Garfield R. Mellema

Clustered Mixture Particle Filter for Underwater Multitarget Tracking in Multistatic Active Sonobuoy Systems

The problem of multitarget tracking in underwater multistatic active sonobuoy systems is challenging because of the large number of false contacts and multiple reflections that reach the receivers. Targeting a robust solution that can track an unknown time-varying number of multiple targets, while keeping continuous tracks even in scenarios with large number of false contacts per ping, a particle filter (PF)-based technique is proposed in this paper. The PF is a nonlinear filtering technique that can accommodate arbitrary sensor characteristics, motion dynamics, and noise distributions. An enhanced version of the PF called the mixture PF is utilized in this paper. While the sampling/importance resampling PF samples from the prior importance density and weights the particles according to the observation likelihood, the mixture PF samples from both importance densities and weights the different groups of particles respectively. The usage of this mixture of importance densities provides better performance and faster convergence to the true targets locations. In order to track an unknown time-varying number of targets, two mixture PFs are used (one for target detection and the other for tracking multiple targets), and a density-based clustering technique. The first filter starts with random uniformly distributed samples over the surveillance area and resets every five pings. Just before the reset, the clustering technique runs to detect the clusters that corresponds to different targets and passes them to the second filter. The performance of the proposed technique is examined and demonstrated by different simulated scenarios and some real datasets from the SEABAR07 trial by the NATO Underwater Research Center.

Reverse-Time Tracking to Enhance Passive Sonar

Passive sonar depends on signals of opportunity to detect, track and localize targets. These signals are typically detected and then tracked using Kalman filter-type signal followers. Target motion analysis (TMA) is then used to estimate the targets range and, from this, its position, course and speed. The accuracy of TMA is strongly dependent on the duration of the available track. Initiating a second tracker in reverse time at the time of detection can reduce or eliminate the delay between target detection and localization. A detection and tracking system for a passive sonar using a towed array receiver is described and an example of reverse-time tracking using real data is provided. Reverse-time tracking is able to significantly increase the amount of track data that can be extracted from already available data, highlighting the need for improved data fusion. Potential improvements to this enhanced system through track association are discussed

A Structured Approach to Passive Sonar Track Association

The presence of multiple, apparently independent track segments originating from the same target complicates the track-level passive sonar picture. If some of these segments can be shown to have common origin, they can be associated into a single composite track, simplifying that picture. The process could also provide additional information about the target, such as range or classification. Track association is typically a manual process, relying on the expertise of a human sonar operator. If a method could be found to reliably apply numerical scores to the degree of apparent relationship between pairs of tracks, those scores could be used to assist an operator in the track association process or as inputs to a fully automated track association process. This paper outlines the construction of a test based on the sample correlation coefficient and describes the result of its application to tracks produced by a probabilistic data association filter (PDAF) from a set of towed array sensor data.

Correlation based testing for passive sonar picture rationalization

Modern passive sonar systems employ a high degree of automation to produce a track-level sonar picture. Further refinement of the track-level information is normally performed by a human operator. Providing automated assistance would reduce the operators workload and is a key enabler for semi- and fully automated sonar systems. The nature of the signals emitted by targets and of the underwater environment typically results in each target being represented by multiple track segments. A tool is required which can numerically describe the relationship between pairs of track segments so that those that apparently share a common origin can be identified automatically. The sample correlation coefficient, is a statistical measure of relatedness. This paper describes the application of a test based on that measure to compare tracks produced by a probabilistic data association filter from a set of towed array sonar data

An operator perspective on net-centric underwater warfare

In March 2007, the Networked Underwater Warfare Technology Demonstration Project at Defence R&D Canada-Atlantic conducted an at-sea antisubmarine trial utilizing Net-centric warfare (NCW) constructs to demonstrate improved technologies for underwater warfare. User feedback was solicited during and after the trial for the purpose of documenting the manner in which the systems were used during the trial and gaining insight into their potential future usage in NCW activities. This paper describes some of the key issues raised and how they might be addressed in the future. Although some of the issues raised can be addressed through adjustments in the communications strategy, the way ahead for NCW will require a redefinition of the concept of operations for each platform and for the team in order to balance the advantages of the team and platform-centric approaches.

Design and Construction of a Specified Rough Interface for Calibrated Acoustic Measurements

Several mechanisms have been proposed to explain the observed penetration of an incident acoustic wave into sediment at angles below the compressional critical angle. A series of well-controlled laboratory measurements were made and compared with the results of a first-order perturbation theory model to evaluate the accuracy of one of these mechanisms, transmission due to interface roughness. An acrylic disc was used to mimic sediment in the laboratory measurements. Properties of the acrylic were determined in the initial measurements and then used to construct a numerical model. The model was validated by a later set of measurements and then extended for the rough interface case through the addition of first-order perturbation theory scattering terms. The extended model was used to select a roughness spectrum for use in a transmission-type scattering experiment. A realization of the roughness spectrum was machined into one face of the acrylic disc. Considerable care was required in the fabrication of the rough interface as its accuracy was essential to the success of the experiment. The results of the transmission-type scattering experiment compared well with the results of the model and clearly demonstrated subcritical scattering at the water-acrylic boundary due to interface roughness