Data-Independent Error Computation for Noisy Analog Front Ends
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communications and storage systems face performance limitations in analog front end components due to noise, which affect the overall system's bit error rate and throughput, especially under noisy conditions.
Innovation Solution
Implementing a data-independent error correction technique that determines signal errors associated with gain and offset without relying on instantaneous signal values, using known properties of the expected signal, and selecting between data-independent and decision-directed error correction based on signal quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If decision-directed error correction is used, then error correction accuracy is improved, but system complexity increases due to requiring detectors and error correction codes
Solution Approach 1:
The patent extracts and removes the requirement for complex detectors and error correction codes by implementing data-independent error computation. The system computes error signals using only known properties of the expected signal (such as mean and variance) rather than requiring full decision-directed processing, thereby reducing system complexity while maintaining error correction capability
Solution Approach 2:
The error correction system serves itself by using inherent statistical properties of the expected signal (mean, variance, higher-order moments) to compute error signals without external assistance from detectors or complex decoding mechanisms. The system is self-sufficient in generating correction signals through data-independent computation
2Device complexity
If data-independent error correction is used, then system complexity is reduced, but error correction performance deteriorates under noisy conditions
Solution Approach 1:
The patent changes the parameters used for error computation from instantaneous signal values to statistical properties (mean, variance, higher-order moments) of the expected signal. By using these robust statistical parameters that are less sensitive to noise, the system maintains reliable error correction performance while using simpler data-independent computation
Solution Approach 2:
The system performs preliminary computation of statistical properties (mean, variance, higher-order moments) of the expected signal in advance. These pre-computed statistical parameters are then used during error correction to maintain performance under noisy conditions without requiring complex real-time decision-making
3Reliability
If traditional error correction methods are used, then noise resilience is improved, but processing latency increases
Solution Approach 1:
The patent replaces the mechanical/sequential process of detection and decoding with a direct computational approach. Instead of using detectors that require multiple processing stages, the system directly computes error signals from statistical properties, substituting a simpler computational mechanism that reduces processing latency while maintaining noise resilience
4Measurement precision
If analog front end processing is optimized, then bit error rate is reduced, but system complexity increases
Solution Approach 1:
The patent changes the approach to analog front end optimization by using data-independent error computation based on statistical properties rather than complex adaptive filtering or equalization. This parameter change in the error computation method reduces bit error rate without requiring increased complexity in the analog front end processing
Data Source
AI summary
A signal error is determined by obtaining a known property of an expected signal. A signal is received and a signal error is determined based at least in part on the received signal and the known property of the expected signal.