FPGA DOA Estimation via Signal Space Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing DOA estimation methods for coherent RF signals in MIMO systems face high computational complexity and resource requirements, especially when applying spatial smoothing to covariance matrices, which is unsuitable for real-time hardware implementation in 5G systems.
Innovation Solution
The method applies forward/backward averaging spatial smoothing to the signal space matrix instead of the covariance matrix, reducing computational complexity and resource consumption by using a pipelined FPGA architecture with stages for covariance matrix computation, signal space extraction, de-correlation, direction matrix calculation, eigenvalue computation, and DOA estimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If spatial smoothing is applied to the covariance matrix for DOA estimation of coherent signals, then estimation accuracy is improved, but computational complexity and resource requirements increase significantly
Solution Approach 1:
The patent extracts the signal space from the covariance matrix using eigenvalue decomposition, separating the signal subspace from the noise subspace. By operating on the extracted signal space matrix rather than the full covariance matrix, the method reduces computational complexity while maintaining DOA estimation accuracy for coherent signals through forward/backward averaging spatial smoothing.
2Measurement precision
If spatial smoothing is applied to the covariance matrix for DOA estimation of coherent signals, then estimation accuracy is improved, but hardware resource requirements increase
Solution Approach 1:
The patent extracts the signal space from the covariance matrix using eigenvalue decomposition, separating the signal subspace from the noise subspace. By operating on the extracted signal space matrix rather than the full covariance matrix, the method reduces computational complexity while maintaining DOA estimation accuracy for coherent signals through forward/backward averaging spatial smoothing.
3Device complexity
If traditional DOA estimation methods are used for real-time processing, then algorithm simplicity is maintained, but processing time increases
Solution Approach 1:
The patent segments the DOA estimation process into distinct pipeline stages: covariance matrix computation, eigenvalue decomposition for signal space extraction, forward/backward averaging spatial smoothing, and DOA angle calculation. This segmentation enables parallel processing and optimization of each stage, significantly reducing overall processing time while maintaining algorithmic clarity through structured organization.
Solution Approach 2:
The patent performs eigenvalue decomposition and signal space extraction as preliminary actions before applying spatial smoothing. By pre-processing the data to extract the signal subspace, subsequent spatial smoothing operations can be applied more efficiently, reducing the computational burden during real-time processing while maintaining accuracy for coherent signals.
Data Source
AI summary
A communications device includes a uniform linear array of M antennas and a field programmable gate array (FPGA) having pipelined stages in which execution of overlapping instructions estimate a direction of arrival of RF signals from multiple sources. A preprocessing stage of the FPGA includes at least one configurable logic block configured to apply forward/backward averaging spatial smoothing to a signal space matrix extracted from a covariance matrix in the preprocessing stage. The FPGA further includes at least one configurable logic block configured to compute the direction of arrival angle for the RF signals using a least squares method.