Title

Self-Adapting Resource Escalation for Resilient Signal Processing Architectures

Authors

Authors

N. Imran; R. F. DeMara; J. Lee;J. Huang

Comments

Authors: contact us about adding a copy of your work at STARS@ucf.edu

Abbreviated Journal Title

J. Signal Process. Syst. Signal Image Video Technol.

Keywords

Survivability; Reconfigurable hardware for video compression; Semiconductor aging; Process variations; Autonomous recovery; Mission-critical systems; FPGAs; UNEQUAL ERROR PROTECTION; FPGA LOGIC BLOCKS; SYSTEMS; DESIGN; RECONFIGURATION; RELIABILITY; TOLERANCE; ROBUST; Computer Science, Information Systems; Engineering, Electrical &; Electronic

Abstract

To deal with susceptibility to aging and process variation in the deep submicron era, signal processing systems are sought to maintain quality and throughput requirements despite the vulnerabilities of the underlying computational devices. The Priority Using Resource Escalation (PURE) online resiliency approach is developed herein to maintain throughput quality based on the output Peak Signal-to-Noise Ratio (PSNR) or other health metric. PURE is evaluated using an H.263 video encoder and shown to maintain signal processing throughput despite hardware faults. Its performance is compared to two alternative reconfiguration algorithms which prioritize the optimization of the number of reconfiguration occurrences and the fault detection latency, respectively. For a typical benchmark video sequence, PURE is shown to maintain a PSNR baseline near 32dB. Compared to the alternatives, PURE maintains a PSNR within a difference of 4.02dB to 6.67dB from the fault-free baseline by escalating healthy resources to higher-priority signal processing functions. The diagnosability, reconfiguration latency, and resource overhead of each approach is analyzed. The results indicate the benefits of priority-aware resiliency over conventional redundancy in terms of fault-recovery, power consumption, and resource-area requirements.

Journal Title

Journal of Signal Processing Systems for Signal Image and Video Technology

Volume

77

Issue/Number

3

Publication Date

1-1-2014

Document Type

Article

Language

English

First Page

257

Last Page

280

WOS Identifier

WOS:000343161400003

ISSN

1939-8018

Share

COinS