FIR-IIR Waveform Equalization for Burst Errors and Gaussian Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Waveform equalizing devices face challenges in achieving high-precision equalization for signals with Gaussian noise and burst errors, and they tend to be large in circuit area due to the need for multiple components and path memories, leading to error propagation and increased size.
Innovation Solution
A waveform equalizing device configuration that includes a finite impulse response (FIR) filter, decision feedback equalization (DFE) section with slicers, an infinite impulse response (IIR) filter with multiple delay portions, and a tap coefficient updating section, which performs convolution operations and updates coefficients based on error detection to achieve high-precision equalization while minimizing circuit area and delay.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a trellis decoder with traceback path is used to reduce error propagation, then waveform equalization performance is improved for signals with Gaussian noise, but error propagation occurs when processing signals with burst errors, causing performance degradation
Solution Approach 1:
The patent divides the equalization process into multiple independent paths: a first path using a trellis decoder for Gaussian noise scenarios, and a second path using a DFE for burst error scenarios. The system selects between these paths based on signal conditions, preventing error propagation in the DFE path when processing burst errors while maintaining performance on Gaussian noise signals.
Solution Approach 2:
The patent implements dynamic switching between different equalization paths based on signal characteristics. The system adapts its behavior by selecting between trellis decoder mode and DFE mode depending on the type of signal interference present, allowing optimal performance across different signal conditions without being compromised by error propagation in any single path.
2Measurement precision
If additional components such as FIFO buffer, delay-imparting filter, and path memory are added to enhance trellis decoder precision, then waveform equalization precision is improved, but circuit area increases significantly
Solution Approach 1:
The patent segments the equalization function into two independent paths with different component sets. The first path uses trellis decoder components for Gaussian noise, while the second path uses DFE components for burst errors. This segmentation allows the system to achieve high precision without requiring all components in both paths simultaneously, thereby reducing overall circuit area.
Solution Approach 2:
The patent creates a multi-functional equalization device that can operate in different modes depending on signal conditions. The same hardware structure can perform trellis decoding when needed and DFE operation when needed, eliminating the need for separate dedicated circuits for each function and reducing total circuit area while maintaining precision.
3Reliability
If a DFE is coupled to multiple path memory outputs of a trellis decoder to receive firm decision results, then error propagation is reduced for signals with Gaussian noise, but the circuit area increases due to increased number of path memories
Solution Approach 1:
The patent segments the decision feedback mechanism into two separate paths: one that uses trellis decoder outputs for Gaussian noise scenarios, and another that uses DFE for burst error scenarios. This segmentation prevents the need for multiple path memories in a single DFE structure, reducing circuit area while maintaining the ability to receive firm decision results where needed.
Solution Approach 2:
The patent extracts the decision feedback function from a single integrated DFE structure and distributes it across two separate paths. The first path extracts firm decision results from the trellis decoder for Gaussian noise processing, while the second path extracts decision results from the DFE for burst error processing. This extraction reduces the number of path memories required in any single structure.
Data Source
AI summary
A waveform equalizing device for performing high-precision waveform equalization for an input signal and outputting the result as an output signal, includes: a FIR filter for performing convolution operation between the input signal and a plurality of tap coefficients; first and second slicers at least one of which makes decision on the value of the output signal; a first delay device for imparting a delay to the output of the second slicer; an IIR filter; and an addition section for summing the output of the FIR filter and the output of the IIR filter and outputting the sum as the output signal. The IIR filter includes first and second delay portions for respectively receiving the outputs of the first slicer and the first delay device and imparting delays, and performs convolution operation between the signal received by the first delay portion and tap coefficients for the first delay portion and between the signal received by the second delay portion and tap coefficients for the second delay portion.