Battery power optimized encryption

Computational power optimization is crucial in the battery power limited secure wireless mobile networks. Therefore, in this paper, we (a) introduce a hardware/software set-up to measure and model the battery power consumption of different encryption algorithms through real-life experimentation and (b) compute optimal power (number of rounds) allocation for encrypting packets such that the constraints on power and security are met. We present the results for three block ciphers: DES, IDEA, and COST though the same analysis can be extended to the other ciphers such as AES, RC4, etc. A new measure called "vulnerability" that quantifies the success of linear cryptanalysis attack is proposed and its relationship with the power consumption is explored. Two mathematical optimization problems are then posed: (a) compute the optimal power allocation to encrypt each packet such that the vulnerability is minimized subject to a total power constraint and (b) to compute the optimal number of encryption rounds for each packet such that a total power constraint is met. The differences in these two formulations are presented. A closed form solution to the first problem is derived while the second optimization formulation is posed as an integer program and is solved numerically. Several numerical results are also provided.