DSP Matrix Inversion via CORDIC for MIMO-OFDM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high computational requirements of Multiple-Input Multiple-Output (MIMO) Orthogonal Frequency Division Multiplexing (OFDM) systems make software implementation in terminal stations difficult, with only custom hardware solutions conventionally available, limiting network capacity and coverage in wireless communication systems.
Innovation Solution
A digital signal processor with a vector unit and registers is provided, using Coordinate Rotation Digital Computer (CORDIC)-based instructions to perform complex matrix operations such as QR decomposition, eigenvalue decomposition, and singular-value decomposition efficiently in software on a multithreaded Single-Instruction Multiple-Data (SIMD) processor, enabling matrix inversion and triangularization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If custom hardware solutions are used for MIMO-OFDM implementation, then computational requirements are met, but device complexity and cost increase
Solution Approach 1:
The patent replaces complex hardware implementations with a software-based solution using standard DSP instructions. Specifically, it uses Coordinate Rotation Digital Computer (CORDIC) algorithms implemented through software on a digital signal processor, eliminating the need for custom hardware circuits while maintaining computational capability for MIMO-OFDM operations.
Solution Approach 2:
The patent changes the implementation approach from hardware-level to software-level by utilizing standard DSP instruction sets and algorithms. It transforms the computational problem into a software implementation using CORDIC-based matrix operations, which can be executed on general-purpose DSP hardware rather than requiring specialized custom hardware.
2Device complexity
If software implementation is used for MIMO-OFDM, then device complexity is reduced, but computational requirements are too high for terminal stations
Solution Approach 1:
The patent segments the computational tasks by implementing matrix operations through modular software functions. It breaks down complex MIMO-OFDM processing into discrete computational steps using standard DSP instructions, making the software implementation manageable and efficient for terminal station processors.
Solution Approach 2:
The patent uses efficient algorithmic approaches that replicate complex hardware functionality through software. By implementing CORDIC-based matrix inversion and decomposition algorithms in software, it creates a functional copy of complex hardware systems using standard processor instructions, achieving comparable performance with reduced complexity.
3Productivity
If more bandwidth is increased for higher data-rate transmission, then data-rate improves, but spectral limitations and cost increase
Solution Approach 1:
The patent applies MIMO technology to add a spatial dimension to communication. Instead of increasing bandwidth to improve data-rate, it uses multiple transmit and receive antennas to create independent spatial channels, enabling higher data-rates within the same spectral resources through spatial diversity and multiplexing.
Data Source
AI summary
A digital signal processor is provided in a wireless communication device, wherein the processor comprises a vector unit, first and second registers coupled to and accessible by the vector unit; and an instruction set configured to perform matrix inversion of a matrix of channel values by coordinate rotation digital computer instructions using the vector unit and the first and second registers.