Multi-Band Noise Generator for Realistic Phase Noise Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing noise generators face challenges in accurately reproducing the real phase noise spectrum over a wide frequency range with high dynamics, requiring high computational resources and memory, making it difficult to achieve realistic simulation and user-friendly operation.
Innovation Solution
A noise generator is designed with multiple noise sources and digital filters, each operating at different sampling rates, with interpolators and combiners to adjust and combine signals, optimizing filter edges for a realistic and efficient simulation across the frequency spectrum.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If IIR filters are used to generate colored noise with high dynamics, then the filtering efficiency is improved, but numerical instability and high filter order requirements worsen the reliability and ease of operation
Solution Approach 1:
The frequency spectrum is divided into multiple frequency bands, with each band processed by a separate digital filter. This segmentation allows each filter to operate within a limited frequency range, reducing the overall filter order and improving numerical stability while maintaining high filtering efficiency across the entire spectrum.
Solution Approach 2:
Each frequency band is assigned customized filter characteristics tailored to its specific requirements. The filter order, cutoff frequencies, and response characteristics are optimized locally for each band, allowing high dynamics and efficiency in each region while avoiding the numerical instability that would result from a single high-order filter covering the entire spectrum.
2Reliability
If FIR filters are used to generate colored noise, then the coefficient estimation reliability is improved, but the large number of coefficients worsens the computing time and memory requirements
Solution Approach 1:
The frequency spectrum is divided into multiple bands, each handled by a separate FIR filter with a relatively small number of coefficients. This segmentation reduces the total number of coefficients from what would be required for a single full-spectrum FIR filter, thereby reducing computing time and memory requirements while maintaining reliable coefficient estimation for each band.
3Measurement precision
If a single arbitrary generator with high sampling rate and large memory depth is used, then the frequency response accuracy is improved, but the memory requirements and calculation time worsen the resource efficiency
Solution Approach 1:
The frequency spectrum is divided into multiple bands, each processed by a separate digital filter operating at an optimized sampling rate for that band. This segmentation allows each filter to use a smaller memory depth appropriate for its specific frequency range, reducing the overall memory requirements while maintaining frequency response accuracy through proper interpolation and combination of the band-limited signals.
4Duration of action of moving object
If two noise signals with relatively prime lengths are generated and combined, then the apparent repetition rate is improved, but the autocorrelation function quality worsens and short repetition times become visible
Solution Approach 1:
Instead of using a single long-period noise signal, the frequency spectrum is divided into bands with each band having its own noise source and filter. The combination of multiple band-limited noise signals with properly designed filter edges produces a composite signal with both long effective period and good autocorrelation properties, avoiding the artifacts that arise from simply combining two short-period signals.
Data Source
Figure 1
Figure 2A
Figure 2B
AI summary
The invention relates to a noise generator for producing a noise signal over a frequency spectrum comprising a first noise source and a first digital filter for a first frequency band, a second noise source and a second digital filter for a second frequency band, an interpolator, and a combiner. The first digital filter has a first sampling rate and the second digital filter has a second sampling rate, wherein the relationship between the second sampling rate and the first sampling rate corresponds to a relationship between the center frequency of the second frequency band and the center frequency of the first frequency band with respect to a sign, wherein a roll-off curve of the second digital filter (34), which roll-off curve determines a lower frequency-band limit, is steeper than a roll-off curve of the first digital filter (24), which roll-off curve determines an upper frequency-band limit. The interpolator is designed to adapt an output signal of the first digital filter in respect of the sampling rate thereof to a sampling rate of the second digital filter, wherein the combiner is designed to combine the adapted output signal of the interpolator and the output signal of the second digital filter.