Payload Damage Estimation for Hybrid ARQ

碩士 === 臺灣大學 === 電信工程學研究所 === 95 === Hybrid automatic retransmission request (HARQ) is an important technique for communications without CSI. When the retransmission arrives, as opposed to simple ARQ, in HARQ all the received transmissions for a payload are used together to determine the common under...

Full description

Bibliographic Details
Main Authors: Wei-Chen Wang, 王偉震
Other Authors: Da-Shan Shiu
Format: Others
Language:en_US
Published: 2007
Online Access:http://ndltd.ncl.edu.tw/handle/88195521981830658820
id ndltd-TW-095NTU05435125
record_format oai_dc
spelling ndltd-TW-095NTU054351252015-10-13T13:55:55Z http://ndltd.ncl.edu.tw/handle/88195521981830658820 Payload Damage Estimation for Hybrid ARQ 用於HybridARQ之資料傳輸損害估計 Wei-Chen Wang 王偉震 碩士 臺灣大學 電信工程學研究所 95 Hybrid automatic retransmission request (HARQ) is an important technique for communications without CSI. When the retransmission arrives, as opposed to simple ARQ, in HARQ all the received transmissions for a payload are used together to determine the common underlying payload. This could allow the transmitter to be more aggressive in its selection of the transmission rate. Intuitively, the amount of resources that a transmitter needs to invest in a retransmission to repair a previously failed transmission depends on how severely the prior transmission is damaged. If the payload damage information (PDI) can be made available at the transmitter, it could signi‾cantly improve the effciency of retransmission. Compared to a typical HARQ scheme, a PDI-available HARQ transmission scheme requires the estimation and description of damage at the receiver and a corresponding resource allocation determination scheme at the transmitter. In this thesis, we investigate the problem of how and how well a receiver can estimate the damage suffered by a failed payload. First, we will show that in order to achieve the ultimate performance in damage estimation, one must estimate damage from the received signal itself. Auxiliary signals and the received payload can be used in the process of channel state estimation. As will be shown in this paper, fundamentally this type of approach cannot reach the optimal level of damage estimation. Next, we look at how the payload damage is to be estimated directly from the received signal. Our main approach is to carefully examine the geometrical space around the received signal to assess potential damage. The constellations near the received signal can be estimated using a multiple-output decoder such as a list decoder. To illustrate how the idea can be applied in practice, we introduce a simple yet effective damage estimator for an convolutionally-coded OFDM link over a frequency selective medium. We can compute the effective minimum distance of the received signal and apply this result to assist in designing retransmission mechanism. We also discuss the output patterns of list decoder in the rest of this thesis. Da-Shan Shiu 許大山 2007 學位論文 ; thesis 74 en_US
collection NDLTD
language en_US
format Others
sources NDLTD
description 碩士 === 臺灣大學 === 電信工程學研究所 === 95 === Hybrid automatic retransmission request (HARQ) is an important technique for communications without CSI. When the retransmission arrives, as opposed to simple ARQ, in HARQ all the received transmissions for a payload are used together to determine the common underlying payload. This could allow the transmitter to be more aggressive in its selection of the transmission rate. Intuitively, the amount of resources that a transmitter needs to invest in a retransmission to repair a previously failed transmission depends on how severely the prior transmission is damaged. If the payload damage information (PDI) can be made available at the transmitter, it could signi‾cantly improve the effciency of retransmission. Compared to a typical HARQ scheme, a PDI-available HARQ transmission scheme requires the estimation and description of damage at the receiver and a corresponding resource allocation determination scheme at the transmitter. In this thesis, we investigate the problem of how and how well a receiver can estimate the damage suffered by a failed payload. First, we will show that in order to achieve the ultimate performance in damage estimation, one must estimate damage from the received signal itself. Auxiliary signals and the received payload can be used in the process of channel state estimation. As will be shown in this paper, fundamentally this type of approach cannot reach the optimal level of damage estimation. Next, we look at how the payload damage is to be estimated directly from the received signal. Our main approach is to carefully examine the geometrical space around the received signal to assess potential damage. The constellations near the received signal can be estimated using a multiple-output decoder such as a list decoder. To illustrate how the idea can be applied in practice, we introduce a simple yet effective damage estimator for an convolutionally-coded OFDM link over a frequency selective medium. We can compute the effective minimum distance of the received signal and apply this result to assist in designing retransmission mechanism. We also discuss the output patterns of list decoder in the rest of this thesis.
author2 Da-Shan Shiu
author_facet Da-Shan Shiu
Wei-Chen Wang
王偉震
author Wei-Chen Wang
王偉震
spellingShingle Wei-Chen Wang
王偉震
Payload Damage Estimation for Hybrid ARQ
author_sort Wei-Chen Wang
title Payload Damage Estimation for Hybrid ARQ
title_short Payload Damage Estimation for Hybrid ARQ
title_full Payload Damage Estimation for Hybrid ARQ
title_fullStr Payload Damage Estimation for Hybrid ARQ
title_full_unstemmed Payload Damage Estimation for Hybrid ARQ
title_sort payload damage estimation for hybrid arq
publishDate 2007
url http://ndltd.ncl.edu.tw/handle/88195521981830658820
work_keys_str_mv AT weichenwang payloaddamageestimationforhybridarq
AT wángwěizhèn payloaddamageestimationforhybridarq
AT weichenwang yòngyúhybridarqzhīzīliàochuánshūsǔnhàigūjì
AT wángwěizhèn yòngyúhybridarqzhīzīliàochuánshūsǔnhàigūjì
_version_ 1717745614835417088