Kazutoshi Nishimura

Continuous-media communication method for minimizing playback interruptions

Video-on-demand service will be carried on the future broadband networks. For this service to gain acceptance, methods of maintaining acceptable quality of service must be implemented. However, this is made difficult by the variation in the delay time caused by other communication traffic on the same network. We propose a continuous-media communication method for minimizing the playback interruptions. In our method, the server predicts the delay time and sends data earlier by an amount equal to the delay time, so that the data is received before it should be played back and the number of playback interruptions is reduced. In this way, our method can guarantee the quality of service using end-to-end communication without reserving the required bandwidth in advance. We evaluated our method by simulation under two conditions In the first simulation, we measured the playback interruption rate when there was background traffic. In the second simulation, we measured the rate when we used the delay times measured in actual networks as the simulated delay time. The results showed that using our method reduced the playback interruption rate by 44 - 82% in the first simulation and by 40 - 100% in the second simulation, compared with using no control.

Video-on-Demand System Using Optical Mass Storage System

Digital video-on-demand systems based on high-speed digital networks and video compression techniques have been proposed recently. One serious problem for these systems is achieving mass storage with quick access. In this paper we propose a video storage hierarchy of magnetic disks and an optical mass storage system (MSS). To solve the problems that occur when optical mass storage systems are used in video-on-demand services, we have proposed techniques such as quick-start and linking of playback, multiple readout, and visual search in a storage hierarchy with magnetic disks. A trial system has verified the feasibility of these new techniques. The average response time for quick-start and linking is less than one second. Another technical problem is the implementation of visual search functions such as fast forward and fast reverse together with multiple access. Average switchover time between visual search functions is also less than one second. In this trial system, which uses the JPEG encoding algorithm, we proposed and verified a new method for synchronizing fixed bit-rate audio and variable bit-rate video.

Storage hierarchy for video-on-demand systems

In this paper, we evaluate the performance of our trial hierarchical storage server by using simulation. Our system is composed of magnetic disk drives and an optical mass storage system (MSS). This is based on three key techniques: (1) multiple readout control using the time slot synchronous method, (2) quick access control of the MSS by using magnetic disk drives, (3) a storage hierarchy method for video programs that has three configurations. To verify the effectiveness of our storage hierarchy, we construct a queuing network simulation model of the system based on experimentally measured values, and analyze the average waiting time for each of the three configurations of video programs. Selecting the appropriate video program configuration shortens the average waiting time to half of previous times, with a quick access of less than 1 second, and with 8 simultaneous readouts. Storage costs are reduced to about half that of magnetic disk drives only.

Multimedia integrated switching architecture for visual information retrieval systems

Advanced visual information retrieval systems supporting both video and images need to have flexible system design so that their system configurations can easily be enhanced. It is therefore desirable to separate the features of a central system into three parts: storage servers, communication servers, and a back-end network that combines these. In this architecture, unscheduled arrivals of data blocks at a back-end network cause two problems: unacceptable fluctuation of video frames and overly long delays of image transfer. To solve these problems, we have designed a new multimedia integrated switching system (MISS) that uses a fully connected crossbar switch to combine servers. MISS treats a time interval of a few hundred microseconds (called a `time-slot) as the basic unit of data block transfer, and allocates appropriate time-slots to all transfer requests in order to simultaneously meet the requirements for each kind of visual information transfer. According to simulation results and estimates based on queuing theory, MISS greatly reduces video frame fluctuation and halves the average image transfer delay. These effects have been confirmed in an experimental visual communication system built around MISS. This system supports JPEG compressed video and images, and six terminals can simultaneously retrieve visual information through an FDDI network.

A large-scale multiple-access method for video-on-demand systems

A multiple-access method is proposed for simultaneously providing video data to more than 100 terminals in a video-on-demand system. The proposed method divides up each video program into segments, stores the segments on multiple disk drives, and performs time-division-multiplex reading based on the time-slot sequences for each disk drive that are periodically phase shifted with respect to each other. This method features : 1) theoretical limit on the maximum number of multiple accesses that can be implemented, and straightforward upscaling of the maximum number of multiple accesses because the amount of buffer memory needed for each terminal is fixed ; 2) no request loss, even when many terminals request the same program because the load for reading each video program is always distributed across all the disk drives ; and 3) system control that guarantees read-request response times within a specified value by allowing the skipping of a very small amount of leading information in each video program. Numerical examples show that this access method makes it possible to design a system having no request losses and a guaranteed response time within one second, even for simultaneous accesses by 900 terminals.

Novel dynamic caching for hierarchically distributed video-on-demand systems

It is difficult to simultaneously serve the millions of video streams that will be needed in the age of Mega-Media networks by using only one high-performance server. To distribute the service load, caching servers should be location near users. However, in previously proposed caching mechanisms, the grade of service depends on whether the data is already cached at a caching server. To make the caching servers transparent to the users, the ability to randomly access the large volume of data stored in the central server should be supported, and the operational functions of the provided service should not be narrowly restricted. We propose a mechanism for constructing a video-stream-caching server that is transparent to the users and that will always support all special playback functions for all available programs to all the contents with a latency of only 1 or 2 seconds. This mechanism uses Variable-sized-quantum-segment- caching technique derived from an analysis of the historical usage log data generated by a line-on-demand-type service experiment and based on the basic techniques used by a time- slot-based multiple-stream video-on-demand server.

Video server technology for high-performance retrieval in video on demand

A powerful new set of video playback control functions is proposed which aids subscribers in finding specific programs or contents from among a vast store of video materials. For conducting title-based searches, repeat and clip capabilities are proposed as ways of previewing or browsing a programs contents. For retrieving information from within a program, skip and fast forward/rewind functions are effective when searching through video materials that are familiar, while midway playback is an effective approach when searching through material that had never been seen before. Most significantly, this set of video playback control permits visual searches without reducing the number of concurrent users that can be supported while preserving a video access response time of under 1 second. The proposed methods are implemented in an experimental system and evaluated.

System architecture for middle-scale video-on-demand services

An experimental video server for middle-scale video-on-demand services that uses a redundant double-layered disk array can read out 100 MPEG-1 1.5-Mbps video streams simultaneously with a response time of under one second through an FDDI-LAN. An exclusive data method that switches between normal data and fast data and a skip-search method are used to provide fast visual search. The gateway connecting the video server LAN to a 6.312-Mbps constant bit-rate line allows broadcast services to be integrated with on- demand services. The protocol implemented in this gateway controls the visual search rate, corrects errors in downloaded data, and accelerates the playback mode changes.

Communication processing techniques for multimedia systems

Key techniques required for communication processing in multimedia systems are (1) high-speed protocol processing of every data packet and (2) simultaneously processing of many sessions. This paper proposes three solutions of high-performance protocol processing: protocol data unit management using page management method, assembling and disassembling of header and contents in hardware, and rescheduling of protocol processing. The feasibility of these techniques have been experimentally verified applying these to the transmission control protocol/internet protocol (TCP/IP) over an fiber-distributed data interface (FDDI) network. As a result, the maximum throughput of 78 Mbps is obtained. The total communication throughput of 64 sessions is 75.6 Mbps, only 3% less than the maximum throughput.<<ETX>>

Video-on-demand System Architecture Using An Optical Mass Storage System

Video-on-demand (VOD) systems, using the communication network t o send requested video programs from video storage centers t o terminals, are applicable t o services such as movies, commodities-guides, and video museums. Against the background of digitization of c o m u nications and progress in video compression techniques, digital VOD has recently been proposed [1,2]. The volume of digital video data, however, is enormous: about 1.1 G B per 100 minutes even at the 1.5 Mbps of the IS0 MPEG standard. The storage of many video programs therefore requires an optical mass storage system (MSS) [3], the use of which is associated with the following problems: (1)Fksponse is slow because of the time required for mounting a disk. (2)When a video program divided into several surfaces or disks, the playback is interrupted while turning over or changing disks, and it is difficult to implement visual search. (3)Waiting time is long when all drives are in use. This paper proposes solutions for these problems and verifies their effectiveness in trial VOD system.