Abstract: GNSS-based passive bistatic radar (PBR) has such advantages as wide coverage, low cost, and low power consumption so that it is potential to silently detect maritime moving target in the waterway once installed on remote islands or buoys. However, the low GNSS transmitting signal power needs a long dwell time to detect the target-reflected signal. It reduces the accuracy and temporal resolution of the estimated maritime moving target state parameters. To solve this problem, a two-step approach based on the tracking-before-detection of Hough Transform and dynamic programming is explored for state detection of passive bistatic radar detection and tracking in GNSS. This approach at first estimates the delay in the 2-dimension delay-slowtime domain, and then estimates the Doppler in the limited delay-Doppler map. The experimental results show that the proposed approach can estimate the delay and Doppler of target-reflected GNSS signal in a long dwell time, and the dynamic programming can be used in the signal with a lower signal-to-noise ratio than the Hough Transform.