Sample Selector Interpolation for Low-Complexity Frequency Offset Resampling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current time domain interpolation methods for compensating frequency offsets in communication systems, such as ADSL, are computationally heavy and inefficient due to the use of polynomial interpolators, which are complex and require high processing speeds, leading to system performance degradation.
Innovation Solution
A signal resampler that employs a combination of a sample selector interpolator and a polynomial interpolator, with oversampling and selective sampling to reduce complexity and increase performance, allowing the polynomial interpolator to operate over a smaller time interval and reducing its complexity or increasing its accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a polynomial interpolator is used for time domain interpolation to compensate frequency offset, then the frequency offset compensation accuracy is improved, but the computational complexity increases
Solution Approach 1:
The patent divides the time domain interpolation process into two distinct stages: a sample selector stage that performs initial sample selection based on frequency offset estimation, and a polynomial interpolator stage that refines the interpolation. This segmentation allows each stage to be optimized independently, reducing the overall computational complexity while maintaining accuracy.
Solution Approach 2:
The patent applies partial action by using the sample selector to perform the majority of the interpolation work for integer sample delays, and only using the polynomial interpolator for the remaining fractional sample delay compensation. This partial application of the complex polynomial method reduces computational burden while maintaining sufficient accuracy.
2Reliability
If a polynomial interpolator operates at high processing speed to maintain synchronization, then the synchronization performance is improved, but the device complexity increases
Solution Approach 1:
The patent segments the synchronization function into frequency offset estimation, sample selection, and polynomial interpolation components. This allows the system to maintain synchronization reliability through coordinated operation of simpler subsystems rather than requiring a single complex high-speed polynomial interpolator.
Solution Approach 2:
The sample selector performs preliminary sample selection based on estimated frequency offset before the polynomial interpolator processes the signal. This preliminary action reduces the workload on the polynomial interpolator and allows the system to maintain synchronization without requiring the interpolator to operate at maximum processing speed.
3Measurement precision
If oversampling is used to improve interpolation accuracy, then the measurement precision is improved, but the quantity of data to be processed increases
Solution Approach 1:
The patent applies partial action by using oversampling selectively only for the sample selection stage, not for the entire interpolation process. This allows the system to gain the accuracy benefits of oversampling where most needed (in sample selection) while avoiding the computational burden of processing oversampled data through the entire signal chain.
Solution Approach 2:
The patent extracts and processes only the necessary samples from the oversampled data stream for the polynomial interpolation stage, rather than processing all oversampled data. This extraction approach maintains interpolation accuracy while reducing the quantity of data that requires complex processing.
Data Source
AI summary
A signal resampler carries out a time domain interpolation of an input signal for compensating for frequency offset, such as found in an ADSL system. A sample selector interpolator carries out part of the interpolation and a second, e.g. polynomial interpolator carries out the rest of the interpolation. The time interval between samples being interpolated, can be effectively divided between a sample selector interpolator and a small second, e.g. polynomial interpolator. The complexity of the second, e.g. polynomial interpolator can be reduced or its accuracy increased if it is effectively interpolating over a much smaller time interval. The sample selector interpolator can be an oversampling arrangement, and enable the order of the second, e.g. polynomial interpolator to be reduced. Selected ones of the oversampled samples are fed to the second, e.g. polynomial interpolator to keep the operating frequency lower. A chain of upsamplers can be used to generate the oversampled samples.