Anshu Saksena

Dynamic ping optimization for surveillance in multistatic sonar buoy networks with energy constraints

In this paper we study the problem of dynamic optimization of ping schedule in an active sonar buoy network deployed to provide persistent surveillance of a littoral area through multistatic detection. The goal of ping scheduling is to dynamically determine when to ping and which ping source to engage in order to achieve the desirable detection performance. For applications where persistent surveillance is needed for an extended period of time, it is expected that the energy available at each ping source is limited relative to the required system lifetime. Hence efficient management of power consumption for pinging is important to support the required lifetime of the network while maintaining acceptable detection performance. Our approach to ping optimization is based on the application of approximate partially observable Markov decision process (POMDP) techniques such as the rollout algorithm. To enable a practical implementation of the policy rollout, we apply sampling-based techniques based on a simplified model that approximates the detailed multistatic model. Using high fidelity sonar simulations, we evaluate the performance of the proposed approach and compare it with the greedy technique in terms of detection performance and system lifetime.

Distributed inference for network localization using radio interferometric ranging

A localization algorithm using radio interferometric measurements is presented. A probabilistic model is constructed that accounts for general noise models and lends itself to distributed computation. A message passing algorithm is derived that exploits the geometry of radio interferometric measurements and can support sparse network topologies and noisy measurements. Simulations on real and simulated data show promising performance for 2D and 3D deployments.

Probabilistic model checking of the next-generation airborne collision avoidance system

We present a probabilistic model checking approach for evaluating the safety and operational suitability of the Airborne Collision Avoidance System X (ACAS X). This system issues advisories to pilots when the risk of mid-air collision is imminent, and is expected to be equipped on all large, piloted aircraft in the future. We developed an approach to efficiently compute the probabilities of generically specified events and the most likely sequences of states leading to those events within a discrete-time Markov chain model of aircraft flight and ACAS X. The probabilities and sequences are computed for all states in the model. Events of interest include near mid-air collisions (NMACs) and undesirable sequences of advisories that affect operational suitability. We have validated numerous observations of the model with higher-fidelity simulations of the full system. This analysis has revealed several characteristics of ACAS Xs behavior.

Improving System-wide Detection Performance for Sonar Buoy Networks using In-Network Fusion

The problem of optimized distributed detection in a system of networked sensors involves a number of design aspects, including balancing probabilities of missed detection and false alarm as well as managing the communication resources through proper in-network information fusion. Moreover, a number of tradeoffs must be exercised, such as the one between the computational requirements for information fusion and sensor control and the communication requirements for information exchange. Therefore, overall system design decisions are best made by jointly considering the impact of design aspects and tradeoffs on the overall system performance. This paper addresses in-network fusion and associated networking algorithms that improve detection performance and energy efficiency for a multistatic sonar application. This is achieved by exchanging and fusing contacts among sonar buoys before transmission out offield. In-network fusion utilizes lower cost buoy-to-buoy communication for the majority of the data communication and enables a reduction in random uncorrelated false alarms by only reporting detections from multiple buoys that present sufficient correlation. The reduction of out-of-field contact transmissions allows a lower signal excess threshold for each buoy, corresponding to an increased probability of detection. We demonstrate the effectiveness of our distributed in-network fusion through both analysis and high fidelity sonar simulations.

Using domain knowledge to constrain structure learning in a Bayesian bioagent detector

A novel procedure for learning a probabilistic model from mass spectrometry data that accounts for domain specific noise and mitigates the complexity of Bayesian structure learning is presented. We evaluate the algorithm by applying the learned probabilistic model to microorganism detection from mass spectrometry data.

2005 Special Issue: Bayesian model selection for mining mass spectrometry data

A procedure for learning a probabilistic model from mass spectrometry data that accounts for domain specific noise and mitigates the complexity of Bayesian structure learning is presented. We evaluate the algorithm by applying the learned probabilistic model to microorganism detection from mass spectrometry data.

Differential Adaptive Stress Testing of Airborne Collision Avoidance Systems

This paper presents a scalable method to efficiently search for the most likely state trajectory leading to an event given only a simulator of a system. Our approach uses a reinforcement learning formulation and solves it using Monte Carlo Tree Search (MCTS). The approach places very few requirements on the underlying system, requiring only that the simulator provide some basic controls, the ability to evaluate certain conditions, and a mechanism to control the stochasticity in the system. Access to the system state is not required, allowing the method to support systems with hidden state. The method is applied to stress test a prototype aircraft collision avoidance system to identify trajectories that are likely to lead to near mid-air collisions. We present results for both single and multi-threat encounters and discuss their relevance. Compared with direct Monte Carlo search, this MCTS method performs significantly better both in finding events and in maximizing their likelihood.

Amplitude based sensor tasking for multi-target search and track

A search and track problem where multiple sensors with a variety of modalities are being used to track an unknown number of targets is considered here. The primary comparison is between a traditional range and bearing measurement model and one that additionally considers amplitude of return using a radar cross section (RCS) target model. Using the UK-PHD filter to handle the nonlinear measurement models, it is shown that using an RCS target model results in more accurate state estimates and a reduced number of track dwells, thus devoting more time to searching for targets. Additionally, the use of an RCS model also results in improved clutter rejection further increasing the observed differences in track accuracy and refresh rate.

Probabilistic risk assessment for comparative evaluation of security features

A systematic approach for comparing the effectiveness of counterfeit deterrence features in banknotes, credit cards, digital media, etc. was previously presented. That approach built a probabilistic model around the expert identification of the most efficient process by which a counterfeiter can gain sufficient information to replicate a particular feature. We have extended the scope and functionality of that approach to encompass the entire counterfeiting process from the learning phase to the production of counterfeits. The extended approach makes determining the probabilities more straightforward by representing a more detailed model of the counterfeiting process, including many probable counterfeiting scenarios rather than just representing the least costly successful scenario. It uses the counterfeiters probability of succeeding and level of effort as metrics to perform feature comparisons. As before, these metrics are evaluated for a security feature and presented in a way that facilitates comparison with other security features similarly evaluated. Based on this representation, the cost and laboratory procedures necessary for succeeding may be recovered by a dynamic programming technique. This information may be useful in forensic profiling of potential counterfeiters.