Search

Abstract: In this paper, we present a novel scalable video transmission strategy over multi-input multi-output (MIMO) wireless systems with time-varying channel capacity. It is a great challenge to simultaneously guarantee the QoS for video delivery and maximize the system throughput over time-varying MIMO channel. We demonstrate that, by making full use of estimated channel state information (CSI) through feedback, a cascade of adaptive operations can be designed to satisfy maximum throughput for scalable video over MIMO systems. These operations include power allocation based on water-filling (WF), adaptive channel selection (ACS), and novel throughput maximizing power reallocation (PR). The proposed ACS transmission scheme enables overall increase in data throughput among enhancement layers by adaptively launching base layer bit-stream to proper sub-channel. Then, after initial power allocation with WF and proper adaptive mode selection, we obtain the surplus power across enhancement layer sub-channels which can be reallocated to some sub-channels by the proposed PR scheme. With such power reallocation, certain enhancement layers will be able to reach new level of QAM modulation through PR so as to maximize the system data throughput. We present in this paper some detailed analysis on these adaptive operations. We also present some simulation results to demonstrate that maximum throughput video transmission over MIMO wireless systems indeed can be achieved based on scalable video coding (SVC) and a sequence of appropriately designed adaptive operations. [Copyright &y& Elsevier]

An optimal joint source-channel coding (OJSCC) scheme is developed for encoding memoryless generalized Gaussian distribution (GGD) sources and transmission over noisy channels, including memoryless channels and bursty channels. In particular, we are able to incorporate both transition probability and bit error rate of a bursty channel model into an end-to-end rate-distortion (R-D) function. This R-D function results in an optimum tradeoff between source coding accuracy and channel error protection under a fixed transmission rate. Based on the results of OJSCC, we further proposed a robust joint source-channel coding (RJSCC) scheme based on a combination of OJSCC with allpass filtering source reshaping. Experimental results based on a broad class of GGD sources suggest that the RJSCC scheme consistently outperforms the OJSCC scheme when the shape factor, a parameter of GGD, is less than 2.0. The comparison of RJSCC with the state-of-the-art robust quantization (RQ) [IEEE Trans. Image Process. 7 (1998) 496-505] shows that RJSCC is comparable with RQ at low bit rate and significantly better than RQ at high bit rate. In addition, an image transmission system using RJSCC demonstrates that the RJSCC system can be efficiently employed in practical applications. [ABSTRACT FROM AUTHOR]

Discrete wavelet transform provides an advantageous framework of multiresolution space-frequency representation with promising applications in image processing. An efficient modeling of the spatial and frequency characteristics of the wavelet transform coefficients has been recognized as a key to designing efficient quantization for wavelet-based image coding. We have proposed a Bayesian estimation framework to provide a unified, simultaneous modeling of both the spatial distribution and the intensity distribution of the wavelet transform coefficients [IEEE Trans. Circuit Syst. Video Technol. 7 (1997) 343], where the spatially localized characteristics of a given subband are modeled by Gibbs random fields (GRF). In this study, we investigate the modeling of individual cluster intensity distribution of the subband coefficients, within the context of a joint scene and signal modeling. This joint modeling decomposes the overall distribution of the coefficients into the superposition of individual cluster distributions. Among Gaussian, Laplacian, and generalized Gaussian distributions, we conclude that the composite of multiple generalized Gaussian distributions is most consistent with the overall distribution of the coefficients for any given high frequency subband. Meanwhile, modeling by Laplacian distributions has its advantage in terms of low computational complexity. The joint modeling enables us to devise a novel scene adaptive and signal adaptive quantization that fully exploits the coding redundancies resulting from wavelet transform. The concept and the mechanism of the proposed Bayesian estimation framework are applicable and tunable to other wavelet-based image processing tasks. [Copyright &y& Elsevier]

In this paper, we present a multiple hierarchical image coding and transmission scheme in which both the fragile structure of the set partitioning in hierarchical trees (SPIHT) source coding and unequal bit error sensitivity in the coded bitstreams have been taken into account. Multiple subsampling is applied to split the wavelet coefficients of the source image into multiple subsources so that each subsource is a crude representation of the original image. Therefore, the sequential dependence in the coded bitstream is broken and the error propagation is limited to a single subsource. Each subsource is coded using the SPIHT algorithm to achieve the high source coding performance. Cyclic redundancy code and rate compatible punctured convolutional (RCPC) code have been employed as the channel coding method to offer unequal error protection to the source coded bits. In each substream, all the bits in the same bit-layer are protected with the same channel coding rate. The higher the bit-layer, the more the channel coding protection is provided. Experimental results show that the proposed scheme has an excellent error resilient performance for image transmission over wireless channels with time varying and bursty error characteristics. In particular, the basic image structure is preserved even in highly noisy channels. [Copyright &y& Elsevier]

Abstract: In this paper, we propose an integrated robust multiple description coding (MDC) scheme for compressed still images over error-prone channels with both packet loss and random bit errors. Multiple descriptions are first generated independently by using multiple description scalar quantizer (MDSQ). To achieve an excellent error resilient capability for each description, multiple bitstreams are then generated based on wavelet trees along the spatial orientations. The spatial orientation trees in the wavelet domain are individually encoded using SPIHT. Error propagation is thus limited within each bitstreams. However, synchronization words are usually needed to avoid error propagation across multiple independent bitstreams. In order to maintain high compression efficiency robust synchronization, we adopt error resilient entropy coding (EREC) to re-organize these variable-length bitstreams into fixed-length data slots before multiplexing and transmission. Therefore, the synchronization of the start of each bitstream can be automatically obtained at the receiver. Finally, to alleviate the devastating image degradation resulted from errors in the beginning of the bitstreams, we propose an error concealment technique to both constrain the EREC/MDC decoding and post-process the decoded wavelet coefficients. With analysis and experimental results, we have demonstrated how each of these components has contributed to the overall robustness of the proposed transmission of compressed images over wireless channels. [Copyright &y& Elsevier]

Abstract: For leaky prediction based FGS (Fine Granularity Scalability), constant quality constrained rate adaptation, i.e., how to optimally truncate/allocate bits given the current channel bandwidth, is still an open problem. The difficulty lies in obtaining accurate R–D (rate–distortion) curves for leaky prediction based FGS (L-FGS) due to the dependency among video frames. In this paper, we propose an accurate R–D model, which considers not only the distortion introduced in the current frame and the propagated distortion from the reference frame due to rate adaptation, but also the correlation between them. An excellent property of our proposed R–D model is that even when applying the model for a long video sequence without any update of the actual distortion values, the estimation error is still negligible and the error is not accumulated. Based on our proposed R–D model, a sliding window technique is further developed to solve the problem of constant quality constrained bit allocation. Experimental results show that the proposed R–D model is very accurate and the corresponding bit allocation algorithm can achieve much more smooth video quality than the traditional uniform bit allocation under both CBR (constant bit rate) and VBR (variable bit rate) channels. [Copyright &y& Elsevier]

Abstract: Current rate control schemes in video coding standards do not have efficient frame-level bit allocation due to the limitation of real-time encoding. In this paper, by taking advantage of offline video encoding, we proposed a frame-level bit allocation scheme based on a latest developed rate–distortion (R-D) model, the ρ-domain R-D model [IEEE Trans. Circuits Syst. Video Technol. (2001) 928], for low bit rate streaming applications. Specifically, at the encoder, we code each video sequence twice. In the first pass, we generate the R-D information of video sequences. Then, in the second pass, according to the available channel bandwidth, by exploiting the pre-generated R-D information we are able to implement frame-level bit allocation in an optimal way so that video sequences can be coded at low bit rate with an improved quality. We apply the proposed bit allocation scheme for both offline video coding and rate-reduction transcoding. Experimental results demonstrate the proposed scheme is able to achieve not only reduced average distortion but also much smoother visual quality. [Copyright &y& Elsevier]