TY - JOUR
T1 - Target localization and tracking in noncoherent multiple-input multiple-output radar systems
AU - Niu, Ruixin
AU - Blum, Rick S.
AU - Varshney, Pramod K.
AU - Drozd, Andrew L.
N1 - Funding Information:
This work was supported in part by U.S. Army Space and Missile Defense Command and the Missile Defense Agency under Contracts HQ0006-06-C-7504 Phase I and W9113M-08-C-0221. This work was presented in part at Asilomar’09, Pacific Grove, CA, November 2009. Authors’ addresses: R. Niu, Department of Electrical and Computer Engineering, Virginia Commonwealth University, 601 West Main Street, Richmond, VA 23284, E-mail: ([email protected]); P. K. Varshney, Department of Electrical and Computer Science, Syracuse University, 4-206 Center for Science and Technology, Syracuse, NY 13244; R. S. Blum, Department of Electrical and Computer Engineering, Lehigh University, 304 Packard Lab, 19 Memorial Drive West, Bethlehem, PA 18015; A. L. Drozd, ANDRO Computational Solutions, Beeches Technical Campus, Building 2, Suite 1, 7902 Turing Road, Rome, NY 13440-2067.
PY - 2012/4
Y1 - 2012/4
N2 - For a noncoherent multiple-input multiple-output (MIMO) radar system, the maximum likelihood estimator (MLE) of the target location and velocity, as well as the corresponding Cramér-Rao lower bound (CRLB) matrix, is derived. MIMO radar's potential in localization and tracking performance is demonstrated by adopting simple Gaussian pulse waveforms. Due to the short duration of the Gaussian pulses, a very high localization performance can be achieved, even when the matched filter ignores the Doppler effect by matching to zero Doppler shift. This leads to significantly reduced complexities for the matched filter and the MLE. Further, two interactive signal processing and tracking algorithms, based on the Kalman filter and the particle filter (PF), respectively, are proposed for noncoherent MIMO radar target tracking. For a system with a large number of transmit/receive elements and a high signal-to-noise ratio (SNR) value, the Kalman filter (KF) is a good choice; while for a system with a small number of elements and a low SNR value, the PF outperforms the KF significantly. In both methods, the tracker provides predictive information regarding the target location, so that the matched filter can match to the most probable target locations, reducing the complexity of the matched filter and improving the tracking performance. Since tracking is performed without detection, the presented approach can be deemed as a track-before-detect approach. It is demonstrated through simulations that the noncoherent MIMO radar provides significant tracking performance improvement over a monostatic phased array radar with high range and azimuth resolutions. Further, the effects of coherent integration of pulses are investigated for both the phased array radar and a hybrid MIMO radar, where only the pulses transmitted and received by colocated transceivers are coherently integrated and the other pulses are combined noncoherently. It is shown that the hybrid MIMO radar achieves significant tracking performance improvement when compared with the phased array radar, by using the extra Doppler information obtained through coherent pulse integration.
AB - For a noncoherent multiple-input multiple-output (MIMO) radar system, the maximum likelihood estimator (MLE) of the target location and velocity, as well as the corresponding Cramér-Rao lower bound (CRLB) matrix, is derived. MIMO radar's potential in localization and tracking performance is demonstrated by adopting simple Gaussian pulse waveforms. Due to the short duration of the Gaussian pulses, a very high localization performance can be achieved, even when the matched filter ignores the Doppler effect by matching to zero Doppler shift. This leads to significantly reduced complexities for the matched filter and the MLE. Further, two interactive signal processing and tracking algorithms, based on the Kalman filter and the particle filter (PF), respectively, are proposed for noncoherent MIMO radar target tracking. For a system with a large number of transmit/receive elements and a high signal-to-noise ratio (SNR) value, the Kalman filter (KF) is a good choice; while for a system with a small number of elements and a low SNR value, the PF outperforms the KF significantly. In both methods, the tracker provides predictive information regarding the target location, so that the matched filter can match to the most probable target locations, reducing the complexity of the matched filter and improving the tracking performance. Since tracking is performed without detection, the presented approach can be deemed as a track-before-detect approach. It is demonstrated through simulations that the noncoherent MIMO radar provides significant tracking performance improvement over a monostatic phased array radar with high range and azimuth resolutions. Further, the effects of coherent integration of pulses are investigated for both the phased array radar and a hybrid MIMO radar, where only the pulses transmitted and received by colocated transceivers are coherently integrated and the other pulses are combined noncoherently. It is shown that the hybrid MIMO radar achieves significant tracking performance improvement when compared with the phased array radar, by using the extra Doppler information obtained through coherent pulse integration.
UR - http://www.scopus.com/inward/record.url?scp=84859833672&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84859833672&partnerID=8YFLogxK
U2 - 10.1109/TAES.2012.6178073
DO - 10.1109/TAES.2012.6178073
M3 - Article
AN - SCOPUS:84859833672
SN - 0018-9251
VL - 48
SP - 1466
EP - 1489
JO - IEEE Transactions on Aerospace and Electronic Systems
JF - IEEE Transactions on Aerospace and Electronic Systems
IS - 2
M1 - 6178073
ER -