Correlation estimation between echo signal and leakage signal is indispensable when designing transceiver distance or processing leakage signal in frequency modulated continuous wave (FMCW) radar system. Although many contributions investigate methods for correlation estimation of leakage signal, almost all of them consider a random process and are difficult to quantify the relation. Differently, in this work, we propose a correlation estimation method utilizing information geometry theory to explore a certain condition that makes the correlation degree of leakage signal and echo signal least. Through this method, the random issue of estimation in signals correlation calculation process could be transformed into a deterministic issue. The validity of proposed method is proved by both simulation and measurement results. The optimal correlation condition, determined conveniently by our proposed method, will benefit for antennas distance design and future signal processing to cancel leakage signal.

A stop condition without prior information, applied to compressed sensing recovery in millimeter-wave detection system considering the deceptive jamming signal, is proposed in this paper. The input observation signal is extended to the case consisting of the interference that is correlated to the recovery signal. By making the sufficient statistics of the input parameters and observation values generated from the iterative process, the recovery algorithm can be stopped more appropriately. The main contribution of this paper is to propose a stop criterion to improve the efficiency of the algorithm for random modulated signal recovery. Furthermore, different modulated parameters of the random modulated signal are investigated to verify the feasibility and adaptability of the proposed method. The comparable results show greater performance and the algorithm efficiency is improved by 53.2.