Low Probability of Intercept (LPI) radar waveform recognition is one of the crucial functions in the electronic intelligence systems. Advances in artificial intelligence promote the performance of the LPI waveform recognition with various signal features defined with analytical expressions. However, noisy LPI waveform recognition is still a challenge for traditional approaches even with noise elimination techniques, especially for the heavily contaminated noisy LPI signals. Recently, the adaptive analysis techniques such as Variational Mode Decomposition (VMD) and Empirical Mode Decomposition (EMD) provide potential methods that explore the inherent features of signals and formulate adaptive features for signal recognitions. In this paper, we propose an adaptive feature construction framework that utilizes both the adaptive features (via VMD and EMD) and pre-defined analytical features (via Wigner-Ville Distribution (WVD), Choi-William Distribution (CWD), and wavelet analysis) to construct the fusion feature, which is further applied on the Convolutional Neural Networks (CNN) based LPI waveform recognition system. The experimental results show that the proposed Feature Adaptive LPI Network-based Exploitation (FALPINE) approach achieves higher probability of correct classification (PCC) than state-of-the-art works, which demonstrates the superior performance of the proposed approach.
This study investigated left ventricular (LV) systolic dysfunction associated with differential strain among myocardial layers in primary hypertension (PH) patients with or without LV hypertrophy (LVH), and normal patients.In 63 PH and 42 healthy patients, two-dimensional speckle tracking echocardiography was used to measure the peak systolic longitudinal and circumferential strain of the myocardial subendocardial, middle and subepicardial layers, and the peak systolic radial strain. To assess LV systolic function, the apical long axis, 4- and 2-chamber views, and parasternal short axis at the basal, middle, and apical levels were acquired by cardiovascular ultrasound (Vivid E9, GE Healthcare, USA).Overall, the pattern in peak systolic longitudinal strain among myocardial layers was subendocardial > middle > subepicardial. In the peak systolic circumferential strain, this was middle > subepicardial > subendocardial. The peak systolic longitudinal strain was normal > NLVH > LVH. Among the groups, the peak systolic circumferential strain at the basal parasternal short-axis level was statistically similar, but at the middle and the apical parasternal short-axis levels were NLVH > normal > LVH. In normal and NLVH patients, the peak radial strain was middle > apical > basal, and in LVH patients was apical > middle > basal. The peak averages of the longitudinal and subendocardial circumferential strains differed significantly when LVH compared with NLVH and normal patients.The systolic function of PH patients was damaged in comparison with normal individuals, which could be detected conveniently and accurately using two-dimensional speckle tracking echocardiography.
30 August 2019 in Advances in Mechanical Engineering
A hardware-in-loop control framework with robot dynamic models, pursuit–evasion game models, sensor and information solutions, and entity tracking algorithms is designed and developed to demonstrate discrete-time robotic pursuit–evasion games for real-world conditions. A parameter estimator is implemented to learn the unknown parameters in the robot dynamics. For visual tracking and fusion, several markers are designed and selected with the best balance of robot tracking accuracy and robustness. The target robots are detected after background modeling, and the robot poses are estimated from the local gradient patterns. Based on the robot dynamic model, a two-player discrete-time game model with limited action space and limited look-ahead horizons is created. The robot controls are based on the game-theoretic (mixed) Nash solutions. Supportive results are obtained from the robot control framework to enable future research of the robot applications in sensor fusion, target tracking and detection, and decision making.
18 November 2024 in IEEE Transactions on Instrumentation and Measurement
Minimum geometric dilution of precision (GDOP) is desired for a high accuracy in the localization of an unknown node. The optimal placements for the access nodes (or sensing nodes) to achieve the minimum GDOP can be configured by combinations of symmetric cones or regular polyhedrons. The minimum GDOP occurs at the tips of the cones or the center of the polyhedrons. These points are referred to as the designated min-GDOP-points. To benefit the minimum GDOP, the unknown node needs to be at the min-GDOP-points. However, that is hardly the case in practice, because the unknown node could be anywhere in a 3-D working volume or a 2-D area. It is shown that the minimum value of GDOP is determined by the number of access nodes and the dimensions of the measurement, and it does not depend on the elevations of access nodes as long as the topology of the placement is not altered. This motivates the proposed design to use the elevations as an extra degree of freedom to improve the overall localization accuracy in the volume. The goal is to seek the optimal elevations of access nodes to reduce the averaged GDOP values over the volume while keeping the theoretical minimum GDOP value unchanged. The obtained solutions are verified by extensive simulations.
Mission planning is one of the core steps to effectively exploit the capabilities of multi-level cooperative control of multiple semi-autonomous entities, such as unmanned aerospace vehicles (UAVs). In this paper, we describe a methodology of Team Dynamics and Tactics (TDT) for mission planning in a military operation. This method for mission planning is implemented in a TDT module for an interconnected system called Strategies for Human-Automaton Resource Entity Deployment (SHARED). The main purpose of TDT is to develop and provide an effective target selection algorithm and an optimal salvo size algorithm to destroy the opposing force combat capabilities. Furthermore, the TDT mission plan will find an optimal assignment of decoys and avoid collateral damage. The proposed mission-planning scheme supplies the optimum degree of force for campaign objectives by using a linear integer programming with fuzzy objective function to allocate the best UAVs and weapons against each target and a parameter Nash game with proportional feedback control to determine optimum salvo size for each UAV. The effectiveness of the proposed scheme is illustrated by a Suppression of Enemy Air Defenses (SEAD) example, and is demonstrated in a simulation environment based on the Boeing C4ISim Open Experimentation Platform (Boeing OEP).
Space superiority requires space protection and space situational awareness (SSA), which rely on rapid and accurate space object behavioral and operational intent discovery. The presence of adversaries in addition to real-time and hidden information constraints greatly complicates the decision-making process in controlling both ground-based and spacebased surveillance assets. This paper develops and implements a solution called Adaptive Markov Inference Game Optimization (AMIGO) for rapid discovery of evasive satellite behaviors. AMIGO is an adaptive feedback game theoretic approach. AMIGO gets information from sensors about the relations between the resident space objects (RSOs) of interest and ground and space surveillance assets (GSAs). The relations are determined by both the RSOs and GSAs. Therefore, AMIGO represents the situation as a game instead of a control problem. The game reasoning utilizes data level fusion, stochastic modeling/propagation, and RSO detection/tracking to predict the future RSOs-GSAs relations. The game engine also supports optional space pattern dictionary/semantic rules for adaptive transition matrices in the Markov game. If no existing pattern dictionary is available, AMIGO builds an initial one and revises it during the game reasoning. The outputs of the AMIGO reasoning include two kinds of control methods: processing of GSA measurements and localization of RSOs. The two sets form a game equilibrium, one for surveillance asset management and the other for the estimation of RSO behaviors. Numerical simulations and visualizations demonstrate the performance of AMIGO.