Energy-Memory-Security Tradeoffs in Distributed Sensor Networks

Abstract

Abstract. Security for sensor networks is challenging due to the resource-constrained nature of individual nodes, particularly their energy limitations. However, designing merely for energy savings may not result in a suitable security architecture. This paper investigates the inherent tradeoffs involved between energy, memory, and security robustness in distributed sensor networks. As a driver for the investigation, we introduce an energy-scalable key establishment protocol called cluster key grouping, which takes into account resource limitations in sensor nodes. We then define a metric (the security leakage factor) to quantify security robustness in a system. Finally, a framework called the security-memoryenergy (SME) curve is presented that is used to evaluate and quantify the multi-metric tradeoffs involved in security design.

A Secure Behavior Modification Sensor System for Physical Activity Improvement

“…Since their initial publication, Design Science has been used as a viable research methodology in a number of doctorial-level studies (Bryant, 2008;Hassan, 2008;Kosaka, 2008;Maheshwari, 2008). (Hwang, Lai, & Verbauwhede, 2004).…”

“…Modeling the energy needed to establish a secure link between two sensor nodes requires the use of a low-energy radio model and the establishment of assumptions of the number of bytes of data to be sent and received between them (Hwang, et al, 2004). Currently, there is active research in the area of low-energy radios used in wireless sensor networks but different assumptions about their radio characteristics, including energy dissipation in a transmit or receive mode, often changes the advantages of a key distribution protocol being modeled (Heinzelman, Chandrakasan, & Balakrishnan, 2000).…”

“…The researcher decided against adopting the E elec and E amp values that were used in other studies (Heinzelman, et al, 2000;Hwang, et al, 2004) and opted instead to use the measured results from Wander et al (2009). They found the energy to transmit and receive one byte of data was equal to 59.2µJ and 28.6µJ, respectively.…”

“…Volatile and non-volatile memory used by a sensor node for program code and software operating requirements is a non-commodity that must be considered when evaluating a key pre-distribution scheme and to ensure that the key distribution algorithm is viable for its practical use in a wireless sensor network (Hwang, et al, 2004). The PKC algorithm posits greater memory savings over the 3-P schemes.…”

Efficient Key Predistribution for Grid-Based Wireless Sensor Networks

“…Much of the literature on key predistribution in wireless sensor networks deals with the case where the physical topology of the network is completely unknown prior to deployment [3,4,5,6,7,9,10,12,13,18,19,21,22,23,24,26,28,29,30,31,35]. In practice, however, many sensor network applications involve networks for which there is some degree of control (indeed, often complete control) over the sensors' locations.…”