weighted centroid localization (WCL), based on the statistical distribution of the ratio of two quadratic forms in normal variables. Specifically, we evaluate the performance of WCL in terms of the root mean square error (RMSE) and cumulative distribution function (CDF). The fourth part of this research focuses on studying the bias of the WCL and also provides solutions for bias correction. Throughout this research, we provide a case study analysis to evaluate the performance of the proposed approaches under changing channel and environment conditions. For the new theoretical framework, we compare analytical and Monte-Carlo simulation results of the performance metric of interest.

A key contribution in our analysis is that we present not only the accurate performance in terms of the RMSE and CDF, but a new analytical framework that takes into consideration the finite nature of the network, overcoming the limitations of asymptotic results based on the central limit theorem. Remarkably, the numerical results unfold that the new analytical framework is able to predict the performance of WCL capturing all the essential aspects of propagation as well as the cognitive radio network spatial topology. Finally, we present conclusions gained from this research and possible future directions.
Degree Doctor of Philosophy (PhD)
Institution RMIT University
School, Department or Centre Engineering
Subjects Computer Hardware not elsewhere classified
Keyword(s) Blind Localization
Performance Analysis
Statistical Signal Processing
Cognitive Radios
Quadratic Forms
Version Filter Type
Access Statistics: 120 Abstract Views, 172 File Downloads  -  Detailed Statistics
Created: Wed, 08 Nov 2017, 12:57:17 EST by Denise Paciocco
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us