Kenneth Ong

Optimized scalable cache management for video streaming system

An important research issue in video streaming is how to efficiently utilize the network resources to provide clients instant access to multiple video objects. Caching strategy and transmission scheme are the two essential points inside the video streaming framework. Recent research efforts on them are not sufficient due to their inflexible support for scalable encoded video streams and heterogeneous requests from clients. In this paper, we propose an optimized caching strategy (OCS) and a scalable transmission scheme (STS) for scalable coded video streaming. By exploring the characteristics of video streaming workload and system design objectives, OCS and STS work efficiently to minimize both network bandwidth cost and user access latency. Firstly, we analyze the caching problem for the proxy-assisted video streaming system and derive a maneuverable caching scenario. Secondly, we develop an efficient transmission scheme for scalable coded videos. Thirdly, we formulate a multi-objective optimization model with closed-form expressions to obtain the optimized caching strategy. Finally, with designed algorithms, an excellent compromise between two competing objectives (minimizing the bandwidth cost and the access latency) is achieved. We start our evaluation by studying the optimized caching strategy for a single video object. Then we apply the strategy to multiple video objects and illustrate the tradeoff between the optimization objectives. Our evaluation results show that compared with other caching strategies, the proposed optimized scalable caching strategy can achieve a significant reduction in bandwidth cost with even a small proxy cache size. Meanwhile, the best performance (in terms of bandwidth cost) is obtained together with the proposed scalable batch-patching transmission scheme.

Optimized cache management for scalable video streaming

Media proxy caching is an efficient method to provide instant access to multiple video objects in video streaming system. Research efforts in this area are not sufficient due to the possible huge number of heterogeneous requests to scalable encoded video objects. In this paper, we propose an optimized caching scheme for scalable video objects so that both transmission cost and access latency are minimized. A multi-objective optimization model is developed and the corresponding algorithms are designed. Our evaluation results show that compared with other caching methods, proposed scheme can achieve a significant reduction in transmission cost with even a small proxy cache size.

Discrete LQ rate control schedule system for multimedia transmission

In this work, we investigated the discrete linear quadratic (DLQ) rate control method for the transmission of MPEG2 streams in IP networks. We derived a transmission scheduling system by solving the DLQ control problem under disturbances. Our study shows that DLQ is superior to conventional proxy scheduling schemes especially in client buffer usage. For some videos that can only achieve 70%/spl sim/80% of buffer usage using basic DLQ, we also propose improvements to attain higher buffer occupancy with no overflow. Transmission capacity and adjust method under heavy traffic are also discussed.

An object-oriented approach to the semantic interpretation of medical text

Abstract A primary problem in the area of text understanding is semantic analysis. This involves both formalizing the general and domain-dependent semantic information relevant to the task involved, and providing an appropriate mapping between the syntactic constituents of a parsed sentence and the semantic representation of the sentence based on the semantic information. This study introduces an object-oriented approach to semantic interpetation that integrates the semantic processing into syntactic structures. The semantic interpreter has two components. The first is a syntactic decomposer. It takes the syntactic description for a sentence as its input, mapping the syntactic structure into an object tree. The other is a semantic merger. It starts from lexical items and merges the semantic frames attached to the objects of syntactic components recursively until a semantic frame net for the sentence is produced. A frame language is designed to formalize the semantic information into semantic frames. This semantic interpreter has been integrated with an automatic knowledge acquisition system that builds the knowledge base for expert systems from medical text. The source of the text was medical diagnosis for heart diseases by ECG. The system is implemented in an object-oriented programming (C + + ). Some results of the application are presented and discussed.

Expert Clinical Interface

We present a process-driven approach that is embedded in the design of a clinical system for patient management in cardiology. It integrates elements of expert systems, multimedia, object-oriented manipulation, instrumentation interfaces and medical telematics into a human-computer interface presentation for making clinical decisions. The Expert Clinical Interface (ECI) is a front-end to our development of a new generation, intelligent clinical system that interacts with the physician in a manner that is in congruence with the diagnosis-therapy process, and that aids in coronary care management. The paper describes the functional requirements for a cardiology workstation, the ECI interface, the data model and architecture, the decision-support system, and the frame-based language used to encode the diagnosis-therapy process.

Discrete LQ Rate Control for MPEG2 Video Streaming System

In this paper, we propose a novel rate control system by applying the discrete linear quadratic rate control method (DLQ) to transmissions of MPEG2 streams in IP networks. We give out the detail deduction for disturbed-DLQ-regulator problem and the corresponding methods to implement the results into media streaming rate control. Study shows that DLQ is superior to conventional rate control schemes especially in client buffer utilization. Besides the basic DLQ method, we also investigate DLQ under heavy-traffic scenarios and provide improvements for this special case. The influences of delay, with respect to different mean values and variances, are discussed carefully.

On Compensation Technique in Multimedia Streaming System

We describe a compensation method for real-time multimedia streaming to improve the user-perceived quality of video. Based on TCP friendly rate control (TFRC), this method takes advantage of the resources available at the proxy server and considers the content-aware principle as well. Using feedback from the client, the proxy server dynamically drops or compensates the enhancement layers of the MPEG-4 stream. We optimize performance by carefully choosing the compensation point to maximize the compensation effect, while at the same time meeting the constraints of the client buffer and realtime requirements. Compared to the traditional prefetch method, the proposed method can further improve the video quality, which is demonstrated through extensive simulations

Optimized Media Filtering Scheme for Layered Video Streaming

This article describes an optimized filtering scheme (OFS) for layered video streaming system to improve the user-perceived quality. Based on content characteristics, this scheme filters the MPEG-4 video objects according to the derived optimized transmission path along the object structure so that the average decoding quality is maximized and quality variance is minimized. With proposed algorithms, OFS can also achieve close target bit rate match with low computational complexity. Compared with three traditional media filters, we demonstrate the effectiveness of OFS by extensive simulations.

A Rule-Based Expert Server System for Multimedia Transmission

In this paper, a novel multimedia transmission server system is designed using rule-based expert system technology. Working parameters and session management methods are separated from the decisionmaking procedure and stored in a knowledge base (XML format). The server fires different transmission strategies inside the knowledge base by runtime inference. Thus it can easily adjust transmission parameters under various environments and incorporate newly developed methods without significantly changing the main body of the server code. Results indicate that the expert system can efficiently deliver smooth streams and save server resources. The overhead brought along with the flexibility is tested to be acceptable.