Radar Signal Detection via Spectrogram Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Radar signal detection in broadband, noisy received signals is hindered by high bandwidth and long detection times, exceeding conventional processor performance limits, necessitating real-time signal processing.
Innovation Solution
The method involves determining a spectrogram of the input signal using an analysis filter bank, selecting sub-frequency and sub-time ranges based on specific criteria such as signal energy, noise ratio, and interference identification, and transforming these ranges using a synthesis filter bank to generate an output signal, while logically linking events and checking for physical correctness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the full bandwidth spectrogram is processed in real time, then complete signal detection capability is maintained, but processor performance limits are exceeded
Solution Approach 1:
The patent divides the full bandwidth spectrogram into multiple sub-frequency ranges and further into sub-time ranges, creating a segmented time-frequency representation. This segmentation allows parallel processing of smaller data chunks, reducing the computational burden on the processor while maintaining comprehensive signal detection capability across the entire bandwidth.
Solution Approach 2:
The patent extracts and processes only those sub-frequency ranges and sub-time ranges that contain relevant signal information, excluding regions with only noise or interference. This selective extraction reduces the amount of data requiring real-time processing while preserving detection reliability for actual radar signals.
2Reliability
If the detection period is extended to improve detection accuracy, then signal detection reliability is improved, but real-time processing capability is lost
Solution Approach 1:
The detection period is divided into multiple sub-time ranges, allowing the system to process shorter time segments in parallel. This maintains real-time processing capability while achieving detection accuracy equivalent to longer integration periods through coherent integration across multiple segments.
Solution Approach 2:
The patent transitions from time-domain processing to time-frequency domain processing by creating a spectrogram. This dimensional transformation allows detection decisions to be made based on energy distribution across both time and frequency, improving detection accuracy for periodic signals without extending the overall detection period.
3Reliability
If all sub-frequency ranges are processed, then complete signal coverage is achieved, but processing complexity increases
Solution Approach 1:
The patent identifies and extracts only those sub-frequency ranges that contain potential signal content, excluding frequency regions dominated by noise or known interference patterns. This selective processing maintains complete signal coverage for relevant frequencies while significantly reducing processing complexity by eliminating unnecessary computational operations in irrelevant frequency bands.
Data Source
Figure 1
Figure 2
AI summary
The method involves determining a spectrogram of an input signal in a certain time-frequency range. The spectrogram of the input signal is selected in an under frequency range and under time range of the time-frequency-range. The output signal is determined from the selected spectrogram. An independent claim is also included for a device for generating a reduced output signal against an input signal in a time range and simultaneously in a frequency range.