Show simple item record



Item Description

dc.contributor.author Miller, Michael Joseph en_US
dc.date.accessioned 2009-07-15T17:42:47Z en_US
dc.date.available 2009-07-15T17:42:47Z en_US
dc.date.issued 1982 en_US
dc.identifier.uri http://hdl.handle.net/10125/9741 en_US
dc.description Thesis (Ph. D.)--University of Hawaii at Manoa, 1982. en_US
dc.description Bibliography: leaves 188-191. en_US
dc.description Microfiche. en_US
dc.description xi, 191 leaves, bound ill. 29 cm en_US
dc.description.abstract Automatic-repeat-request (ARQ) systems have been the most popular means for error control in digital transmission systems. They provide a relatively simple and highly reliable means for eliminating transmission errors. However, the throughput of an ARQ system may deteriorate badly with increasing bit-error rates especially if there are significant transmission delays such as experienced in satellite or long terrestrial circuits. This dissertation first proposes a class of mixed-mode protocols which incorporate a selective-repeat mode of retransmission. This is combined with a secondary mode to prevent receiver buffer overflow. The throughput analysis for these schemes is presented and shows that they can significantly outperform the conventional Go-Back-N procedure for transmission over circuits with delay. The analysis also shows how throughput is related to the size of buffer provided at the receiver. It is also demonstrated that the choice of secondary retransmission mode does not have a significant effect on the throughput but has a bearing on complexity. Further improvement in performance may be achieved by use of a hybrid ARQ system incorporating forward-error correction as well as retransmission. The dissertation considers some parity-retransmission schemes in which blocks of parity bits are used for retransmissions rather than repetition of the original information block. This enables the system to adaptively incorporate error correction as well as detection when channel bit-error rates increase. An analysis procedure is presented which permits comparison of throughput efficiency for a variety of ARQ retransmission protocols and forward-error correction codes. Particular attention is focused on the use of half-rate convolutional codes for error correction. A trellis algorithm is shown to be useful for computation of the error correction capability of modestly powerful convolutional codes with sliding-block feedback decoding. An alternative approach using combinatorial procedures is also presented. New rate one-half codes are found which are related to optimum orthogonalizable rate one-third codes. These related code pairs can be used in parity-retransmission schemes to provide more powerful error correction when channel bit-error rates deteriorate badly. Finally the throughput and reliability performance of the hybrid schemes is outlined for various combinations of retransmission protocols and error-correction systems. Results are presented for convolutional codes and block codes and indicate the possible tradeoffs between complexity and performance. It is concluded that convolutional codes with relatively simple sliding-block decoding can ensure high throughput is maintained on a hybrid ARQ system despite significant bit-error rates and transmission delays. en_US
dc.language.iso en-US en_US
dc.relation Theses for the degree of Doctor of Philosophy (University of Hawaii at Manoa). Electrical Engineering; no. 1554 en_US
dc.rights All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner. en_US
dc.subject Data transmission systems en_US
dc.subject Digital communications en_US
dc.subject Error-correcting codes (Information theory) en_US
dc.title Automatic-repeat-request systems for error control in digital transmission en_US
dc.type Thesis en_US
dc.type.dcmi Text en_US

Item File(s)

Description Files Size Format View
Restricted for viewing only uhm_phd_8220042_r.pdf 4.303Mb PDF View/Open
For UH users only uhm_phd_8220042_uh.pdf 4.247Mb PDF View/Open

This item appears in the following Collection(s)

Show simple item record

Search


Advanced Search

Browse

My Account

Statistics

About