Optimal Design of Large Dimensional Adaptive Subspace Detectors

Atitallah, Ismail Ben; Kammoun, Abla; Alouini, Mohamed‐Slim; Al-Naffouri, Tareq Y.

doi:10.1109/tsp.2016.2573750

Cited by 7 publications

(1 citation statement)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If the steering vector is unknown or mismatched, then the detection performance of these methods will be degraded [15], [16]. For the case of unknown target steering vector, [17], [18] proposed the generalized likelihood ratio test (GLRT) detection method in Gaussian and compound Gaussian environment, as well as studied the influence of mismatched steering vector on detection performance. In addition, the totally blind matrix CFAR detector based on information geometry obtains good detection performance for the problem of detecting moving target in sea clutter [19]- [22].…”

Section: Introductionmentioning

confidence: 99%

Maximum Eigenvalue Matrix CFAR Detection Using Pre-Processing in Sea Clutter

Zhao

Jin

2019

IEEE Access

View full text Add to dashboard Cite

For the pulse Doppler radar system with short pulse sequence, maximum eigenvalue-based matrix CFAR detector (MEMD) provides an effective solution to the problem of detecting moving targets in the background of K distribution sea clutter when the target Doppler frequency is within clutter spectrum. However, its performance is inferior to adaptive normalized matched filter method (ANMF) when the target Doppler frequency heavily deviates from the clutter spectrum. In essence, the reason for the better performance of ANMF method is the clutter is effectively suppressed by whitening and matched filter. Motivated by this, the fusion idea of eigenvalue detection and pre-processing via fast Fourier transform is considered taking advantage of the priori information of target steering vector. A pre-processing-based maximum eigenvalue matrix CFAR detection (P-MEMD) method is proposed to suppress the influence of clutter on detection performance. Finally, the numerical experimental results based on simulated data and real sea clutter data show that the proposed method achieves better detection performance and robustness than some conventional methods. INDEX TERMSMoving target detection, pre-processing, sea clutter.

show abstract