Optimal power/rate trade-off for internet of multimedia things lifetime maximization under dynamic links capacity

International audienceToday, wireless cameras are widely deployed in various applications such as in commercial and military domains (eg. patients monitored in smart hospitals, multimedia surveillance deployed in smart cities, intelligent multimedia monitoring of an ecological system, etc.). The design of such systems can be achieved by allowing the interaction between all their component objects (i.e., internet of things), which is one of the preoccupations of the researchers this last decade. However, using the multimedia sensor nodes for the internet of things (IoT) is problematic in term of power consumption, since these latter would be wireless and battery powered. Within the wider context of IoT, Wireless Multimedia Sensor Networks (WMSNs) can be a promising technology for the aforementioned domains. However, in a such network, the captured data is usually voluminous and requires local processing, and thus, entailing a higher energy requirement. Therefore, the most important issue is to balance the video data coding versus the perceived video quality at the sink, while maximizing the overall network lifetime. To this end, we propose a fully distributed algorithm, that considers the problem of allocating the resources in wireless video sensor network, in a way to ensure a trade-off between both: the encoding power and the source rate at each node, while serving the desired video quality requirement. In opposition to existing approaches, our algorithm, of distributed nature, is provided to find such a trade-off whatever the initial network configuration is. In addition, a dynamic change of links capacity is considered with the limited transmission and reception powers. A new problem is then formulated to cope with the channel transmission error and it is solved