IEEE Transactions on Aerospace and Electronic Systems - new TOC TOC Alert for Publication# 7
- IEEE Aerospace and Electronic Systems Society Informationel febrero 8, 2024 a las 1:18 pm
- Guest Editorial for the TAES Special Section on Deep Learning for Radar Applicationsel febrero 8, 2024 a las 1:18 pm
It has been roughly a decade since the first papers using deep neural networks (DNNs) for radar applications were published. Deep learning has revolutionized almost every technical area, from computer vision and natural language processing to health, finance, and biology—any field where data can be analyzed to provide insight. However, in radar applications, deep learning faces unique challenges due to the phenomenology of radio frequency (RF) propagation that creates essential differences in the data itself and impacts the design of DNNs for radar signal analysis , , .
- Table of Contentsel febrero 8, 2024 a las 1:18 pm
- Joint Multierror Calibration by Merging Errors in Distributed Coherent Aperture Radar Using Strong Scatter Echoesel noviembre 30, 2023 a las 1:18 pm
Full coherence needs strict phase synchronization in distributed coherent aperture radar (DCAR). However, DCAR suffers from various types of errors, including gain-phase, time alignment, and antenna position errors (APEs), which seriously damage phase synchronization. In addition, they are coupled and difficult to estimate individually. Thus, calibrating these coupled errors is still an intractable problem in improving coherence performance. In this article, we propose a joint multierror calibration method by merging errors using echoes of several inaccurate-position strong scatters. First, orthogonal waveforms are transmitted in DCAR to obtain transmitting degrees of freedom (DoF), and its echoes from strong scatters are investigated to estimate the coupled errors jointly. Second, parameter extraction is applied to acquire time delays (TD) and gain-phase of echo peaks as measurements in error estimation. In particular, gain-phase and time alignment errors are merged as equivalent gain-phase errors, avoiding the high precision requirement of time alignment errors. Then, a practical approximation of the maximum a posteriori (MAP) estimator and singular value decomposition (SVD) are employed to obtain estimation solutions for balancing efficiency and precision. Finally, the Cramér–Rao lower bounds (CRLB) of the required parameters and critical unknowns (e.g., antenna positions, radar positions, and attitude angles) are derived to analyze the estimation performance. Simulations are employed to validate the necessity of joint calibration, the advantages of merging errors, and the effectiveness of the proposed method.
- Predicting Hypersonic Glide Vehicle Behavior With Stochastic Grammarsel noviembre 29, 2023 a las 1:21 pm
Hypersonic glide vehicles are a new class of vehicles that fly at hypersonic speeds and have high maneuverability. These fast-moving targets exhibit different flight characteristics compared to conventional vehicles, so traditional tracking and defense systems require new methods to contend with them. In this article, we propose a machine learning method for predicting the behavior of hypersonic glide vehicles. Our method is based on a stochastic grammar, which is a mathematical framework that describes the possible transition patterns of sequences. We use the stochastic grammar to predict the transition patterns in hypersonic glide vehicle trajectories. Given a partial trajectory, our method uses the grammar to predict the hypersonic glide vehicle's future kinematics, such as its altitude, velocity, and acceleration. We evaluate our method on two datasets of simulated hypersonic glide vehicle trajectories and show that it can successfully predict hypersonic glide vehicle behavior, even in the presence of noise. We also show that our method can predict several minutes into the future and can accurately predict future hypersonic glide vehicle behavior based on shorter observation times. Our results suggest that our method has the potential to be a valuable tool for predicting the behavior of hypersonic glide vehicles.