Transmission of JPEG2000 images over frequency‐selective channels with unequal power allocation. Issue 1 (1st February 2013)

• Record Type: Journal Article
• Authors: Shayegannia, Moein; Hajshirmohammadi, Atousa; Muhaidat, Sami; Torki, Mahin
• Is Part Of: IET image processing. Volume 7:Issue 1(2013)
• Page Start: 33
• Page End: 41
• Publication Date: 2013-02-01
• Subjects: distortion -- fading channels -- image coding -- OFDM modulation; JPEG2000 images -- unequal power allocation -- UPA -- orthogonal frequency division multiplexing -- block fading frequency selective channels -- distortion model -- coding pass -- channel state information -- cyclic prefixes -- channel memory length -- image quality -- multitap channel scenario -- optimisation algorithm; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/iet-ipr.2012.0179
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract : In this study, transmission of JPEG2000 images using an unequal power allocation (UPA) scheme and orthogonal frequency division multiplexing (OFDM) over block‐fading frequency‐selective channels is presented. A distortion model is provided to evaluate the contribution of each coding pass (CP) in the construction of the received image. The optimisation algorithm exploits the hierarchical structure of the JPEG2000 images and uses the distortion model along with the channel state information for allocating optimal values of power for each CP to minimise the end‐to‐end distortion. Furthermore, the actual total power consumed for transmission is measured and compared with the total power initially assigned. For the purpose of simulations, the authors set the number of OFDM subcarriers to be 16, the length of the cyclic prefixes equal to the channel memory length and analyse the quality of the received image in a 2‐tap and 3‐tap frequency‐selective channel, with and without our proposed UPA technique in an OFDM system. The results show an improvement of up to 10.5 dB in the decoded image quality when the UPA scheme is used. In addition, our system manages to maintain similar quality for the received image in a multi‐tap channel scenario.