Distributed Acoustic Sensor Systems with Local Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current monitoring systems for surveillance areas are complex, logistically impractical, and require large data transmission and processing, lacking a simple and cost-effective solution for timely and spatial process monitoring.

Innovation Solution

A device with distributed acoustic sensor systems that classify audio signals locally and transmit classification results to a central evaluation device, enabling efficient data transmission and precise monitoring of activities, processes, and events in surveillance areas.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If distributed acoustic sensor systems with local classification are used, then data transmission volume is reduced and monitoring efficiency is improved, but device complexity increases due to multiple sensor systems and processors

Engineering Contradiction:
Improvemonitoring efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The monitoring system is divided into multiple independent sensor systems, each with its own processor for local audio signal classification. Each sensor system operates autonomously to classify sounds at its location, then transmits only classification results to a central evaluation device. This segmentation reduces the total data transmission volume while distributing processing tasks across multiple units, improving overall monitoring efficiency without requiring one centralized complex system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Audio signals are classified locally at each sensor system before transmission to the central evaluation device. This preliminary classification action filters out unnecessary data, so only relevant classification results are transmitted rather than raw audio signals. This reduces data transmission volume and processing burden on the central system while maintaining monitoring effectiveness.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If multiple sensor systems are distributed at different locations, then spatial monitoring capability is improved, but logistical implementation difficulty increases

Engineering Contradiction:
Improvespatial monitoring precisionVSAvoidlogistical implementation ease
Core Design Contradiction:
Measurement precisionVSEase of manufacture

Solution Approach 1:

The system uses multiple independent sensor systems that can be individually installed at different locations within the monitoring area. Each sensor system is a self-contained unit with acoustic sensor, processor, and transmission device, making them modular and easier to deploy distributed across space without requiring complex integrated infrastructure.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Each sensor system independently performs audio signal detection, local classification, and transmission of results without requiring centralized control for these functions. This self-service capability simplifies deployment logistics, as each unit operates autonomously and can be installed independently at its designated location.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If complete audio signals are transmitted to central evaluation device, then analysis accuracy is maintained, but data transmission volume and processing load increase significantly

Engineering Contradiction:
Improveanalysis accuracyVSAvoiddata transmission volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

Instead of transmitting complete audio signals, the system extracts only the essential classification results from each local sensor system. The processor at each sensor system analyzes the audio signal locally and extracts only the classification outcome (e.g., type of sound event, location, time), discarding the redundant raw audio data. This extraction maintains the necessary information for central evaluation while dramatically reducing data transmission volume.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

Audio signal classification is performed preliminarily at each distributed sensor system before transmission. This preliminary processing action filters and condenses the data, so only classification results are sent to the central evaluation device rather than complete audio recordings. This maintains analysis capability while minimizing data transmission requirements.

Inventive Principle:
Principle #10Preliminary action

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

The system provides a cost-effective, logistically feasible, and time-accurate monitoring solution with reduced data transmission, allowing for real-time analysis and spatial localization of sound events, thereby improving process monitoring efficiency.

Implementation Method 1

a first acoustic sensor (122), a first processor (124), and a first transmission device (126), which can be attached at a first location (129) in the monitoring area (110)... a first audio signal (142) detected by the first acoustic sensor (122)

Methodology Applied
Scientific EffectAcoustic detection: Sound

Data Source

PatentEP3791370B1Apparatus, method, and computer program for acoustically monitoring a monitoring region
Publication Date: 2023.07.26 FRAUNHOFER GESELLSCHAFT ZUR FORDERUNG DER ANGEWANDTEN FORSCHUNG EV
  • EP3791370B1 patent drawingFigure 1
  • EP3791370B1 patent drawingFigure 2
  • EP3791370B1 patent drawingFigure 3

AI summary

The invention relates to an apparatus (100) for acoustically monitoring a monitoring region (110), having: a first sensor system (120) with a first acoustic sensor (122), a first processor (124) and a first transmission device (126) which can be attached to a first point in the monitoring region; and a second sensor system (130) with a second acoustic sensor (132), a second processor (134) and a second transmission device (136) which can be attached to a second point in the monitoring region, said point being different from the first point. The first processor can be designed to classify a first audio signal detected by the first acoustic sensor in order to obtain a first classification result, and the second processor can be designed to classify a second audio signal detected by the second acoustic sensor in order to obtain a second classification result. The first transmission device can be designed to transmit the first classification result to a central evaluation device (150), and the second transmission device can be designed to transmit the second classification result to the central evaluation device. Furthermore, the apparatus can have the central evaluation device, wherein the central evaluation device can be designed to receive the first classification result and to receive the second classification result and to generate a monitoring output for the monitoring region on the basis of the first classification result and the second classification result.