This paper summarizes research done in the area of electronic system reliability and assesses the approaches used in the calculation of electronic system failure rates. A detailed literature survey is conducted to investigate the various available reliability prediction models.The paper starts with a definition of reliability, briefly discusses various regions of system failure rate in time, justifies the role of reliability prediction methods, provides a historical overview, classifies the traditional models into easy to understand categories and discusses the advantages and disadvantage, reviews the key models that are currently in use, and compares the first and most widely used model (i.e., MIL-HDBK-217) with the most recently introduced model (i.e., PRISM).Index Terms-Electronic component reliability, empiricalbased models, failure rate, MIL-HDBK-217, physics-of-failure, prediction models.

Wireless sensor networks are poised to revolutionize our abilities in sensing and controlling our environment. Power conservation is a primary research concern for these networks. Often, the single most important savings can be obtained by switching off the wireless receiver when not needed. In this paper, we describe an algorithm which allows the nodes to learn the behavior of each other by only observing the transmission behaviors, and from this derive the schedule without external help. Our approach is robust to statistical variations in the nodal transmission periods. We draw important conclusions on the effect of quasi-periodicity on the scalability of the solution. We provide results of numerical simulations that show the effectiveness of our approach.

No abstract