This paper addresses the Internet of Things (IoT) paradigm, which is gaining substantial ground in modern wireless telecommunications. The IoT describes a vision where heterogeneous objects like computers, sensors, Radio-Frequency IDentification (RFID) tags or mobile phones are able to communicate and cooperate efficiently to achieve common goals thanks to a common IP addressing scheme. This paper focuses on the reliability of emergency applications under IoT technology. These applications' success is contingent upon the delivery of high-priority events from many scattered objects to one or more objects without packet loss. Thus, the network has to be selfadaptive and resilient to errors by providing efficient mechanisms for information distribution especially in the multi-hop scenario. As future perspective, we propose a lightweight and energyefficient joint mechanism, called AJIA (Adaptive Joint protocol based on Implicit ACK), for packet loss recovery and route quality evaluation in the IoT. In this protocol, we use the overhearing feature, characterizing the wireless channels, as an implicit ACK mechanism. In addition, the protocol allows for an adaptive selection of the routing path based on the link quality.

Node failures represent a fundamental problem in wireless sensor networks. Such failures may result in partitioned networks and lose of sensed information. A network recovery approach is thus necessary in order to ensure continuous network operations. In this paper, we propose CoMN2 a scalable and distributed approach for network recovery from node failures in wireless sensor networks. CoMN2 relies on a new concept called network mapping which consists in partitioning the network into several regions of increasing criticality. The criticality is set according to the energy, the traffic distribution and the deployment of nodes. Using this network mapping, our solution CoMN2 ensures the continuous network activity by efficiently swapping nodes from low critical area to highly critical area when required. Simulation results prove the effectiveness of our approach and show that the obtained improvement in terms of lifetime is in the order of 40%.