Cooperative and relay wireless communication networks use one or more distinct wire-
less nodes as relays to combat performance impairments, which include signal atten-
uation due to power loss over long distances and the signal power fluctuation due to
multi-path fading. The second one is the primary focus of this thesis.
Diversity is an effective weapon to combat the fading effects and using multiple
antennas at both the transmitter and receiver is a way of achieving it. However, using
multiple antennas may not be practical for some mobile terminals due to the limited
physical size, battery power and computational capability of these mobile terminals.
Therefore, cooperative schemes have been proposed in which several wireless nodes
cooperate with each other to achieve diversity even though each node has only one an-
tenna. Many cooperative schemes have been proposed including Amplify-and-Forward
(AF) and Decode-and-Forward (DF) [1] to achieve this diversity. This thesis primarily
proposes and studies a DF scheme.
For the DF schemes, imperfect decoding at the relay is one of the major limitations
to the achievable performance. To address this problem and yet remain simple to
implement, a novel DF based cooperative scheme is proposed in this thesis. This
proposed scheme uses a distributed Turbo coding (DTC) scheme and is resilient to relay
errors. It is analyzed by using the approximate Gaussian density evolution method.
The analysis shows how the iterative Turbo decoding performs with different bit error
rates (BERs) at the relay. Based on the results, a novel scheme, which adaptively
changes the code used at the relay, is then proposed to further improve performance.
The diversity offered by the proposed scheme is achieved at the cost of decreased
spectral efficiency compared to a non-cooperative network. To recover some of the
spectral efficiency loss due to cooperation, the proposed single source node cooperative
scheme is extended to a two source nodes (users) scheme. The relay node forwards a
combined packet which is formed by combining multiple users' information packets into
a single packet. The results show that combining multiple users' information packets
using superposition modulation can achieve better performance than that of using
"XOR" operation (network coding) when two users have different channel qualities,
which is very common in a wireless communication environment.
Identifer | oai:union.ndltd.org:canterbury.ac.nz/oai:ir.canterbury.ac.nz:10092/5979 |
Date | January 2011 |
Creators | Lin, Rui |
Publisher | University of Canterbury. Electrical and Computer Engineering |
Source Sets | University of Canterbury |
Language | English |
Detected Language | English |
Type | Electronic thesis or dissertation, Text |
Rights | Copyright Rui Lin, http://library.canterbury.ac.nz/thesis/etheses_copyright.shtml |
Relation | NZCU |
Page generated in 0.0019 seconds