Space Applications," 2016 IEEE 22nd International Symposium On On-Line Testing and Robust
Space Applications," 2016 IEEE 22nd International Symposium On On-Line Testing and Robust
https://ieeexplore.ieee.org/document/7604689
LDPC block codes, created by R.G. Gallager in 1960, initially started to succumb to
oblivion, due to the current era’s technology limitations, which means that it could not be
reasonable anymore. This paper, further on goes to explain the absolute basics of LDPC and
the two method’s i.e., (i) by Gaussian elimination and (ii) through the Richardson – Urbanke
(RU) method.
Here in this paper, the authors have introduced a encoder architecture, which they
claim to be suitable for Consultative Committee for Space Data System (CCSDS) which is an
a multi-national forum for the development of communications and data systems standards
for spaceflight. The paper claims that the Quasi – Cyclic Low – Density Parity – Check Codes
(QC-LDPC) have been adopted by the above-mentioned authority. The motivation for them
to develop this new encoder since the already existing LDPC codes are not efficient
implementations for CCSDS and are more optimized for other standards such as IEEE
802.11e, 802.11n, 802.11 ac, Flash memory Storage, etc. The aim and motivation of this
paper is to propose an architecture which is based on the direct method of multiplication of
an input information block vector with the generator matrix of the code and actual encoder
implementations are demonstrated for the special case of the two classes of LDPC codes
adopted in the CCSDS space communication protocols. The authors have also compared it to
the existing and market solutions but state claim that their designed encoders outperform
the already designed one’s with state – of – the – art throughput performance, while at the
same time keeping resource utilization at a minimum.
The paper covered up the absolute basics of space communication and its protocols
which enlightened me into digging deeper into the same. The main aim of this paper was to
cover the Space Communication Protocol Specifications (SCPS), which is nothing but the data
link layer of CCSDS and more specifically the synchronization and channel – coding sub –
layer of telemertry Space Data – Link Protocol (TM – SDLP). Based on the author’s research,
for deep – space communications, nine AR4JA codes are definied in the above mentioned
standard which are the result of the combination of three block length sizes (1024, 4096 and
16384) over three code rates: ½, 2/3 and 4/5. The authors have cerated a encoder
architecture through all possibilities. The diagram of the proposed encoder is given in the
figure below:
Based on the communication standard, the output of the encoder is the channel access daa
unit (CADU), which is nothing but the optionally randomized codeword, for the AR4JA and
C2 LDPC code. Each data bus is Lm and La bits wide. What these are the number of parallel
outputs created by a single cursive convolutional encoder better known as RCE. These two
parameters help us to find out the latency given by
Latency=
(m)
( L a−1 ) Lm
La
Further on, the author’s have made a conclusion table based on other paper’s values
and compared them to theirs. Based on their proposed architecture, they managed to
achieve a throughput over 9 times higher than conventional method for AR4JA and over 7
times higher for C2 codes. The paper was very well explained, although the language could
have been a lot simpler to understand as there were times, I had to refer other sources to
understand the mindset of the authors.
2. R. Adhikary, J. N. Daigle and L. Cao, "Protocol Design for Content Transfer in Deep Space
Network," GLOBECOM 2017 - 2017 IEEE Global Communications Conference, Singapore,
2017, pp. 1-6.
https://ieeexplore.ieee.org/document/8255096
This paper mainly focuses on the protocol design for data transfer using RaptorQ
codes and turbo codes with a practical channel prediction mode. The main motivation of this
paper was to standardize and effectively help content transfer for the new at – that – time
Mars landings.
Every deep space mission has a communication system to send and to receive
commands to and fro the spacecraft from earth. The paper says, that on a daily basis,
telemetry files are prepared and need to be telemetered to the deep space network
stations. The space communication channel over which the content transfer takes place is
characterized by large round – trip times, intermittent connectivity and highly variable
propagation channels. One important finding mentioned in this paper is that, for a real time
channel condition prediction mechanism is not utilized for content ransxfer from space
vehicles to DSN stations. The standard usually does a background sequencing where in, a
spacecraft is made to reconfigure it’s paramteres once every 28 days. Here in this paper, the
authors have proposed a new content delivery protocol using RaptorQ codes, turbo codes
and a channel prediction model.
The paper proposed to use RaptorQ codes at the application layer of the CCSDS
protocol stack mentioned in the previous paper. Moreover, they also use a dynamic turbo
coding system for data telemetry between orbiter to DSN staiotns. Reason why turbo codes
is because they are the most advanced forward error correction (FEC) codes and can achieve
near Shannon – limit error correction performance with reasonable coding and decoding
complexity. The framze sizes is the same as mentioned in the previous paper. RaptorQ codes
are the most advanced and efficnet fountain codes designed to date. Reason being is
because it can generate theoretically infinite number of encoded symbols. It removes the
need of Negative Acknowledgement (NAK) and hence that reduces delay. The methodology
was very straightforwardly put for both RaptorQ and Turbo code with it’s maximum SNR.
The channel prediction model achieved a 80% positive rate.
The paper concludes with the simulation results of the channel prediction model.
Based on the other papers, reviewed by the author, the result was very upto the mark. The
paper could have focused more towards explaining the equations given although just the
basic idea of proposing this method was interesting to read about.