期刊名称：International Journal of Advanced Computer Science and Applications(IJACSA)
印刷版ISSN：2158-107X
电子版ISSN：2156-5570
出版年度：2015
卷号：6
期号：6
DOI：10.14569/IJACSA.2015.060640
出版社：Science and Information Society (SAI)
摘要：Over the years, several techniques for improving throughput of wireless communication have been developed in order to cater for the ever increasing demand of high speed network service. However, these techniques can only give little improvement in performance because packets have to be delivered as is. As such researchers have begun thinking outside the box by proposing ideas that require relay nodes to temper packets’ contents in order to improve the throughput of a network. One of the state of the art techniques in this field is called Network Coding (NC). NC is a state of the art technique that allows relay nodes linearly combine two or more packets in a way they can be recovered upon reaching their destination. However, increasing packet size increases possibility of error affecting it. In this paper, the authors decide to investigate whether adding data recovery technique can improve the performance of a network that uses network coding, if it can, by how much can it? Is it worth the trouble? In order to answer these questions, the authors carried out a quantitative analysis of throughput in a Stop-and-Wait Automatic Repeat reQuest (SW-ARQ) data transmission system with Network Coding (NC) and Forward Error Correction (FEC). Vandermonde matrix is chosen as the coding technique for this research because it has both NC and data recovery characteristics. Python programming language is used to develop three Discrete Event Simulations: SW-ARQ without any NC, SW-ARQ with NC and SW-ARQ with NC and FEC. The obtained results show that SW-ARQ with NC and FEC is superior to traditional SW-ARQ in terms of throughput, especially in channels with high error rates.
