V.N. Zhurakovsky – Ph. D. (Eng.), Associate Professor, Bauman Moscow State Technical University
A.Yu. Byldin – Engineer, Bauman Moscow State Technical University
K.S. Kondrashov – Engineer, Bauman Moscow State Technical University
Survey active radar stations scan the space by mechanical rotation of the antenna. In case of use by the phased lattice Stations with phased antenna array also can do electronic scanning by means of the antenna's diagram. The information on targets and hindrances is formed on the signals reflected from surrounding objects. The results are transferred to the operator. The data from survey stations is often transferred to external systems as targeting. There are contradictory requirements imposed to survey radar station. Station have to provide the consumer with information on surrounding space rather often and in the wide range of corners and ranges, but also it has to have sufficient accuracy.
Nowadays there are several tens of the methods of secondary processing suitable for application in survey radar stations. Most of them suffer from lack of systematic synthesis (even without maneuvers and hindrances) in the conditions of existence of a set of the targets. The approach based on the theory of stochastic point processes or random finite sets solves this problem, however the algorithms realized with his help are connected with a large number of restrictions and approximations.
The algorithm proposed in this paper is developed on the basis of the stochastic point processes theory and the decisions theory. The most important innovation in that algorithm is use of specific function of losses. It allows considering the characteristics of the purposes especially important for survey stations, in particular – the speed and arrival time. Another advantage of the algorithm is small quantity of assumptions at its synthesis.
First part of this paper includes the mathematical description of probabilistic multi-target model and model of coordinate points and false alarms. Second part is devoted to algorithm's synthesis and contains application of a Bayesian filtration to stochastic point processes, application of a decision theory to target data obtaining and description of developed algorithm. The developed algorithm requires serious computing expenses, but modern hardware in onboard complexes already possesses sufficient computing resources for its implementation.
- Pulford G. Taxonomy of multiple target tracking methods // IEE Proceedings – Radar Sonar and Navigation. 2005. T. 152. S. 291.
- Tuong V.B. Random Finite Sets in Multi-Object Filtering. Perth: s.n. 2008. 254 c.
- Choppala P.B. Bayesian Multiply Target Tracking. PhD Thesis. Wellington. 2014. 239 c.
- Bol’shakov I.A., Rakovicz V.S. Prikladnaya teoriya sluchajny’x potokov. M.: Sov. radio. 1978. 250 c.
- Mahler R., Goodman I. Mathematics of Data Fusion. New-York: Kluwer Academic. 1997. 423 c.
- Bol’shakov I.A. Statisticheskie problemy’ vy’deleniya potoka signalov iz shuma. M.: Sov. radio. 1969. 464 c.
- Bga-Ngu Vo A closed form solution to the Probability Hypothesis Density Smoother // Information Fusion. 2010. № 3. S. 1−8.
- Streit R. The Pointillist Family of Multitarget Tracking Filters // IEEE T-AES. 2015. S. 300−320.
- Bar-Shalom Y., Xiao-Rong L. Multitarget-Multisensor Tracking: Principles and Techniques. Denver: YBS. 1995. 629 c.
- Bakut P.A., Zhulina Yu.V., Ivanchuk N.A. Obnaruzhenie dvizhushhixsya ob’‘ektov. M.: Sov. radio. 1980. 288 c.
- Tixonov V.I., Kul’man N.K. Nelinejnaya fil’tracziya i kvazikogerentny’j priem signalov. M.: Sov. radio. 1975. 704 c.
- Bar-Shalom Y., Willet P., Tian X. Tracking and datafusion: A handbook of algorithms. New York: YBS Press. 2011. 1235 c.