(Back to Session Schedule)

The 23rd International Technical Conference on Circuits/Systems, Computers and Communications

Session G6  Signal Processing for Communications
Time: 9:00 - 10:30 Wednesday, July 9, 2008
Location: 8F 804 Room
Chairs: Masaaki Yamanaka (Hiroshma International University, Japan), Sung-Bum Park (Samsung Electronics, Republic of Korea)

G6-1 (Time: 9:00 - 9:18)
TitleA Packet Loss Concealment Algorithm Robust to Burst Packet Loss for CELP-type Speech Coders
AuthorChoong Sang Cho (SoC Research Center, Korea Electronics Technology Institute, Republic of Korea), *Nam In Park, Hong Kook Kim (Department of Information and Communications, Gwangju Institute of Science and Technology, Republic of Korea)
Pagepp. 941 - 944
KeywordG.729, G.729.1, PLC
AbstractIn this paper, we proposed a packet loss concealment algo-rithm for a CELP-type speech coder that reduces the degra-dation of speech quality due to packet loss. The proposed PLC consists of a voicing probability, a periodic/random excitation generation, and speech amplitude control. We evaluated the performance of the proposed algorithm on G.729 under random and burst packet loss rates of 3% and 5% and compare it with PLC employed in G.729. From the PESQ measure, an A-B preference tests and a waveform comparison, it was shown that the proposed PLC algorithm provided better speech quality than the PLC employed in G.279 under packet loss rates of 3% and 5%.

G6-2 (Time: 9:18 - 9:36)
TitlePredictor Order and Error Distribution of MMAE Predictors for Lossless Image Coding
Author*Hirokazu Ueno, Yoshitaka Morikawa (Okayama University, Japan)
Pagepp. 945 - 948
Keywordlossless image coding, MMAE predictor, predictor order, error distribution
AbstractThis paper investigates the relation between an error distribution and a predictive order of minimum mean abusolute error predictors(MMAE predictors) designed for lossless coding of greyscale images. Design of MMAE predictors reduces to the linear programming problem. Let k is the number of coefficients in a predictor(predictor order), we imagine that predictor order k have a distribution shaping effect. Main purpose of this paper is to ensure that k have such an effect.

G6-3 (Time: 9:36 - 9:54)
TitleHigh Dynamic Range Image Watermarking
AuthorFabrizio Guerrini (University of Brescia, Italy), Masahiro Okuda (The University of Kitakyushu, Japan), *Nicola Adami, Riccardo Leonardi (University of Brescia, Italy)
Pagepp. 949 - 952
Keywordwatermarking , high dynamic range images
AbstractHigh Dynamic Range (HDR) images represent the future format for digital images since they allow more sensitive rendering. However, nowadays special types of pre-processing collectively known as tone mapping operators are needed to adapt HDR images to currently existing displays. Tone mapped images, although of reduced dynamic range, have nonetheless high quality and hence retain some commercial value. In this paper we propose a solution to the problem of HDR watermarking, e.g. for copyright embedding, that should survive tone mapping. Therefore, the requirements imposed on the watermark encompass imperceptibility, a certain degree of security and robustness to tone mapping operators. The proposed watermarking system belongs to the blind, detectable category, it is based on the QIM paradigm and employs a higher order statistic as feature. Preliminary experimental analysis shows positive results; more work is needed and currently underway.

G6-4 (Time: 9:54 - 10:12)
TitleLossless Compression of Color Halftone Images using Color Channel Adaptive Templates
Author*Sung-Bum Park, Woo-Sung Shim, Young-Ho Moon, Jong-Bum Choi, Dai-Woong Choi, Jae-Won Yoon (Samsung Electronics Co., Ltd., Republic of Korea)
Pagepp. 953 - 956
Keywordlossless compression, color halftone image, template design, line pattern
AbstractA novel lossless compression scheme is presented for color halftone images dithered by a clustered dot ordered screen. In the encoder, the context template is designed according to the line pattern of each color channel image. Based on the designed template, then, each channel image is compressed by a context-based arithmetic encoder. Finally, both the compressed image data and the template information are transmitted to the decoder. In the decoder, the template shape information is reconstructed, and then, the halftone image is decoded losslessly. Based on the adaptiveness of the template to the input image, the proposed system yields better compression performance than the conventional JBIG, which saves 35% of the JBIG bitstream.

G6-5 (Time: 10:12 - 10:30)
TitleSynthesis of a Complex Filter Using Lossy Transformers
Author*Kazuhiro Shouno, Yukio Ishibashi (University of Tsukuba, Japan)
Pagepp. 957 - 960
Keyworddissipative, passive, analog, complex, filter
AbstractIn this paper, the synthesis of a dissipative complex filter is proposed. The proposed circuit is realized by using resistors, capacitors, and lossy transformers. As an example, a third-order complex filter is designed and its validity of the proposed method was confirmed through computer simulation.