Abstract:
The MIMO (Multiple Input/Multiple Output) concept has been very successful in telecommunications. It is based on the use of several antennas for transmission and
several antennas for reception, which enabled the telecommunications system to
significantly increase its performance. This concept is well suited to radars where
there are several transmitting and receiving antennas, it offers several advantages over
conventional SISO radar, in terms of detection, location, identification and tracking
performance of targets. In the first part of our work, we are interested in the problem
of CFAR detection in a MIMO radar in a Pareto environment. To do this, we
introduce new MIMO-CFAR detectors using binary decision rules and the relation
between the exponential and Pareto laws, and by analogy to the case of the
exponential distribution, we generalize for a MIMO radar the analytical expressions
of the probabilities of false alarm of MIMO-CA, MIMO-SO1, MIMO-SO2, MIMOGO1, MIMO-GO2, MIMO-OS1, and MIMO-OS2-CFAR detectors. Finally, we
perform an analysis of the proposed detectors via Monte Carlo simulations for
homogeneous and non-homogeneous clutter situations. In the second part, we propose
to use fuzzy fusion rules to improve the performance of the FCA-CFAR and FOSCFAR detectors for non-coherent MIMO radars in a homogeneous and nonhomogeneous Pareto environment. First, the membership function for each individual
detector is calculated. At the fusion center, the membership functions are combined
using four fuzzy fusion rules, namely; the “MIN”, “MAX”, “algebraic product” and
“algebraic sum” to give a binary decision after defuzzification.
