|
Today, the growing interest in Wireless Sensor Networks has led to their deployment at large scale in several fields ranging from environmental monitoring to smart transportation, industrial systems, health and biomedical systems, intelligent environments, etc. Sensors being a key component of the Internet of Things, many IoT solutions entail their deployment in order to gather information from the surrounding environment, which will be forwarded for remote data processing (at sink level, fog/edge level or by a cloud-processing service). Sensor nodes have to cooperate in order to ensure a reliable forwarding for the different types of gathered data while respecting the QoS requirements. For some applications, other concerns have to be considered, such as security issues and energy limitation, as sensors have to operate for many years with limited batteries that cannot easily be replaced. To enhance network performances, additional contextual information can also be handled to adjust node and network behaviors. However, with the significant growth in the use of IoT and distributed sensor systems, many challenges arise and need to be addressed to support such large deployment, mainly in the context of smart cities and environments.
In this context, the recourse to distributed algorithms in which computation is distributed among all sensor nodes allows overcoming inherent WSN limitations related to the short range communication over a wireless medium, resource limitations in terms of energy, computational resources, unreliability of links, etc. Further, distributed algorithms are required to comprehend the aspects of the large implementations that lead to certain characteristics, i.e., scalability, efficiency, and optimization.
|
