Fault-tolerant adaptive FIR filters using variable detection threshold

L. K. Lin; G. Robert Redinbo

doi:10.1117/12.190881

28 October 1994 Fault-tolerant adaptive FIR filters using variable detection threshold

L. K. Lin, G. Robert Redinbo

Proceedings Volume 2296, Advanced Signal Processing: Algorithms, Architectures, and Implementations V; (1994) https://doi.org/10.1117/12.190881
Event: SPIE's 1994 International Symposium on Optics, Imaging, and Instrumentation, 1994, San Diego, CA, United States

Abstract

Adaptive filters are widely used in many digital signal processing applications, where tap weight of the filters are adjusted by stochastic gradient search methods. Block adaptive filtering techniques, such as block least mean square and block conjugate gradient algorithm, were developed to speed up the convergence as well as improve the tracking capability which are two important factors in designing real-time adaptive filter systems. Even though algorithm-based fault tolerance can be used as a low-cost high level fault-tolerant technique to protect the aforementioned systems from hardware failures with minimal hardware overhead, the issue of choosing a good detection threshold remains a challenging problem. First of all, the systems usually only have limited computational resources, i.e., concurrent error detection and correction is not feasible. Secondly, any prior knowledge of input data is very difficult to get in practical settings. We propose a checksum-based fault detection scheme using two-level variable detection thresholds that is dynamically dependent on the past syndromes. Simulations show that the proposed scheme reduces the possibility of false alarms and has a high degree of fault coverage in adaptive filter systems.

Citation Download Citation

L. K. Lin and G. Robert Redinbo "Fault-tolerant adaptive FIR filters using variable detection threshold", Proc. SPIE 2296, Advanced Signal Processing: Algorithms, Architectures, and Implementations V, (28 October 1994); https://doi.org/10.1117/12.190881

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available