Routing Protocols to Maximize Battery Efficiency

In this paper we propose a routing protocol for wireless ad hoc networks whose nodes are largely battery powered. The battery capacity of the nodes is viewed as a common resource of the system and its use is to be optimized. Results from a previous study on battery management have shown that: (1) pulsed current discharge outperforms constant current discharge, (2) battery capacity can be improved by using a bursty discharge pattern due to charge recovery effects that take place during idle periods, (3) given a certain value of current drawn off the battery, higher current impulses degrade battery performance, even if the percentage of higher current impulses is relatively small. We develop a network protocol based on these findings. This protocol favors routes whose links have a low energy cost. We also distribute multihop traffic in a manner that allows all nodes a good chance to recover their battery energy reserve. INTRODUCTION In this paper we consider an ad hoc network with battery powered nodes. Several schemes for efficient routing algorithms in wireless ad hoc networks have been proposed in the literature [1, 2, 3]. In particular, the problem of energy efficient routing has been addressed in [4, 5, 6]. While in [4, 5] the goal is to minimize the energy consumed to transport a packet from the source to the destination, in [6] the objective is to maximize the lifetime of a network with static or slowly varying topology by finding optimal traffic splits. The routing protocol proposed here aims at maximizing the network lifetime of a network with dynamic topology. This work was supported by NSF under grant CCR 9714651 and by the Italian National Research Council. We consider the battery capacity of the nodes as a common resource of the system and we improve the network lifetime by efficiently exploiting the available battery capacity. In [7, 8] it was shown that the lifetime of a battery depends on the discharge demand profile and on the level of power (current) that is drawn off. If a battery is discharged for short time intervals followed by idle periods, significant improvements in delivered energy seem possible. During the idle periods the battery can partially recover the capacity lost while delivering current impulses. This mechanism is called recovery effect. Moreover, if the requested current exceeds a rated current specification of the battery, the battery delivers a smaller amount of energy. This phenomenon is called rate capacity effect. We exploit these findings to develop a routing protocol, called BEE (Battery Energy Efficient) protocol, that aims to balance the battery consumption among all network nodes. The scheme selects routes whose links have a low energy cost; e.g., a route with 3 hops may be preferred to a route with 2 hops if the maximum level of power per link is lower. Also, the proposed scheme favors the nodes with low battery status by routing the multihop traffic through nodes with high battery capacity and allowing the others to benefit of the recovery effect. Results show that the BEE scheme greatly extends the network lifetime; in comparison with the MTE scheme, which minimizes the total transmission power, a factor of improvement up to 1.8 can be obtained. When a simpler version of the BEE algorithm is used in order to reduce its complexity, an improvement with respect to the MTE protocol is still possible. BATTERY BEHAVIOR The maximum energy that can be obtained from real bat-