Power optimization for wireless sensor networks security based on an FPGA implementation

Xiong, X

Abstract

Security is a key consideration when deploying Wireless Sensor Networks (WSNs). Due to the constrained hardware resources in sensor network nodes, lightweight cryptographic primitives are often implemented for fast execution and small memory usage. Apart from the computational complexity of cryptographic algorithms, the technique of implementing them in practical terms is another crucial aspect in WSN security development, which directly reflects to the power efficiency. Since the battery life confines the lifetime of a sensor node, energy conservation is normally set as the first priority in developing security solution. This means that the optimal security operation for WSNs has to consume the smallest amount of energy possible while it is active.

A novel power optimization methodology is proposed intended to provide a guideline to evaluate cryptographic primitives and implementation techniques for constructing the power optimal security solution. Noting the inevitable limitations of traditional security implementation techniques, a FPGA-based hybrid technique was innovated to offer high efficiency with flexibility. An interesting impact of the operating frequency on the power consumption in security development was also identified.

Using this methodology, the most suitable cryptographic primitive and hardware implementation configurations were suggested for building environment monitoring application from procedural experiments. This produced an optimal security solution that proved the feasibility and effectiveness of the proposed methodology as its actual measurements fully satisfied the predetermined criteria. The outcome showed that the FPGA implementation technique was the most power efficient implementation solution for flexibility-essential applications and the adjustment of operating frequency could be a useful tool to further optimize the security solution for low power.

• Thesis Type: Thesis
• Deposit Date: Oct 3, 2012
• Award Date: Jan 1, 2009