InWaterSense Dataset: Data from a wireless sensor network on water quality monitoring in a river in Kosovo
The dataset presented here contains data gathered as part of the InWaterSense (inwatersense.uni-pr.edu) project implemented by University of Prishtina supported by the IPA EU grant. The water quality monitoring in Sitnica river in Kosova is performed through a Wireless Sensor Network (WSN) which supports remote, continuous and real-time measurements for the water quality parameters through its corresponding static sensors. These data are those emitted by static sensors of the WSN installed in the river bank Sitnica located in village Plemetin near Kosovo’s capital city Prishtina. Measurements in two points are performed: sensing node 1 (housing) and sensing node 2 (manhole) in a distance of around 100 m from each other. The coordinates of each of the two sensing nodes are given in Table 1. The measurement period covered is almost eight months of continuous measurements in total starting May 2015 to beginning of January 2016. The frequency of measurements is configured to be real-time in intervals of every 10 minutes. Measured data are transmitted via GPRS to the remote server at University of Prishtina. The raw dataset contains among others the location and the time of a given observation, followed by the value of the single individual parameter observed in the same raw in a structure similar to (location, timestamp, parameter, value). The water quality parameters measured include temperature, electrical conductivity, pH, and dissolved oxygen. Raw data are also available in another structure better suited for performing analysis: per given location and timestamp, there are values of all parameters observed listed in the same raw in separate individual columns instead of being spread among separate rows (rows-to-columns transformation performed on parameters). The transformed dataset is of structure similar to (location, timestamp, parameter 1, parameter 2, .., parameter n), namely (Node Id, Timestamp, Timestamp as DateTime, Temperature, Conductivity, pH, DissolvedOxygen). The dataset is useful for the broad scientific community, from environmental engineering to artificial intelligence to the health sector just to mention few. Moreover, practitioners might benefit from this dataset in driving forth the pollution prevention policies and techniques.