Please use this identifier to cite or link to this item:
Title: Neural networks for direction of arrival estimation in the presence of array sensor failures
Authors: Vigneshwaran Subbaraju
Keywords: DRNTU::Engineering::Electrical and electronic engineering
Issue Date: 2007
Source: Vigneshwaran, S. (2007). Neural networks for direction of arrival estimation in the presence of array sensor failures. Master’s thesis, Nanyang Technological University, Singapore.
Abstract: Direction of Arrival (DoA) estimation is an important step in the implementation of adaptive/smart antennas which are an integral part of mobile communication systems. Conventional array signal processing algorithms have been widely used for DoA estimation. Among the conventional methods eigen-decomposition based methods such as MUSIC, ESPRIT etc have been found to be accurate, and provide higher resolution when compared to other methods. But these methods have been found to be computationally complex and are hence unsuitable for real-time operation. They are also sensitive to sensor failures, noise and other antenna effects like mutual coupling, non-uniformities in the positions of the sensors etc and require calibrated antennas with uniform features. A sensor is said to have failed when it can no longer receive the signals from the sources and its output consists only of the noise produced by the other circuitry involved. Mutual coupling is a phenomenon where the signal at a particular sensor induces a signal at others sensors in its close proximity. In the existing methods, sensor failure is tackled by introducing some corrections to the wrong data acquired after sensor failure and mutual coupling effects are usually overcome by calibrating the array before making the measurements.
Description: 123 p.
DOI: 10.32657/10356/39065
Fulltext Permission: open
Fulltext Availability: With Fulltext
Appears in Collections:EEE Theses

Files in This Item:
File Description SizeFormat 
EEE_THESES_NEW_130.pdf11.47 MBAdobe PDFThumbnail

Page view(s) 1

checked on Oct 27, 2020

Download(s) 1

checked on Oct 27, 2020

Google ScholarTM




Items in DR-NTU are protected by copyright, with all rights reserved, unless otherwise indicated.