We present the design and analysis of a family of codes based on the efflciently-encodable rate-compatible irregular LDPC codes (E2RC codes) introduced by Kim, Ramamoorthy and Mclaughlin '06. The basic idea is to utilize the structured parity part (of the parity check matrix) of E2RC codes and design optimal degree distributions for the systematic part based on EXIT chart analysis. In this work, we propose a new technique for computing EXIT functions of the structured parity part without resorting to Monte Carlo simulations. Our method provides smoother EXIT functions in substantially lesser time. Furthermore, we pose and solve the problem of designing good codes using linear programming. Next, we consider the performance of rate-compatible codes under puncturing using EXIT charts. Finally, we propose joint optimization of our codes at any specified code rate(s) which has not been explored in previous design of rate-compatible punctured codes.

We consider the design and analysis of the efficiently-encodable rate-compatible (E2RC) irregular LDPC codes proposed in previous work. In this work we introduce semi-structured E2RC-like codes and protograph E2RC codes. EXIT chart based methods are developed for the design of semi-structured E2RC-like codes that allow us to determine near-optimal degree distributions for the systematic part of the code while taking into account the structure of the deterministic parity part, thus resolving one of the open issues in the original construction. We develop a fast EXIT function computation method that does not rely on Monte-Carlo simulations and can be used in other scenarios as well. Our approach allows us to jointly optimize code performance across the range of rates under puncturing. We then consider protograph E2RC codes (that have a protograph representation) and propose rules for designing a family of rate-compatible punctured protographs with low thresholds. For both the semi-structured and protograph E2RC families we obtain codes whose gap to capacity is at most 0.3 dB across the range of rates when the maximum variable node degree is twenty.

We propose a construction of rate-compatible punctured codes based on protographs that have a special E2RC structure in their parity part (E2RC codes were introduced in Kim, Ramamoorthy and McLaughlin '06) . The protograph representation of these codes facilitates their asymptotic performance analysis and allows the implementation of high speed decoders. The construction process starts with a good high rate protograph. The protographs of lower rate codes are derived from the higher rate protographs via the process of check-splitting. The check-splitting is done in a specific manner so that the parity nodes in the protograph have the special E2RC structure. We also present additional design rules that ensure that the gap to capacity remains low across the range of rates. Using our approach we exhibit protographs that have a gap of at most 0.27 dB to capacity across the range of rates 1/2 to 8/9. These conclusions are supported by our simulation results. Our work, therefore presents a systematic method for the design of E2RC-like codes.

The rapid development of wireless networks and mobile devices has made mobile video communication a particularly promising service. We previously proposed an effective video form, scalable portrait video. In low-bandwidth conditions, portrait video possesses clearer shape, smoother motion, and much cheaper computational cost than discrete cosine transform (DCT)-based schemes. However, the bit rate of portrait video cannot be accurately modeled by a rate-distortion function as in DCT-based schemes. How to effectively control the bit rate is a hard challenge for portrait video. In this paper, we propose a novel model-based rate-control method. Although the coding parameters cannot be directly calculated from the target bit rate, we build a model between the bit-rate reduction and the percentage of less probable symbols (LPS) based on the principle of entropy coding, which is referred to as the LPS-rate model. We use this model to obtain the desired coding parameters. Experimental results show that the proposed method not only effectively controls the bit rate, but also significantly reduces the number of skipped frames. The principle of this method can also be applied to general bit plane coding in other image processing and video compression technologies.

In videoconferencing, the image quality is significantly affected by the illumination condition. Unsatisfactory illumination conditions may lead to underexposure or overexposure of the area of interest, in particular a human face. To resolve this issue, we propose a solution to improve image quality automatically by correcting exposure and enhancing contrast. Our work is characterized by a method for automatically building a skin-color model and a novel contrast enhancement approach. Some techniques that can reduce the computational cost are also introduced. Experimental results show that obvious improvement in image quality is achieved while the computation overhead is very small. The proposed solution can be integrated into videoconferencing systems and is especially suitable for scenarios where low-complexity computing is required.