Particle Filter Track-Before-Detect Implementation on GPU
Track-before-detect (TBD) based on the particle filter (PF) algorithm is known for its outstanding performance in detecting and tracking of weak targets. However, large amount of calculation leads to difficulty in real-time applications. To solve this problem, effective implementation of the PF-based TBD on graphics processing units (GPU) is proposed in this chapter. By recasting the particles propagation process and weights calculating process on the parallel structure of GPU, the running time of this algorithm can be greatly reduced. Simulation results in the infrared scenario are demonstrated to compare the implementation on two types of the graphic card with the CPU-only implementation.
KeywordsTrack-before-detect Particle filter GPU
The work is supported by the Fundamental Research Funds for the Central Universities of China (ZYGX2011J012).
- 2.Salmond DJ, Birch H (2001) A particle filter for track-before-detect. In: Proceedings of the American control conference, vol 5. Arlington. pp 3755–3760. doi: 10.1109/ACC.2001.946220
- 3.Rollason M, Salmond D (2001) A particle filter for track-before-detect of a target with unknown amplitude. Target tracking: algorithms and applications, vol 1. Enschede: 14/1–4. doi: 10.1049/ic:20010240
- 4.Rutten MG, Gordon NJ, Maskell S (2004) Efficient particle based track before detect in Rayleigh noise. In: Proceedings of Signal and Data Processing of Small Targets, International Society for Optical Engineering, Bellingham, pp 509–519Google Scholar
- 5.Rutten MG, Ristic B, Gredon NJ (2005) A comparison of particle filters for recursive track-before-detect. In: Proceedings of the 8th international conference on information fusion, vol 1. Piscataway, pp 169–175. doi: 10.1109/ICIF.2005.1591851
- 8.The resource for CUDA developers (2010) http://www.nvidia.com/object/cuda_home.html. Accessed Jan 2010
- 9.Shu Zhang, Yanli Chu (2009) GPU computing for high performance-CUDA. China Water & Power Press, BeijingGoogle Scholar