Please use this identifier to cite or link to this item:
|Title:||RAPTOR-CODED FREE-SPACE OPTICAL COMMUNICATION EXPERIMENT|
|Department:||Electrical and Computer Engineering|
|Abstract:||Free-space optical (FSO) communications are attractive in a host of applications, such as last-mile connections with backbone networks, due to their high data rates, license-free operation, low interference and ease-of-deployment. However, FSO communication systems suffer from availability issues mostly caused by bad weather conditions, thus preventing the widely deployment of FSO communication technology. Raptor codes are a class of fountain codes which adapt to varying channel conditions. The rateless property of Raptor codes makes them ideal for channels with high vulnerability such as FSO channels. This work presents a performance evaluation of two classes of Raptor codes in FSO communications: Raptor 10 and RaptorQ codes. Both of the Raptor codes are capable of improving the reliability of FSO communications and adapting FSO data rates regardless of channel conditions. Compared with the Raptor 10, RaptorQ codes are capable of offering more robust communication performance. This work presents the software implementation of soft-switching hybrid FSO/LAN link using a Raptor 10 code. Preliminary experimental measurements show that the hybrid link could exploit the bandwidth of both links and avoids availability issues caused by harsh weather conditions. This work also includes an FPGA-based hardware implementation of a Raptor 10 coded FSO link with focus on the receiver architecture. A complete hardware architecture of Raptor 10 decoder is presented in this work. The hardware structure of a pipelined receiver is proposed and evaluated, showing the important role of Raptor codes in FSO communications. This work serves as the first step towards the FPGA-based hardware implementation of soft-switching hybrid FSO/RF link using Raptor codes.|
|Appears in Collections:||Open Access Dissertations and Theses|
Files in This Item:
|Lu_Min_finalsubmission201512_MSc_degree.pdf.pdf||Thesis||3.46 MB||Adobe PDF||View/Open|
Items in MacSphere are protected by copyright, with all rights reserved, unless otherwise indicated.