
Automated Safety & Security Systems for Industrial Application using Self-Healing Approach
Author(s) -
B Sri SaiCharan
Publication year - 2021
Publication title -
international journal for research in applied science and engineering technology
Language(s) - English
Resource type - Journals
ISSN - 2321-9653
DOI - 10.22214/ijraset.2021.35640
Subject(s) - spare part , self healing , computer science , redundancy (engineering) , overhead (engineering) , scalability , embedded system , fault tolerance , reliability engineering , computer hardware , distributed computing , engineering , mechanical engineering , medicine , alternative medicine , pathology , database , operating system
The concept of self-healing for hardware systems is studied in this study, and a new approach is offered. In the way they offer healing and recovery abilities, key technologies have been offering imitations to biological entities. Digital systems with inspired uniform architecture offer better fault-tolerance capabilities. The ability of a system to detect and repair flaws or failures is known as self-healing. Given that current self-healing systems are focused on redundancy and spare blocks, one of the key issues with current self-healing approaches is area overhead and scalability for complicated structures. This research proposes a new method for self-healing based on embryonic hardware that does not require the use of spare cells. When compared to alternative techniques that rely on spare cells, the area overhead is reduced. The suggested method entails time multiplexing two functions in a single cell inside a single clock cycle. The proposed technique's reliability is investigated and compared to that of a traditional system with varying failure rates. This method can heal up to 50% of the cells, and each cell can only cover one neighbouring failed cell at a time. The proposed solution has a 9 percent area overhead, which is significantly lower than prior spare cell approaches. The proposed method is tested on two case studies: an ALU array and a neural network.