Small-Signal Sonar Filtering with Dual-Threshold Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Sonar systems face challenges due to absorption and spreading losses that result in system noise, compromising detection accuracy, particularly in multibeam systems where sidelobes and noise distribution patterns hinder target detection and introduce false positives.
Innovation Solution
A sonar device with dual noise filters and a selector that apply different thresholds to filter sonar data, ensuring accurate target detection by preserving genuine signals while reducing sidelobes and noise, using a binary mask to select appropriate filtered data based on target presence.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single noise filter with a fixed threshold is used, then the filtering process is simple, but it cannot simultaneously preserve weak target signals and remove noise effectively
Solution Approach 1:
The patent divides the filtering system into two separate noise filters, each with different thresholds. The first noise filter uses a lower threshold to preserve weak target signals, while the second noise filter uses a higher threshold to remove noise more aggressively. This segmentation allows the system to handle different signal conditions with specialized filters, improving overall detection accuracy without requiring a single complex adaptive filter.
Solution Approach 2:
The patent implements dynamic threshold selection by using a binary mask to automatically choose between two different filtering results. The binary mask is generated based on the detected target positions, and it dynamically selects which filter output to use in each spatial location. This dynamic approach allows the system to adapt to local signal conditions without requiring complex real-time threshold adjustment mechanisms.
2Object-affected harmful factors
If a high detection threshold is used to remove noise, then noise reduction is improved, but weak target signals are lost
Solution Approach 1:
The patent applies partial filtering actions by using two different thresholds - a lower threshold for the first filter that preserves weak signals, and a higher threshold for the second filter that removes more noise. The binary mask selectively applies each filtering level where appropriate, ensuring that weak target signals are not lost while still achieving effective noise reduction in regions where targets are absent.
Solution Approach 2:
The patent applies different filtering characteristics to different spatial locations through the binary mask mechanism. In regions where targets are detected (indicated by the binary mask), the first filter with the lower threshold is applied to preserve weak signals. In regions where no targets are present, the second filter with the higher threshold is applied to maximize noise removal. This local differentiation allows optimal noise suppression without sacrificing weak target detection.
3Measurement precision
If a low detection threshold is used to identify small objects, then detection sensitivity is improved, but false-positive noise targets increase
Solution Approach 1:
The patent introduces a binary mask as an intermediary element that mediates between the two filtering results. The binary mask is generated based on target detection and serves as a selector that determines which filter output to use in each spatial location. This intermediary mechanism allows the system to use the sensitive first filter where targets are present while using the more restrictive second filter elsewhere, thereby reducing false positives without sacrificing detection sensitivity.
Solution Approach 2:
The patent segments the filtering output into two distinct paths with different threshold characteristics. The first filtering path uses a low threshold to maintain high sensitivity for small objects, while the second filtering path uses a high threshold to suppress false positives. The binary mask segments the final output by selecting from these two paths based on local target presence, effectively separating the detection sensitivity function from the false-positive suppression function.
4Object-affected harmful factors
If conventional sidelobe filtering is applied, then sidelobe suppression is improved, but genuine target signals near the sidelobe limit are distorted
Solution Approach 1:
The patent implements dynamic filter selection based on the binary mask that indicates genuine target positions. In regions where the binary mask detects genuine targets, the system dynamically selects the first filtering result with the lower threshold, preserving signal integrity. In regions where no genuine targets are present, the system dynamically selects the second filtering result with the higher threshold, providing stronger sidelobe suppression. This dynamic adaptation prevents distortion of genuine targets while maintaining effective sidelobe filtering.
Solution Approach 2:
The patent applies different filtering characteristics to different spatial locations based on target presence. At locations where genuine targets are detected (indicated by the binary mask), a milder filtering approach is applied to preserve signal integrity. At locations where no genuine targets are present, a more aggressive filtering approach is applied to suppress sidelobes. This local differentiation ensures that genuine target signals are not distorted while still achieving effective sidelobe suppression in appropriate regions.
Data Source
AI summary
A sonar device including reception circuitry and processing circuitry is disclosed. The reception circuitry acquires sonar data from one or more targets in an underwater environment. The processing circuitry includes a target detector, a first noise filter, a second noise filter, and a selector. The target detector generates a binary mask that indicates presence or absence of target based on the sonar data. The first and second noise filters receive the sonar data, perform first and second noise filtering on the sonar data to reduce noise in the sonar data respectively, and output first and second filtered sonar data respectively. The first noise filtering is based on a first threshold. The second noise filtering is based on a second threshold different from the first threshold. The selector selects one of the first filtered sonar data and the second filtered sonar data as output based on the binary mask.