Extremely low-overhead security for wireless sensor networks: Algorithms and implementation

Recent advances in the development of compact microprocessors have brought forth new applications in the field of data acquisition and wireless communication. One of these applications is a compact sensor mote that has the ability to perform both data acquisition and transmission within a Wireless Sensor Network (WSN). Wireless sensor communication is susceptible to the same data security threats as traditional wireless networks. In an environment where sensors are broadcasting battlefield intelligence or patient biometrics, data confidentiality must be enforced utilizing some form of encryption.;Unlike traditional wireless networks, where the communicating devices have unrestricted access to power and memory, a wireless sensor has very limited resources. A wireless sensor consists of a battery designed to last an extended amount of time, therefore it is critical that the computation and transmission overhead involved in enforcing data security be optimized to preserve battery life.;The research presented in this thesis details a nesC/TinyOS implementation of the NTRUEncrypt PKCS executing on a Crossbow MICAz mote. Algorithm details regarding message encryption and decryption are analyzed along with optimization techniques that improve execution time and reduce memory size. A summary of performance metrics including execution time, power consumption and code size relative to the comparable ECIES-160 PKCS is also provided.