Measures of spatial autocorrelation like Moran’s I do not take into account information about the reliability of observations. In a context of mobile sensors, however, this is an important aspect to consider. Mobile sensors record data asynchronously and capture different contexts, which leads to considerable heterogeneity. In this paper we propose two different ways to integrate the reliability of observations with Moran’s I. These proposals are tested in the light of two case studies, one based on real temperatures and movement data and the other using synthetic data. The results show that the way reliability information is incorporated into the Moran’s I estimates has a strong impact on how the measure responds to volatile available information. It is shown that absolute reliability information is much less powerful in addressing the problem of differing contexts than relative concepts that give more weight to more reliable observations, regardless of the general degree of uncertainty. The results presented are seen as an important stimulus for the discourse on spatial autocorrelation measures in the light of uncertainties.

Estimation of Moran's I in the Context of Uncertain Mobile Sensor Measurements.

Along with the rise of the smart city movement, Internet of Things is an upcoming phenomenon. Objects and devices are becoming more and more wirelessly interconnected, communicating information between themselves and to human beings. As an addition to static sensor networks that gather real-time environmental data, the feasibility of implementing a dynamic sensor network based on LoRa communication is researched. To achieve such a dynamic system, a self-developed sensor platform was constructed, based on the microcontroller LoPy, measuring temperature and humidity. The emphasis of the research is on the localisation of the sensor platforms. A WiFi fingerprinting radiomap was constructed based on available MAC-addresses, their signal strengths, and GPS coordinates. In this method the GPS module is only used for the composition of the radiomap. The quality of the radiomap methodology was assessed by constructing it of measurements gathered in four days, and testing it for the remaining three days. This test gave a correctness of 50% while another 38% of measurements were localised in a neighbouring cell. The quality of the collected sensor data turned out to be dependent on the weather conditions and the placement location on the carrier vehicle. Another topic of research was LoRa communication, which was deemed as very limited for dynamic implementations, as the sending of location-related data takes up a large part of the already limited message size.

/pdf/using-a-dynamic-sensor-network-to-obtain-spatiotemporal-data-519w431h3j.pdf

Using a Dynamic Sensor Network to Obtain Spatiotemporal Data in an Urban Environment

Along with the rise of the smart city movement, Internet of Things is an upcoming phenomenon. Objects and devices are becoming more and more wirelessly interconnected, communicating information between themselves and to human beings. As an extension on static sensor networks that gather real-time environmental data, the feasibility of implementing a dynamic sensor network based on LoRa communication is researched. To achieve such a dynamic system, a self-developed sensor platform was constructed, based on the microcontroller LoPy. Sensors attached to it include a hygrometer, thermometer and microphone. The emphasis of the research was on localisation of the sensors, to put the gathered sensor data into geographical context. A WiFi fingerprinting radiomap was constructed based on available MAC-addresses, their signal strengths, and GPS coordinates. The GPS module was only used for composing the radiomap. When the radiomap is completed, the module can be switched off, only to be switched on for periodical updates of the radiomap. The quality of the radiomap methodology was evaluated by constructing it of measurements gathered in four days, and testing it for the remaining three days. This test gave a correctness of 50% while another 38% of measurements were localised in a neighbouring cell. The correctness can be improved by having a longer training period. The quality of the collected sensor data turned out to be dependent on the weather conditions and the placement location on the carrier vehicle. Vehicle requirements were specified as driving through the city centre and having a schedule and route producing as little noise, heat and air pollution as possible. Another topic of research was LoRa communication, which was deemed as very limited for dynamic implementations, as the sending of location-related data takes up a large part of the already limited message size. To decrypt the sent message and store it in a meaningful database, Node-RED was used. Despite visualisation of measurements showed promising results, there is margin for improvement as far as data capturing is concerned.

Puck Flikweert

Papers

Using a Dynamic Sensor Network to Obtain Spatiotemporal Data in an Urban Environment

DynamIoT - Geomatics Synthesis Project on IoT: Using a dynamic sensor network to obtain spatiotemporal data in an urban environment