Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Underdetermined DOA Estimation of Off-Grid Sources Based on the Generalized Double Pareto Prior

Yongfeng Huang, Zhendong Chen, Kun Ye, Lang Zhou, Haixin Sun

TL;DR

The paper tackles underdetermined DOA estimation with dictionary mismatch by introducing a generalized double Pareto prior based sparse Bayesian learning framework. It combines a three-stage hierarchical GDP prior, EM-based Bayesian inference, and an innovative off-grid refinement that updates grid points and the dictionary to continuously reduce modeling error. The key contributions include leveraging a GDP prior for enhanced sparsity, modeling off-grid effects via a first-order Taylor expansion, and iteratively refining the grid to achieve superior DOA accuracy, especially with coarse grids and few snapshots. Empirical results demonstrate notable performance gains over state-of-the-art methods, underscoring the method's potential for robust DOA estimation in challenging scenarios.

Abstract

In this letter, we investigate a new generalized double Pareto based on off-grid sparse Bayesian learning (GDPOGSBL) approach to improve the performance of direction of arrival (DOA) estimation in underdetermined scenarios. The method aims to enhance the sparsity of source signal by utilizing the generalized double Pareto (GDP) prior. Firstly, we employ a first-order linear Taylor expansion to model the real array manifold matrix, and Bayesian inference is utilized to calculate the off-grid error, which mitigates the grid dictionary mismatch problem in underdetermined scenarios. Secondly, an innovative grid refinement method is introduced, treating grid points as iterative parameters to minimize the modeling error between the source and grid points. The numerical simulation results verify the superiority of the proposed strategy, especially when dealing with a coarse grid and few snapshots.

Underdetermined DOA Estimation of Off-Grid Sources Based on the Generalized Double Pareto Prior

TL;DR

The paper tackles underdetermined DOA estimation with dictionary mismatch by introducing a generalized double Pareto prior based sparse Bayesian learning framework. It combines a three-stage hierarchical GDP prior, EM-based Bayesian inference, and an innovative off-grid refinement that updates grid points and the dictionary to continuously reduce modeling error. The key contributions include leveraging a GDP prior for enhanced sparsity, modeling off-grid effects via a first-order Taylor expansion, and iteratively refining the grid to achieve superior DOA accuracy, especially with coarse grids and few snapshots. Empirical results demonstrate notable performance gains over state-of-the-art methods, underscoring the method's potential for robust DOA estimation in challenging scenarios.

Abstract

In this letter, we investigate a new generalized double Pareto based on off-grid sparse Bayesian learning (GDPOGSBL) approach to improve the performance of direction of arrival (DOA) estimation in underdetermined scenarios. The method aims to enhance the sparsity of source signal by utilizing the generalized double Pareto (GDP) prior. Firstly, we employ a first-order linear Taylor expansion to model the real array manifold matrix, and Bayesian inference is utilized to calculate the off-grid error, which mitigates the grid dictionary mismatch problem in underdetermined scenarios. Secondly, an innovative grid refinement method is introduced, treating grid points as iterative parameters to minimize the modeling error between the source and grid points. The numerical simulation results verify the superiority of the proposed strategy, especially when dealing with a coarse grid and few snapshots.
Paper Structure (10 sections, 22 equations, 3 figures)

This paper contains 10 sections, 22 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Probelm Formulation
  3. Proposed Method
  4. Three-stage Hierachical Prior Model
  5. Bayesian Inference
  6. New Off-Grid Refinement
  7. Performance of Evaluation
  8. Simulation of Setup
  9. DOA Estimation Performance
  10. Conclusion

Figures (3)

  • Figure 1: Single DOA estimation result obtained by GDP-OGSBL, where the dotted lines represent the actual incident angles of the impinging signals. (a) Normalized power spectrum; (b) Spatial spectrum.
  • Figure 2: Comparison of DOA estimation performance of different algorithms. (a) SNRs; (b) the number of snapshots.
  • Figure 3: Comparison chart of algorithms under different search step