Katsuyoshi Iida

Path selection using active measurement in multi-homed wireless networks

The mobile Internet is built upon a number of different wireless access networks with widely varying features in terms of coverage area, bandwidth, packet loss, and delay. To move across these different networks smoothly, issues associated with the changing features need to be addressed. In this paper, a path selection method for the coverage overlap area is proposed in which the mobile host actively measures the round trip time (RTT) and bottleneck bandwidth for each path and a path is selected based on four rules.

Performance evaluations of DCCP for bursty traffic in real-time applications

The Datagram Congestion Control Protocol (DCCP) has been proposed as a transport protocol which supports real-time traffic using window-based flow control. We investigate the DCCP performance for various traffic flows, focusing on how DCCP flows affect TCP flows and vice versa. Through those simulations, we examine an unfair bandwidth distribution problem caused by the incompatibility of DCCP with the fast recovery algorithm of TCP.

Sender-initiated multicast forwarding scheme

We propose a sender-initiated multicast (SIM), specifically designed for small group communications such as teleconferencing and file distribution. Contrary to traditional IP multicast, SIM reduces the cost of allocating a global multicast address by attaching receiver addresses to the packet header. SIM routers route packets according to these addresses; therefore, the cost of the control traffic between routers can be lessened by applying the existing unicast routing table. The key feature of SIM is its preset mode, which uses SIM forwarding information base (FIB) entries on routers to achieve cost-efficient packet forwarding. Another feature is an automatically created SIM tunnel, which provides the ability to maintain SIM FIB only on routers that act as multicast branching points. We describe the SIM mechanism in detail, and present results evaluated through simulations. We show how SIM can achieve low cost in maintaining state information, cost-efficient packet forwarding, and incremental deployment.

Performance evaluation of inter-domain IP traceback

IP traceback is technology used to find the true source address of a Distributed Denial of Service (DDoS) attack with source address spoofing. We focus on IP option traceback (IP-OPT) for inter-domain IP traceback. In the Passive Detection Packet (PDP) method, which is a basic mechanism of IP-OPT, there is a trade off between the amount of trace traffic and the detection time for the path of attack time. However, no analysis of this condition has been made at this time. Thus, we mathematically analyze the tradeoff of PDP, and show that 1.1/spl times/10/sup -4/ is the optimal value of the pacekt generation probability for IP-OPT through numerical experiments.

Performance evaluation of the architecture for end-to-end quality-of-service provisioning

Real-time communications services over the Internet need a new architecture to meet their required quality. From a viewpoint of quality of service provisioning architecture, the Internet can mainly be divided into three types of subnetworks: domain networks, access networks, and stub networks. In this article we focus on issues arising in the former two networks for end-to-end QoS provisioning. First, the access networks are of rather low-speed links, so delay is still of major concern. We examine the statistical delay bound through numerical results derived from our analysis. Schemes to reduce delay are proposed, and their performance is evaluated. Next, domain networks are likely to be of very high-speed links, which can accommodate a huge number of voice flows of low bit rates. Thus, effective flow management will be of major concern because per-flow management is a very costly proposition. Therefore, we pay attention to a flow aggregation scheme, and evaluate its performance by analyzing its blocking probability.

Delay analysis for CBR traffic in static-priority scheduling: single-node and homogeneous CBR traffic case

The quality of the real-time traffic severely depends on its loss rate and delay time. In the integrated service networks, the real-time traffic and the non-real-time traffic share the network resources so that one can affect the quality of another and vice versa. In this context, it is very crucial to develop some mechanism to guarantees the quality-of-service required by the real-time traffic. In this paper, we analyze the delay time of CBR packets from real-time sources when CBR packets have the priority over UBR packets in a manner that UBR packets are serviced only if no CBR packets are waiting in the buffer. Furthermore, we obtain various numerical results on the statistical bound on delay time such as the 99.9-percentile delay and compare it with the deterministic bound. By the comparison, we show that CAC (Call Admission Control) based upon the statistical bound is very effective in using the network resources efficiently when CBR packets can tolerate some loss due to late arrival.

Multi-path Transmission Algorithm for End-to-End Seamless Handover across Heterogeneous Wireless Access Networks

In mobile networks, new technologies are needed to enable mobile hosts to move across various kinds of wireless access networks. In the past, many researchers have studied handover in IP networks. In almost all cases, special network devices are needed to maintain the host’s mobility. However, in these technologies, mobile hosts cannot move across different wireless access networks without closing the connection and degrading the goodput. To overcome these, we propose here a multi-path transmission algorithm for end-to-end seamless handover. The main purpose of this algorithm is to improve the goodput during handover by sending the same packets along multiple paths, reducing unnecessary consumption of network resources. We evaluate our algorithm through simulations and show that mobile hosts gain a better goodput.

Delay analysis for CBR traffic in static-priority scheduling: single-node and heterogeneous CBR traffic case

In the integrated service networks, the real-time traffic and the non-real-time traffic share the network resources so that one can affect the quality of another and vice versa. In this context, it is very crucial to develop some mechanism to guarantee the quality-of-service (QoS) required by the real-time traffic. In this paper, we analyze the delay time of CBR packets from real-time sources when CBR packets have the priority over UBR packets in a manner that UBR packets are serviced only if no CBR packets are waiting in the buffer. There are two types of CBR packets; i.e., they are different in their transmission rate and/or packet length. In this sense, the case treated is called the heterogeneous case. We obtain various numerical results on the statistical bound on delay time such as the 99.9-percentile delay and compare it with the deterministic bound. By the comparison, we show that call admission control (CAC) based upon the statistical bound is very effective in using the network resources efficiently when CBR packets can tolerate some loss due to late arrival. We also examine whether the delay time distribution in the heterogeneous case can be approximated by that of some homogeneous cases.

Receiver-based flow control mechanism with interlayer collaboration for real-time communication quality in W-CDMA networks

Mobile networks are becoming increasingly prevalent, and this has led to an increase in the bandwidth available over wireless links in IMT-2000. Both non-real-time forms of communication, such as e-mail and web browsing, and real-time forms of communication, such as audio and video applications, are well suited to wireless networks. However, wireless networks are subject to relatively long transmission delays because of the need to recover lost packets caused by high bit error rates. This degrades the quality of real-time communications. Therefore, in the present paper, we propose a receiver-based flow control mechanism employing an interlayer collaboration concept to improve the quality of real-time communications without adversely affecting the performance of non-real-time communications on IMT-2000 networks. In addition, simulations are performed in order to evaluate the performance of the proposed mechanism and demonstrate its effectiveness.

Performance analysis of flow aggregation of constant bit rate type traffic at ingress router

Flow aggregation will be a key technology for offering real-time communication services in large scale networks, because it can greatly reduce the flow management cost. In this paper, we exactly analyze the performance of flow aggregation of constant bit rate type traffic, especially the call blocking probability at the ingress routers. Through our numerical results, we have shown the characteristics of flow aggregation. As a result, although the flow aggregation scheme, in fact, needs more capacity than the per-flow management to achieve the same blocking probability, it enables us to manage a great number of flows of relatively small bandwidth requirements, e.g., voice flows in high speed backbone networks without much redundant capacity. Our analysis is so general that it can treat various kinds of heterogeneous cases, and helps to determine the capacity allocation to different kinds of aggregated flows.