Viterbi Decoder Plain Sailing Design for TCM Decoders

International Journal of Trend in Scientific Research and Development (IJTSRD)
Volume 3 Issue 5, August 2019 Available Online: www.ijtsrd.com e-ISSN: 2456 – 6470
@ IJTSRD | Unique Paper ID – IJTSRD26710 | Volume – 3 | Issue – 5 | July - August 2019 Page 1794
Viterbi Decoder Plain Sailing Design for TCM Decoders
Suman Chandel1, Manju Mathur2
1Schollar, 2Professor
1,2Rajasthan College of Engineering for Women, Jaipur, Rajasthan, India
How to cite this paper: Suman Chandel |
Manju Mathur "Viterbi Decoder Plain
Sailing Design for TCM Decoders"
Published in
International
Journal of Trend in
Scientific Research
and Development
(ijtsrd), ISSN: 2456-
6470, Volume-3 |
Issue-5, August
2019, pp.1794-1797,
https://doi.org/10.31142/ijtsrd26710
Copyright © 2019 by author(s) and
International Journalof Trendin Scientific
Research and Development Journal. This
is an Open Access article distributed
under the terms of
the Creative
CommonsAttribution
License (CC BY 4.0)
(http://creativecommons.org/licenses/by
/4.0)
ABSTRACT
Convolutional codes are error correction technique used in noisy channels.
Viterbi Algorithm is the most widely-used decodingAlgorithm,which decodes
the sequence in a maximum likelihood sense. But the complexityof theViterbi
decoder increases with the coding rate of the system. Viterbi decoder is the
most power hungry module in the Trellis coded modulation system. Viterbi
decoding is the best technique for decoding the convolutional codes but it is
limited to smaller constraint lengths. The basic building blocks of Viterbi
decoder are branch metric unit, add compare and select unit and survivor
memory management unit. From the simulation results it is observed that the
proposed Viterbi decoder architecture with modified Branch metric
calculation can reducesignificant amountof computations in ordertodecrease
the hardware usage and to simplify the proceedings.
KEYWORDS: Convolution Encoder, Trace Back, Trellis Coded Modulation (TCM),
Viterbi Decoder, VLSI.
I. INTRODUCTION
For reliable communicationindigitalcommunicationsystem,
error correction is very important. Convolutional codes are
frequently used to correct errors in noisy channels. They
perform well even on very bad channels (with error
probabilities of about 10−3). Convolutional codes represent
one technique within the general class of channel codes,
permits reliable communication of an information sequence
over a channel that adds noise, alters the data, or otherwise
distorts the transmitted signal. Although convolutional
encoding is a simple process, decoding of a convolutional
code is much more complex task. Viterbi Algorithm is the
most widely used decoding Algorithm, which decodes the
sequencein a maximum likelihood sense. Butthe complexity
of the Viterbi decoder increases with the coding rate of the
system.
Trellis coded modulation (TCM) schemes are used in many
bandwidth- efficient systems. A TCM systemusuallyemploys
a high-rate convolutional code, which leads to a high
complexity of the Viterbi decoder (VD) for the TCM decoder,
even if the constraint length of the convolutional code is
moderate. Rate ¾ convolutional encoder is taken to analyse
the efficiency of using proposed architecture in the Viterbi
decoder. T-Algorithm is employed to achieve power
reduction. Branch metric calculation is modified to
reducesignificant amount of computations in order to
decrease the hardwareusageandtosimplifytheproceedings.
The precomputation steps are reduced thereby reducing the
amount of calculations.
The remainder of the paper provides the theoretical
background of Viterbi algorithm and viterbi decoder. Abrief
description of the modified calculation techniques used in
the Viterbi decoder design is also given in this paper.
Conclusions are provided in the last Section.
II. VITERBI ALGORITHM
Viterbi algorithm was introduced in 1967 by Viterbi. Viterbi
algorithm is called as optimum algorithm because it
minimizes the probability of error. The algorithm can be
broken down into the following three steps. 1. Weigh the
trellis; that is, calculate the branch metrics. 2. Recursively
computes the shortest paths to time n, in terms of the
shortest paths to time n-1. In this step, decisions are used to
recursively update the survivor path of the signal. This is
known as add-compare-select(ACS)recursion.3.Recursively
finds the shortest path leading to each trellis state using the
decisionsfrom Step2. The shortestpathiscalledthesurvivor
path for that state and the process is referred to as survivor
path decode. Finally, if all survivor paths are traced back in
time, they merge into a unique path, which is the most likely
signal path. The flowchartgiven infig 1 shows theworkingof
Viterbi algorithm. The algorithm for Viterbi decoder can be
explained by using the example below. Let us consider the
input 11010001.
IJTSRD26710

International Journal of Trend in Scientific Research and Development (IJTSRD) @ www.ijtsrd.com eISSN: 2456-6470
Fig.1. Viterbi Algorithm
III. VITERBI DECODER
The basic units of Viterbi decoder are branchmetricunit,add
compare and select unit and survivor memory management
unit. The first unit is called branch metric unit. Here the
received data symbols are compared to the ideal outputs of
the encoder from the transmitter and branch metric is
calculated. Hamming distance or the Euclidean distance is
used for branch metric computation. The second unit, called
path metriccomputation unit, calculatesthepathmetricsofa
stage by adding the branch metrics, associated with a
received symbol, to the path metrics from the previous stage
of the trellis. The final unit is the trace-back process or
register exchange method, where the survivor path and the
output data are identified. The trace back (TB) and the
register- exchange (RE) methods are the two major
techniques usedfor the pathhistorymanagement in thechip
designs of Viterbi decoders.TheTBmethodtakesuplessarea
but requires much more time as compared to RE method
because it needs to search or trace the survivor path back
sequentially. Also, extra hardware is required to reverse the
decoded bits. The major disadvantage of the RE approach is
that its routingcost is very high especially in the case of long-
constraint lengths and it requiresmuchmoreresources.Fig2
shows the block diagram of Viterbi decoder which containsa
BMU,PMM, ACSU and SMU.
decision decoder, a branch metric is a Hamming distance
between the received pair of bits and the “ideal” pair.
Therefore, a branch metriccan take values of 0, 1 and 2.Thus
for every input pair we have 4 branch metrics (one for each
pair of “ideal” values). For a soft decision decoder, a branch
metric is measured using the Euclidean distance. Let xbe the
first received bit in the pair, y – the second, x0 and y0 – the
“ideal” values. Then branch metric is
Mb=(x-x0)2 (y-y0)2
Fig.2. Diagram of Viterbi decoder
IV. IMPLEMENTATION
A. Convolutional encoder
Rate ¾ , high rate convolutional encoder is taken to analyse
the efficiency of proposed Viterbi decoder architecture.
Specifications of this encoder are given in the table 1. Fig 3
shows the structure of rate ¾ convolutional encoder
Table 1: Specifications of rate ¾ convolution encoder
Parameter Symbol Value
Input Length K 3
Output length N 4
Coding rate k/n 3/4
Memory register count M 6
Constraint length K 7
Fig.3. Rate ¾ convolution encoder
B. Viterbi decoder
A Viterbi algorithm consists of the following three major
parts:
Branch metric calculation – calculation of a distance
between the input pair of bits and the four possible
“ideal” pairs (“00”, “01”, “10”, “11”).
Path metric calculation – for every encoder state,
calculate a metric for the survivor path ending in this
state (a survivor path is a path with the minimum
metric).
Traceback – this step is necessary for hardware
implementations that don't store full information about
the survivor paths, but store only one bit decision every
time when one survivor path is selected from the two.
C. Branch metric calculation
Methods of branch metric calculation are different for hard
decision and soft decision decoders. For a hard Furthermore
when we calculate 4 branch matrics for a soft decision
decoder, we don’t actually need to know the absolute metric
values- only the differencebetweenthemmakesthesense.So
nothing will change if we subtract one value from all four
branch metrics.
Mb= (x2-2x x0 + x02) + (y2 – 2y y0 + y02)
Mb * = Mb – x2 – y2 = (x02 – 2 x x0) + (y02 – 2 y y0)
Note that the secondformula,Mb *,canbecalculatedwithout
hardware multiplication: x02 and y02 can be pre- calculated,
and multiplication of x by x0 and y by y0 can be done very
easily in hardware given that x0 and y0 are constants. It
should be alsonoted that Mb * is asigned variableandshould
be calculated in 2's complement format.

D. Path metric calculation
Path metrics are calculated using a procedure called ACS
(Add-Compare-Select). This procedure is repeated for every
encoder state.
Add –for a given state, we know two states on the
previous step which can move to this state, and the
output bit pairs that correspond to these transitions. To
calculate new path metrics, we add the previous path
metrics with the corresponding branch metrics.
Compare, select– we now have two paths, ending in a
given state. One of them (with greatermetric) is dropped.
As there are 2K-1 encoder states, we have 2K-1 survivor paths
at any given time. It is important that the difference between
two survivor path metrics cannot exceed δ log(K−1),where δ
is a difference between maximum and minimum possible
branch metrics. The problem with path metrics is that they
tend to grow constantly and will eventually overflow. But,
sincethe absolute values of path metricdon'tactuallymatter,
and the difference between them is limited,adatatypewitha
certain number of bits will be sufficient. There are two ways
of dealing with this problem.Sincetheabsolutevaluesofpath
metric don't actually matter, we can at any time subtract an
identical value from the metric of every path. It is usually
done when all path metricsexceedachosenthreshold(inthis
case the threshold value is subtracted from every path
metric). This method is simple, but not very efficient when
implemented in hardware. The second approach allows
overflow, but uses a sufficient number of bits to be able to
detect whether the overflow took place or not. The compare
procedure must be modified in this case.
E. Trace back method
The final stepis the trace back procedure, wherein the allthe
values are consolidated to obtain the final output. If the
position of minimum value is 00 or 01, a 0 is obtained
in the output. For any other values, a 1 is obtained at the
output. In this calculation, the minimum value of the lastrow
is taken and based on its position a 1 or 0 is assigned as one
of the output bits. The corresponding path is noted and
converted to binary.
The minimum value corresponding to that particular path, is
noted and once again a position based output bit assignment
is made as explained previously, for the row 3. This is
repeated for all the remaining rows. Thus,4 output bits are
obtained. These bits are then concatenated and a process of
bit reversal is made .thus a 4 bit output is derived from an 8
bits input that was given to the Viterbi decoder. In the TB
method, the storage can be implementedasRAMandiscalled
the path memory. Comparisons in the ACS unit and not the
actual survivors are stored. After at least L branches have
been processed, the trellis connections are recalled in the
reverse order and the path is traced back through the trellis
diagram. The TB method extractsthedecodedbits,beginning
from the state with the minimum PM. Beginning at this state
and tracing backward in timeby following the survivor path,
which originallycontributed to thecurrentPM,auniquepath
is identified. While tracing back through the trellis, the
decoded output sequence, corresponding to the traced
branches, is generated in the reverse order. Trace back
architecture has a limited memory bandwidth in nature, and
thus limits the decoding speed.
F. Register Exchange Method
The register exchange (RE) method is the simplest
conceptually and a commonlyusedtechnique. Becauseof the
large power consumption and large area required in VLSI
implementations of the RE method, the trace back method
(TB) method is the preferred method in the design of large
constraint length, high performance Viterbi decoders. In the
register exchange, a register assigned to each state contains
information bits for the survivor path fromtheinitialstate to
the current state. In fact, the register keeps the partially
decoded output sequence along the path, as illustrated in
Figure 6. The register of state S1 at t=3 contains ’101’.This is
the decoded output sequence along the hold path from the
initial state. The register- exchange method eliminates the
need to trace back since the register of the final state
contains the decoded output sequence. However, this
method results in complex hardware due totheneedtocopy
the contents of all the registers in a stage to the next stage.
The survivor path information is applied to the least
significant bit of each register, and alltheregistersperforma
shift left operation at each stage to make room for the next
bits. Hence, each register fills in the survivor path
information from the least significant bit toward the most
significant bit. The scheme is called shift update. The shift
update method is simple in implementation but causes high
switching activity due to the shift operation and, hence,
results in high power dissipation. Fig 4 shows the register
exchange method.
Fig.4. Exchange method
V. CONCLUSION
Viterbi algorithm is the most commonly used algorithm to
decode the convolution code. The main drawback is its high
computational complexity and power hungry
implementation for large coding rate since its constraint
length should be high. In this project, Viterbi decoder with
modified Branch metic calculation is designed in order to
decrease the hardware usage and to simplify the
proceedings.
ACKNOWLEDGEMENT
I would like to express my specialthanksand appreciation to
my family and my teachers for their support and
encouragement throughout this work.
REFERENCES
[1] J. Nargis1, D. Vaithiyanathan2, R.Seshasayanan3,
“Design of High Speed Low Power Viterbi Decoder for
TCM System”, IEEE Transactions on Information
Communications and Embedded Systems, Feb 2013
[2] Jinjin He, Huaping Liu, Zhongfeng Wang, Xinming
Huang, and Kai Zhang, “High-Speed Low-PowerViterbi
Decoder Design for TCM Decoders”, IEEE Transactions
on VLSI, VOL. 20, NO. 4,APRIL 2012
[3] Bandwidth-efficient modulations, Consultative
Committee For SpaceDataSystem, Matera,Italy,CCSDS
401(3.3.6) Green Book, Issue 1, Apr. 2003.

[4] J. B. Anderson and E. Offer, Reduced-state sequence
detection with convolutional codes, IEEE Trans. Inf.
Theory, vol. 40, no. 3, pp. 965972, May 1994.
[5] Vladimir Stojanovic,Viterbialgorithmimplementation,
Massachusets Institute of Technology.
[6] S. J. Simmons, Breadth-first trellis decoding with
adaptive effort, IEEE Trans. Commun., vol. 38, no. 1,pp.
312, Jan.1990.
[7] F. Chan and D. Haccoun, Adaptive viterbi decoding of
convolutional codes over memoryless channels, IEEE
Trans. Commun., vol. 45, no. 11, pp. 13891400, Nov.
1997.
[8] R. A. Abdallah and N. R. Shanbhag, Error-resilient low-
power viterbi decoder architectures, IEEE Trans.Signal
Process., vol. 57, no. 12, pp. 49064917, Dec. 2009
[9] Anubhuthi Khare, Jagdish Patel, Manish Saxena, FPGA
Based Efficient implementation of Viterbi Decoder,
International Journal of Engineering and Advanced
Technology (IJEAT) ISSN: 2249 – 8958, Volume-1,
Issue-1, October 2011

Viterbi Decoder Plain Sailing Design for TCM Decoders

More Related Content

What's hot (20)

Similar to Viterbi Decoder Plain Sailing Design for TCM Decoders (20)

More from ijtsrd (20)

Recently uploaded (20)

Viterbi Decoder Plain Sailing Design for TCM Decoders