TY - GEN
T1 - Improved Toeplitz algorithms to coherent sources DOA estimation
AU - Bai, Jun
AU - Shen, Xiaohong
AU - Wang, Haiyan
AU - Liu, Yi
PY - 2010
Y1 - 2010
N2 - In this paper, an improved Toeplitz matrix construction algorithm of covariance matrix is applied. This covariance matrix can be used to estimate both the incoherent and the coherent source direction with the Multiple Signal Classification ("MUSIC") algorithm. First, the proposed algorithm uses the correlation sequences of output narrowband signals between each sensor and the reference sensor instead of the corresponding output sequence; Second, the inner product operation is carried on to the new sequences of each sensor and the reference sensor; Last, we construct a Toeplitz matrix using the inner product sequence. Obviously, the constructed covariance matrix for "MUSIC" algorithm is a non-singular matrix no matter the source is coherent or not. Furthermore, a comparative Monte Carlo simulation test among MUSIC, TOP and this paper's improved Correlation Toeplitz (CTOP) Method is presented on linear equipspaced array (LEA) carrying on to both the incoherent and the coherent source signals under each kind of signal-to-noise ratio (SNR). All results show that the CTOP has more accurate estimation performance than MUSIC and TOP method to both the incoherent and the coherent source, the CTOP has superb performance under the low SNR condition, and what is more commendable is that the CTOP is an unbiased estimation method and can overcome the influence of sensor's error also.
AB - In this paper, an improved Toeplitz matrix construction algorithm of covariance matrix is applied. This covariance matrix can be used to estimate both the incoherent and the coherent source direction with the Multiple Signal Classification ("MUSIC") algorithm. First, the proposed algorithm uses the correlation sequences of output narrowband signals between each sensor and the reference sensor instead of the corresponding output sequence; Second, the inner product operation is carried on to the new sequences of each sensor and the reference sensor; Last, we construct a Toeplitz matrix using the inner product sequence. Obviously, the constructed covariance matrix for "MUSIC" algorithm is a non-singular matrix no matter the source is coherent or not. Furthermore, a comparative Monte Carlo simulation test among MUSIC, TOP and this paper's improved Correlation Toeplitz (CTOP) Method is presented on linear equipspaced array (LEA) carrying on to both the incoherent and the coherent source signals under each kind of signal-to-noise ratio (SNR). All results show that the CTOP has more accurate estimation performance than MUSIC and TOP method to both the incoherent and the coherent source, the CTOP has superb performance under the low SNR condition, and what is more commendable is that the CTOP is an unbiased estimation method and can overcome the influence of sensor's error also.
KW - Coherent
KW - Correlation
KW - DOA estimation
KW - Toeplitz matrix
UR - http://www.scopus.com/inward/record.url?scp=77953160511&partnerID=8YFLogxK
U2 - 10.1109/ICMTMA.2010.592
DO - 10.1109/ICMTMA.2010.592
M3 - 会议稿件
AN - SCOPUS:77953160511
SN - 9780769539621
T3 - 2010 International Conference on Measuring Technology and Mechatronics Automation, ICMTMA 2010
SP - 442
EP - 445
BT - 2010 International Conference on Measuring Technology and Mechatronics Automation, ICMTMA 2010
T2 - International Conference on Measuring Technology and Mechatronics Automation, ICMTMA 2010
Y2 - 13 March 2010 through 14 March 2010
ER -