Gunshot Detection Network Using Local ML and Multilateration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing gunshot detection systems face challenges with high bandwidth requirements, processing power consumption, and inaccurate location identification due to echo and obstruction issues, especially in large urban areas.
Innovation Solution
A network of recording devices mounted on light fixtures uses machine learning and multilateration techniques to process audio data locally and transmit spectrograms only when gunshots are detected, enabling precise gunshot location determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If continuously streaming audio data from each recording device to the server, then gunshot detection capability is improved, but bandwidth consumption increases significantly
Solution Approach 1:
The patent extracts and processes only the relevant information (gunshot detection results and spectrograms) from the audio data stream, rather than transmitting the entire continuous audio stream. This selective extraction significantly reduces bandwidth consumption while maintaining gunshot detection capability.
Solution Approach 2:
The recording devices perform preliminary gunshot detection and spectrogram generation locally before transmitting data to the server. This preliminary processing filters out non-gunshot audio data, reducing the volume of data that needs to be transmitted over the network.
2Reliability
If continuously analyzing audio streams from multiple recording devices, then gunshot detection accuracy is improved, but processing power requirements increase significantly
Solution Approach 1:
The patent divides the gunshot detection task into segments: local devices perform initial detection and spectrogram generation, while the server performs multilateration and final location determination. This segmentation distributes processing power requirements across multiple nodes, reducing the burden on any single device.
Solution Approach 2:
The system performs partial analysis locally (spectrogram generation and gunshot detection) and partial analysis centrally (multilateration), rather than performing complete analysis at one location. This partial distribution of processing tasks optimizes the balance between local and central processing power requirements.
3Ease of operation
If placing recording devices sporadically around the city, then deployment complexity is reduced, but location accuracy decreases due to echo and obstruction issues
Solution Approach 1:
The patent introduces spectrograms as an intermediary representation of audio data, which enhances the detectability of gunshot signals even in challenging acoustic environments with echoes and obstructions. Spectrograms transform temporal audio signals into time-frequency representations that make gunshot detection more robust.
Solution Approach 2:
The system replaces reliance on ideal acoustic propagation conditions (mechanical sound wave transmission) with signal processing techniques (spectrogram analysis and machine learning) that can compensate for acoustic distortions caused by echoes and obstructions.
4Ease of manufacture
If mounting recording devices at lower heights, then installation ease is improved, but recording accuracy decreases due to impedance from passerby objects
Solution Approach 1:
The patent moves the recording devices from the horizontal plane (ground level) to the vertical dimension (elevated mounting positions). This dimensional change allows devices to be mounted on vertical surfaces such as light poles and building facades, elevating them above ground-level obstructions while maintaining relatively simple installation procedures.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach reduces bandwidth and processing demands while accurately and quickly notifying authorities of gunshot locations, allowing for timely intervention and evidence collection.
Implementation Method 1
A microphone inside or mounted to the device housing
Implementation Method 2
The sound waves of gunshots can be loud and can ccho off of buildings surrounding the streets in a city
Data Source
AI summary
A system may include a first computing device comprising a first processor configured to receive a set of audio data from a microphone; execute a first sound detection machine learning model using the set of audio data as input to determine whether the set of audio data corresponds to an actionable sound, the set of audio data comprising a plurality of sounds; and responsive to determining the set of audio data corresponds to the actionable sound, transmit the set of audio data to a second computing device. The system may include the second computing device comprising a second processor configured to execute a time detection machine learning model using the set of audio data as input to determine an instance and time of the actionable sound within the set of audio data; and determine a location of the actionable sound based at least on the time of the actionable sound.