Adaptive Two-Step Bayesian Generalized Likelihood Ratio Test Algorithm for Low-Altitude Detection

Hao ZHOU  Guoping HU  Junpeng SHI  Bin XUE  

IEICE TRANSACTIONS on Communications   Vol.E102-B   No.3   pp.571-580
Publication Date: 2019/03/01
Online ISSN: 1745-1345
DOI: 10.1587/transcom.2017EBP3418
Type of Manuscript: PAPER
Category: Antennas and Propagation
low-altitude,  multipath,  clutter,  target detection,  generalized likelihood ratio test,  

Full Text: PDF>>
Buy this Article

The low-altitude target detection remains a difficult problem in MIMO radar. In this paper, we propose a novel adaptive two-step Bayesian generalized likelihood ratio test (TB-GLRT) detection algorithm for low-altitude target detection. By defining the compound channel scattering coefficient and applying the K distributed clutter model, the signal models for different radars in low-altitude environment are established. Then, aiming at the problem that the integrals are too complex to yield a closed-form Neyman-Pearson detector, we assume prior knowledge of the channel scattering coefficient and clutter to design an adaptive two-step Bayesian GLRT algorithm for low-altitude target detection. Monte Carlo simulation results verify that the proposed detector has better performance than the square law detector, GLRT detector or Bayesian GLRT detector in low-altitude environment. With the TB-GLRT detector, the maximum detection probability can reach 70% when SNR=0dB and ν=1. Simulations also verify that the multipath effect shows positive influence on detection when SNR<5dB, and when SNR>10dB, the multipath effect shows negative influence on detection. When SNR>0dB, the MIMO radar, which keeps a detection probability over 70% with the proposed algorithm, has the best detection performance. Besides, the detection performance gets improved with the decrease of sea clutter fluctuation level.