Providing time- and space- efficient procedure calls for asynchronous software thread integration
Proceedings of the 2004 international conference on Compilers, architecture, and synthesis for embedded systems
ACM Transactions on Embedded Computing Systems (TECS)
Integrating real-time hybrid task scheduling into a sensor node platform
Proceedings of the 2010 ACM Symposium on Applied Computing
Proceedings of the 8th International Conference on Ubiquitous Information Management and Communication
Hi-index | 0.00 |
This paper presents methods to add efficient cryptographicsupport to low-performance embedded processorswith embedded networks (e.g. sensor networks). Softwarethread integration (STI) is used to create efficient threadswhich can perform cryptographic operations during time-slice(TDMA) communication, eliminating most context-switchingoverhead. The AvrX kernel is enhanced to automaticallyselect the most efficient threads based upon availablework, saving processor cycles and power.The results show that an STI-based implementation enablescommunication at higher rates while also performingmore cryptographic work compared with traditional ISR(interrupt service routine) or busy-wait schemes. Significantperformance improvements are found for both the RC4 andRC5 ciphers. First, STI enables cryptographic processing tooccur during communication at a bit rate of f{cpu}/8, whichis not possible with an ISR approach. Second, cryptographicthroughput at lower communication rates increases by up to200% for both RC4 and RC5.