Coding Schemes for Implementation of Fault Tolerant Parrallel Filter

International Journal of Advances in Applied Sciences (IJAAS)
Vol. 7, No. 4, December 2018, pp. 353~360
ISSN: 2252-8814, DOI: 10.11591/ijaas.v7.i4.pp353-360  353
Journal homepage: http://iaescore.com/online/index.php/IJAAS
Coding Schemes for Implementation of Fault Tolerant Parrallel
Filter
Nutenki Siddhartha, G Renuka
Department of ECE, SR Engineering College, Hasanparthy, Warangal, Telangana, India.
Article Info ABSTRACT
Article history:
Received Mei 28, 2018
Revised Jul 10, 2018
Accepted Aug 18, 2018
Digital filters are utilized as a one of flag handling and correspondence
frameworks. At times, the unwavering quality of those frameworks is basic,
and blame tolerant channel executions are needed. Throughout the years,
numerous systems that endeavor the channels' structure and properties to
accomplish adaptation to internal failure have been proposed. As innovation
scales, it empowers more unpredictable frameworks that join many channels.
In those perplexing frameworks, it is regular that a portion of the channels
work in parallel. A plan in view of big rectification coding has been as of late
proposed to protect parallel channels. In that plan, each channel is deal with as
a bit, and excess channels that go about as equality check bits are acquainted
with distinguish and rectify blunders. In this short, applying coding systems to
secure parallel channels is tended to in a broader manner. This decreases the
assurance overhead and makes the quantity of excess channels autonomous of
the quantity of parallel channels. The proposed technique is first described and
then illustrated with two case studies. Finally, both the effectiveness in
protecting against errors and the cost are evaluated for a field-programmable
gate array implementation.
Keyword:
Error Correction Codes (ECCs)
Filters
Soft Errors
Copyright © 2018 Institute of Advanced Engineering and Science.
All rights reserved.
Corresponding Author:
Nutenki Siddhartha,
Department of ECE,
SR Engineering College,
Hasanparthy, Warangal, Telangana, India.
Email: siddharthanutenki@gmail.com
1. INTRODUCTION
Electronic circuits are progressively present in and space applications where dependability is basic.
In those applications, the circuits need to give some level of adaptation to non-critical failure. This need is
additionally expanded by the inherent unwavering quality difficulties of cutting edge CMOS innovations that
incorporate, e.g., producing varieties and delicate blunders. Various strategies can be utilized to shield a circuit
from blunders. Those range from adjustments in the assembling procedure of the circuits to decrease the
quantity of mistakes to including repetition at the rationale or framework level to guarantee that blunders don't
influence the framework usefulness.
Channels are regularly utilized as a part of electronic frameworks to underline motions in certain
recurrence ranges and reject motions in other recurrence ranges. In circuit hypothesis, a channel is an electrical
system that adjusts the abundancy as well as stage qualities of a flag regarding recurrence. In a perfect world,
a channel won't add new frequencies to the info flag, nor will it change the segment frequencies of that flag,
yet it will change the relative amplitudes of the different recurrence segments as well as their stage connections.
Today channels are generally utilized as a part of number of uses which in view of car, restorative, and space
where unwavering quality of segments in computerized electronic circuits is basic. Channels or the like are
basic in the operation of most electronic circuits. There are various bases of characterizing channels and these
cover in a wide range of ways; there is no straightforward progressive grouping. As the behavioral properties

 ISSN: 2252-8814
IJAAS Vol. 7, No. 4, December 2018: 353 – 360
354
of flag changes the strategies of separating it will be contrast. Being particular with channel, the computerized
channels have huge applications in advanced flag preparing. Sifting is likewise a class of flag preparing, the
characterizing highlight of channels being the total or incomplete concealment of some part of the flag. It is in
this manner in light of a legitimate concern for anybody associated with electronic circuit configuration to be
able to create channel circuits fit for meeting a given arrangement of particulars. In flag handling, an advanced
channel is a gadget or process that expels some undesirable segment or highlight from a flag. Advanced
channels are utilized for two general purposes; division of signs that have been joined, and reclamation of signs
that have been contorted somehow. Frequently, this implies expelling a few frequencies and not others with a
specific end goal to smother meddling signs and decrease foundation commotion.
Parallel Processing and correspondence frameworks. As a rule, Parallel channels are normally found
in current flag the channels play out an indistinguishable preparing on various approaching signs from there is
an inclination to utilize different input–multiple yield frameworks.
The security of computerized channels has been broadly contemplated. For instance, blame tolerant
executions in light of the utilization of deposit number frameworks or number-crunching codes have been
proposed. The utilization of decreased accuracy replication or word-level insurance has been likewise
examined another alternative to perform blunder remedy is to utilize two distinctive channel usage in parallel.
Every one of those systems concentrate on the security of a solitary channel.
This concise examinations the assurance of parallel channels utilizing more broad coding strategies.
Specifically, a key contrast with ECCs is that both channel sources of info and yields are numbers. Hence, not
just a zero or a one can be utilized for the coding (as finished with ECCs). This can be abused, as appeared in
whatever is left of this brief, to give mistake remedy by including just two repetitive channels paying little heed
to the quantity of parallel channels. The decreased number of repetitive channels does not influence the capacity
of the plan to revise mistakes however diminishes the usage cost. In whatever is left of this short, to begin with,
the parallel channels and the current ECC-based security plot are depicted. At that point, the proposed coding
plan is given and delineated a couple of useful contextual analyses. At last, the contextual analyses are assessed
for a field-programmable door cluster (FPGA) usage and contrasted and the already proposed ECC-based
system.
1.1. Concept of Fault Tolerance
Various procedures can be utilized to shield a circuit from mistakes. Those range from changes in the
assembling procedure of the circuits to lessen the quantity of blunders to including repetition at the rationale
or framework level to guarantee that mistakes don't influence the framework usefulness. Computerized Filters
are a standout amongst the most generally utilized flag handling circuits and a few procedures have been
proposed to shield them from blunders. There are number of techniques used to recognize flaws and the
activities important to adjust the issues inside circuit. Advanced channels are broadly utilized as a part of flag
handling and correspondence frameworks. There are distinctive adaptation to internal failure ways to deal with
customary computational circuits and the DSP circuits. Now and again, the unwavering quality of those
frameworks is basic, and blame tolerant channel executions are required. Throughout the years, numerous
methods that endeavor the channels structure and properties to accomplish adaptation
2. LITERATURE REVIEW
[1] In this paper, adaptation to non-critical failure construct framework based with respect to Error
Correction Codes (ECCs) utilizing VHDL is outlined, executed, and tried. It recommends that with the
assistance of ECCs i.e. Blunder Correction Codes there will be more ensured Parallel channel circuit has been
conceivable. The channel they have utilized for blunder location and revision are for the most part limited
motivation reaction (FIR) channels. They have been utilized Hamming Codes for blame adjustment in which
they takes a square of k bits and produces a piece of n bits by including n−k equality check bits. The equality
check bits are XOR mixes of the k information bits. By appropriately planning those mixes it is conceivable to
distinguish and revise blunders. In this plan they have utilized repetitive module in which the information and
equality check bits are store d and can be recuperated later regardless of the possibility that there is a blunder
in one of the bits. This is finished by re - registering the equality check bits and contrasting the outcomes and
the qualities put away. Along these lines utilizing hamming codes mistake can be recognized and revised inside
the circuit.
[2] In this paper, Triple Modular Redundancy (TMR) and Hamming Codes have been utilized to
ensure distinctive circuits against Single Event Upsets (SEUs). In this paper, the utilization of a Novel
Hamming approach on FIR Filters is considered and executed with a specific end goal to give low
unpredictability, lessen deferral and territory proficient assurance methods for higher bits information. A novel
Hamming code is proposed in this paper, to build the proficiency of higher information bits. In this paper, they

IJAAS ISSN: 2252-8814 
Coding Schemes for Implementation of Fault Tolerant Parrallel Filter (Nutenki Siddhartha)
355
have proposed method used to illustrate, how the part of overhead because of mixing the repetition bits, their
consequent evacuation, cushion to cushion postpone in the decoder and utilization of aggregate range of FIR
channel for higher bits are decreased. These depend on the novel hamming code usage in the FIR channel
rather than regular hamming code used to ensure FIR channel. In this plan Hamming code utilized for
transmission of 7-bit information thing.
[3] In this paper, the outline of a FIR channel with self checking capacities in view of the deposit
checking is broke down. Typically the arrangement of deposits used to check the consistency of the aftereffects
of the FIR channel are based of theoretic contemplations about the dynamic range accessible with a picked set
of buildups, the math qualities of the blunders caused by a blame and on the normal for the channel execution.
This examination is frequently hard to perform and to acquire adequate blame scope the arrangement of picked
buildups is overestimated. Gotten result and consequently requires that Instead, in this paper they have
demonstrated how utilizing a comprehensive blame infusion battles permits to proficiently choose the best
arrangement of buildups. Test comes about originating from blame infusion crusades on a 16 taps FIR channel
showed that by watching the happened mistakes and the identification modules relating to various deposit has
been conceivable to decrease the quantity of discovery module, while paying a little lessening of the level of
SEUs that can be distinguished. Paired rationale overwhelms the equipment execution of DSP frameworks
[4] In this paper they have proposed engineering for the execution of blame - tolerant calculation
inside a high throughput multirate equalizer for a deviated remote LAN. The zone overhead is limited by
abusing the logarithmic structure of the Modulus Replication Residue Number System (MRRNS). They had
exhibited that for our framework the zone cost to amend a blame in a solitary computational channel is 82.7%.
Adaptation to non-critical failure inside MRRNS design is executed through the expansion of excess channels.
This paper has displayed a point by point investigation of the cost of executing single blame adjustment ability
in a FIR channel utilizing the MRRNS. The blame tolerant design makes utilization of the mathematical
properties of the MRRNS, and has been appeared to give critical range funds when contrasted and general
systems. This engineering likewise requires couple of extra parts to be outlined, as indistinguishable excess
channels are utilized, and the polynomial mapping stages are just extended from the first segments.
Main Objectives of proposed method:
There are various objectives over the protection of digital signal processing circuits. The main
objectives are given as follows-
1). To achieve fault free digital circuit.
2). To detect and correct errors in digital circuit with more accuracy.
3). To reduce the overhead, needed for protection from error.
4). To improve efficiency.
5). To develop the application area from lower to higher order application.
3. ECC-BASED PROTECTION OF PARALLEL FILTERS
The impulse response h[n] completely defines a discrete time filter that performs the following
operation on the incoming signal x[n]:
(1)
This property can be exploited in the case of parallel filters that operate on different incoming signals,
as shown on Figure. 1. In this case, four filters with the same response process the incoming signals x 1[n], x
2[n], x 3[n], and x 4[n] to produce four outputs y 1[n], y 2[n], y 3[n], and y 4[n].
Figure 1. Block diagram for Parallel FIR Filters

 ISSN: 2252-8814
356
To detect and correct errors, each filter can be viewed as a bit in an ECC, and redundant filters can be
added to form parity check bits. This is also illustrated in Figure. 2, where three redundant filters are used to
form the parity check bits of a classical single error correction Hamming code [14]. Those correspond to the
outputs z1 [n], z2 [n], and z3 [n]. Errors can be detected by checking if
(2)
When some of those checks fail, an error is detected. The error can be corrected based on which specifi
checks failed. For example, an error on filter y 1 will cause errors on the checks of z1, z2, and z3
Figure 2. ECC-based scheme for four filters and a Hamming code.
4. CODING FOR FAULT-TOLERANT PARALLEL FILTERS
The proposed scheme is illustrated in Figure. 3 for the case of four parallel filters. The input signals
are encoded using a matrix with arbitrary coefficients to make the signals that enter the four original and two
redundant filters. In its more general form, this coding matrix A can be formulated as
(3)

IJAAS ISSN: 2252-8814 
357
Figure 3. Proposed coding scheme in general form
With this coding scheme, the outputs of the filters, i.e., y1[n], y2[n], y3[n], and y4[n], can be obtained as
follows:
(4)
where A1235 is a sub matrix of A, including the first, second, third, and fifth rows. This process can
be repeated with different submatrixes of A, for example, with A1236, A2345, and A2346. In the error-free
case, all the recovered versions of y1 [n], y2 [n], y3 [n], and y4 [n] will be the same. When there are differences,
an error is detected. For xample, suppose that
(5)
Which means that there is an error among filters {1 2 3 5 6} and that filters {2 3 4 5 6} are correct.
Therefore, the faulty filter is filter 1. Then, the error can be corrected by taking the final outputs from a set that
does not include filter 1.
The error correction and detection logic can be simplified assuming that there is only a single error.
In that case, checking only that, for each recovered set, the sums of the values y1 [n] + y2 [n] + y3 [n] + y4 [n]
are equal is enough. In more detail, four checks are needed, each involving five filters and excluding one. For
example, if branch 1 is excluded, the error checking would be
(6)
in which w¯2345 = [1111](A2345) −1, and w¯2346 = [1111] (A2346) −1.

 ISSN: 2252-8814
358
5. RESULTS
The written Verilog HDL Modules have successfully simulated and verified using Modelsim III 6.4b
and synthesized using Xilinx ISE 13.2.
Simulation Result
Figure 4. Simulation Result
In Figure 5 we observed that simulation results and Figure.6 and Figure.7 shows that RTL schematic and
Technology schematic respectively.
Table 1. Design Summary
Device Utilization Summary (Estimated Values)
Logic Utilization Used Available Utilization
Number of Slices 225 4656 4%
Number of Slice
FlipFLOPS
229 9312 2%
Number of 4 input
LUTs
421 9312 4%
Number of Bonded
IOBs
234 232 100%
Number of GCLKs 1 24 4%
RTL Schematic
Figure 5. RTL Schematic

IJAAS ISSN: 2252-8814 
359
Technology Schematic
Figure 6. Technology Schematic
Timing Report
Figure 7. Timing Report
5. CONCLUSION
In this paper a new scheme to protect parallel filters that are commonly found in modern signal
processing circuits has presented. The approach is based on applying ECCs to the parallel filters outputs to
detect and correct errors. The technique can be used for parallel filters that have the same response and process
different input signals. An objectives has also been discussed to show the effectiveness of the scheme in terms
of error correction and problem definition also shows the overheads. The proposed scheme can also make
system price lower. Proposed work will result in more efficient fault tolerant system using parallel IIR filters
based on ECCs, which will meet the goal to achieve low power consumption, increase area of application and
high speed.
REFERENCES
[1] P. P. Vaidyanathan, Multirate Systems and Filter Banks, Englewood Cliffs, N.J., USA: Prentice Hall, 1993.
[2] A. Sibille, C. Oestges and A. Zanella, MIMO: From Theory to Implementation, New York, NY, USA: Academic,
2010.
[3] N. Kanekawa, E. H. Ibe, T. Suga and Y. Uematsu, Dependability in Electronic Systems: Mitigation of Hardware
Failures, Soft Errors, and ElectroMagnetic Disturbances, New York, NY, USA: Springer Verlag, 2010.
[4] M. Nicolaidis, “Design for soft error mitigation,” IEEE Trans. Device Mater. Rel., vol. 5, no. 3, pp. 405–418, Sep.
2005.
[5] C. L. Chen and M. Y. Hsiao, “Error-correcting codes for semiconductor memory applications: A state-of-the-art
review,” IBM J. Res. Develop.vol. 28, no. 2, pp. 124–134, Mar. 1984.
[6] A. Reddy and P. Banarjee “Algorithm-based fault detection for signal processing applications,” IEEE Trans.
Comput., vol. 39, no. 10, pp. 1304–1308, Oct. 1990.
[7] T. Hitana and A. K. Deb, “Bridging concurrent and non-concurrent error detection in FIR filters,” in Proc. Norchip
Conf., 2004, pp. 75–78.

 ISSN: 2252-8814
360
[8] Y.-H. Huang, “High-efficiency soft-error-tolerant digital signal processing using fine -grain subword-detection
Processing,” IEEE Trans. Very Large Scale Integr. (VLSI) Syst., vol. 18, no. 2, pp. 291–304, Feb. 2010.
[9] S. Pontarelli, G. C. Cardarilli, M. Re, and A. Salsano, “Totally fault tolerant RNS based FIR filters,” in Proc. IEEE
IOLTS, Jul. 2008, pp. 192–194.
BIOGRAPHIES OF AUTHORS
Nutenki Siddhartha is pursuing Master of Technology in Electronic Design Technology
in SR Engineering College, Warangal, and Telangana. He has completed his B.Tech
from Kamala Institute of Technology and Science, Karimnagar, Telangana in 2014. His
areas of interests are VLSI, Digital Signal Processing and Computer Networks. He
Published 1 Poster in ICRTEECT-2017.
G. Renuka working as assistant professor of Electronics and Communication
Engineering, S R Engineering College, Warangal. She has 12 years of teaching
experience. She has obtained B.Tech (ECE) Degree from Ramappa Engineering
College Warangal, Andhra Pradesh, India in 2003 and M.Tech (Digital
Communications) Degree from Kakatiya University, Andhra Pradesh, India in 2009.
Pursuing her Ph.D under JNTUH. She is a member of IETE, and her research areas
include VLSI, Communications.

Coding Schemes for Implementation of Fault Tolerant Parrallel Filter

More Related Content

What's hot (17)

Similar to Coding Schemes for Implementation of Fault Tolerant Parrallel Filter (20)

More from IJAAS Team (20)

Recently uploaded (20)

Coding Schemes for Implementation of Fault Tolerant Parrallel Filter