Efficient Off-Grid Bayesian Parameter Estimation for Kronecker-Structured Signals
This work studies the problem of jointly estimating unknown parameters from Kronecker-structured multidimensional signals, which arises in applications like intelligent reflecting surface (IRS)-aided channel estimation. Exploiting the Kronecker structure, we decompose the estimation problem into smaller, independent subproblems across each dimension. Each subproblem is posed as a sparse recovery problem using basis expansion and solved using a novel off-grid sparse Bayesian learning (SBL)-based algorithm. Additionally, we derive probabilistic error bounds for the decomposition, quantify its denoising effect, and provide convergence analysis for off-grid SBL. Our simulations show that applying the algorithm to IRS-aided channel estimation improves accuracy and runtime compared to state-of-the-art methods through the low-complexity and denoising benefits of the decomposition step and the high-resolution estimation capabilities of off-grid SBL.
Code (1)
Tasks
Denoisingparameter estimationSimilar Papers 제목 키워드 기반
Bayesian Algorithms for Kronecker-structured Sparse Vector Recovery With Application to IRS-MIMO Channel Estimation
We study the sparse recovery problem with an underdetermined linear system characterized by a Kronecker-structured dictionary and a Kronecker-supported sparse vector. We cast this problem into the sparse Bayesian learnin…
Stochastic Neural Network with Kronecker Flow
Recent advances in variational inference enable the modelling of highly structured joint distributions, but are limited in their capacity to scale to the high-dimensional setting of stochastic neural networks. This limit…
Multi-Armed BanditsThompson SamplingVariational InferenceScalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)
We introduce a kernel approximation strategy that enables computation of the Gaussian process log marginal likelihood and all hyperparameter derivatives in O(p) time. Our GRIEF kernel consists of p eigenfunctions fo…
Bayesian InferenceGaussian ProcessesScalable Gaussian Processes with Grid-Structured Eigenfunctions (GP-GRIEF)
We introduce a kernel approximation strategy that enables computation of the Gaussian process log marginal likelihood and all hyperparameter derivatives in $\mathcal{O}(p)$ time. Our GRIEF kernel consists of $p$ eigenfun…
Bayesian InferenceGaussian ProcessesSuccessive Linear Approximation VBI for Joint Sparse Signal Recovery and Dynamic Grid Parameters Estimation
For many practical applications in wireless communications, we need to recover a structured sparse signal from a linear observation model with dynamic grid parameters in the sensing matrix. Conventional expectation maxim…
Bayesian Inferencecompressed sensing