Loading [MathJax]/extensions/MathZoom.js
A Novel Quaternion DOA Estimation Algorithm Based on Magnetic Loop Antenna | IEEE Conference Publication | IEEE Xplore

A Novel Quaternion DOA Estimation Algorithm Based on Magnetic Loop Antenna


Abstract:

Almost all of the DOA (direction-of-angle) estimation methods which are based on long vector data model generally need a high-dimensional searching process or complex com...Show More

Abstract:

Almost all of the DOA (direction-of-angle) estimation methods which are based on long vector data model generally need a high-dimensional searching process or complex computation process. Since the process is complex, it is limited in the practical applications. A novel estimation algorithm of DOA is proposed based on quaternion data model in this paper. The signal output model based on quaternion is set for electromagnetic loop sensors. Based on quaternion data model, there are only DOA parameters but not polarization parameters in the direction matrix. Thus, the high-dimensional searching process is replaced by low-dimensional process in the proposed DOA estimation algorithm, so that the computation is heavily reduced. The performance of the proposed algorithm is confirmed through numerical examples.
Date of Conference: 23-25 September 2010
Date Added to IEEE Xplore: 14 October 2010
ISBN Information:

ISSN Information:

Conference Location: Chengdu, China

I. INTRODUCTION

The polarization sensitive array is composed of multiple components polarization sensitive antenna elements which are spatially and uniformly placed. Multiple components polarization sensitive antenna includes tripole with three magnetic loop[1]; tripole or three magnetic loops[2] two pole with two magnetic loops[3] two magnetic loop with one pole or two magnetic loops[4]. In recent years, the DOA estimation of polarization signals based on polarization sensitive sensor array have been investigated extensively [1]–[4], But nearly all of these methods are based on long vector data model. In this approach, the vector dimension of recorded signals data was unfolded along the distance dimension of the antenna so that the output data of different components can form a long vector. The main advantage of this way is that it can use the well-known matrix algebra technique over the real or the complex field. However, the “long vector” approach has the drawback of destroying the vector-type of signal locally because of recognition of the data into a large vector, where the steering vector includes the direction information and the polarization information at the same time. When the DOA of the signal is estimated by the searching function, we need to do more dimensions search because the direction information and polarization information can't be separated, the searching process is complicated and limit the practical usage.

Contact IEEE to Subscribe

References

References is not available for this document.