رویکرد جدید برای جایگزینی یک گره خرابی در شبکه حسگر بی سیم Novel approach for replacement of a failure node in wireless sensor network

1 Introduction Recent technological advances have led to the emergence of pervasive networks of both small and low power devices that integrate sensors and actuators with limited on-board processing and wireless communication capabilities [1]. The wireless communications play a crucial role in dataprocessing networks. They offer open solutions to provide mobility as well as essential services where the installation of infrastructures is difficult or not possible. These networks are under active development because of their interface flexibility allowing user mobility. The control of mobility is a critical issue in communication field, because the mobile environment is characterized by several critical aspects: a frequent disconnection, a modest debit of communication, and especially the limited source of energy. Since the majority of the low power devices have batteries with limited lifetime and the replacement of these batteries on thousands of these devices is infeasible, especially in areas where access is difficult or sometimes even impossible, it is well approvable that a sensors network should be deployed with a strong density in order to extend the network lifetime [2]. In a high density network, if all the sensor nodes act in an active mode, then an excessive quantity of energy will be wasted. In one hand, the data of sensors gathered are likely to be strongly correlated and redundant. In the other hand, an excessive collision of packages can occur, because sensors send simultaneous packages with the presence of some releases events. Consequently, it is neither necessary nor desirable that all nodes act simultaneously in an active mode. One of the emerged questions in such a high density of sensors networks is the density control. In [3], the authors choose the prolongation of the operating time system while keeping only a necessary set of sensors in an active mode and putting the remained sensors in sleep mode. Another category of approaches for the connectivity maintenance in wireless sensor networks (WSN) is the restoration of the connectivity after a sensor node failure [4–7]. Supposing a WSN deployed in a difficult accessing zone, the lifetime of this network depends strongly on the connectivity factor between its nodes. Several factors can be at the origin of the connectivity rupture, between the nodes of the considered network, such as the lack of energy on the significant node level, infection of a vital node by a malevolent code and a logical/physical failure of a primary node. The failure of a sensor node can leave its entire zone (or a part of it) without coverage, and it can generate the partition of the network if it is a gateway node (a relay node). This means that the network will be divided into two or several small networks where some nodes can be disconnected from the network. This implies a loss of connectivity between the parts of the network. Our objective is to restore connectivity after failure of a sensor node by taking into account the constraint of energy.