Millimeter Wave Channel Estimation Using Coupled Sparse Bayesian Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing channel estimation methods for millimeter wave (mmWave) communication systems are not accurate due to differences in physical characteristics compared to lower frequency radio waves, and current methods like sparse recovery formulations do not effectively represent mmWave channel properties.
Innovation Solution
The proposed solution involves a coupled sparse Bayesian learning algorithm that exploits joint angle-of-departure (AoD) and angle-of-arrival (AoA) angular spread to improve channel estimation performance by treating channel gains as random variables and using a two-dimensional block sparse model, which iteratively updates prior variance through the expectation-maximization algorithm.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If sparse recovery formulation is used for mmWave channel estimation, then the estimation process can be simplified, but the accuracy of channel estimation deteriorates because it does not accurately represent mmWave channel properties
Solution Approach 1:
The patent changes the parameters of the sparse recovery formulation by incorporating specific mmWave channel properties including angular spread parameters (AoA and AoD spreads), path loss exponents, and shadowing factors. This transforms the generic sparse recovery approach into a specialized model that maintains computational simplicity while achieving accurate mmWave channel estimation through parameter customization
Solution Approach 2:
The patent extends the sparse recovery formulation from a one-dimensional approach to a two-dimensional angular domain by incorporating both angle of arrival (AoA) and angle of departure (AoD) spreads. This dimensional expansion creates a more comprehensive channel representation that captures the spatial characteristics of mmWave propagation while maintaining the computational efficiency of sparse recovery methods
2Ease of manufacture
If conventional channel estimation methods designed for lower frequency radio waves are used, then the estimation process is well-established, but the accuracy deteriorates due to differences in physical characteristics of mmWaves
Solution Approach 1:
The patent applies local quality by adapting specific components of the channel estimation model to match mmWave characteristics while retaining the overall structure of conventional methods. It incorporates localized adjustments for angular spread, path loss, and shadowing effects that are specific to mmWave propagation, thereby maintaining implementation simplicity while achieving frequency-appropriate accuracy
Solution Approach 2:
The patent introduces dynamic parameters into the channel estimation model, including time-varying angular spreads and path loss exponents that adapt to different mmWave propagation conditions. This dynamic approach allows the estimation method to respond to changing channel characteristics while maintaining a structured framework similar to conventional methods
Data Source
Figure 1A
Figure 1B
Figure 1C
AI summary
Devices and methods for decoding a symbol transmitted over a millimeter wave (mmWave) channel. Receiving a test symbol transmitted over the mmWave channel. Estimating channel state information (CSI) of the mmWave channel using a block sparse signal recovery on the test symbol, according to a multi-dimensional spreading model with statistics on multi-dimensional paths. The multi-dimensional spreading model with statistics on multi-dimensional paths include an angle of departure (AoD), angle of arrival (AoA), and a path spread for the AoD and a path spread for the AoA, propagating in the mmWave channel. Receiving a symbol over the mmWave channel, and decoding the symbol using the CSI, wherein steps of the method are performed by a processor of a receiver.