V.N. Zhurakovsky - Ph.D. (Eng.), Associate Professor, Bauman Moscow State Technical University. E-mail: zhurakovsky@sm.bmstu.ru А.Yu. Byldin - Engineer, Bauman Moscow State Technical University К.S. Kondrashov - Engineer, Bauman Moscow State Technical University. E-mail: sm2-2@inbox.ru

In modern survey radar-tracking systems various methods of multi-target tracking are widely used, multi-hypotheses methods of tracking are often applied. Most of such methods demand essential computing expenses and can't define the number of targets. Authors of article have proposed the solution of the stated problems by introduction of special approximation for the Bayesian filter. Instead of estimation of a posteriori density of probability estimation of the statistical moment - the intensity presented as random finite set is carried out. Intensity is such value, that the integral of intensity over any region gives the expected number of targets in this area. This article presents filter realization in case of Gaussian birth and dynamic models. It is shown that if initial prior intensity is a mixture of normal distributions, then a posterior intensity will also be a mixture of normal distributions. Moreover, analytical expressions for recursive estimation of weights, means and covariances of normal component of intensity are obtained. Thus, the developed algorithm allows to get an estimation of conditions of a prior unknown number of the targets, and also to define this number, and spends significantly less system resources, than other algorithms of multi-hypotheses tracking. Researches of algorithm application to a filtration of range and speed are carried out, results of modeling are given.

 

1. Coraluppi S., Carthel C. Generalizations to the track-oriented MHT recursion // Information Fusion (Fusion). 18th International Conference. 6-9 July 2015. P. 346-350.
2. Crouse D., Willet P., Svensson L., Svennson D., Guerriero M. The Set MHT // Information Fusion (FUSION), Proceedings of the 14th International Conference. 5-8 July 2011. P. 1-8.
3. Zeng T., Zheng L., Li Y., Chen X., Long T. Offline Performance Prediction of PDAF With Bayesian Detection for Tracking in Clutter // IEEE Transactions on Signal Processing. 2013. V. 61. № 3. P. 770-781.
4. Habtemariam B., Tharmarasa R., Thayaparan T., Mallick M., Kirubarajan T. A Multiple-Detection Joint Probabilistic Data Association Filter // IEEE Journal of Selected Topics in Signal Processing. 2013. V. 7. № 3. P. 461-471.
5. Mikaehljan S.V. Metody filtracii na osnove mnogotochechnojj approksimacii plotnosti verojatnosti ocenki v zadache opredelenija parametrov dvizhenija celi pri pomoshhi izmeritelja s nelinejjnojj kharakteristikojj // Nauka i obrazovanie. MGTU im. N.EH. Baumana. № 10. Oktjabr 2011.
6. Mahler R. Random set theory for target tracking and identification // Data Fusion Hand Book, D. Hall and J. Llinas (eds.). CRC press Boca Raton. 2001. P. 14/1-14/33.
7. Vo B., Singh S., Doucet A. Sequential Monte Carlo methods for multi-target filtering with random finite sets // IEEE Transactions on Aerospace and Electronic Systems. 2005. V. 41. № 4. P. 1224-1245.
8. Shicang Zhang, Xinmei Hu, Liangbin Wu Multiple manoeuvring targets tracking via GM-PHD and IMM-SB/MHT // Radar Conference 2013. IET International. 14-16 April 2013. P. 1-5.
9. Tang X., Chen X., McDonald M., Mahler R., Tharmarasa R., Kirubarajan T. A Multiple-Detection Probability Hypothesis Density Filter // IEEE Transactions on Signal Processing. 2015. V. 63. № 8. P. 2007-2019.
10. Mahler R. A survey of PHD filter and CPHD filter implementations // Signal Processing, Sensor Fusion, and Target Recognition XV. SPIE Defense & Security Symposium. April 2007.
11. Kondrashov K.S., ZHurakovskijj V.N. Avtozakhvat traektorijj v rezhime avtonomnogo obzora v uslovijakh nizkojj tochnosti vkhodnykh dannykh // Nauka i obrazovanie. MGTU im. N.EH. Baumana. № 11. Nojabr 2013.
12. ZHurakovskijj V.N., Kondrashov K.S. Algoritm razdelenija podvizhnykh i malopodvizhnykh obektov v malogabaritnojj obzornojj RLS // Spectekhnika i svjaz. ROSNOU. Nauchno-tekhnicheskijj zhurnal. 2015. №2.
13. Bolshakov I.A. Statisticheskie problemy vydelenija potoka signalov iz shuma. M.: Sov. radio. 1969.