Advancement Through Research
The reliability of values detected and reported by sensors is of paramount importance in achieving accurate context awareness within multi-sensor systems. Context awareness refers to the ability of a system to understand its environment based on data collected from various sensors. Unfortunately, difficulties and errors in sensor detection and reporting can significantly undermine the dependability of these values, leading to potential misinterpretations or incorrect decisions based on faulty data. When data obtained from sensors is of low reliability, it inevitably compromises not only the accuracy but also the overall trustworthiness of the context information being gathered. Such inaccuracies can have serious implications in applications ranging from environmental monitoring to autonomous navigation and smart city initiatives. To address this critical issue, the objective of this study is to propose a robust method for evaluating the reliability of sensor values acquired during specific time intervals. This reliability assessment is performed before implementing data fusion techniques, which combine input from multiple sensors to enhance the overall understanding of the context. The reliability of the sensor data is determined by analyzing occurrence rates of sensor events, as well as identifying change patterns over defined periods. By focusing on these metrics, the proposed method aims to enhance the quality of the context information processed by the system. Ultimately, only those sensor values that have been verified to meet certain reliability standards will be utilized in further analyses, thereby ensuring that the resultant context awareness is both accurate and reliable. This systematic approach not only aims to mitigate the risks associated with unreliable data but also improves the overall performance and effectiveness of multi-sensor systems.
Variable Multi-Sensor Signals; Context Awareness; Context Inference; Data fusion
