John C. Sciortino

Evolving sensor suites for enemy radar detection

Designing optimal teams of sensors to detect the enemy radars for military operations is a challenging design problem. Many applications require the need to manage sensor resources. There is a tradeoff between the need to decrease the cost and to increase the capabilities of a sensor suite. In this paper, we address this design problem using genetic algorithms. We attempt to evolve the characteristics, size, and arrangement of a team of sensors, focusing on minimizing the size of sensor suite while maximizing its detection capabilities. The genetic algorithm we have developed has produced promising results for different environmental configurations as well as varying sensor resources.

Comparison of time of arrival vs. multiple parameter based radar pulse train deinterleavers

This paper provides a comparison of the two main techniques currently in use to solve the problem of radar pulse train deinterleaving. Pulse train deinterleaving separates radar pulse trains into the tracks or bins associated with the detected emitters. The two techniques are simple time of arrival (TOA) histogramming and multi-parametric analysis. TOA analysis uses only the time of arrival (TOA) parameter of each pulse to deinterleave radar pulse trains. Such algorithms include Cumulative difference (CDIF) histogramming and Sequential difference (SDIF) histogramming. Multiparametric analysis utilizes any combination of the following parameters: TOA, radio frequency (RF), pulse width (PW), and angle of arrival (AOA). These techniques use a variety of algorithms, such as Fuzzy Adaptive Resonance Theory (Fuzzy-ART), Fuzzy Min-Max Clustering (FMMC), Integrated Adaptive Fuzzy Clustering (IAFC) and Fuzzy Adaptive Resonance Theory Map (Fuzzy-ARTMAP) to compare the pulses to determine if they are from the same emitter. Good deinterleaving is critical since inaccurate deinterleaving can lead to misidentification of emitters. The deinterleaving techniques evaluated in this paper are a sizeable and representative sample of both US and international efforts developed in the UK, Canada, Australia and Yugoslavia. Mardia [1989] and Milojevic and Popovich [1992] shows some of the early work in TOA-based deinterleaving. Ray [1997] demonstrates some of the more recent work in this area. Multi-parametric techniques are exemplified by Granger, et al [1998] and Thompson and Sciortino [2004]. This paper will provide an analysis of the algorithms and discuss the results obtained from the referenced articles. The algorithms will be evaluated for usefulness in deinterleaving pulse trains from agile radars.

Team-based resource allocation using a decentralized social decision-making paradigm

We examine the use of local decentralized decision-making methods for solving the problem of resource allocation. Specifically, we study the problem of frequency coverage given a team of cooperating receivers. The decision making process is decentralized in that receivers can only communicate locally. We use an extension of the minority game approach to allocate receivers to current frequency coverage tasks.

Optimizing a Search Strategy for Multiple Mobile Agents

In this paper, we propose a rule-based search method for multiple mobile distributed agents to cooperatively search an area for mobile target detection. The collective goals of the agents are (1) to maximize the coverage of a search area without explicit coordination among the members of the group, (2) to achieve suffcient minimum coverage of a search area in as little time as possible, and (3) to decrease the predictability of the search pattern of each agent. We assume that the search space contains multiple mobile targets and each agent is equipped with a non-gimbaled visual sensor and a range-limited radio frequency sensor. We envision the proposed search method to be applicable to cooperative mobile robots, Unmanned Aerial Vehicles (UAVs), and Unmanned Underwater Vehicles (UUVs). The search rules used by each agent characterize a decentralized search algorithm where the mobility decision of an agent at each time increment is independently made as a function of the direction of the previous motion of the agent, the known locations of other agents, the distance of the agent from the boundaries of the search area, and the agents knowledge of the area already covered by the group. Weights and parameters of the proposed decentralized search algorithm are tuned to particular scenarios and goals using a genetic algorithm. We demonstrate the effectiveness of the proposed search method in multiple scenarios with varying numbers of agents. Furthermore, we use the results of the tuning processes for different scenarios to draw conclusions on the role each weight and parameter plays during the execution of a mission.

High-accuracy, low-ambiguity emitter classification using an advanced Dempster-Shafer algorithm

High-accuracy, low-ambiguity emitter classification based on ESM signals is critical to the safety and effectiveness of military platforms. Many previous ESM classification techniques involved comparison of either the average observed value or the observed limits of ESM parameters with the expected limits contained in an emitter library. Signal parameters considered typically include radio frequency (RF), pulse repetition interval (PRI), and pulse width (PW). These simple library comparison techniques generally yield ambiguous results because of the high density of emitters in key regions of the parameter space (X-band). This problem is likely to be exacerbated as military platforms are more frequently called upon to conduct operations in littoral waters, where high densities of airborne, sea borne, and land based emitters greatly increase signal clutter. A key deficiency of the simple techniques is that by focusing only on parameter averages or limits, they fail to take advantage of much information contained in the observed signals. In this paper we describe a Dempster-Shafer technique that exploits a set of hierarchical parameter trees to provide a detailed description of signal behavior. This technique provides a significant reduction in ambiguity particularly for agile emitters whose signals provide much information for the algorithm to utilize.

Designing teams of unattended ground sensors using genetic algorithms

Improvements in sensor capabilities have driven the need for automated sensor allocation and management systems. Such systems provide a penalty-free test environment and valuable input to human operators by offering candidate solutions. These abilities lead, in turn, to savings in manpower and time. Determining an optimal team of cooperating sensors for military operations is a challenging task. There is a tradeoff between the desire to decrease the cost and the need to increase the sensing capabilities of a sensor suite. This work focuses on unattended ground sensor networks consisting of teams of small, inexpensive sensors. Given a possible configuration of enemy radar, our goal isto generate sensor suites that monitor as many enemy radar as possible while minimizing cost. In previous work, we have shown that genetic algorithms (GAs) can be used to evolve successful teams of sensors for this problem. This work extends our previous work in two ways: we use an improved simulator containing a more accurate model of radar and sensor capabilities for out fitness evaluations and we introduce two new genetic operators, insertion and deletion, that are expected to improve the GAs fine tuning abilities. Empirical results show that our GA approach produces near optimal results under a variety of enemy radar configurations using sensors with varying capabilities. Detection percentage remains stable regardless of changes in the enemy radar placements.

Implementation of battlespace agents for network-centric electronic warfare

In the Network-Centric Warfare (NCW) paradigm, battlespace agents autonomously perform selected tasks delegated by actors/shooters and decision-makers including controlling sensors. Network-Centric electronic warfare is the form of electronic combat used in NCW. Focus is placed on a network of interconnected, adapting systems that are capable of making choices about how to survive and achieve their design goals in a dynamic environment. The battlespace entities: agents, actor/shooters, sensors, and decision-makers are tied together through the information and sensors grids.

Distributed task allocation in dynamic environments

This work investigates the behavior of a distributed team of agents on a dynamic distributed task allocation problem. Previous work finds that a distributed decision making process can effectively assign tasks appropriately to team members even when agents have only local information. We study this problem in a distributed environment in which agents can move, thus causing local neighborhoods to change over time. Results indicate that a higher level of adaptation is clearly required in the dynamic environment. Despite the increased difficulty, the distributed team is able achieve comparable behavior in both static and dynamic environments.

Using a multi-objective genetic algorithm for developing aerial sensor team search strategies

Finding certain associated signals in the modern electromagnetic environment can prove a difficult task due to signal characteristics and associated platform tactics as well as the systems used to find these signals. One approach to finding such signal sets is to employ multiple small unmanned aerial systems (UASs) equipped with RF sensors in a team to search an area. The search environment may be partially known, but with a significant level of uncertainty as to the locations and emissions behavior of the individual signals and their associated platforms. The team is likely to benefit from a combination of using uncertain a priori information for planning and online search algorithms for dynamic tasking of the team. Two search algorithms are examined for effectiveness: Archimedean spirals, in which the UASs comprising the team do not respond to the environment, and artificial potential fields, in which they use environmental perception and interactions to dynamically guide the search. A multi-objective genetic algorithm (MOGA) is used to explore the desirable characteristics of search algorithms for this problem using two performance objectives. The results indicate that the MOGA can successfully use uncertain a priori information to set the parameters of the search algorithms. Also, we find that artificial potential fields may result in good performance, but that each of the fields has a different contribution that may be appropriate only in certain states.

Classifying and evolving multi-agent behaviors from animal packs in search and tracking problems

This work investigates the efforts behind defining a classification system for multi-agent search and tracking problems, specifically those based on relatively small numbers of agents. The pack behavior search and tracking classification (PBSTC) we define as mappings to animal pack behaviors that regularly perform activities similar to search and tracking problems, categorizing small multi-agent problems based on these activities. From this, we use evolutionary computation to evolve goal priorities for a team of cooperating agents. Our goal priorities are trained to generate candidate parameter solutions for a search and tracking problem in an emitter/sensor scenario. We identify and isolate several classifiers from the evolved solutions and how they reflect on the agent control systemss ability in the simulation to solve a task subset of the search and tracking problem. We also isolate the types of goal vector parameters that contribute to these classified behaviors, and categorize the limitations from those parameters in these scenarios.