Wideband Spectrum Sensing Using Wavelet Edge Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Cognitive radios face challenges in identifying frequency bands in a large frequency range without known center frequencies and bandwidths, as conventional multiband sensing is not applicable, and existing methods struggle with noise uncertainty and sharp changes within occupied bands.
Innovation Solution
The method involves calculating a wavelet product of the received signal, setting a threshold for local maxima, and detecting edges to identify frequency bands, using a multi-scale wavelet transform and energy comparison to classify bands as occupied or vacant, without requiring prior information on signal or noise power.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiband sensing is used with known band policies, then sensing can be performed efficiently in predefined bands, but it becomes invalid when signals can be at any center frequency with any bandwidth
Solution Approach 1:
The patent transforms the sensing approach from fixed band parameters to variable parameter detection. By using wavelet transform with multiple scales, the system can detect signals with any center frequency and bandwidth, adapting to unknown band policies while maintaining sensing effectiveness through parameter-based analysis rather than fixed channel structures.
Solution Approach 2:
The wavelet-based wideband sensing method serves as a universal sensing mechanism that encompasses both multiband sensing (when band policies are known) and detects arbitrary signals (when policies are unknown). The same wavelet transform framework adapts to different scenarios, making the system multi-functional without requiring separate sensing mechanisms.
2Ease of operation
If conventional multiband sensing is applied, then sensing is simplified for predefined bands, but it fails to detect signals with unknown center frequencies and bandwidths
Solution Approach 1:
The wavelet transform acts as an intermediary tool that bridges the gap between simple multiband sensing and complex arbitrary signal detection. By transforming the signal into the time-frequency domain, the wavelet method provides a middle ground that maintains operational simplicity while significantly improving detection reliability for signals with unknown parameters.
3Adaptability or versatility
If wideband sensing is performed without prior information, then the system can detect arbitrary signals, but it faces challenges with noise uncertainty and sharp changes within bands
Solution Approach 1:
The wavelet transform inherently segments the signal analysis into different scales and frequency bands. This segmentation allows the system to handle noise uncertainty by analyzing different frequency components separately and to detect sharp changes within bands by examining local maxima at various scales, thereby reducing the overall detection difficulty.
Solution Approach 2:
The patent moves the analysis from a single frequency dimension to a two-dimensional time-frequency space using wavelet transform. This dimensional change provides additional information about signal localization and structure, making it easier to distinguish actual signals from noise and to detect sharp transitions within bands by examining the time-frequency representation.
Data Source
AI summary
According to embodiments of the present invention, a method for identifying one or more frequency bands in a received signal. The method comprising: calculating a wavelet product of the received signal; setting a threshold for local maxima of the wavelet product; and detecting one or more edges in the received signal based on local maxima of the wavelet product which are greater than the threshold to identify one or more frequency bands in the received signal. A corresponding apparatus and computer program product are also provided. Embodiments also relate to a method for evaluating performance of a method for identifying one or more frequency bands in a received signal. A corresponding apparatus and computer program product are also provided.