An Ultra-Low Energy Solution for Large-Scale Distributed Audio Monitoring

Abstract

This paper presents a flexible, low-cost, and ultra-low-energy acoustic sensor network solution. This enables both highly distributed and real-time audio measurements to be employed in a wide variety of scenarios. These types of distributed measurements can be useful in environmental monitoring, healthcare settings, smart cities, and security surveillance systems. However, they are typically inflexible in their applications and require specialized deployment. The proposed solution is nonintrusive and uses long-range radio to enable low-power distributed audio monitoring. The system can recognize and be configured for different types of audio measurement and analysis tasks (such as sound pressure level, annoyance, direction of arrival, and roughness), and is shown to be as accurate r=0.998 as a high-cost measurement system. Remote control, time of day, and grid capture mechanisms can also be realized. The proposed solution provides the ability for an over-the-air configuration, leading to minimal physical access required once deployed. In addition, higher data rates are possible, offering real-time monitoring metrics.

Keywords

• Acoustic sensor networks
• edge computing
• energy efficiency
• remote health monitoring
• smart cities