Professional Documents
Culture Documents
2.fault Tolerant Parallel Filters Based On Error Correction Codes
2.fault Tolerant Parallel Filters Based On Error Correction Codes
2, FEBRUARY 2015
Abstract— Digital filters are widely used in signal processing and processing circuits for which specific techniques have been proposed
communication systems. In some cases, the reliability of those systems is over the years [2].
critical, and fault tolerant filter implementations are needed. Over the
Digital filters are one of the most commonly used signal processing
years, many techniques that exploit the filters’ structure and properties
to achieve fault tolerance have been proposed. As technology scales, it circuits and several techniques have been proposed to protect them
enables more complex systems that incorporate many filters. In those from errors. Most of them have focused on finite-impulse response
complex systems, it is common that some of the filters operate in parallel, (FIR) filters. For example, in [3], the use of reduced precision replicas
for example, by applying the same filter to different input signals. was proposed to reduce the cost of implementing modular redundancy
Recently, a simple technique that exploits the presence of parallel filters
to achieve fault tolerance has been presented. In this brief, that idea in FIR filters. In [4], a relationship between the memory elements of
is generalized to show that parallel filters can be protected using error an FIR filter and the input sequence was used to detect errors. Other
correction codes (ECCs) in which each filter is the equivalent of a bit schemes have exploited the FIR properties at a word level to also
in a traditional ECC. This new scheme allows more efficient protection achieve fault tolerance [5]. The use of residue number systems [6]
when the number of parallel filters is large. The technique is evaluated
and arithmetic codes [7] has also been proposed to protect filters.
using a case study of parallel finite impulse response filters showing the
effectiveness in terms of protection and implementation cost. Finally, the use of different implementation structures of the FIR
filters to correct errors with only one redundant module has also been
Index Terms— Error correction codes (ECCs), filters, soft proposed [8]. In all the techniques mentioned so far, the protection
errors. of a single filter is considered.
I. I NTRODUCTION However, it is increasingly common to find systems in which
Electronic circuits are increasingly present in automotive, medical, several filters operate in parallel. This is the case in filter banks [9]
and space applications where reliability is critical. In those applica- and in many modern communication systems [10]. For those systems,
tions, the circuits have to provide some degree of fault tolerance. the protection of the filters can be addressed at a higher level by
This need is further increased by the intrinsic reliability challenges considering the parallel filters as the block to be protected. This
of advanced CMOS technologies that include, e.g., manufacturing idea was explored in [11], where two parallel filters with the same
variations and soft errors. A number of techniques can be used to response that processed different input signals were considered. It was
protect a circuit from errors. Those range from modifications in the shown that with only one redundant copy, single error correction can
manufacturing process of the circuits to reduce the number of errors be implemented. Therefore, a significant cost reduction compared
to adding redundancy at the logic or system level to ensure that with TMR was obtained.
errors do not affect the system functionality [1]. To add redundancy, In this brief, a general scheme to protect parallel filters is presented.
a general technique known as triple modular redundancy (TMR) can As in [11], parallel filters with the same response that process
be used. The TMR, which triplicates the design and adds voting different input signals are considered. The new approach is based on
logic to correct errors, is commonly used. However, it more than the application of error correction codes (ECCs) using each of the
triples the area and power of the circuit, something that may not be filter outputs as the equivalent of a bit in and ECC codeword. This
acceptable in some applications. When the circuit to be protected has is a generalization of the scheme presented in [11] and enables more
algorithmic or structural properties, a better option can be to exploit efficient implementations when the number of parallel filters is large.
those properties to implement fault tolerance. One example is signal The scheme can also be used to provide more powerful protection
using advanced ECCs that can correct failures in multiples modules.
Manuscript received May 19, 2013; revised November 3, 2013; accepted The rest of this brief introduces the new scheme by first summariz-
February 23, 2014. Date of publication March 11, 2014; date of current ing the parallel filters considered in Section II. Then, in Section III,
version January 30, 2015. This work was supported in part by China’s 863
Plan Program Research on the Key Technology for the Base Band Signal the proposed scheme is presented. Section IV presents a case study
Processing for Onboard Payload, in part by the Sino-Japan Joint Fund Key to illustrate the effectiveness of the approach. Finally, the conclusions
Technique Research for GSS Integrated Mobile Satellite Communications, are summarized in Section V.
in part by the Tsinghua University Initiative Scientific Research Program
under Grant 2010THZ03-Key Technologies of Sky-Earth Integration Wireless
Communication Network, in part by the National Basic Research Program of II. PARALLEL F ILTERS W ITH THE S AME R ESPONSE
China under Grant 2012CB316000, and in part by the Spanish Ministry of A discrete time filter implements the following equation:
Science and Education under Grant AYA2009-13300-C03.
Z. Gao, W. Pan, M. Zhao, and J. Wang are with the Tsinghua National ∞
Laboratory for Information Science and Technology, Tsinghua University, Bei- y[n] = x[n − l] · h[l] (1)
jing 100084, China (e-mail: zgao@tsinghua.edu.cn; panwen@tsinghua.edu.cn; l=0
zhaoming@tsinghua.edu.cn; wangj@tsinghua.edu.cn).
P. Reviriego and J. A. Maestro are with the Universidad Antonio de Nebrija, where x[n] is the input signal, y[n] is the output, and h[l] is the
Madrid E-28040, Spain (e-mail: previrie@nebrija.es; jmaestro@nebrija.es). impulse response of the filter [12]. When the response h[l] is nonzero,
Z. Xu is with the School of Information and Communication Engineering, only for a finite number of samples, the filter is known as a FIR filter,
Beijing Information Science and Technology University, Beijing 100085, otherwise the filter is an infinite impulse response (IIR) filter. There
China.
Color versions of one or more of the figures in this paper are available
are several structures to implement both FIR and IIR filters.
online at http://ieeexplore.ieee.org. In the following, a set of k parallel filters with the same response
Digital Object Identifier 10.1109/TVLSI.2014.2308322 and different input signals are considered. These parallel filters are
1063-8210 © 2014 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission.
See http://www.ieee.org/publications_standards/publications/rights/index.html for more information.
IEEE TRANSACTIONS ON VERY LARGE SCALE INTEGRATION (VLSI) SYSTEMS, VOL. 23, NO. 2, FEBRUARY 2015 385
TABLE I
E RROR L OCATION IN THE H AMMING C ODE
For example, an error on filter y1 will cause errors on the checks of TABLE II
z 1 , z 2 , and z 3 . Similarly, errors on the other filters will cause errors R ESOURCE C OMPARISON FOR F OUR PARALLEL FIR F ILTERS
on a different group of z i . Therefore, as with the traditional ECCs,
the error can be located and corrected.
The overall scheme is illustrated on Fig. 2. It can be observed that
correction is achieved with only three redundant filters.
For the filters, correction is achieved by reconstructing the
erroneous outputs using the rest of the data and check outputs.
For example, when an error on y1 is detected, it can be corrected TABLE III
by making R ESOURCE C OMPARISON FOR E LEVEN PARALLEL FIR F ILTERS