Probability of Target Presence for Multistatic Sonar Ping Sequencing
ABSTRACT In this communication, the problem of determining effective pinging strategies in multistatic sonar systems with multiple transmitters is addressed. New algorithms are presented to determine effective pinging strategies for generalized search scenarios. An important part of this work is the development of metrics to be used in the optimization procedures. For maintaining search coverage, a ldquoprobability of target presencerdquo metric formulation is used. This formulation utilizes sonar performance prediction and a Bayesian update to incorporate negative information (i.e., searching an area but finding no targets) into the optimization procedure. The possibility of targets moving into previously searched areas is accounted for by using a Fokker-Planck (FP) drift/diffusion formulation. Monte Carlo simulations are used to show the accuracy and efficiency of this formulation. This formulation is shown to be computationally efficient compared to Monte Carlo simulations. It is also demonstrated that by choosing the ping sequence intelligently, the field performance can be improved compared to random or sequential ping sequencing.
- SourceAvailable from: isif.org[Show abstract] [Hide abstract]
ABSTRACT: This paper describes the forecasting of probability of target presence in a search area (also referred to as the PT map) considering both detection and non-detection conditions. Tracking results are also incorporated to obtain a more accurate PT map under the detection condition. The probability of target presence is a suitable metric for real-time ping control for a submarine search mission, whose objective is to quickly identify and localize as many targets as possible within the search area. Existing formulations of the probability of target presence metric for ping control include an open-loop approach in which measurements are ignored or a semi-adaptive approach in which measurements are considered but without the true/false target investigation. Since false contacts are inevitable in practical applications and the true/false target investigation of the contacts is not immediate, tracking results must be considered in the PT map generation to obtain an accurate assessment of the present and projected operational pictures. We develop an approach to obtain the current and forecasted PT maps by incorporating a measurement model, a sonar performance model, Bayes theorem and a centralized Kalman-Filter based tracker. The PT map is composed of two portions: the portion which contains detected target probability and the portion which contains missed target probability. Each portion of the PT map is updated and propagated separately. The forecasted PT map at the next ping time is obtained by combining the two propagated PT maps. It will be demonstrated by simulations that the combined forecasted PT map represents an accurate multistatic operational picture and can be used with a sonar performance model to obtain a field metric for ping control optimization for the area search mission.Information Fusion (FUSION), 2011 Proceedings of the 14th International Conference on; 01/2011
- [Show abstract] [Hide abstract]
ABSTRACT: The typical ping control objective for a multistatic network is to optimize a sonar performance metric within a search area over the mission scenario time horizon subject to available ping energy at sources. Assuming sonar performance models and near real-time multistatic target trackers are available, sonar performance metric predictions for near-future time window can be obtained. However, for Anti-Submarine Warfare applications, it is not realistic to accurately predict performance metrics involving an unknown number of evasive targets over a long scenario time horizon. Therefore, effective and efficient ping control methods must consider effective strategies to obtain desired performance over the spectrum of operational objectives and time frames. In this paper, we develop four integer-linear goal programming models to provide intelligent ping control decisions for various operational modes that depend on remaining ping energy and remaining scenario time. We incorporate the multiple objectives of: maximizing the sonar performance metric, judicious use of energy-limited sources and maintaining a certain level of ping activity. We show that the constraint matrix of each relaxed linear model possesses the total unimodularity property, guaranteeing optimal integer ping control solutions. Therefore, computationally efficient linear programming methods can be used in the implementation of these models. We simulate multiple operational scenarios and demonstrate the properties of the resulting ping strategies in terms of the performance metric and individual source and network lifetime. Results are compared to the baseline ping strategy, which considers the sonar performance metric alone.Information Fusion (FUSION), 2012 15th International Conference on; 01/2012