The high dimension and complexity of feature descriptor of Scale Invariant Feature Transform (SIFT), not only occupy the memory spaces, but also influence the speed of feature matching. We adopt the statistic feature point’s neighbor gradient method, the local statistic area is constructed by 8 concentric square ring feature of points-centered, compute gradient of these pixels, and statistic gradient accumulated value of 8 directions, and then descending sort them, at last normalize them. The new feature descriptor descend dimension of feature from 128 to 64, the proposed method can improve matching speed and keep matching precision at the same time.

Correctly sorting the staggered pulse trains each station received is one of the key technologies in the location effect of multi-station time difference passive detection system. According to the problem of straight grid division and the difficulty in sorting two emitters with the same one-dimensional time difference in the histogram method, a time difference sorting algorithm based on natural clustering is proposed. Simulation results show that the algorithm can overcome the defects of histogram method above, and solve the pulse miss-sorting problem, offering better sorting results.

In order to overcome the influence of proning to pitch of platform, in this paper, a novel Algorithm used for preprocessing navigation data is proposed. By means of a characteristic coordinate transformation and two steps delay based on self-installed registration initialization point. Accurate flight parameters of platform are acquired for forward-looking SAR motion compensation precisely. Simulation test proves the validity and serviceability to performance.