This brief proposes a joint error information theoretic criterion (JET) assisted blind calibration method for multichannel compressive sampling systems under complete blindness. Addressing gain-phase errors and mutual coupling, we first estimate these errors, then leverage an information theoretic criterion (ITC) to estimate signal sparsity, enabling blind calibration and signal reconstruction. An optimal ITC penalty term is derived through joint error model analysis. Simulation and hardware experiments validate the method's effectiveness. Simulation experiments and hardware experiments confirm that our method achieves over 99.9% accuracy in sparsity estimation within the range [0-0.5] of gain-phase errors and mutual coupling errors.

As a sparse-based direction of arrival (DOA) estimation algorithm, the L1-singular value decomposition (SVD) algorithm is widely used to measure the orientation of targets. In real measurements, the coherent environment that often arises due to multipath propagation leads to the deterioration of the noise immunity and estimation accuracy of the L1-SVD algorithm. Although the decoherence of L1-SVD can be enhanced by introducing spatial smoothing after SVD, which is called SS-L1-SVD, the algorithm does not fully utilize the available information in the observed data. In this article, we propose a new method called L1-enhanced spatial smoothing decomposition (ESSD). ESSD combines spatial smoothing with matrix decomposition by utilizing the relationship among the covariance matrix and the left singular matrix and the singular value matrix. ESSD not only improves the decoherence ability of the algorithm but also makes full use of the information in the observed data and reduces the computational complexity, which makes the algorithm more practical than the traditional algorithms in real measurements. In order to further verify the performance of the new algorithm, we not only performed simulation experiments but also designed a physical experimental platform that can be used for DOA estimation and constructed a real coherent environment caused by multipath propagation and performed physical experiments. The results of simulation and physical experiments show that the L1-ESSD algorithm reduces the error by about 1 degrees and the computation time by about 8 s compared with the conventional L1-SVD algorithm.

Single snapshot direction of arrival (DOA) estimation gains traction in automotive MIMO radar. Gridless methods based on atomic norm show superiority in single snapshot DOA estimation. However, the atomic norm is a convex relaxation of the atomic l0 norm, which leads to resolution limitations. To avoid the disadvantage of resolution limitation, we propose a multi-objective DOA estimation model with atomic l0 norm and measurement errors as optimization objectives. It can estimate the angle and the number of sources simultaneously and has the advantage of directly exploiting sparsity through the atomic l0 norm. Then, we design a cooperative co-evolution crayfish optimization algorithm (CO3) to solve this model. The algorithm contains two innovations, one of which is the proposal of a new multi-population cooperative co-evolutionary decomposition strategy that efficiently decomposes a multi-objective DOA estimation model into multiple single-objective problems without having to consider the fitness allocation problem. Each single-objective problem is then solved using a crayfish optimization algorithm. The other is to propose a variable-length neighbor-hood orthogonal crossover operator to carry out the work of information exchange between populations, which can effectively speed up the convergence of the algorithm. Simulation results and actual data verify the superiority of the method in this paper in terms of source number selection and DOA estimation. © 2025 IEEE.

Sparse direction-of-arrival (DOA) estimation methods can be formulated as a group-sparse optimization problem. Meanwhile, sparse recovery methods based on nonconvex penalty terms have been a hot topic in recent years due to their several appealing properties. Herein, this paper studies a new nonconvex regularized approach called the trimmed lasso for DOA estimation. We define the penalty term of the trimmed lasso in the multiple measurement vector model by ℓ2,1-norm. First, we use the smooth approximation function to change the nonconvex objective function to the convex weighted problem. Next, we derive sparse recovery guarantees based on the extended Restricted Isometry Property and regularization parameter for the trimmed lasso in the multiple measurement vector problem. Our proposed method can control the desired level of sparsity of estimators exactly and give a more precise solution to the DOA estimation problem. Numerical simulations show that our proposed method overperforms traditional approaches, which is more close to the Cramer-Rao bound. © 2024 Elsevier Inc.

The L1-SVD is widely used to solve direction of arrival (DOA) estimation problem in sparse manner. However, due to the influence of the coherent environment often caused by multipath propagation in practical applications, the noise immunity and estimation accuracy of the traditional L1-SVD algorithm deteriorate. In this paper, to improve its performance in correlated environment, we propose a new method called L1-SSD. We introduce spatial smoothing processing into the singular value decomposition (SVD) process and complete the DOA estimation by using the new spatial smoothing decomposition (SSD) with l1-norm minimization. The hardware experimental results in real coherent environment verify that the L1-SSD algorithm can have higher estimation accuracy and better noise immunity than the traditional L1-SVD algorithm with slightly faster computation speed. © 2024 IEEE.

With the rapid increase in signal frequency and bandwidth faced by communication and radar systems, compressive sampling systems have received attention. However, these systems face many nonideal situations. The calibration of gain errors is critical and more difficult in the presence of mutual coupling. Therefore, in this article, we focus on the joint blind calibration of modulated wideband converter (MWC) systems with the unknown gain and mutual coupling. Differing from previous studies on array blind calibration, a novel joint blind calibration method for compressed sampling systems that does not depend on Vandermonde matrix properties of array manifolds is proposed. We model this calibration process as a multilinear inverse problem. By transforming the multilinear inverse problem into a linear inverse problem, a general joint blind calibration algorithm for compressed sampling systems is proposed. For the MWC system, we propose an optimized algorithm to meet the identifiability of joint blind calibration and improve the algorithm performance by using the sparse multiband signal characteristics. Simulation experiments and hardware prototype experiments verify our approach.

This paper presents rank-awareness algorithms to solve sparse blind deconvolution using modulated input. We consider sparse blind deconvolution as a rank-one column sparse matrix recovery problem, so the proposed algorithms can use both the rank-one property and the sparsity of the unknowns. Unknown input s is first multiplied by a random sign sequence r and then convolved with an arbitrary filter h to obtain the measurements y. The unknown signal s is assumed to have a sparse representation. Sparse blind deconvolution using modulated input has unique applications, such as the blind calibration of the random demodulation system. When the number of measurements has satisfied certain conditions, blind deconvolution can be solved without considering signal sparsity. This paper mainly studies how to use signal sparsity to reduce the number of measurements required for sparse blind deconvolution. We propose two methods to solve this problem. The first method uses the l(1)-norm regularization to promote the unknown signal to iterate in the direction of sparsity. The second method transforms the sparse blind deconvolution problem into a rank-one constrained block-sparse signal recovery problem, and we propose the rank-awareness sparse blind demodulation algorithm to solve it. Our proposed methods could effectively reduce the number of measurements required for sparse blind deconvolution. Under certain conditions, our proposed sparse blind deconvolution algorithms required 320 and 160 measurements, while 400 measurements were required when signal sparsity was not considered. The simulation results verify the effectiveness of the proposed algorithms.

The trimmed lasso is a new nonconvex regularized approach for sparse signal recovery, which shows better estimation results in the single measurement vector problem. In this paper, we define the penalty term of the trimmed lasso in the multiple measurement vector model by £2,1-norm. We then present an approach to apply the trimmed lasso to on-grid based direction-of-arrival estimation problems by the alternating directions method of multipliers (ADMM) and majorization-minimization algorithm, which change the nonconvex objective to the convex weighted problem. Numerical simulations show that the proposed method improves the performance over £2,1 minimization algorithm. © 2023 IEEE.

传统的频谱估计首先要对信号进行奈奎斯特采样,未考虑信号的稀疏先验知识,造成了资源和时间大量的浪费。随机解调系统是一种基于压缩感知理论的模拟信息转换结构,能够以低于奈奎斯特频率的采样率对频域稀疏的多频点信号进行压缩采样,在采样的同时压缩信号,保证信号信息的完整性。这样就极大地减少了采样资源和存储资源的浪费,同时,在频谱感知领域,很多时候对精度和实时性的要求很高,需要采用硬件的方法进行频谱感知的实现。因此,本文围绕基于FPGA的随机解调实时频谱估计方法展开研究,主要内容如下:1、研究随机解调的原理。对压缩感知、随机解调和信号重构算法的原理进行深入研究,用数学的方法解释随机解调,分析信号模型、混频、低通滤波、均匀采样和信号重构对于随机解调系统实现的作用。通过仿真实验验证了基于随机解调的频谱估计的可行性,为后续基于硬件平台实现随机解调频谱估计提供参考。2、基于FPGA的OMP算法实现。基于仿真实现的OMP算法实时性不够,因此基于FPGA设计了OMP算法的实时实现。在充分研究随机解调和OMP算法的基础上,结合近些年提出的逐步QR分解的方法,实现了基于FPGA的针对随机解调复数域的OMP算法。3、搭建基于硬件的随机解调频谱估计系统。结合硬件平台作为随机解调压缩采样前端,基于FPGA的OMP算法作为信号处理后端,实现了基于硬件的随机解调频谱估计。随机解调压缩采样前端包括上位机、Analog Discovery2、信号调理电路和电源。上位机设置信号参数,Analog Discovery2生成模拟信号和m序列,调理电路进行压缩采样,最后将观测值送到FPGA中进行处理。4、利用随机解调频谱估计系统进行硬件实验。利用硬件的随机解调系统产生信号和伪随机序列,进行压缩采样产生实际数据,设计典型实验对基于随机解调的OMP模块进行验证和研究,基于FPGA的OMP算法实现速度为35.192μs,恢复频谱的最低信噪比为28.25d B。另外还通过硬件实验验证了恢复信噪比与信号频点数、频点间隔以及定点数格式的关系。

The fixed-dictionary based gridded frequency sparse representation in random demodulation systems faces the grid mismatch problem, which may lead to severe degradation of the compression algorithm to the extent that high precision signal parameters and signal reconstruction cannot be obtained. Gridless compressed sensing introduces the concept of atomic norm, which solves the grid problem caused by spectrum discretization and greatly improves the accuracy of signal frequency estimation. However, in practice, solving large-scale meshless compressed sensing problems directly can consume a lot of time and storage resources. In this paper, we borrow the idea of segmented random sampling to compress the signal by generating segmented pseudo-random sequences in a random demodulation system. Then each segment of the compressed sampled signal is reconstructed and frequency estimated by atomic norm separately, and the obtained frequency estimates of several segments are averaged to obtain the frequency estimates of the original signal. The algorithm can achieve a recovery signal-to-noise ratio of 68 dB for a signal with 256 points per segment, and the recovery accuracy is only related to the number of points within the segment, independent of the number of segments. The algorithm reduces the computational complexity and improves the solution size and the accuracy of frequency estimation. © 2023 IEEE.