International Conference on Innovative Computing and Communication (ICICC 2020)

Channel Coding for Wireless Communication Systems: A Review

PoojaPathaka* and Dr. Richa Bhatiab

aDepartment of Electronics and Communication Engineering
Inderprastha Engineering College Ghaziabad and Research Scholar, AIACTR, GGSIPU, Delhi, India
[email protected]

b
Department of Electronics and Communication Engineering
Ambedkar Institute of Advanced Techonology and Research, GGSIPU, Delhi, India
[email protected]

Abstract: In the present era the wireless communication demanded for higher throughput. There are numerous conventional methods to increase the
throughput, but many of them requires larger bandwidth and or levels of modulation used in them are limited, therefore new methods has been acquainted
to improve the system performance of various wireless networks. Channel Coding is an example of such one method which is utilized with certain
variations to improve the performance of the system.

This paper is the study of present adopted channel coding schemes for wireless communication system and then those schemes are being looked at based
on different parameters such as bit error rate, packet error rate, computational complexity etc. Considering these parameters into account, it is found that
for short packet communication all modern channel coding schemes perform well as compared to turbo codes while turbo codes provide better
performance for larger message lengths with small code rates.

Keywords: Channel Coding, LPDC, Convolution Code, Turbo Code, Polar Code, Space time codes, Block error rate.

modern coding schemes for wireless communication system. These coding

schemes are compared in Section III, whereas concluding remarks along
with some potential research opportunities are presented in Section IV.
1 Introduction

With the advent of the technology, there is a rapid development and 2 Coding Schemes
expanded interest in the field of wireless communication. Demanded by
high performance different channel coding schemes are used in various The concept of channel capacity given by equation 1 was proposed by
wireless networks. Generally, channel coders and decoders work on blocks
of data i.e. whole sequence is given as input which introduces delay [1]. Shannon in 1950s in order to describe the maximum rates of dependable
Moreover, communication system demands the higher data rate leads to information transmission in noisy channel.
higher bandwidth which further leads to higher bit rate error (BER) [2].The
usage of coding techniques is the best solution to reduce the BER without C= B*log2(1+ S/N)---------------1
increasing the bandwidth. However, the coding used should be initially
optimized and then used. where C is the channel capacity in bits per second, B is the bandwidth in
To reduce the delay, message block is divided into smaller parts and Hertz, S and N noise and signal power in watt respectively.
each of the parts is encoded separately. In wireless communication different Many classical coding schemes have been broadly applied but their highest
channel models are used for transmission. Each channel shows different coding gain is still far away from the Shannon limit. Consequently
behaviour and also the performance of the channel models vary for the researchers have been forced to discover more efficient coding scheme
different channel coding schemes. This paper makes the theoretical during the previous decades.
comparison of different channel coding schemes in terms of BER/BLER and In this section modern channel coding schemes, which have achieved
complexity of the decoding algorithm for different channels. near Shannon limit performance are briefly introduced. For particular code
The paper is structured as follows: Section II presents the introduction to message length is denoted by k and codeword length is denoted by n.

POOJA PATHAK 1

Electronic copy available at: https://ssrn.com/abstract=3565791

 International Conference on Innovative Computing and Communication (ICICC 2020)

2.1 Turbo Codes parallelization of the decoding algorithm. These codes are the class of sparse
linear codes which were first introduced by R. Gallager in 1963. They are
Turbo codes are high performance forward error correction codes defined by sparse parity check matrix H of (n-k) X n. The sparse nature of
invented in 1993 by Berrou and his colleagues [3]. These were the first to the parity check matrix allows iterative decoding using message passing
approach channel capacity at which communication is possible for a given algorithm. [1] [6] [7]. LDPC codes supports low complexity iterative
noise level. These codes are built by concatenating two convolution codes in decoding algorithm. The most promising decoding algorithm for LDPC
parallel via an interleaver. Figure 1 illustrates the block diagram of a turbo codes is belief propagation decoding which shows its performance near to
encoder with code rate 1/3[4 ] [5]. the Shannon limit. Extension of binary LDPC codes is non-binary LDPC
codes. Non binary LDPC codes perform better than their binary counterpart
over a non-binary Galois field.

2.3 Polar Codes

Polar codes are another class of linear block codes that brings the great
development in the field of coding theory. This coding scheme was
proposed by Arikan in 2009 with the concept of achieving channel capacity
which was proven mathematically as well. These codes are built on the
theory of channel polarization which says that when the channel are
combined at infinite length and split, the input channels are converted into
Figure1. Turbo encoder with code rate1/3 virtual bit channels which means channel are polarized. Channel so created
after splitting could be of high capacity or low capacity. The bits are
transmitted over high capacity also known as reliable channels and zero
Figure 2 illustrates the decoding circuitry of turbo codes which estimates the information called frozen bits are transmitted over low capacity or unreliable
transmitted values by performing iterative decoding steps. The received channel. These reliable and unreliable channels are decided on the basis of
signal is different from the transmitted signal because of the addition of Bhattacharya parameter. Figure 3 illustrates coder of the Polar encoder for
channel noise. code length 2

Figure3. Polar encoder for code length 2

Successive Cancellation (SC) algorithm is the decoding algorithm used

for polar codes. List decoder used along with SC algorithm improves the
performance of Polar decoder [1] [4] [5] [8].
Figure 2. Iterative Turbo Decoder

2.2 Low Density Parity Check Codes ( LDPC Codes)

LDPC codes have got a lot of attention from the past few years because
of its real time and high throughput communication. The interest in these
codes also increases due to their near Shannon limit performance and in

POOJA PATHAK 2

Electronic copy available at: https://ssrn.com/abstract=3565791

 International Conference on Innovative Computing and Communication (ICICC 2020)

2.4 Convolution Codes done at the cost of increasing the complexity and increase power
consumption, which is critical in 5G [3]. For very short message length
Convolution coding is an error correcting coding technique which was first (k<=128) LTE turbo codes can be replaced by any of the modern channel
introduced by Elias in 1955.The convolution codes (n,k,m) are characterized coding schemes as they show the significant improvement in the
by the parameters input length k, output length n and number of memory performance over LTE turbo codes [8]. BER and BLER are improved using
elements m. These codes are allowed to be processed while receiving bits as LDPC codes but, as the code length increases for different modulation level
they contain the memory which makes them different from the block the decoding complexity increases as well. When the coding length is
codes.Terminated convolution codes are the candidate coding scheme for doubled the complexity LTE decoder is increased by the factor of 2 [12].
short block length, although it suffers with the rate loss due to zero tailing BER in case of STC decrease as the number of transmitting and receiving
bits [8]. In LTE standard, Convolution codes used to transmit control antenna increases, but more energy is required to get the dependable
information bits are known as tailing bit convolution codes (TBCC). TBCC wireless link [13].
reduces the rate loss compared to the other channel codes for short message
length. TBCC can start and end at any state which is known to the decoder 3.2 Complexity
[1].
The complexity of the modern channel codes is compared on the
2.5 Space Time Codes basis of the computational complexity associated with encoding and
decoding process. The polar code works on successive cancellation and has
Space time codes were originally created in 1998 by Tarkoh et al. to define a the lowest decoding complexity as compared to all other coding schemes. In
new method of encoding and decoding the signals using multiple antennas. comparison to LDPC and Convolution Codes, Turbo codes have much more
STC term is generally used for transmit diversity techniques in which spatial complex decoding algorithm.
diversity is used by the transmitter and its corresponding receivers though Min-sum algorithm (MSA) works well for LDPC decoding as compared to
there are many other schemes where the concept of multiple antennas is sum product algorithm (SPA) as MSA is much simpler than SPA but
used [9].This scheme banks upon transmitting the multiple copies of the data number of iterations are more in MSA if some reference complexity is
stream so that few of them survive the physical path in good state and there assumed. In case of convolution codes, terminated decoding is the better
can be reliable decoding of the data stream at the receiver. STC codes are nT approach because it has more number of state decoders as compared to
X p codes where nT denotes number of transmitting antennas and p denotes tailing bit decoding. For polar codes successive cancellation list (SCL)
total number of time periods required for transmitting one block of coded algorithm performs better than successive cancellation [14]. For more than 8
symbol [10]. Space time block codes were introduced in order to achieve iterations Log Map Algorithm is used for turbo code decoding, but as the
high capacity with simple decoding scheme. STBC code provides full number iteration increases, the performance reduces [15].
diversity and full transmission rate. Alamouti introduced STBC scheme With the increased number of antenna the performance of STC codes
which was simply a transmit diversity technique employing two antennas increases with very little increase in the complexity of the decoding
[11]. algorithm. It is due the fact that linear processing is used for decoding [10].

3 Comparison between The Channel codes 3.2 Performances Analysis for different Channel Models

In this section the performance of the channel coding schemes are compared The LDPC channel codes produce better results using log likelihood
for different parameters. decoding in case of AWGN over Racian and Rayleigh fading channel.
Amongst the three Rayleigh shows the worst performance. For higher values
3.1 Bit Error Rate (BER)/Block Error rate (BLER) of SNR the BER decreases maximum in AWGN channel [16]. LDPC still
performs well with the increase in the block length sophisticated decoder
For short message length k=64,128,256 and 512 convolution codes circuitry is considered [2]. The polar coded OFDM performs well at low
and polar codes outperforms LTE turbo codes and LDPC whereas tail bit SNR on AWGN, Rayleigh and Racian channel, but as the SNR increases the
convolution code (TBCC) 2 performs better than LTE turbo codes for the turbo code’s performance is far better than the polar codes. SC decoding
message length k<= 128. But as the message length increases the considered for polar produces the worst results on AWGN channel. Table 1
performance of the convolution code decrease the most as compared to all gives the comparison of various modern coding schemes on different
other coding schemes. parameters.
The performance of convolution codes can be improved by increasing the
number of memory elements (6 for TBCC1 and 10 for TBCC2). But it is

Table1: Comparison of Various Channel Coding Schemes

POOJA PATHAK 3

Electronic copy available at: https://ssrn.com/abstract=3565791

 International Conference on Innovative Computing and Communication (ICICC 2020)

S.No. Parameter Polar Turbo LDPC STC

1 Computational Low for most Higher for most Low for most coding rate Low decoding complexity
complexity[17] coding rate coding rate but low but high at low coding
for low coding rate rate

2 Decoding Algorithm[18] BP,SC,SCL Max log Min sum algo Maximum likelihood
Algorithm
3 Hardware Efficiency[18] Lower Lower but higher Higher High
than Polar

4 Energy Efficiency17] High for all High for low coding Low at lower coding rate Low
coding rate rate

5 Throughput[17] High for pipeline Low but can be High for higher coding High for higher number of
structure enhanced using rate & Parallel structure antenna
parallel structure

4 Conclusion International Journal of Advanced Research in Computer and

Communication Engineering Vol. 4, Issue 10, October 2015, pp 151-159
This paper reviews the recent progress taking place for better transmission [4] Ming Zhan, Zhibo Pang, DacfeyDzung and Ming Xiao , “Channel
of information using appropriate channel coding schemes. All the five Coding for High Performance Wireless Control in Critical Applications:
modern coding schemes are compared on the basis of BER and Survey and Analysis, Special Section on Advances in Channel Coding for
computational complexity over different channel. The study of different 5G and Beyond, IEEE Access, Open Journal, 2018.
modulation schemes concludes that the Polar and convolution codes [5]AldrinClaytusVaz, C GurudasNayak and DayanandaNayak,
performs well for short message length whereas turbo codes and LDPC “Performance Comparison between Turbo and Polar Codes”, Proceedings of
codes can be used for larger information block length and decoding IEEE International Conference on Electronics Communication and
complexities of the channel codes. General conclusion says that there is no Aerospace Technology [ICECA 2019], pp 1072-1075.
single coding scheme which fulfils all the requirements of the [6] Tom Richardson and ShrinivasKudekar, “Design of Low-DensityParity
communication system. One may provide the smaller BER but higher Check Codes for 5G New Radio”, IEEE Communications Magazine
complexity and so. The performance of the system can significantly improve 2018,Volume 56 Issue 3.
by using better codes which can be done by improving the properties of the [7] Hamid R. Barzegar and Luca Reggiani, “Channel Coding for Multi-
codes In any case, on the off chance that we look in to the present situation, Carrier Wireless Partial Duplex”, Journal of Wireless Communications and
in 4G and 5G, OFDMA is utilized as the major multiple access technique Mobile Computing, Volume 2019, Article ID 9891821
and structuring the channel code for these systems is yet a challenging task [8]Mahyar Shirvanimoghaddam et.al, “Short Block-Length Codes for Ultra-
as reducing the number of symbols reduces the latency of the system. Reliable Low Latency Communications”,IEEE Communications Magazine
Enhancing the existing codes for having better outcomes and bringing down 2018,Volume 57 Issue 2.
the decoding complexities of the present coding schemes can be an [9]Bijun Zhang, HuiShen et.al, “A 5G Trial of Polar Code”, IEEE
interesting future work. Globecom Workshops: 2016.
[10] K. MeenaAnusha and Dr. Ch. Santhi Rani, “Comparison of STBC
REFERENCES System and STTC SystemUsing Various Doppler Spectra”, International
Journal of Electronics and Communication Engineering and Technology
(IJECET), Volume 7, Issue 6, November-December 2016, pp. 01–13.
[1]Onurcan ˙Iscan, Diego Lentner, Wen Xu, “A Comparison of Channel [11] Priyanka Mishra,C.K. Shukla,Anil Kumar, “A Review Paper On
Coding Schemes for 5G Short Message Transmission”, IEEE Globecom Orthogonal Space Time Block Coded
Workshops: 2016. MIMO Systems with Full Rate and Full Diversity Code”, International
[2] KulvirKaur , HeenaDogra and Gurjot Singh Gaba, “A Theoretical Journal of Electronics, Electrical and Computational System, Volume
Review of Channel Coding Techniques Over Different Channel Models”, 6,Issue 12, December 2017.
International Journal of Control and Automation Vol.10, No.9 (2017), [12] Walled Khalid Abdulwahab and AbdulkareemAbdulrahmanKadhim, “
pp.37-42. Comparative Study of Channel Coding Schemes for 5G”, IEEE
[3]AkankshaTiwari and Vikas Kumar Sahu, “Error Correction Channel International Conference on Advanced Science and Engineering (ICOASE-
Coding Practices for Wireless Communication Systems - A review”, 2018), Kurdistan Region, Iraq, pp 239-243.

POOJA PATHAK 4

Electronic copy available at: https://ssrn.com/abstract=3565791

 International Conference on Innovative Computing and Communication (ICICC 2020)

[13] Shakti Raj Chopra, NavjotKaur and HimanshuMonga, “Space Time [16] Ravi Kurariya, Rashika Gupta and Ravimohan, “A Review on Space
Coding Techniques in MIMO: A Review”, Indian Journal of Science and Time Coding Techniques”, Journal of Emerging Technologies and
Technology, Vol 9(47), December 2016. Innovative Research, December 2015, Volume 2, Issue 12.
[14]MichałSybis et al., “Channel coding for ultra-reliable low- [17] Shuai Shao et al., “Survey of Turbo, LDPC and Polar Decoder
latencycommunication in 5G systems”, IEEE 84th Vehicular Technology ASICImplementations”, IEEE Communications Surveys& Tutorials,
Conference., 2016. 2019Volume: 21, Issue 3.
[15]Duy-Huy Nguyen and Hang Nguyen, “An improved Log-MAP [18] AlexiosBalatsoukas-Stimming, Pascal Giard, and Andreas Burg,
algorithm based onpolynomial regression function for LTE Turbodecoding”, “Comparison of Polar Decoders with ExistingLow-Density Parity-Check
IEEE ICC 2015- Worshop on advance PHY and MAC Techniques for Super and Turbo Decoders”, IEEE Wireless Communications and Networking
dense Wireless Networks. Conference Workshops (WCNCW), 2017

POOJA PATHAK 5

Electronic copy available at: https://ssrn.com/abstract=3565791