The University of Southampton
University of Southampton Institutional Repository

Iterative Source-and-channel decoding aided video communications

Iterative Source-and-channel decoding aided video communications
Iterative Source-and-channel decoding aided video communications
Shannon’s source-and channel-coding separation theorem states that reliable near-capacity transmission can be accomplished by separate source coding using lossless entropy codes and channel coding under the idealized assumption of Gaussian channels and potentially infinite encoding/decoding delay as well as complexity. However, it is impossible to remove all source redundancy with the aid of practical finite-delay and finite-complexity source encoders. As a remedy, joint source-channel coding (JSCC) has been used for achieving an improved system performance by exploiting the residual source correlation.

We propose a novel tree-structured multiple description coding (T-MDC) scheme that may be combined with arbitrary video codecs for the sake of creating multiple video descriptions. The technique advocated splits the original video signal into an appropriately chosen number of correlated descriptions in the time-domain, while retaining the correlation among the video frames within each description. Each description may be encoded using arbitrary video compression tools into a bitstream. T-MDC was also employed in the scenario of multiple description coding for multiview video communications. Furthermore, our proposed scheme is also capable of splitting the video stream into multiple descriptions of unequal importance.

Then, a novel inter-layer forward error correction (IL-FEC) coded video scheme is proposed, where the information of the base layer (BL) is incorporated into the systematic bits of the enhancement layers (ELs) with the aid of an exclusive or (XOR) operation. When the BL can be successfully decoded in its own right, the systematic bits of the ELs can be extracted by flipping the sign of the check information received without introducing any degradation, where the check information is generated by performing IL XOR operations on the BL and the ELs. However, when the BL cannot be correctly decoded without the assistance of the ELs, the IL-FEC decoding philosophy exchanging information between the BL and the ELs will be activated to assist in decoding the BL.

We then conceive a two-dimensional (2D) iterative Markov process aided decoder for a video receiver, which may be combined with channel decoding. Furthermore, a reduced complexity first-order Markov model based source decoder will be derived. Iterative decoding is performed by exchanging extrinsic information between two source decoders. Explicitly, we propose the first-order Markov process aided three-dimensional (3D) iterative source channel decoding (ISCD) concept relying on an recursive systematic convolutional (RSC) codec invoked for uncompressed video transmissions, where both the horizontal and vertical intra-frame correlations as well as the inter frame correlations are exploited. The proposed technique is capable of exploiting both the intra-frame and inter-frame correlations for iterative source-channel decoding.

Finally, we study the application of ISCD conceived for distributed video coding (DVC), where the video signals are modelled by a first-order Markov process. A horizontal and a vertical source decoder are employed for exchanging their information using the iterative decoding philosophy. This scheme may be combined with the entire suite of classic FEC codecs employed in state-of-the-art DVC systems. We benchmark the attainable system performance against that of the existing pixeldomain Wyner-Ziv (PDWZ) video coding systems. Finally, we exploit the inter-view correlation with the aid of inter-view motion search in distributed multi-view video coding (DMVC). We rely on the system architecture of WZ coding invoked for multiview video. We construct a novel mesh-structured pixel-correlation model from the inter-view motion vectors (MVs) and derive its decoding rules for joint source-channel decoding (JSCD). The proposed system was benchmarked against the existing PDWZ coding based DMVC scheme.
Huo, Yongkai
f2be4331-3c36-48c4-b4df-f5fa29a45e94
Huo, Yongkai
f2be4331-3c36-48c4-b4df-f5fa29a45e94
Hanzo, L.
66e7266f-3066-4fc0-8391-e000acce71a1

Huo, Yongkai (2013) Iterative Source-and-channel decoding aided video communications. University of Southampton, Physical Sciences and Engineering, Doctoral Thesis, 280pp.

Record type: Thesis (Doctoral)

Abstract

Shannon’s source-and channel-coding separation theorem states that reliable near-capacity transmission can be accomplished by separate source coding using lossless entropy codes and channel coding under the idealized assumption of Gaussian channels and potentially infinite encoding/decoding delay as well as complexity. However, it is impossible to remove all source redundancy with the aid of practical finite-delay and finite-complexity source encoders. As a remedy, joint source-channel coding (JSCC) has been used for achieving an improved system performance by exploiting the residual source correlation.

We propose a novel tree-structured multiple description coding (T-MDC) scheme that may be combined with arbitrary video codecs for the sake of creating multiple video descriptions. The technique advocated splits the original video signal into an appropriately chosen number of correlated descriptions in the time-domain, while retaining the correlation among the video frames within each description. Each description may be encoded using arbitrary video compression tools into a bitstream. T-MDC was also employed in the scenario of multiple description coding for multiview video communications. Furthermore, our proposed scheme is also capable of splitting the video stream into multiple descriptions of unequal importance.

Then, a novel inter-layer forward error correction (IL-FEC) coded video scheme is proposed, where the information of the base layer (BL) is incorporated into the systematic bits of the enhancement layers (ELs) with the aid of an exclusive or (XOR) operation. When the BL can be successfully decoded in its own right, the systematic bits of the ELs can be extracted by flipping the sign of the check information received without introducing any degradation, where the check information is generated by performing IL XOR operations on the BL and the ELs. However, when the BL cannot be correctly decoded without the assistance of the ELs, the IL-FEC decoding philosophy exchanging information between the BL and the ELs will be activated to assist in decoding the BL.

We then conceive a two-dimensional (2D) iterative Markov process aided decoder for a video receiver, which may be combined with channel decoding. Furthermore, a reduced complexity first-order Markov model based source decoder will be derived. Iterative decoding is performed by exchanging extrinsic information between two source decoders. Explicitly, we propose the first-order Markov process aided three-dimensional (3D) iterative source channel decoding (ISCD) concept relying on an recursive systematic convolutional (RSC) codec invoked for uncompressed video transmissions, where both the horizontal and vertical intra-frame correlations as well as the inter frame correlations are exploited. The proposed technique is capable of exploiting both the intra-frame and inter-frame correlations for iterative source-channel decoding.

Finally, we study the application of ISCD conceived for distributed video coding (DVC), where the video signals are modelled by a first-order Markov process. A horizontal and a vertical source decoder are employed for exchanging their information using the iterative decoding philosophy. This scheme may be combined with the entire suite of classic FEC codecs employed in state-of-the-art DVC systems. We benchmark the attainable system performance against that of the existing pixeldomain Wyner-Ziv (PDWZ) video coding systems. Finally, we exploit the inter-view correlation with the aid of inter-view motion search in distributed multi-view video coding (DMVC). We rely on the system architecture of WZ coding invoked for multiview video. We construct a novel mesh-structured pixel-correlation model from the inter-view motion vectors (MVs) and derive its decoding rules for joint source-channel decoding (JSCD). The proposed system was benchmarked against the existing PDWZ coding based DMVC scheme.

Text
Huo.pdf - Other
Download (10MB)

More information

Published date: November 2013
Organisations: University of Southampton, Southampton Wireless Group

Identifiers

Local EPrints ID: 361905
URI: http://eprints.soton.ac.uk/id/eprint/361905
PURE UUID: 274bf646-26fa-4034-8173-1e2ea10f3c3e
ORCID for L. Hanzo: ORCID iD orcid.org/0000-0002-2636-5214

Catalogue record

Date deposited: 10 Feb 2014 10:25
Last modified: 15 Mar 2024 05:02

Export record

Contributors

Author: Yongkai Huo
Thesis advisor: L. Hanzo ORCID iD

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×