In order to cope with the challenges of non-cooperative targets such as stealth targets to modern radar, especially when traditional threshold detection and tracking methods can hardly detect fast-moving stealth targets, technological innovation has been long waited. In this paper we have proposed a new algorithm which can reduce the computational consumption and improve tracking precision. Firstly, the number of particles in the traditional particle filter is reduced and a small number of sampling points are sampled from the possible distribution of the target to be tracked, each given a proper weight. Then, the transformed sampling points are sequentially smoothed, and finally the target positions are estimated. The simulation results show that the proposed algorithm is more accurate than the existing particle filter algorithm and has lower computational costs. In the case when SNR is between 0dB to 15dB, a total of 100 Monte Carlo simulations are carried out, obtaining a high detection probability. The detection probability of the improved algorithm is better than that of the existing particle filter at 7dB, the computational cost is also lower than the existing particle filter algorithm.