TY - JOUR
T1 - Knowledge-aided parametric tests for multichannel adaptive signal detection
AU - Wang, Pu
AU - Li, Hongbin
AU - Himed, Braham
PY - 2011/12
Y1 - 2011/12
N2 - In this paper, the problem of detecting a multi-channel signal in the presence of spatially and temporally colored disturbance is considered. By modeling the disturbance as a multi-channel auto-regressive (AR) process with a random cross-channel (spatial) covariance matrix, two knowledge-aided parametric adaptive detectors are developed within a Bayesian framework. The first knowledge-aided parametric detector is developed using an ad hoc two-step procedure for the estimation of the signal and disturbance parameters, which leads to a successive spatio-temporal whitening process. The second knowledge-aided parametric detector takes a joint approach for the estimation of the signal and disturbance parameters, which leads to a joint spatio-temporal whitening process. Both knowledge-aided parametric detectors are able to utilize prior knowledge about the spatial correlation through colored-loading that combines the sample covariance matrix with a prior covariance matrix. Computer simulation using various data sets, including the KASPPER dataset, show that the knowledge-aided parametric adaptive detectors yield improved detection performance over existing parametric solutions, especially in the case of limited data.
AB - In this paper, the problem of detecting a multi-channel signal in the presence of spatially and temporally colored disturbance is considered. By modeling the disturbance as a multi-channel auto-regressive (AR) process with a random cross-channel (spatial) covariance matrix, two knowledge-aided parametric adaptive detectors are developed within a Bayesian framework. The first knowledge-aided parametric detector is developed using an ad hoc two-step procedure for the estimation of the signal and disturbance parameters, which leads to a successive spatio-temporal whitening process. The second knowledge-aided parametric detector takes a joint approach for the estimation of the signal and disturbance parameters, which leads to a joint spatio-temporal whitening process. Both knowledge-aided parametric detectors are able to utilize prior knowledge about the spatial correlation through colored-loading that combines the sample covariance matrix with a prior covariance matrix. Computer simulation using various data sets, including the KASPPER dataset, show that the knowledge-aided parametric adaptive detectors yield improved detection performance over existing parametric solutions, especially in the case of limited data.
UR - http://www.scopus.com/inward/record.url?scp=81455148135&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=81455148135&partnerID=8YFLogxK
U2 - 10.1109/TSP.2011.2168220
DO - 10.1109/TSP.2011.2168220
M3 - Article
AN - SCOPUS:81455148135
SN - 1053-587X
VL - 59
SP - 5970
EP - 5982
JO - IEEE Transactions on Signal Processing
JF - IEEE Transactions on Signal Processing
IS - 12
M1 - 6020816
ER -