Soft-Bit LLR Normalization for Finite-Bit FEC Decoding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in properly quantizing log-likelihood ratio (LLR) values for forward error correction (FEC) decoding, as these values have a wide range due to varying signal-to-noise ratios (SNR) and channel gains, leading to performance losses in FEC decoding.
Innovation Solution
The method involves normalizing LLR values using an average SNR to confine their range, allowing for efficient finite-bit representation, which is applicable to both OFDM and single-carrier systems, and includes a process of calculating normalized LLR values and subsequent quantization for effective decoding.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If LLR values are calculated using conventional methods, then soft bit information is obtained for FEC decoding, but the values have a wide range due to varying SNR and channel gains, leading to performance loss in decoding
Solution Approach 1:
The patent applies parameter changes by normalizing LLR values using the average SNR of the channel. The normalization factor is calculated as 1/(average SNR), and each LLR value is multiplied by this factor to confine its range. This transforms the LLR values from having a wide range dependent on channel conditions to having a confined range suitable for finite-bit quantization, thereby resolving the contradiction between maintaining decoding reliability and achieving quantization accuracy.
2Productivity
If LLR values are quantized with finite bits, then efficient representation is achieved, but performance loss occurs due to the wide range of LLR values
Solution Approach 1:
The patent changes the parameter scale of LLR values through normalization by average SNR. This parameter transformation allows the LLR values to be represented efficiently with finite bits (e.g., 4 bits) while maintaining decoding accuracy, because the normalized values are confined to a predictable range that can be accurately quantized without significant information loss.
Solution Approach 2:
The patent performs preliminary normalization of LLR values before quantization. By pre-processing the LLR values to confine their range using the average SNR calculation, the system prepares the data in advance for efficient finite-bit representation, ensuring that subsequent quantization does not cause performance loss.
3Manufacturing precision
If LLR values are normalized using average SNR, then the range is confined and finite-bit representation is enabled, but additional calculation steps are required
Solution Approach 1:
The patent implements self-service by calculating the average SNR from the received signal characteristics and using it to normalize the LLR values. The system uses its own received signal measurements to determine the normalization factor, eliminating the need for external calibration or complex lookup tables, thereby achieving precise LLR representation with relatively simple processing.
Data Source
AI summary
A method of generating normalized bit log-likelihood ratio (LLR) values. A signal is received in a frequency band after transmission over a media, wherein the signal includes at least one complex data symbol having a plurality of information bits, and the complex data symbol is transmitted on at least one frequency channel. Initial LLR values are calculated for each of the plurality of information bits based on bit-to-symbol mapping of modulation and noise variance information from the complex data symbol. An average signal to noise ratio (SNR) of the frequency channel is calculated. Each initial LLR value is normalized by dividing by the average SNR to generate a plurality of normalized LLR values. The normalized LLR values may be quantized to provide a finite-bit representation.