End-to-end data reliability in optical flow switching
Author(s)Li, Rui, Ph. D. Massachusetts Institute of Technology
Massachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Vincent W.S. Chan and John M. Chapin.
MetadataShow full item record
Ever since optical fiber was introduced in the 1970s as a communications medium, optical networking has revolutionized the telecommunications landscape. With sustained exponential increase in bandwidth demand, innovation in optical networking needs to continue to ensure cost-effective communications in the future. Optical flow switching (OFS) has been proposed for future optical networks to serve large transactions in a cost-effective manner, by means of an all-optical data plane employing end-to-end lightpaths. Due to noise added in the transmission and detection processes, the channel has non-zero probability of bit errors that may corrupt the useful data or flows transmitted. In this thesis, we focus on the end-to-end reliable data delivery part of the Transport Layer protocol and propose effective and efficient algorithms to ensure error-free end-to-end communications for OFS. We will analyze the performance of each algorithm and suggest optimal algorithm(s) to minimize the total delay. We propose four classes of OFS protocols, then compare them with the Transport Control Protocol (TCP) over Electronic Packet Switching (EPS) and indicate under what values of the parameters: file size, bit error rate (BER), propagation delay and loading factor is OFS better than EPS. This analysis can serve as important guidelines for practical protocol designs for end-to- end data transfer reliability of OFS.
Thesis (S.M.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 2011.Cataloged from PDF version of thesis.Includes bibliographical references (p. 217-220).
DepartmentMassachusetts Institute of Technology. Dept. of Electrical Engineering and Computer Science.
Massachusetts Institute of Technology
Electrical Engineering and Computer Science.