Per-Tone Equalizer with Adaptive Filter Decimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current multi-carrier systems face challenges with intersymbol and intercarrier interference (ISI-ICI) due to varying delays in the transmission path, which complicates equalization and increases complexity, especially with the advent of newer standards like ADSL+ that double the number of tones, leading to increased computational load and memory requirements.
Innovation Solution
The proposed solution involves an equalizer that adaptively filters multi-carrier signals by applying decimation to difference terms and dynamically adjusting the filter size based on noise measurements for each carrier or group of carriers, using Fourier transformation and recursive least squares (RLS) data structures to optimize performance while reducing complexity, allowing for per-tone or per-group equalization and decimation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If per-tone equalization is applied to each carrier individually, then equalization performance is improved, but device complexity increases significantly
Solution Approach 1:
The patent merges the equalization of multiple carriers into a single unified filter structure. Instead of implementing separate equalizers for each carrier tone, the invention combines them into one filter that processes multiple carriers simultaneously, thereby maintaining per-carrier equalization performance while dramatically reducing the number of filters and computational operations required.
Solution Approach 2:
The patent creates a universal filter structure that can equalize multiple carrier tones with a single device. This multi-functional filter replaces numerous specialized per-tone equalizers, allowing one filter to perform the work of many while reducing overall system complexity and resource requirements.
2Productivity
If the number of carriers is doubled in ADSL+, then transmission capacity is improved, but memory requirements and processing power increase
Solution Approach 1:
The patent combines the processing of doubled number of carriers into a single filter structure, reducing memory requirements by eliminating the need for separate filter coefficients for each carrier. The unified filter uses shared memory resources to store coefficients that serve multiple carriers, thereby supporting increased transmission capacity without proportionally increasing memory requirements.
3Productivity
If the number of carriers is doubled in ADSL+, then transmission capacity is improved, but processing power requirements increase
Solution Approach 1:
The patent merges multiple carrier processing operations into a single filter computation, reducing the total number of processing operations required. By combining the equalization functions into one filter, the system processes doubled transmission capacity with approximately the same processing power as the original system, rather than requiring double the processing power.
Data Source
AI summary
An equalizer for a multi carrier signal for carrying out equalization adapted to each carrier or group of carriers, Fourier transforms the multi carrier signal, and obtains difference terms of the multi carrier signal. Both are input to an adaptive filter, to output equalized signals, wherein decimation is applied to at least some of the difference terms input to the filter. This enables the complexity to be reduced for a given performance level. In particular since only non-zero filter taps need to be stored and updated, coefficient memory and coefficient calculation capacity can be reduced. Another way to reduce complexity involves measuring noise for at least some of the carriers, and dynamically adapting the size of the filter on a per carrier basis according to global optimizion heuristic algorithms which adapt this filter size based on the comparison between this noise and an optimal performance figure achieved in a previous ISI-ICI free measurement phase.