In this paper we present a benchmark problem for data association based on a real-world networked surveillance system, and com-pare the behavior of several multidimensional assignment (MDA) algorithms. The problem consists of a set of N_s observers which transmit track/event reports to a fusion center through a particular (real-world based) communication network among one of N_n networks. The network discards the observer’s track identity (ID), replacing it by a network-generated ID and the observer ID, thus losing the information on the origin of the tracks sent by each observer. The solution approach developed in this paper consists of a hierarchical decomposition of the problem. This hierarchical approach first eliminates the redundancy introduced by the communication network by using an MDA algorithm per each observer present, and then using another MDA algorithm to choose which ‘non-redundant’ reports to fuse. This decomposition drastically re-duces the dimensionality of the problem from N_s £ N_n to N_s problems of dimension N_n and one of dimension N_s.

A comparison of two association criteria, normalized distance squared (NDS) and likelihood ratio (LR), is carried out. It is shown that the LR yields significantly superior results. Also the selection of certain parameters in the likelihood ratio is discussed. Finally, to evaluate their performance, three different MDA algorithms are used in this setup, Lagrangean Relaxation based MDA, Sequential m-best 2D and Linear Programming. A thorough comparison of these algorithms in terms of the quality of their solutions as well as their run times is done showing some pitfalls and advantages of each.