Dependable Monitoring with Wireless Sensor Networks in Water Environments

Continuous monitoring of aquatic environments using water sensors is important for several applications related to the prevention of accidents, to water resources and aquaculture management and recreational activities. Thus, it is fundamental to ensure the quality of the monitoring data in order to avoid false alarms or ignoring relevant events.

However, operating these sensors in the water environment presents several challenges with clear consequences on data quality. For instance, sensors are constantly being subjected to factors that directly interfere with data quality, such as potentially strong currents and debris accumulation, and communication with sensors, affected by waves and more interferences.

AQUAMON will develop a dependable monitoring platform for application in aquatic environments using wireless sensor networks, addressing some of these challenges. In particular, it will address data communication quality problems over water surfaces, due to waves and propagation characteristics over a water surface, transmission predictability, due to shared medium access contention, and data quality, caused by faults that affect both sensors and communication, creating data errors and data loss.

The following approaches will be followed to achieve the project goals. To obtain communication quality guarantees for the wireless sensor network, a higher transmission power in several operation scenarios will be used in 802.15.4 networks. To improve communication reliability, protocols will be designed and implemented to limit the amount of packet collisions, using communication synchronization techniques. For processing sensor data, the project will resort to a combination of methods, aiming at achieving enhanced data quality. On the one hand, faults will be detected using signal analysis and processing techniques, as well as inference techniques based on data series. On the other hand, using multiple redundant sensors enables the application of data fusion techniques and characterizing the quality of resulting data. Finally, collected data will also be analysed by considering specific water-process and predictive models for the dynamics of the monitored environments, which will provide additional redundancy and contribute to data quality assurance.

To illustrate the validity of the proposed solution, the platform will be tested in the Seixal bay, using the real time hydrodynamic and water quality monitoring and forecast system of the Tagus estuary, developed in the scope of the projects FP7 PREPARED, AdI SI-GeA and FCT MOLINES. This application will highlight the effectiveness of the AQUAMON platform and allow quantifying the gains relative to conventional sensor networks, using a forecast system based on the best process-based numerical models.