Detection and States Estimation of Multiple Mobile Targets in Wireless Sensor Network
ABSTRACT By introducing a wireless fading model, we rebuild the perception model of a wireless sensor network. The profiles of a multi-target and multi-sensor mixing matrix at each frequency are estimated as samples of the spectrum superposition of multiple targets. A differential evolution approach is employed to separate multiple targets, at the same time, to decouple path fading and Doppler shifts in the frequency domain. Each column of the mixing matrix preserves the waveform that is formed by the effects of corresponding targets on nodes. Based on this modeling, the states of multiple targets, including the location, velocity, and motion direction are estimated.
- SourceAvailable from: Yu Hen Hu[Show abstract] [Hide abstract]
ABSTRACT: A maximum likelihood (ML) acoustic source location estimation method is presented for the application in a wireless ad hoc sensor network. This method uses acoustic signal energy measurements taken at individual sensors of an ad hoc wireless sensor network to estimate the locations of multiple acoustic sources. Compared to the existing acoustic energy based source localization methods, this proposed ML method delivers more accurate results and offers the enhanced capability of multiple source localization. A multiresolution search algorithm and an expectation-maximization (EM) like iterative algorithm are proposed to expedite the computation of source locations. The Crame´r-Rao Bound (CRB) of the ML source location estimate has been derived. The CRB is used to analyze the impacts of sensor placement to the accuracy of location estimates for single target scenario. Extensive simulations have been conducted. It is observed that the proposed ML method consistently outperforms existing acoustic energy based source localization methods. An example applying this method to track military vehicles using real world experiment data also demonstrates the performance advantage of this proposed method over a previously proposed acoustic energy source localization method.IEEE Transactions on Signal Processing 02/2005; 53(1-53):44 - 53. DOI:10.1109/TSP.2004.838930 · 3.20 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: This paper considers the problem of the estimation of the motion parameters of multiple targets moving linearly in a three-dimensional (3-D) observation area contaminated by clutter. The measurements are limited to time differences of arrival (TDOAs) of short-duration acoustic emissions, or transients, generated by the targets. This problem can arise in situations where the level of continuous broadband target-related noise is very low. Owing to the fact that transient emissions are nonstationary and can have low signal-to-noise ratio (SNR), the corresponding TDOA measurement errors are usually non-Gaussian. Therefore, Gaussian mixture distributions are used to appropriately model these errors. An iterative maximum-likelihood optimization technique based on a modified deterministic annealing expectation-maximization (MDAEM) algorithm is applied to this problem. In each iteration, the algorithm uses a nonlinear least-squares (LS) technique in computing the motion parameters for each target. It generalizes the variance deflation method previously used for the initialization of target tracking algorithms and increases the possibility of attaining a globally optimal solution for random initial conditions. Simulation results are presented for several different numbers of targets, clutter densities, and probabilities of gross error of the target related measurements and are found to be comparable to the estimates obtained when the measurement-to-target assignments are exactly knownIEEE Transactions on Signal Processing 03/2007; 55(2-55):424 - 436. DOI:10.1109/TSP.2006.885745 · 3.20 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Traditional multihypothesis tracking methods rely upon an enumeration of all the assignments of measurements to tracks. Pruning and gating are used to retain only the most likely hypotheses in order to drastically limit the set of feasible associations. The main risk is to eliminate correct measurement sequences. The probabilistic multiple hypothesis tracking (PMHT) method has been developed by Streit and Luginbuhl in order to reduce the drawbacks of "strong" assignments. The PMHT method is presented in a general mixture densities perspective. The Expectation-Maximization (EM) algorithm is the basic ingredient for estimating mixture parameters. This approach is then extended and applied to multitarget tracking for nonlinear measurement models in the passive sonar perspective.IEEE Transactions on Aerospace and Electronic Systems 11/1997; 33(4-33):1242 - 1257. DOI:10.1109/7.625121 · 1.39 Impact Factor