Frequency-Domain Filtering for Channel Estimation in OFDM Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing channel estimation methods in OFDM systems require high complexity and memory due to the need for different filter coefficient vectors for each sub-carrier, especially for edge tones, leading to high implementation costs.
Innovation Solution
Implementing a single frequency-domain filter with a common filter coefficient vector that combines frequency-domain samples from neighboring sub-carriers to generate refined channel estimates for all sub-carriers, including edge tones, reducing memory requirements and implementation costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If different filter coefficient vectors are used for each sub-carrier (especially edge tones) to achieve accurate channel estimates, then measurement precision is improved, but device complexity and memory requirements increase
Solution Approach 1:
The patent combines multiple filter coefficient vectors into a single unified filter coefficient vector that is applied to all sub-carriers. This merging approach reduces the memory storage requirement from storing multiple separate coefficient vectors to storing just one, while still achieving accurate channel estimation by processing each sub-carrier's received signal through the same filter structure.
Solution Approach 2:
The single filter coefficient vector serves as a universal solution for all sub-carriers, including edge tones. Instead of having specialized filter coefficients for each sub-carrier, the same filter coefficients are used universally across all sub-carriers, reducing complexity while maintaining functionality through the filter's application to each sub-carrier's signal.
2Measurement precision
If optimal MMSE-based estimation algorithms are used to achieve good channel estimation performance, then measurement precision is improved, but device complexity and implementation cost increase
Solution Approach 1:
The patent merges the complex optimal MMSE-based estimation algorithms into a simpler implementation by using a single filter coefficient vector that replicates the effect of multiple sub-carrier-specific filters. This combining approach maintains the accuracy benefits of MMSE estimation while reducing the algorithmic complexity and memory requirements associated with storing and processing multiple separate coefficient vectors.
3Device complexity
If a single filter coefficient vector is used for all sub-carriers to reduce memory requirements, then device complexity is reduced, but measurement precision may be compromised
Solution Approach 1:
The patent merges the functionality of multiple sub-carrier-specific filter coefficient vectors into a single unified filter coefficient vector. This single vector is then applied to all sub-carriers through the filter structure, achieving both the memory reduction benefit of using a single coefficient vector and the accuracy benefit of having filtered channel estimates for each sub-carrier.
Solution Approach 2:
While using a single filter coefficient vector, the patent segments the processing by applying the filter separately to each sub-carrier's received signal. This segmentation in the processing stage allows the single filter to effectively handle each sub-carrier independently, maintaining measurement precision while using a unified coefficient set.
Data Source
AI summary
A method of channel estimation includes receiving a signal after transmission over a media having a plurality of sub-carriers in a frequency band. The signal is preprocessed including performing a fast Fourier transform (FFT) to generate a plurality of frequency-domain samples. Channel estimating is applied to the plurality of frequency-domain samples using (i) least squares (LS) estimation, wherein the LS estimation generates intermediate LS channel estimates for each of the sub-carriers, and (ii) frequency-domain filtering and scaling the intermediate LS channel estimates. The frequency-domain filtering uses a common frequency-domain filter consisting of a single filter coefficient vector having a plurality of frequency-domain filter coefficients to generate refined channel estimates for each of the plurality of sub-carriers.