Backhyun Kim

VOD service using web-caching technique on the head-end- network

To provide fluent on-demand services on Internet, the VOD servers have to possess a powerful processing ability and the network needs a vast bandwidth. The more clients request service, the more these resources are required. This paper presents a web-caching technique for providing fluent VOD services on Internet. This technique adopts some caches in the Head-End-Network that consists of a Switching-Agent (SA), some Head-End-Nodes and many clients. This technique distributively stores video streams in the order of request on some HENs which request an identical video item. Thus, storing streams on each HEN are different. When the client requests service for cached video items, connecting HEN receives streams from HENs which stored them and transmits them to client. These procedures are performed under the control of SA. Therefore this technique distributes the load of HENs, leads to equiloaded HEN and protects data duplication among HENs.

Overlay multicast routing architecture in mobile wireless network

In this paper we propose overlay multicast (Mcast) routing architecture that uses basic trees and their neighboring mobile nodes (NMN) in mobile wireless network. The basic trees are generated according to hop-counts from mobile nodes (MN) toward base node. NMNs are 1-hop away from a MN and exist in other basic trees. NMNs identify only whether they have Mcast group MNs in mapping table, and never broadcast MNs’ Mcast request. But MNs on an identical basic tree must broadcast Mcast request to parent or child MNs when they don’t have Mcast group in mapping table. MNs don’t have Mcast routing table but have only mapping table to convert logical address into physical IP address. The proposed Mcast routing architecture has effectiveness to solve shortcomings in mobile wireless network and sensor networks since it can decrease effectively flooding traffic restricted to MNs on a basic tree.

Web-Cached Multicast for On-Demand Video Distribution

In this paper, we propose multicast technique in order to reduce the required network bandwidth by n times, by merging the adjacent multicasts depending on the number of HENs (Head-End-Nodes) n that request the same video. Allowing new clients to immediately join an existing multicast through patching improves the efficiency of the multicast and offers services without any initial latency. A client might have to download data through two channels simultaneously, one for multicast and the other for patching. Each video stream is divided into blocks which are the same size of multicast grouping interval I m . Blocks then are evenly distributed into different HENs according to their popularity and the order of requests. Only when the playback time exceeds the amount of cached video data, server generates new multicast channel. Since the interval of multicast can be dynamically expanded according to the popularity of videos, it can be reduced the servers workload and the network bandwidth. We adopt the cache replacement strategy as LFU (Least-Frequently-Used) for popular videos, LRU (Least-Recently-Used) for unpopular videos, and the method for replacing the first block of video last to reduce end-to-end latency. We perform simulations to compare its performance with that of conventional multicast. From simulation results, we confirm that the proposed multicast technique offers substantially better performance.

Content delivery with spatial caching scheme in mobile wireless networks

Future wireless/mobile system will be served streaming service to support various applications. In this environment, users can access all the required information whenever and wherever they may be. In this paper, we proposed a new scheme that minimizes the network bandwidth consumption and the service blocking rate in mobile network. Our proposed scheme consists of location estimation and caching strategy. To estimate nodes current location, we use hexagonal cellular based plane and cascading dynamic address scheme which make mobile equipments, mobile router (MR) and mobile node (MN), calculate the distance from sender and estimate the existence possibility of the alternative route toward sender. The service blocking can be minimized by using caching strategy. In evaluation, we examined our proposed scheme in the view of the total network bandwidth consumption and the connecting probability as well as the impact of each cache capacity. From simulation results, we confirm that the proposed scheme offers substantially better performance.

Multicast for multimedia delivery in wireless network

In this paper we propose a novel technique for multimedia service using multicast delivery in wireless network. It supports seamless multimedia service that uses basic tree and their neighboring mobile nodes (NMNs) in wireless network. The basic trees are generated based on hop-counts from each mobile node (MN) toward base node (BN). Each MN has some NMNs (1-hop away) which are composing the upper, lower or peer MNs within its transmission range and are located at different basic trees. MN joins to a specific multicast (Mcast) group through its basic tree. For protecting seamless service according to mobility of MNs that have already joined Mcast group, those MNs send periodically Mcast join message to their NMNs when they move to the other basic trees, and registers to those basic trees. In this case NMNs that received Mcast join message send it to their upper MN over the same basic tree when they don’t join Mcast group just like wired Mcast. But the MNs on the new basic tree do not send Mcast streams until the moved MN registers to a new basic tree for reducing the amount of traffic.

Multimedia Data Transmission Using Multicast Delivery in Digital Library

This paper presents the method for multimedia data transmission using multicast delivery in digital library. It is not conventional server-based multicast delivery but client-based on-demand one. The Multicast-Agent Scheduler for providing on-demand multimedia service using multicast in digital library generates immediately a multicast group address and port number when clients request an identical multimedia data, then it sends them to media server and clients. And then media server transmits requested multimedia streams to multicast group and the clients join the group automatically. This technique reduces media server load in digital library and maximizes the efficiency of network resources. And this paper implements multimedia transmission system in digital library with WMS (Window Media Services) and its API.

Region-Based multicast routing protocol with dynamic address allocation scheme in mobile ad-hoc networks

Mobile Ad hoc Network (MANET) is a network that mobile nodes can freely and dynamically communicate with each others without the support of any infrastructure. Due to the nodes movement, the network topology can be changed randomly and unpredictably. Previously, address as identifier is used for establishing a delivery route. But it is not easy that each mobile node has a unique address due to the lack of centralized address protocol like DHCP. In this paper, we propose region-based dynamic address allocation scheme that can simplify the route establishment and maintenance. Routing is done by a short distance vector algorithm based on the proactive and the flat topology routing protocol. Basic idea of our address allocation scheme is to separate node identity and node address. Node address indicates nodes current location and is allotted by its parent node with a cascading address scheme. In evaluation, we examined our proposed scheme in a point of the view of the number of control packets to establish delivery trees and the connecting probability. From simulation results, we confirm that the proposed scheme offers substantially better performance.

Web-Cached Multicast Technique for on-Demand Video Distribution

In this paper, we propose multicast technique in order to reduce the required network bandwidth by n times, by merging the adjacent multicasts depending on the number of HENs (Head-End-Nodes) n that request the same video. Allowing new clients to immediately join an existing multicast through patching improves the efficiency of the multicast and offers services without any initial latency. A client might have to download data through two channels simultaneously, one for multicast and the other for patching. The more the frequency of requesting the video is, the higher the probability of caching it among HENs increases. Therefore, the requests for the cached video data can be served by HENs. Multicast from server is generated when the playback time exceeds the amount of cached video data. Since the interval of multicast can be dynamically expanded according to the popularity of videos, it can be reduced the server`s workload and the network bandwidth. We perform simulations to compare its performance with that of conventional multicast. From simulation results, we confirm that the Proposed multicast technique offers substantially better performance.

Multicast VOD System for Interactive Services in the Head-End-Network

This paper proposes an interactive VOD system to serve truly interactive VCR services using multicast delivery, client buffer and web-caching technique which implements the distributed proxy in Head-End- Network(HNET). This technique adopts some caches in the HNET that consists of a Switching Agent(SA), some Head-End-Nodes(HEN) and many clients. In this model, HENs distributively store the requested video under the control of SA. Also, client buffer dynamically expands to support various VCR playback rate. Thus, interactive services are offered with transmitting video streams from network, HENs and stored streams on buffer. Therefore this technique makes the load of network occur In the limited area, minimizes the additional channel allocation from server and restricts the transmission of duplicated video contents

Web Proxy Caching Mechanism to Evenly Distribute Transmission Channel in VOD System

The proxy caching technique can significantly reduce server’s load but is still suffered from the shortage of the network bandwidth and from the traffic imbalance among proxies. In this paper, we propose the equivalent loaded proxy caching mechanism to minimize the required transmission bandwidth by using the distribution mechanism determined by the request order and by the popularity of videos calculated with the request frequency. In our technique, videos transmitted from the server are fragmented by Switching Agent and are distributively stored among Head-End-Nodes according to the request order, and SA stores some popular videos to reduce the retransmission of an identical video from the server. From simulation results, we acquire that the proposed web caching mechanism has equivalent loaded among HENs and it can reduce VOD server’s bandwidth significantly.