High-precision Acoustic Gunshot Localization System

Background: 

In public safety and emergency response, rapidly and accurately identifying the origin of gunfire is critical. Our acoustic gunshot localization system, built on multi-microphone arrays and advanced time-difference-of-arrival (TDOA) algorithms, can perform spatial localization in complex urban environments and lock onto a gunshot source within seconds, providing powerful technical support for law enforcement and security teams.

In the realm of public safety and emergency response, shooting incidents rank among the most urgent and high-risk emergencies. The ability to detect and pinpoint a gunshot source as quickly as possible directly affects the response speed and effectiveness of security forces. Traditional methods struggle to rapidly locate gunshot origins and cannot make fast, reliable determinations of source location—placing significant pressure on security and counter-terrorism operations.

Why Do We Need Acoustic Gunshot Localization?

Gunshots are instantaneous, high-intensity pulse signals, typically accompanied by strong sound pressure and distinct spectral features. In urban areas, campuses, or large venues, relying solely on human hearing for sound source identification often results in large errors and slow response times. By leveraging distributed microphone arrays and intelligent algorithms for spatial localization and real-time alerts, response speed and accuracy can be significantly improved.

With the ability to effectively detect gunfire within an 800-meter range, HERTZINNO’s Airsonar gunshot localization technology offers the following advantages over traditional methods:

• Independent of light and weather: Operates reliably day or night, in any weather conditions.

• Wide coverage: Configurable for different ranges; multi-point deployment enables a large-scale monitoring network. A single device integrates a 16-channel circular microphone array, capable of rapidly localizing gunshots and explosions in full 360° coverage.

• High real-time performance: From sound acquisition to alert triggering, latency can be less than 1 second.

• Easy deployment: Compact design with edge-computing architecture supports both standalone and clustered deployment.

• Seamless integration: Compatible with ONVIF protocol, enabling linkage with mainstream PTZ cameras to create an integrated audio-visual detection solution.

  

Technical Principles:

The core of acoustic gunshot localization lies in Time Difference of Arrival (TDoA) and beamforming algorithms.

When a gunshot propagates through space, microphones positioned at different locations capture the signal at slightly different times. By accurately calculating these time differences, the system can reverse-engineer the geometric location of the sound source.

CRNN Deep Learning Algorithm

The sonar system employs one of the most robust and reliable methods in array signal processing—beamforming technology. In parallel, it integrates a CRNN (Convolutional Recurrent Neural Network)-based deep learning algorithm for gunshot recognition, enabling precise detection and localization.

In real-world deployment, beyond rapidly identifying and localizing gunshots, the system must handle diverse background noises such as human activity, vehicles, horns, and whistles. To accurately distinguish gunfire and explosions from irrelevant sounds, the system applies state-of-the-art acoustic feature extraction techniques. These include Mel-scaled spectral features (closely aligned with human auditory perception), combined with parameters such as sound variance, formants, harmonicity, frequency, and kurtosis to form a multi-dimensional feature set.

By training on these features with a deep-learning CRNN network, the system achieves a high recognition rate for gunshot events, ensuring both accuracy and reliability in complex environments.

The figure below shows the training loss and validation loss of the CRNN neural network based on deep learning.

Beamforming

Airsonar uses a noise-recognition system based on a controlled beamforming technique that maximizes output power. The system is built with a modular design to facilitate functional expansion and is composed of modules such as calibration, data acquisition, and data analysis. The beamforming block diagram is shown in below.

The system first performs parameter configuration — adjusting settings such as the expected gunshot amplitude, duration, and saturation level as required. The sensor array then collects the acoustic signals, which are filtered and subjected to spatial scanning.

A reconstruction algorithm is used to rebuild the acoustic field on the surface of the object, producing a visualized sound-field image that enables localization and identification of the noise source.

The sonar system collects signals from each array element and generates a simulated beam pattern based on the acoustic field information. From the resulting distribution map, beams of varying intensities can be observed. The height of each beam represents the strength of the acoustic field at that point. When the highest-intensity point is correctly identified as a gunshot, it indicates the shooter’s position.

Application Scenarios

• Urban Public Safety: Assists police in rapidly pinpointing gunshot locations across wide urban areas, reducing response times.

• Campuses and Venues: Provides fast alerts during school shootings or large-scale events, ensuring personnel safety.

• Industrial Parks and Border Areas: Monitors potential threats in sensitive zones, enhancing security systems.

• Military and Critical Infrastructure Protection: Enables round-the-clock acoustic monitoring around borders, airports, or energy facilities.

Summary

HERTZINNO’s acoustic gunshot detection and localization system delivers fast, accurate, and reliable gunshot recognition using advanced microphone arrays, TDoA positioning algorithms, and deep learning technology. Unlike traditional methods, it provides real-time alerts, 360-degree coverage, and second-level localization of gunshots in complex urban environments.

This intelligent gunshot detection solution integrates seamlessly with security infrastructure, supporting ONVIF protocol and PTZ cameras, making it an ideal choice for public safety, campus security, border protection, and critical infrastructure monitoring.

By enabling acoustic surveillance and visualized sound-field mapping, HERTZINNO empowers law enforcement, emergency responders, and security teams to reduce response times, improve situational awareness, and protect lives.