Range-scattering Information and Range Resolution of Multiple Target Radar

Radar super-resolution technology studies have shown that the resolution of radar systems can exceed the traditional physical limit resolution. However, the super-resolution technology lacks analysis from aspects of information theory. In this paper, we propose a new definition of range resolution according to a series of information- theoretic analysis of the radar system for multiple target detection. With the definition, we prove the achievability of the super-resolution technology from the perspective of information theory. First, we derive a theoretical expression of range information and closed-form expression of scattering information for multiple complex Gaussian scattering targets in the presence of complex additive white Gaussian noise (CAWGN). Then, we define the new range resolution in the view of information theory and derive the upper bound of the range resolution 2/ρβ, which is different from the traditional range resolution 1/B, where ρ
²
denotes the signal-to-noise ratio (SNR), B denotes the bandwidth and β
²
denotes the root mean square bandwidth. Finally, We use two-dimensional maximum likelihood estimation (MLE) to verify that the range resolution can be less than 2/ρβ in the high SNR regime, which is different from the one-dimensional MLE used in traditional range resolution.