Viterbi Decoder SNR Estimation Using Path Convergence Length
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional Viterbi decoders lack the capability to estimate current signal-to-noise ratio (SNR) values at their input, which is essential for assessing communication channel quality and optimizing decoding processes.
Innovation Solution
The proposed solution involves additional circuitry that estimates SNR by calculating the decoding path length where all paths converge and using functional dependencies of the average number of symbols mismatched with the maximum likelihood symbol, allowing for the generation of a binary control signal to reset matrices when the SNR exceeds a threshold, thereby enabling current SNR estimation independent of input signal magnitude.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional Viterbi decoding algorithm is used, then decoding function is achieved, but capability of generating current input SNR values is lost
Solution Approach 1:
The Viterbi decoder is enhanced to perform dual functions: standard decoding and SNR estimation. The path metrics calculation unit simultaneously computes decoding path metrics and accumulates squared metric values for SNR estimation, allowing one device to serve multiple purposes without requiring separate dedicated hardware for SNR measurement
Solution Approach 2:
The decoder utilizes its own internal path metrics calculation results to generate SNR estimates. By accumulating the squared path metrics within the existing decoding structure, the system serves its own measurement needs without external assistance or additional independent measurement equipment
2Measurement precision
If direct bit error rate estimation method is used, then SNR measurement is achieved, but estimation range is limited
Solution Approach 1:
The invention changes the measurement parameter from bit error rate (discrete, limited range) to path metric values (continuous, wide range). By utilizing the continuous path metrics accumulated during Viterbi decoding and applying nonlinear transformation, the system achieves SNR estimation across a much broader range while maintaining accuracy through the mathematical relationship between path metrics and SNR
Data Source
AI summary
A method of estimating signal-to-noise ratio in a Viterbi decoder comprising: setting a threshold SNR value; determining a dependence on SNR of the average decoding path length; filling branch metrics matrix, minimal path metrics matrix, path metrics matrix and paths matrix with initial values; receiving packets from a communication channel; calculating the matrices that contains paths stored during operation of Viterbi algorithm in its rows, and a minimal path metrics matrix, including calculating an estimate of a decoding path length, where all the paths converge, based on the paths matrix; calculating current SNR estimate using an estimate of a decoding path length, based on results of previous steps; setting a decoder control signal to an active state if the current estimated SNR does not exceed the threshold, and to an inactive state otherwise; if the decoder control signal is in active state, the branch metrics matrix, the minimal path metrics matrix, the paths metrics matrix and the paths matrix are filled with the initial values; generating a decoded symbol; and repeating the steps on a next packet if one is available.