FM Modulation Detection Using Autocorrelation Under Noise
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Estimating modulation in frequency modulated radio communication is challenging, especially under weak signal conditions and in the presence of noise, as peak-based methods tend to overestimate frequency deviation, leading to suboptimal input filter settings that degrade Signal-to-Noise Ratio (SNR) and Signal-to-Noise-And-Distortion (SINAD).
Innovation Solution
The proposed solution involves calculating modulation by measuring the autocorrelation of the recovered audio signal, applying a peak hold detector, and using a leaky integrator to track clean peaks, thereby providing a more accurate estimate of signal power and rejecting noise power, which is then used to adjust the input filter bandwidth.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If peak-based modulation measurement is used, then the measurement process is simple, but the measurement precision deteriorates under weak signal conditions and noise
Solution Approach 1:
The patent extracts the useful signal component from the noisy signal by using autocorrelation to identify periodic structures. The autocorrelation function R(τ) = E[x(t)x(t+τ)] isolates the periodic signal components while suppressing random noise, allowing accurate peak detection even in weak signal conditions.
Solution Approach 2:
The patent introduces autocorrelation as an intermediary processing step between the raw noisy signal and the final modulation measurement. This intermediary transformation converts the difficult problem of peak detection in noise into a simpler problem of finding maxima in the autocorrelation function, which has enhanced signal-to-noise properties.
2Adaptability or versatility
If input filter bandwidth is widened to accommodate frequency deviation variations, then the adaptability improves, but the signal-to-noise ratio deteriorates under weak signal conditions
Solution Approach 1:
The patent implements a feedback mechanism where the measured modulation index from autocorrelation analysis is used to dynamically adjust the input filter bandwidth. The system continuously monitors the signal characteristics and adapts the filter settings in real-time, optimizing the trade-off between accommodating frequency deviation and minimizing noise admission.
Solution Approach 2:
The patent transforms the static filter bandwidth setting into a dynamic parameter that adapts to changing signal conditions. The filter bandwidth is made variable based on the measured modulation index, allowing the system to optimize performance for different frequency deviation scenarios while minimizing noise admission in each specific case.
Data Source
AI summary
An FM receiver is unaware of the modulation level (frequency deviation) of the signal and has to make an estimate of it, or some reasonable time-average of it, and accordingly set the input filter's bandwidth. We calculate modulation by measuring the autocorrelation of the recovered audio signal instead of its peaks, and then applying a peakhold detector. Since FM noise can be modeled to be somewhat uncorrelated, we can expect to get an accurate estimate of signal power while rejecting noise power substantially if we measure a one-sample delayed autocorrelation estimate. Since the above measurement is alike a power measurement, we compute its square root, gain adjust it to obtain a cleaner peak measurement, and then track these clean peaks using a leaky integrator. This gives an estimate of modulation that subdues the effect of the noise.