Abstract
This study investigates the information–theoretic waveform design problem to improve radar performance in the presence of signal-dependent clutter environments. The goal was to study the waveform energy allocation strategies and provide guidance for radar waveform design through the trade-off relationship between the information theory criterion and the signal-to-interference-plus-noise ratio (SINR) criterion. To this end, a model of the constraint relationship among the mutual information (MI), the Kullback–Leibler divergence (KLD), and the SINR is established in the frequency domain. The effects of the SINR value range on maximizing the MI and KLD under the energy constraint are derived. Under the constraints of energy and the SINR, the optimal radar waveform method based on maximizing the MI is proposed for radar estimation, with another method based on maximizing the KLD proposed for radar detection. The maximum MI value range is bounded by SINR and the maximum KLD value range is between 0 and the Jenson–Shannon divergence (J-divergence) value. Simulation results show that under the SINR constraint, the MI-based optimal signal waveform can make full use of the transmitted energy to target information extraction and put the signal energy in the frequency bin where the target spectrum is larger than the clutter spectrum. The KLD-based optimal signal waveform can therefore make full use of the transmitted energy to detect the target and put the signal energy in the frequency bin with the maximum target spectrum.
Highlights
Cognitive radar can be used to adaptively investigate the radar scene and determine following actions based on previous measurements
Many metrics were proposed for radar waveform optimization, such as the mutual information (MI) [5], the minimum mean squared error (MMSE) [6,7], the Kullback–Leibler divergence (KLD) [8], and signal-to-noise ratio (SNR) [9,10]
Bell [11] first proposed that information–theoretic tools are important for radar waveform design and considered the problems of optimal target detection and optimal target information extraction
Summary
Cognitive radar can be used to adaptively investigate the radar scene and determine following actions based on previous measurements. Transmitted waveform design is an important factor affecting radar system performance, which is attracting increasing attention [1,2,3,4]. One of the most important aspects of signal waveform design is the choice of an optimization metric. Many metrics were proposed for radar waveform optimization, such as the mutual information (MI) [5], the minimum mean squared error (MMSE) [6,7], the Kullback–Leibler divergence (KLD) [8], and signal-to-noise ratio (SNR) [9,10]. Our focus in this work is on designing waveforms based on information–theoretic metrics to maximize target estimation and detection performance. Bell [11] first proposed that information–theoretic tools are important for radar waveform design and considered the problems of optimal target detection and optimal target information extraction
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
