Heterogeneous Wireless Sensor Network for Real Time Remote Monitoring of Sand Dynamics on Coastal Dunes

In this paper the authors describe the architecture of a heterogeneous Wireless Sensor Network (WSN) to be deployed on coastal sand dunes: the aim is to provide real time measurements of physical parameters to better define sediment transport in connection with aeolian processes. The WSN integrates different typologies of sensors and is provided with both local and remote connection. In particular, three different typologies of sensors are integrated in the network: a multilayer anemometric station, a sensor developed ad-hoc to measure the sand dune level and a sand collector capable of measuring the weight of trapped sand and its quantity. Each sensor node is composed at least by an XBee Series 2 transmission module that is able to transmit the data collected by the sensor at a distance of about 100 meters: while the sand level sensor and the sand collector are provided only with this transmission module, the anemometric station also integrates an Arduino Uno board in charge of data processing. A Gateway node composed by an Arduino Uno Board integrated with a GMS Shield for remote data transmission and an XBee transmission module for Local Area communication has also been developed: this node is in charge of collecting all the data packets sent by the Sensor Nodes and transmit them to a remote server through GPRS connection. A Glassfish server has been set up to collect these packets and store them in a MySQL database. The anemometric station is composed by three anemometer/anemoscope couples positioned 40cm, 120cm and 200cm from the ground. During the tests, the station data were sampled every 20 minutes, wind speed and direction was calculated directly on the Arduino Uno Board and then a packet composed by the six data (three speeds and three directions) was sent to the Gateway. The sand level sensor is composed by an array of 24 photo resistors (LDRs) mounted on a plastic tube 5cm apart from each other (reaching a total length of 120cm). Sunk LDRs do not sense sun light and send a 0 value. Surfacing LDRs detect sunlight and send a higher value. By counting sunk LDRs it is possible to measure the current level of the dune. During the tests, the sensor was sampled once per hour, three data packets were sent every time, each packet with the reading of 8 LDRs; the level value was calculated on the Gateway before being transmitted to the remote server. The sand collector is mainly a plastic cylinder about one meter high, able to orientate according to the wind direction. The wind-blown sand flows inside the cylinder and is collected on its bottom, where the load cell can measure weight variations. During the tests, the sensor was sampled once per hour, the value of the Load Cell was transmitted to the Gateway that calculated the sand weight before transmitting this value to the remote server. The proposed WSN can provide both a static and a dynamic framework of sand transport processes acting on coastal dunes.