The rapid growth of wireless systems has shown little sign of ceasing, due to increased
consumer demand for reliable interactive services. A key component of the development has
centered on satellite networks, which allows provision of services in scenarios where terrestrial
systems are not viable. The Digital Video Broadcasting-Satellite Second Generation (DVB-S2)
standard was developed for use in satellite broadcast applications, the foremost being video
broadcasting. Inherent to DVB-S2 is a powerful forward error correction (FEC) module, present
in both the Physical and Data Link Layer. Improving the error correcting capability of the FEC
is a natural advent in improving the quality of service of the protocol. This is more crucial in
real time satellite video broadcast where retransmission of data is not viable, due to high
latency.
The Physical Layer error correcting capability is implemented in the form of a concatenated
BCH-LDPC code. The DVB-S2 standard does not define the decoding structure for the receiver
system however many powerful decoding systems have been presented in the literature; the
Belief Propagation-Chase concatenated decoder being chief amongst them. The decoder utilizes
the concept of soft information transfer between the Chase and Belief Propagation (BP)
decoders to provide improved error correcting capability above that of the component decoders.
The following dissertation is motivated by the physical layer (PL) FEC scheme, focused on the
concatenated Chase-BP decoder. The aim is to generate results based on the BP-Chase decoder
in a satellite channel as well as improve the error correcting capability.
The BP-Chase decoder has shown to be very powerful however the current literature provides
performance results only in AWGN channels. The AWGN channel however is not an accurate
representation of a land-mobile satellite (LMS) channel; it does not consider the effect of
shadowing, which is prevalent in satellite systems. The development of Markov chain models
have allowed for better description of the characteristics of the LMS channel. The outcome
being the selection of a Ku band LMS channel model. The selected LMS channel model is
composed of 3 states, each generating a different degree of shadowing. The PL system has been
simulated using the LMS channel and BP-Chase receiver to provide a more accurate
representation of performance of a DVB-S2 network. The effect of shadowing has shown to
reduce coding performance by approximately 4dB, measured over several code lengths and
decoders, when compared with AWGN performance results.
The second body of work aims to improve the error correcting capability of the BP-Chase
decoder, concentrating on improving the LDPC decoding module performance. The LDPC
system is the basis for the powerful error correcting ability of the concatenated scheme. In
attempting to improve the LDPC decoder a reciprocal improvement is expected in the overall
decoding performance of the concatenated decoder. There have been several schemes presented
which improve BP performance. The BP-Ordered statistics decoder (OSD) was selected
through a process of literary review; a novel decoding structure is presented incorporating the
BP-OSD decoder into the BP-Chase structure. The result of which is the BP-OSD-Chase
decoder. The decoder contains two stages of concatenation; the first stage implements the BPOSD
algorithm which decodes the LDPC code and the second stage decodes the BCH code
using the Chase algorithm. Simulation results of the novel decoder implementation in the DVBS2
PL show a coding gain of 0.45dB and 0.15dB versus the BP and BP-Chase decoders
respectively, across both the AWGN and LMS channel. / Thesis (M.Sc.Eng.)-University of KwaZulu-Natal, Durban, 2012.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:ukzn/oai:http://researchspace.ukzn.ac.za:10413/9670 |
| Date | January 2012 |
| Creators | Naidoo, Theran. |
| Contributors | Takawira, Fambirai., Xu, Hong-Jun. |
| Source Sets | South African National ETD Portal |
| Language | en_ZA |
| Detected Language | English |
| Type | Thesis |
Page generated in 0.0139 seconds