Premium
A parameter estimation method for time‐frequency‐overlapped frequency hopping signals based on sparse linear regression and quadratic envelope optimization
Author(s) -
Wang Liandong,
Liu Zhipeng,
Feng Yuntian,
Liu Xiaoguang,
Xu Xiong,
Chen Xiang
Publication year - 2020
Publication title -
international journal of communication systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.344
H-Index - 49
eISSN - 1099-1131
pISSN - 1074-5351
DOI - 10.1002/dac.4463
Subject(s) - computer science , frequency hopping spread spectrum , robustness (evolution) , frequency domain , carrier frequency offset , estimation theory , jamming , estimator , outlier , algorithm , optimization problem , frequency offset , channel (broadcasting) , statistics , telecommunications , mathematics , artificial intelligence , orthogonal frequency division multiplexing , computer vision , gene , thermodynamics , biochemistry , chemistry , physics
Summary The frequency hopping (FH) signals have well‐documented merits for commercial and military fields due to near‐far resistance and robustness to jamming. Therefore, the parameter estimation of FH signals is an important task for subsequent information acquisition and autonomous electronic countermeasure or attack. However, under the complex electromagnetic environment, there always exist overlaps in the time‐frequency domain among multiple signals, which bring poor signal sparsity and make the estimation more challenging. In this paper, a novel parameter estimation approach is developed for the time‐frequency‐overlapped FH signals under single‐channel reception. The exact solution is mainly composed of the sparse linear regression‐based matrix optimization (SLR‐MO) and quadratic envelope optimization (QEO). SLR‐MO highlights the removal of noise and distortion features for improving the overall sparsity and time‐frequency resolution. QEO further eliminates parts of the interfering signal features and outliers and then extracts and optimizes the average time‐frequency ridge to complete the parameter estimation (hopping instants, period, and carriers). Simulation results demonstrate that the developed estimator outperforms the traditional methods in the scope of application, estimation accuracy, and the robustness under low signal‐to‐noise ratio (SNR).