Wavelet Bandpass Sampling for Sparse Signal Recovery Without Aliasing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing compressed sensing methods face challenges in efficiently sampling sparse multi-band signals at low rates, particularly due to aliasing issues and high power consumption, especially when dealing with wideband signals and interference outside the signal's support.

Innovation Solution

A non-uniform wavelet bandpass sampling method that projects signals onto waveforms from a Gabor or wavelet frame at a lower bandpass sampling rate, followed by non-uniform sampling, which reduces the sampling rate while minimizing aliasing and noise interference by selecting waveforms based on the signal's bandwidth characteristics.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If non-uniform sampling is used to reduce sampling rate, then power consumption is reduced and sampling rate approaches information rate, but aliasing occurs and signal recovery is impeded

Engineering Contradiction:
Improvesampling rateVSAvoidsignal recovery accuracy
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The signal is pre-processed through projection onto waveforms from a Gabor or wavelet frame before sampling. This preliminary transformation concentrates the signal energy into a fewer number of coefficients, creating a sparse representation that can be sampled at lower rates without losing critical information, thus preventing aliasing while enabling reduced sampling rates

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the representation parameters of the signal by transforming it from the time domain to a time-frequency domain using wavelet or Gabor frames. This parameter transformation allows the signal to be represented in a sparse form where only a few coefficients carry significant information, enabling sub-Nyquist sampling while maintaining signal recovery accuracy

Inventive Principle:
Principle #35Parameter changes

2Reliability

If wideband signal sampling is performed at Nyquist rate, then complete signal capture is achieved, but power consumption increases

Engineering Contradiction:
Improvesignal capture completenessVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent transforms the sampling approach by changing from uniform time-domain sampling to non-uniform sampling in the time-frequency domain. By projecting the wideband signal onto wavelet or Gabor frames, the signal representation is changed to a sparse form where only essential frequency components need to be captured, reducing the effective sampling rate and thus power consumption while maintaining complete signal capture

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

Before sampling, the signal undergoes preliminary projection onto a dictionary of waveforms. This pre-processing step identifies and concentrates the essential signal energy into specific time-frequency atoms, allowing the system to focus sampling resources only on the relevant signal components rather than uniformly sampling the entire wideband spectrum, thereby reducing power consumption

Inventive Principle:
Principle #10Preliminary action

3Use of energy by moving object

If sampling rate is reduced below Nyquist frequency, then power consumption is reduced, but aliasing occurs

Engineering Contradiction:
Improvepower consumptionVSAvoidaliasing
Core Design Contradiction:
Use of energy by moving objectVSObject-affected harmful factors

Solution Approach 1:

The patent changes the sampling parameters by moving from uniform time-domain sampling to non-uniform sampling in the time-frequency domain. The projection onto wavelet or Gabor frames transforms the signal into a representation where aliasing is minimized because the sampling occurs in a transformed domain that better preserves signal structure even at reduced rates

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The signal is pre-transformed into a sparse representation before sampling. This preliminary action of projection onto a dictionary concentrates signal energy into fewer coefficients, creating a representation that is more robust to undersampling. The sparsity induced by the transformation prevents aliasing because the essential signal information is concentrated in specific coefficients that can be captured even at sub-Nyquist rates

Inventive Principle:
Principle #10Preliminary action

4Productivity

If non-uniform sampling is used, then sampling rate is reduced, but timing jitter sensitivity increases

Engineering Contradiction:
Improvesampling rate reductionVSAvoidtiming jitter sensitivity
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent changes from time-domain sampling parameters to time-frequency domain sampling parameters. By projecting the signal onto wavelet or Gabor frames with specific time-frequency localization properties, the sampling becomes less sensitive to timing jitter because the waveforms provide inherent time-frequency resolution that can tolerate small timing variations without significant loss of measurement accuracy

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3319236B1Method of non-uniform wavelet bandpass sampling
Publication Date: 2020.01.29 COMMISSARIAT A LENERGIE ATOMIQUE ET AUX ENERGIES ALTERNATIVES
  • EP3319236B1 patent drawingFigure 1
  • EP3319236B1 patent drawingFigure 2
  • EP3319236B1 patent drawingFigure 3

AI summary

The invention relates to a compressed sensing method based on non-uniform wavelet bandpass sampling. A K-sparse signal of interest is projected onto a sequence of waveforms succeeding one another at the bandpass sampling rate, the waveforms belonging to an overcomplete dictionary, the parameters of the waveforms depending on the characteristics of the bands of the signal. The correlation values are then non-uniformly sampled to provide a compressed representation of the signal.