Shigehiro Ano

Rapid Channel Zapping for IPTV Broadcasting with Additional Multicast Stream

This paper proposes a novel approach for improving channel zapping delay on IPTV broadcasting services. Channel zapping delay is a crucial issue for digital TV broadcasting, which entails audio/video data buffering before reproduction. Such delay could be mitigated in an IP environment if a receiver accelerated this buffering using additional burst transmissions from dedicated servers when a channel zap occurs. However, most conventional solutions are based on the unicast burst, which may cause an impulsive server/network load because channel zaps tend to happen simultaneously when programs are finished or suspended by commercial messages. To reduce this load, we propose a new multicast-based solution that takes into consideration the timing variation in channel zap on each receiver. We have confirmed a maximum 1-second reduction in the zapping delay using commercial multicast streams.

IP-based HDTV broadcasting system architecture with non-stop service availability

This paper presents an architectural concept for IP-based TV broadcasting infrastructure on ISPs multicast network. As broadcasting grows into an indispensable service menu for ISP, its service availability and independency of content format are crucial issues for smooth business deployment. Hence we newly propose some practical solutions to realize a non-stop IP broadcasting service with simple and content format free server implementation. We also implemented a prototype as an HDTV broadcasting system. The verification results show that our system can compensate about 27,500 packet losses and archive non-stop HDTV broadcasting, even in the event of network layer trouble, without any changes in router implementation.

Stream Mining for Network Management

Network management is an important issue in maintaining the Internet as an important social infrastructure. Finding excessive consumption of network bandwidth caused by P2P mass flows is especially important. Finding Internet viruses is also an important security issue. Although stream mining techniques seem to be promising techniques to find P2P and Internet viruses, vast network flows prevent the simple application of such techniques. A mining technique which works well with extremely limited memory is required. Also it should have a real-time analysis capability. In this paper, we propose a cache based mining method to realize such a technique. By analyzing the characteristics of the proposed method with real Internet backbone flow data, we show the advantages of the proposed method, i.e. less memory consumption while realizing real-time analysis capability. We also show the fact that we can use the proposed method to find mass flow information from Internet backbone flow data.

Locating congested segments over the Internet by clustering the delay performance of multiple paths

Abstract Delay variation-based detection and location of congestion in a large network is considered. Since the Internet is still highly prone to performance deterioration due to transient large delays, locating a part of the network (segments) responsible is vital to ensure that Internet Service Providers can mitigate or prevent such performance deterioration. In the proposed method, the end-to-end packet delays from multiple origins to multiple destinations are actively and continuously measured. By analyzing those data on delay variation along each monitored path, congestion is detected by finding a delay performance deterioration worse than a predefined criteria and a congested segment responsible could be inferred by finding a set of paths among which delay variations are strongly correlated. This is a network tomographic approach based on a clustering technique that effectively tackles the correlation among packet delay variation along individual paths. The proposed method was evaluated through a real-world long-term experiment on the Japan’s commercial Internet, and was shown to have considerable potential to promptly locate congested segments through various analyses on the experimental results.

Inferring the origin of routing changes based on preferred path changes

Previous studies on inferring the origin of routing changes in the Internet are limited to failure events that generate a large number of routing changes. In this paper, we present a novel approach to origin inference of small failure events. Our scheme focuses on routing changes imposed on preferred paths of prefixes and not on transient paths triggered by path exploration. We first infer the preferred path of each prefix and measure the stability of each inter-AS link over this preferred path. The stability is measured based on routing changes of specific prefixes that regularly use the link and are advertised by the AS adjacent to the link. We then correlate the stability of other links over this path and infer the instability boundary as the origin. Our analysis using Oregon RouteViews data and trouble tickets from operational networks shows that our inference scheme can identify the origins of small failure events with very high accuracy.

TCP Throughput Estimation by Lightweight Variable Packet Size Probing in CDMA2000 1x EV-DO Network

TCP throughput is one of the major communication quality metrics. In order to grasp this metric by active measurement while imposing a lower load, various estimation techniques have been proposed. These techniques are categorized as either equation-based or history-based; here we focus on the latter. Conventional history-based estimation techniques target a wired environment and estimate temporally. By contrast, we target a wireless environment and estimate spatially. For TCP throughput estimation using lightweight probing, we propose a new modeling whereby TCP throughput is determined by capacity and delay fluctuation. We evaluate our proposed modeling by conducting experiments in a commercial CDMA2000 1x EV-DO network. This evaluation empirically verified the validity of our model to estimate TCP throughput in a wireless environment with lightweight probing.

OSPF-Based Fast Reroute for BGP Link Failures

The Border Gateway Protocol (BGP) is one of the key components of todays Internet infrastructure. Many ISPs rely on BGP for convergence of inter-domain routes upon BGP link failures. However, the convergence of BGP requires several seconds of downtime, which may cause critical performance degradation of applications. In this paper, we propose the use of an intra-domain routing protocol to provide IP fast reroute for BGP link failures. The proposed scheme extends OSPF to restore the nexthop IP address associated with the BGP link failure. Since the reachability of the nexthop is restored at the IGP layer, no convergence is required at the BGP layer. We also introduce a recursive lookup mechanism for forwarding of packets to achieve fast update of the forwarding table. Evaluation using a prototype and real BGP prefixes observed in the Internet showed that the proposed scheme is capable of reducing the current recovery time by 13.40 seconds. The proposed scheme performs recovery with limited routers and is effective where Route Reflectors are in place.

Empirical Study on Inferring BGP Routing Instability and its Location Based on Single Point Observation

As the Internet has become a communication infrastructure, inter-domain routing instabilities have become prone to serious problems. Some studies have proposed methods that infer instabilities and their locations by clustering BGP update messages collected at multiple observation points and are based on post-processing using public messages. However, to realize real-time processing, which is vital for network operations, single point observation is a feasible option with costs and todays competitive environments in mind. Therefore, we propose an inference method including a new clustering algorithm that is suitable for messages collected at a single observation point. We also describe the empirical results used to validate our method.

Locating Congested Segments on the Internet by Clustering the Delay Performance of Multiple Paths

A practical method of locating congested segments on the Internet by periodic end-to-end packet delay measurements along multiple paths is presented. This method is a network tomographic approach based on a clustering technique that effectively tackles the correlation among delay variations. Since the Internet is still highly prone to performance deterioration due to transient delays, locating the segments responsible is vital to ensure that Internet service providers can mitigate or prevent such performance deterioration. Our concern is how to detect and locate such segments promptly in a large network. The proposed method was implemented and evaluated through a real-world experiment by actively measuring packet delays from multiple origins to multiple destinations over the Japanese commercial Internet, and was shown to have considerable potential to precisely locate congested segments.

A Scalable Approach to Tomography-based Internet Measurement System

Network tomography enables network operators to detect and mitigate congestion by inferring internal states from multiple end-to-end active performance measurements. In previous studies, network tomography was applied to either small networks, e.g., a network which allows measurements in a full-mesh manner, or a reasonably large network, e.g., overlay-network with 100 end nodes. In contrast, our study was conducted to develop a large-scale tomography-based measurement system that scales with the nation-wide Internet. In this paper, we propose a novel architecture which consists of a probability measurement path selection method and a distributed database building method. The architecture provides network operators with look up services for measurement results so that they can achieve network-tomography applications.