Design of Selective Detector for Distributed Targets Through Stochastic Characteristic of the Fictitious Signal

Area of Science:

• Signal Processing
• Statistical Detection Theory
• Radar and Sensor Systems

Background:

• Detecting distributed targets in Gaussian noise with an unknown covariance matrix is challenging, especially when signals do not perfectly match the expected model (signal mismatch).
• Existing methods may struggle with unknown noise characteristics and signal variations, impacting detection accuracy.

Purpose of the Study:

• To develop a robust detector for distributed targets under unknown covariance Gaussian noise, specifically addressing the issue of signal mismatch.
• To introduce a novel approach using a fictitious signal to improve the plausibility of the null hypothesis during signal mismatch.

Main Methods:

• Modeling the fictitious signal as a Gaussian component with a specific covariance structure (stochastic factor multiplied by a rank-one matrix).
• Employing the generalized likelihood ratio test (GLRT) to formulate the modified detection problem.
• Deriving the detector and proving its constant false alarm rate (CFAR) property.

Main Results:

• The proposed detector, GLRT-SL, demonstrates effectiveness in performance analysis.
• Under a signal-to-noise ratio (SNR) of 23 dB, the detector shows a rapid decrease in detection probability (to 0.65) as signal mismatch increases (generalized cosine squared from 1 to 0.83).
• This decline is faster compared to other methods like G-ABORT-HE (falls to 0.98) and GW-ABORT-HE (decreases to 0.85), indicating sensitivity to mismatch.

Conclusions:

• The proposed GLRT-SL detector effectively handles distributed target detection in unknown Gaussian noise with signal mismatch.
• The CFAR property ensures stable performance across different noise conditions.
• The detector's performance characteristics highlight its utility in scenarios where signal variations are present.