Hide/Show Apps

An energy efficient hierarchical approach using multimedia and scalar sensors for emergency services

Kızılkaya, Burak
Recently, environment monitoring and detection systems became more accessible with the help of IoT applications. Furthermore, connecting smart devices makes monitoring applications more accurate and reliable. On the other hand, optimizing the energy requirement of smart sensors especially while transmitting data has always been very important, and there are different applications to create energy efficient IoT systems. Detailed analysis of lifetimes of various types of sensors (survival analysis) has therefore become essential. For the environment monitoring scenarios, with the help of smart multimedia sensors, more precise and accurate real-time information can be extracted. Video and audio sensors can be used as complementary mechanisms to have more accurate information. However, transmission of visual data is known to be one of the most costly operations for wireless multimedia sensor networks. To minimize energy consumption, visual data transmission should be minimized. In this thesis, a novel hierarchical approach is presented for emergency applications. Proposed framework makes use of multimedia and scalar sensors hierarchically to minimize the visual data transmission and in turn energy consumption. In addition, edge computing is introduced where lightweight machine learning algorithms are used for edge processing to prevent unnecessary data transmission. The heterogeneous sensor network architecture is applied within the domain of forest fire detection systems. Proposed framework is evaluated in terms of accuracy of detection as well as energy efficiency. The results are quite promising with validation accuracy of 98.20% and 29.94% energy saving compared to multimedia sensor based surveillance systems.