Energy Reduced Software-Based Self-Testing for Wireless Sensor Network Nodes

Reliable operation of networks consisting of many embedded wireless nodes operating under strict cost and energy constraint is a major requirement for their widespread acceptance. Towards this goal, we consider self-testing of complete wireless nodes in the field through a low-energy software-based self-test (SBST) method. The energy consumption of SBST is optimized both for individual components such as a CPU, embedded memories, and an RF module, as well as at the system level, considering the interplay between module tests. We derive an SBST scheme that utilizes existing CPU instructions efficiently by taking the least amount of cycles and selecting operands with least Hamming distance and weight to minimize the overall energy consumption of the test code. Time interleaving of module tests at the system level is used to further reduce the overall test energy consumption. The efficacy of the proposed methods is evaluated experimentally, using current measurement circuitry integrated in a wireless sensor network node. Finally, we provide a significant amount of data from wireless network performance testing that can be efficiently utilized to build statistical models for protocol, algorithm and hardware design of wireless sensor network.