N.P. Bogomolov, V.G. Sidorov, G.Y. Shaidurov, I.V. Lyutikov

One of the most promising areas, allowing to increase the accuracy of the trajectory measurement of moving targets as well as resolution, noise immunity and vitality of radar systems under suppression of fire and using electronic warfare by the enemy , is the development and creation of multi-radar systems. The work is devoted to solving the acute issues of improving the accuracy of estimating the parameters of the target movement for multi-radar systems using decentralised trajectory information processing. At the receiving point of multi-radar systems primary and secondary processing of radar images are performed. Designed and well-known decentralised systems of trajectory information processing vary in application of the predicted state vector estimation using algorithms of the receiving point. The formation of the resulting estimate of the state vector under centralised information processing is carried out using known operations of statistical matching (unification) estimates of the state vectors for target at each receiving point (FD-algorithm). A characteristic feature of the synthesised algorithm is an availability of inverse association from centralised information processing to receiving point. Along the lines of communication extrapolated estimate of the state vector, formed in centralised information processing system on the basis of the resulting estimate of the state vector is transferred to the receiving point. The algorithm of decentralised system of trajectory information processing at each receiving point given estimate is the predicted estimate in filtering algorithms (FOOS-algorithm). As a result the accuracy of estimating the state vectors calculated by secondary processing at the receiving point is increased which in turn improves the accuracy of the resulting estimates of the state vector under centralised information processing. Due to the fact that the predicted estimate of the resulting state vector is applied in the algorithm of decentralised system of trajectory information processing at every receiving point as results of the simulation showed, it is necessary to take into account the mutual correlation between the estimates of state vectors calculated at the receiving point. The analysis of the functioning of the synthesised FOOS and known FD-algorithms was implemented on the basis of simulation of statistical modelling in the bi-static radar system. The results of the research showed that the developed decentralised system of trajectory information processing allows to track targets with different patterns of movement with less measurement errors than the known algorithm. Standard errors of measurement for Foos-algorithm are 15-20% less than for FD algorithm.