Gary B. Lamont’s research while affiliated with Air Force Institute of Technology and other places

The congested exosphere continues to contain more satellites and debris, raising the potential for destructive collisions. The Special Perturbations (SP) Tasker algorithm currently assigns the ground sensors tasks to track object locations. Accurate locations help avoid collisions. However, the SP Tasker ignores priority, which is the satellite’s importance factor. This article introduces the Evolutionary Algorithm Tasker (EAT) to solve the Satellite Sensor Allocation Problem (SSAP), which is a hybrid Evolutionary Strategy and Genetic Algorithm concept including specific techniques to explore the solution space and exploit the best solutions found. This approach goes beyond the current method, which does not include priority and other methods from the literature that have been applied to small-scale simulations. The SSAP model implementation extends Multi-Objective Evolutionary Algorithms (MOEAs) from the literature while accounting for priorities. Multiple real-world factors are considered, including each sensor’s field-of-view, the orbital opportunities to track a satellite, the capacity of the sensor, and the relative priority of the satellites. The single objective EAT is statistically compared to the SP Tasker algorithm. Simulations show that both the EAT and MOEA approaches effectively use priority in the core tasking algorithms to ensure that higher priority satellites are tracked.

This paper provides a resource management (RM) framework for electronic support (ES) receivers. The resource manager estimates the number of interference signals and the bandwidths of each interference signal in the electromagnetic (EM) environment. Environment estimates are used to select an appropriate adaptive digital beamforming (DBF) algorithm from a predefined look-up table (LUT) of adaptive DBF algorithms. Algorithms are selected from the LUT based on their ability to increase the signal-to-interference plus noise level of the desired signal by a desired amount; the algorithms are also selected based upon their computational complexity. A study of the resource manager computational complexity demonstrates that applying the new architecture does not increase the receiver computational complexity above standard ES receivers. The resulting adaptive receiver is capable of operating with reduced computational loading over standard support receivers. The framework allows for the use of a single receiver for both narrowband and wideband operation without imposing unrequited computational complexity in the narrowband environments.

Receiver systems designed to act on multiple signals of interest at the same time must be able to differentiate between the different signals to properly form simultaneous beams toward each signal. The problem of detecting simultaneous beams is further complicated by the addition of directional interference sources. When interference sources are co-located in frequency with the signals of interest they are confused with additional signals of interest. This paper presents a signal disambiguation algorithm that is able to act in a congested electromagnetic environment and distinguish between multiple signals of interest while rejecting noise co-located in frequency with the signals of interest.

A framework for creating a digital representation of physical structural components is investigated. A model updating scheme used with an artificial neural network to map updating parameters to the error observed between simulated experimental data and an analytical model of a turbine-engine fan blade. The simulated experimental airfoil has as-manufactured geometric deviations from the nominal, design-intent geometry on which the analytical model is based. The manufacturing geometric deviations are reduced through principal component analysis, where the scores of the principal components are the unknown updating parameters. A range of acceptable scores is used to devise a design of computer experiments that provides training and testing data for the neural network. This training data is composed of principal component scores as inputs. The outputs are the calculated errors between the analytical and experimental predictions of modal properties and frequency-response functions. Minimizing these errors will result in an updated analytical model that has predictions closer to the simulated experimental data. This minimization process is done through the use of two multiobjective evolutionary algorithms. The goal is to determine if the updating process can identify the principal components used in simulating the experiment data.

This study applies the NSGA-II, SPEA2, and MOEA/D Multi-Objective Evolutionary Algorithms (MOEAs) to the radar phase coded waveform design problem. The MOEAs are used to generate a series of radar waveform phase codes that have excellent range resolution and Doppler resolution capabilities, while maintaining excellent autocorrelation properties. The study compares the ability of NSGA-II, SPEA2, and MOEA/D to generate a Pareto front of phase code solutions, and then improve upon the quality of the solutions while maintaining a sufficient diversity of available radar phase codes. Results demonstrate that for solving moderate to large instances of the radar phase code problem all three MOEAs generate a diverse set of Pareto optimal radar phase codes. The phase codes generated by NSGA-II have overall better autocorrelation properties than those generated by SPEA2 and MOEA/D, however, all three MOEAs produce useable phase codes.

Solving complex real world Multi-Objective Optimization problems is the forte of Multi-Objective Evolutionary Algorithms (MOEA). Such algorithms have been part of many scientific and engineering endeavors. This study applies the NSGA-II and SPEA2 MOEAs to the radar phase coded waveform design problem. The MOEAs are used to generate a series of radar waveform phase codes that have excellent range resolution and Doppler resolution capabilities. The study compares the ability of NSGA-II and SPEA2 to continually evolve (phase code) solutions on the Pareto front for the problem while maintaining a diversity of solutions (phase codes). Results demonstrate that for the radar phase code problem NSGA-II provides a more diverse population of acceptable solutions and therefore a greater number of different viable phase codes when compared to the solutions provided by SPEA2.

Intrusion Detection (ID) is essential for protecting contemporary computer networks from a range of threats. Modern ID techniques must cope with increasingly sophisticated attacks as well as rapidly rising network line speeds. Signature- based ID is forced to sample sparsely, increasing the likelihood of malicious traffic entering the network without scrutiny. Con- sequently, flow-based ID is gaining attention as an effective complement. ID systems are furthermore often characterized as either network-based or host-based. The autonomous multi agent design paradigm is a scalable, attractive alternative for its potential to leverage the strengths of both architectures: the broad perspective and visibility into distributed malicious activity provided by network-based ID, and the comprehensive view of the local node provided by host-based ID. This paper therefore develops an architecture for a new multi agent, flow-based intrusion detection sysem. The architecture is designed in two iterations of increasing complexity. These innovative ID designs use a "repuation" system to permit agents to dynamically find nodes that are most effective for classifying malicious network ac- tivity. Furthermore, each system design includes the development of an innovative classifier that uses multi objective evolutionary algorithms to aid in the search for effective operational parameter values. Evaluation using an extensive agent simulation framework highlights the conditions under which the reputation system provides a significant classification benefit.

Trends in computing and networking, in terms of physical capability, attack surface, and attacker sophistication, call for automated, fault-tolerant response systems. Military networks present such environments with unique authorities, critical systems, and threats. Within such environments, multi agent systems may make special contributions regarding recognisance and attack scenarios. We survey three multi agent systems designed for cyber operations, with particular emphasis on our classifier for flow-based attacks, which demonstrates the effectiveness of reputation for distributing classifying agents effectively.