We previously proposed a new version of TCP, called Inline measurement TCP (ImTCP), in [2],[3]. The ImTCP sender adjusts the transmission intervals of data packets and then utilizes the arrival intervals of ACK packets for available bandwidth estimation. This type of active measurement is preferred because the obtained results are as accurate as those of other conventional types of active measurement, even though no extra probe traffic is injected onto the network. In the present research, we develop a new capacity measurement function and combine it with ImTCP in order to enable simultaneous measurement of both capacity and available bandwidth in ImTCP. The capacity measurement algorithm is a new packet-pair-based measurement technique that utilizes the estimated available bandwidth values for capacity calculation. This new algorithm promises faster measurement than current packet-pair-based measurement algorithms for various situations and works well for high-load networks, in which current algorithms do not work properly. Moreover, the new algorithm provides a confidence interval for the measurement result.

In ad hoc wireless networks, wireless terminals can autonomously construct and can maintain the network. They communicate with some neighbor terminals, exchange network information and determine routes for packets on the multi-hop wireless network. Flexible Radio Network (FRN), one of the ad hoc wireless network systems, adopts a proprietary protocol that provides a multiple routes management and a packet retransmission mechanism against packet transmission errors. This system is a commercial product that has been in use in a recent few years. In this paper, we first evaluate the performance through simulations for data-link protocol and routing protocol of the FRN to clarify its basic properties. Furthermore, we propose some techniques that enhance its performance and solve problems on the protocols. We show how they improve the system performance through simulations and analyses.

Recently problems with commercial IP telephony systems have been reported one after another, in Japan. One of the important causes is congestion in the control plane. It has been recognized that with the current Internet it is important to control not only congestion caused by overload of the data plane but also congestion caused by overload of the control plane. In particular, "retry traffic," such as repeated attempts to set up a connection, tends to cause congestion. In general, users make repeated attempt to set up connections not only when the data plane is congested but also when the control plane in the network is overloaded. The latter is caused by user behavior: an increase in the waiting time for the processing of connection establishment to be completed tends to increase his or her initiation of reattempts. Thus, it is important to manage both data plane and control-plane resources effectively. In this paper, we focus on RSVP-based communication services including IP telephony, and introduce a model that takes account of both data-plane and control-plane systems, and we examine the behavior of retry traffic. In addition, we compare the system stability achieved by two different resource management methods, the hard-state method and the soft-state method.

The term “flash crowd” describes a situation in which a large number of users access a Web service simultaneously. Flash crowds, in particular, constitute a critical problem in e-commerce applications because of the potential for enormous economic damage as well as difficulty in management. Flash crowds can become more serious depending on users' behavior. When a flash crowd occurs, the delay in server response may cause users to retransmit their requests, thereby adding to the server load. In the present paper, we propose to use the psychological factors of the users for flash crowd mitigation. We aim to analyze changes in the user behavior by presenting feedback information. To evaluate the proposed method, we performed subject experiments and stress tests. Subject experiments showed that, by providing feedback information, the average number of request retransmissions decreased from 1.33 to 0.09, and the subjects that abandoned the service decreased from 81% to 0%. This confirmed that feedback information is effective in influencing user behavior in terms of abandonment and retransmission of requests. Stress tests showed that the average number of retransmissions decreased by 41%, and the proportion of abandonments decreased by 30%. These results revealed that the presentation of feedback information could mitigate the damage caused by flash crowds in real websites, although the effect is limited. The proposed method can be used in conjunction with conventional methods to handle flash crowds.

As IoT services become more popular, mobile networks will have to accommodate a wide variety of devices that have different requirements such as different bandwidth limitations and latencies. This paper describes edge distributed mobile network architectures for the IoT era based on dedicated network technology and multi-access edge computing technology, which have been discussed in 3GPP and ETSI. Furthermore, it describes two context-aware control methods that will make mobile networks on the network architecture more efficient, reliable, and real-time: autonomous and distributed mobility management and bandwidth-guaranteed transmission rate control in a networked control system.

In this paper, we survey the fairness issues in the congestion control mechanisms of TCP, which is one of most important service aspects in the current and future Internet. We first summarize the problems from a perspective of the fair service among connections. Several solution methods are next surveyed. Those are modifications of TCP congestion control mechanism and router support for achieving the fair service among TCP connections. We finally investigate the fair share of resources at endhosts.

Per-flow unfairness of TCP throughput in the IEEE 802.11 wireless LAN (WLAN) environment has been reported in past literature. A number of researchers have proposed various methods for alleviating the unfairness; most require modification of MAC protocols or queue management mechanisms in access points. However, the MAC protocols of access points are generally implemented at hardware level, so changing these protocols is costly. As the first contribution of this paper, we propose a transport-layer solution for alleviating unfairness among TCP flows, requiring a small modification to TCP congestion control mechanisms only on WLAN stations. In the past literature on fairness issues in the Internet flows, the performance of the proposed solutions for alleviating the unfairness has been evaluated separately from the network bandwidth utilization, meaning that they did not consider the trade-off relationships between fairness and bandwidth utilization. Therefore, as the second contribution of this paper, we introduce a novel performance metric for evaluating trade-off relationships between per-flow fairness and bandwidth utilization at the network bottleneck. We confirm the fundamental characteristics of the proposed method through simulation experiments and evaluate the performance of the proposed method through experiments in real WLAN environments. We show that the proposed method can achieve better a trade-off between fairness and bandwidth utilization, regardless of vendor implementations of wireless access points and wireless interface cards.

Recent research on overlay networks has revealed that user-perceived network performance could be improved by an overlay routing mechanism. The effectiveness of overlay routing is mainly a result of the policy mismatch between the overlay routing and the underlay IP routing operated by ISPs. However, this policy mismatch causes a "free-riding" traffic problem, which may become harmful to the cost structure of Internet Service Providers. In the present paper, we define the free-riding problem in the overlay routing and evaluate the degree of free-riding traffic to reveal the effect of the problem on ISPs. We introduce a numerical metric to evaluate the degree of the free-riding problem and confirm that most multihop overlay paths that have better performance than the direct path brings the free-riding problem. We also discuss the guidelines for selecting paths that are more effective than the direct path and that mitigate the free-riding problem.

Many ATM switching modules with high performance have been proposed and analyzed. A development of a large scale ATM switching system (e.g., used as a central switch) is the key to realization of the broadband ISDN. However, the dimension of ATM switching ICs is limited by the technological and physical constraints on VLSI. A multistage switching configuration is one of the promising configurations for a large scale ATM switch. In this paper, we treat a 3-stage switching configuration with no internal bufferes; i.e., bufferless switches are employed at the first and second stages, and output buffered switches at the third stage. A short-term cell loss probability is analyzed in order to examine the influence of bursty traffic on performance of the bufferless switch used at the first two stages. Furthermore, we propose a 4-stage switching configuration with traffic distributors added at the first stage. This switch provides more paths between a pair of input and output ports than the 3-stage switching configuration mentioned above. A few schemes to distribute cells are compared. It is shown that the distributor successfully reduces the deterioration of cell loss probability due to bursty traffic by splitting incoming cells into several switching modules.

It is now becoming important for mobile cellular networks to accommodate all kinds of Internet of Things (IoT) communications. However, the contention-based random access and radio resource allocation used in traditional cellular networks, which are optimized mainly for human communications, cannot efficiently handle large-scale IoT communications. For this reason, standardization activities have emerged to serve IoT devices such as Cellular-IoT (C-IoT). However, few studies have been directed at evaluating the performance of C-IoT communications with periodic data transmissions, despite this being a common characteristic of many IoT communications. In this paper, we give the performance analysis results of mobile cellular networks supporting periodic C-IoT communications, focusing on the performance differences between LTE and Narrowband-IoT (NB-IoT) networks. To achieve this, we first construct an analysis model for end-to-end performance of both the control plane and data plane, including random access procedures, radio resource allocation, establishing bearers in the Evolved Packet Core network, and user-data transmissions. In addition, we include the impact of the immediate release of the radio resources proposed in 3GPP. Numerical evaluations show that NB-IoT can support more IoT devices than LTE, up to 8.7 times more, but imposes a significant delay in data transmissions. We also confirm that the immediate release of radio resources increases the network capacity by up to 17.7 times.

In CDMA mobile cellular systems, wireless quality is improved by soft handoff techniques. However, it requires to hold multiple channels of cells, which is likely to increase call blocking at wired channels. It is therefore necessary to consider the entire system including the wired and wireless portions of systems for investigating an effectiveness of the soft handoff. In this paper, we also clarify the effect of interference power from mobile stations that are not in the soft handoff because of lack of wired channels. In the analysis, we model three-way soft handoff which has not been considered in past researches. We also show the effect of a call admission control to wireless quality.

Rate based control is a promising way to achieve an efficient packet transmission especially in high speed packet switching networks where round trip delay is much larger than packet transmission time. Although inappropriate tuning for the parameters, increasing and decreasing factors, of the rate control function causes the performance degradation, most of the previous works so far have not studied the effect of the parameters on the performance. In this paper, we investigate the effect of the rate control parameters on the throughput under the condition that the packet loss probability is kept below a specific value, say 10-6. For this purpose, we build a queueing model and carry out a transient analysis to examine the dynamic behavior of the queue length at an intermediate node in a high speed network suffering from large propagation delay. Numerical examples exploit the optimal value of the parameters when one or two source-destination pairs transmit packets. We also discuss the effect of the propagation delay on the performance. Our model can be applicable to investigate the performance of various kinds of rate-based congestion control when the relation between the congestion measure and the rate control mechanism is given explicitly.

In the present paper, ImTCP-bg, a new background TCP data transfer mechanism that uses an inline network measurement technique, is proposed. ImTCP-bg sets the upper limit of the congestion window size of the sender TCP based on the results of the inline network measurement, which measures the available bandwidth of the network path between the sender and receiver hosts. ImTCP-bg can provide background data transfer without affecting the foreground traffic, whereas previous methods cannot avoid network congestion. ImTCP-bg also employs an enhanced RTT-based mechanism so that ImTCP-bg can detect and resolve network congestion, even when reliable measurement results cannot be obtained. The performance of ImTCP-bg is investigated through simulations, and the effectiveness of ImTCP-bg in terms of the degree of interference with foreground traffic and the link bandwidth utilization is also investigated.

Overlay networks are expected to be a promising technology for the realization of QoS (Quality of Service) control. Overlay networks have recently attracted considerable attention due to the following advantages: a new service can be developed in a short duration and it can be started with a low cost. The definition and necessity of the overlay network is described, and the classification of various current and future overlay networks, particularly according to the QoS feature, is attempted. In order to realize QoS control, it is considered that routing overlay and session overlay are promising solutions. In particular, session and overlay networks are explained in detail since new TCP protocols for QoS instead of current TCP protocols that control congestion in the Internet can be used within overlay networks. However, many open issues such as scalability still need further research and development although overlay networks have many attractive features and possess the potential to become a platform for the deployment of new services.

When computing resources are consolidated in a few huge data centers, a massive amount of data is transferred to each data center over a wide area network (WAN). This results in increased power consumption in the WAN. A distributed computing network (DCN), such as a content delivery network, can reduce the traffic from/to the data center, thereby decreasing the power consumed in the WAN. In this paper, we focus on the energy-saving aspect of the DCN and evaluate its effectiveness, especially considering traffic locality, i.e., the amount of traffic related to the geographical vicinity. We first formulate the problem of optimizing the DCN power consumption and describe the DCN in detail. Then, numerical evaluations show that, when there is strong traffic locality and the router has ideal energy proportionality, the system's power consumption is reduced to about 50% of the power consumed in the case where a DCN is not used; moreover, this advantage becomes even larger (up to about 30%) when the data center is located farthest from the center of the network topology.

We investigate performance of TCP protocol over ATM networks by using a simulation technique. As the ATM layer, we consider (1) rate-based control of the ABR service class and (2) an EPD (Early Packet Discard) technique applied to the UBR service class and (3) and EPD with per-VC accounting for fairness enhancement applied to the UBR service class. In comparison, we adopt a multi-hop network model where the multiple ATM switches are interconnected. In such a network, unfairness among connections is a possible cause of the problem due to differences of the number of hops and/or the round trip times among connections. Simulation results show that the rate-based control method of ABR achieves highest throughput and best fairness in most circumstances. However, the performance of TCP over ABR is degraded once the cell loss takes place due to the inappropriate control parameter setting. To avoid this performance degradation, we investigate the appropriate parameter set suitable to TCP on ABR service. As a result, parameter tuning can improve the performance of TCP over ABR, but limited. We therefore consider TCP over ABR with EPD enhancement where the EPD technique is incorporated into ABR. We last consider the multimedia network environment, where the VBR traffic exists in the network in addition to the ABR/UBR traffic. By this, we investigate an applicability of the above observations to a more generic model. Through simulation experiments, we find that the similar results can be obtained, but it is also shown that parameters of the rate-based congestion control must be chosen carefully by taking into account the existence of VBR traffic. For this, we discuss the method to determine the appropriate control parameters.

In this paper, we propose a new multicast routing algorithm for constructing the delay-constrained minimal spanning tree in the VP-based ATM networks, in which we consider the efficiency even in the case where the destination dynamically joins/departs the multicast connection. For constructing the delay-constrained spanning tree, we first generate a reduced network consisting of only VCX nodes from a given ATM network, originally consisting of VPX/VCX nodes. Then, we obtain the delay-constrained spanning tree with a minimal tree cost on the reduced network by using our proposed heuristic algorithm. Through numerical examples, we show that our dynamic multicast routing algorithm can provide an efficient usage of network resources when the membership nodes frequently changes during the lifetime of a multicast connection, while the existing multicast routing algorithm may be useful for constructing the multicast tree with a static nature of destination nodes. We also demonstrate that more cost-saving can be expected in dense networks when applying our proposed algorithm.

This article introduces a self-organizing model which builds the topology of a DHT mapping system for ICN. Due to its self-organizing operation and low average degree of maintenance, the management overhead of the system is reduced dramatically, which yields inherent scalability. The proposed model can improve latency by around 10% compared to an existing approach which has a near optimal average distance when the number of nodes and degree are given. In particular, its operation is simple which eases maintenance concerns. Moreover, we analyze the model theoretically to provide a deeper understanding of the proposal.

Content-centric networking (CCN) is an innovative network architecture that is being considered as a successor to the Internet. In recent years, CCN has received increasing attention from all over the world because its novel technologies (e.g., caching, multicast, aggregating requests) and communication based on names that act as addresses for content have the potential to resolve various problems facing the Internet. To implement these technologies, however, requires routers with performance far superior to that offered by today's Internet routers. Although many researchers have proposed various router components, such as caching and name lookup mechanisms, there are few router-level designs incorporating all the necessary components. The design and evaluation of a complete router is the primary contribution of this paper. We provide a concrete hardware design for a router model that uses three basic tables — forwarding information base (FIB), pending interest table (PIT), and content store (CS) — and incorporates two entities that we propose. One of these entities is the name lookup entity, which looks up a name address within a few cycles from content-addressable memory by use of a Bloom filter; the other is the interest count entity, which counts interest packets that require certain content and selects content worth caching. Our contributions are (1) presenting a proper algorithm for looking up and matching name addresses in CCN communication, (2) proposing a method to process CCN packets in a way that achieves high throughput and very low latency, and (3) demonstrating feasible performance and cost on the basis of a concrete hardware design using distributed content-addressable memory.

In a multi-tenant data center, nodes and links of tenants' virtual networks (VNs) share a single component of the physical substrate network (SN). The failure of a single SN component can thereby cause the simultaneous failures of multiple nodes and links in a single VN; this complex of failures must significantly disrupt the services offered on the VN. In the present paper, we clarify how the fault tolerance of each VN is affected by a single SN failure, especially from the perspective of VN allocation in the SN. We propose a VN allocation model for multi-tenant data centers and formulate a problem that deals with the bandwidth loss in a single VN due a single SN failure. We conduct numerical simulations (with the setting that has 1.7×108bit/s bandwidth demand on each VN, (denoted by Ci)). When each node in each VN is scattered and mapped to an individual physical server, each VN can have the minimum bandwidth loss (5.3×102bit/s (3.0×10-6×Ci)) but the maximum required bandwidth between physical servers (1.0×109bit/s (5.7×Ci)). The balance between the bandwidth loss and the required physical resources can be optimized by assigning every four nodes of each VN to an individual physical server, meaning that we minimize the bandwidth loss without over-provisioning of core switches.