Designing AI to Complement Human Tasks in Wireless Acoustic Sensor Networks for Environmental Monitoring

Abstract

This research investigates AI design to assist human operations in Wireless Acoustic Sensor Networks (WASN) for environmental monitoring. WASNs are becoming an invaluable asset for real-time data gathering and processing, particularly in applications related to illegal logging, wildlife tracking, and natural disaster surveillance, such as landslides and avalanches. With proper harvesting and processing of acoustic signals during natural phenomena, WASNs generate useful information that contributes to total knowledge of various environmental parameters. The networks are, however, plagued with some limitations like energy management, data transmission reliability, and immunity against any environmental factors. Aside from this, AI technology is also possible for storing and processing vast volumes of acoustic information in a form conducive to even more accurate event identification and characterization within the system. AI technology, along with WASN technology, is also possible to bring about more accurate decision-making, predictive monitoring, and auto-sensing with lesser human participation. This chapter explores WASN deployment techniques like network planning and data mining and uses empirical evidence to show their effectiveness. This chapter also explores future potential and future research directions in the field regarding the use of WASNs for effective, robust, and sustainable environmental monitoring.