Journal
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS
Volume 29, Issue 12, Pages 2076-2085Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TVLSI.2021.3110135
Keywords
Latches; Robustness; Feedback loop; Transient analysis; Inverters; Resilience; Single event upsets; Double-node upset; radiation-hardened latch; robustness; self-healing; soft errors
Ask authors/readers for more resources
This paper introduces a single event double node upset self-healing latch designed to meet the high-robustness requirement of applications in a harsh radiation environment. Through fault injection simulation and analysis, it is shown that the proposed latch can self-heal from radiation events, with superior performance in speed, power consumption, and area compared to existing SEDNU resilient latches.
This work presents a single event double node upset (SEDNU) self-healing (DNUSH) latch to meet the high-robustness requirement of the applications used in a harsh radiation environment. The DNUSH latch is based on dual modular redundancy and mainly employs C-elements and inverters, forming multi-feedback interlocked loops to retain the correct data even after the radiation event. The self-healing capability of the proposed latch is successfully shown by the fault injection simulation using Synopsys HSPICE. Simulation results show that the proposed latch can self-heal from all SEDNUs, consumes low power even for high-speed operations, and has the least power-delay-area product (PDAP) compared to the existing SEDNU resilient latches. The proposed latch offers on average 51.25% improvement in speed, 22.67% saving in power consumption, and 59.74% lower PDAP compared to the existing SEDNU resilient latches. In addition, the sensitivity assessment of the proposed latch against the process, voltage, and temperature (PVT) variations are found to be either low or equivalent to the reference latches.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available