A discretization-free sparse and parametric approach for linear array signal processing

Abstract

Direction of arrival (DOA) estimation in array processing using uniform/sparse linear arrays is concerned in this paper. While sparse methods via approximate parameter discretization have been popular in the past decade, the discretization may cause problems, e.g., modeling error and increased computations due to dense sampling. In this paper, an exact discretization-free method, named as sparse and parametric approach (SPA), is proposed for uniform and sparse linear arrays. SPA carries out parameter estimation in the continuous range based on well-established covariance fitting criteria and convex optimization. It guarantees to produce a sparse parameter estimate without discretization required by existing sparse methods. Theoretical analysis shows that the SPA parameter estimator is a large-snapshot realization of the maximum likelihood estimator and is statistically consistent (in the number of snapshots) under uncorrelated sources. Other merits of SPA include improved resolution, applicability to arbitrary number of snapshots, robustness to correlation of the sources and no requirement of user-parameters. Numerical simulations are carried out to verify our analysis and demonstrate advantages of SPA compared to existing methods.

15 pages, double column, 8 figures, 1 table. To appear in IEEE Transactions on Signal Processing, accepted in July 2014

References in corpus (5)

• On Gridless Sparse Methods for Line Spectral Estimation From Complete and Incomplete Data
• Iterative Sparse Asymptotic Minimum Variance Based Approaches for Array Processing
• Continuous Compressed Sensing With a Single or Multiple Measurement Vectors
• Exact Joint Sparse Frequency Recovery via Optimization Methods
• Adaptive matching pursuit for off-grid compressed sensing

Cited by in corpus (21)

• Deep Networks for Direction-of-Arrival Estimation in Low SNR
• A joint precoding framework for wideband reconfigurable intelligent surface-aided cell-free network
• On Gridless Sparse Methods for Line Spectral Estimation From Complete and Incomplete Data
• On the Performance of One-Bit DoA Estimation via Sparse Linear Arrays
• Continuous Compressed Sensing With a Single or Multiple Measurement Vectors
• Exact Joint Sparse Frequency Recovery via Optimization Methods
• New Methods for MLE of Toeplitz Structured Covariance Matrices with Applications to RADAR Problems
• Direction-of-Arrival Estimation for Constant Modulus Signals Using a Structured Matrix Recovery Technique
• Subspace Representation Learning for Sparse Linear Arrays to Localize More Sources than Sensors: A Deep Learning Methodology
• Alternating projections gridless covariance-based estimation for DOA
• Demixing Sines and Spikes Using Multiple Measurement Vectors
• Off-the-Grid Line Spectrum Denoising and Estimation with Multiple Measurement Vectors
• Enhancing Sparsity and Resolution via Reweighted Atomic Norm Minimization
• Expectation Propagation based Line Spectral Estimation
• Gridless Evolutionary Approach for Line Spectral Estimation with Unknown Model Order
• A Comparative Study of Invariance-Aware Loss Functions for Deep Learning-based Gridless Direction-of-Arrival Estimation
• Grid-less Variational Bayesian Inference of Line Spectral from Quantized Samples
• A Block Alternating Optimization Method for Direction-of-Arrival Estimation with Nested Array
• Gridless Line Spectral Estimation with Multiple Measurement Vector via Variational Bayesian Inference
• Gridless Variational Bayesian Channel Estimation for Antenna Array Systems with Low Resolution ADCs
• Multiobjective Bilevel Evolutionary Approach for Off-Grid Direction-of-Arrival Estimation