Atsuo Tachibana

An Enhanced Available Bandwidth Estimation Technique for an End-to-End Network Path

This paper presents a unique probing scheme, a rate adjustment algorithm, and a modified excursion detection algorithm (EDA) for estimating the available bandwidth (ABW) of an end-to-end network path more accurately and less intrusively. The proposed algorithm is based on the well-known concept of self-induced congestion and it features a unique probing train structure in which there is a region where packets are sampled more frequently than in other regions. This high-density region enables our algorithm to find the turning point more accurately. When the dynamic ABW is outside of this region, we readjust the lower rate and upper rate of the packet stream to fit the dynamic ABW into that region. We appropriately adjust the range between the lower rate and the upper rate using spread factors, which enables us to keep the number of packets low, and we are thus able to measure the ABW less intrusively. Finally, to detect the ABW from the one-way queuing delay, we present a modified EDA from PathChirps original EDA to better deal with sudden increase and decrease in queuing delays due to cross traffic burstiness. For the experiments, an Android OS-based device was used to measure the ABW over a commercial 4G/LTE mobile network of a Japanese mobile operator, as well, as real testbed measurements were conducted over fixed and WLAN network. Simulations and experimental results show that our algorithm can achieve ABW estimations in real time and outperforms other stat-of-the-art measurement algorithms in terms of accuracy, intrusiveness, and convergence time.

Multi-dimensional sensor data aggregator for adaptive network management in M2M communications

This paper proposes a method of aggregating tempo-spatial data generated by sensors deployed in buildings or houses. The size of each sensor data such as temperature is usually small, but it often involves many additional data to represent its attribute values like time, location, data type and data precision. This would often increase the traffic volume between sensor gateway at building/home side and service providers at server side. In our method, such sensor data are packed into multidimensional matrices indexed by those attribute values for more compact representation, and the compressed sensing technique is adaptively applied to further reduce the data size. The method was applied to a field trial with KDDI corporation to collect data from 29 community facilities, and the traffic volume was reduced to 50% with reasonable precision of data restoration.

NEXT: New enhanced available bandwidth measurement technique, algorithm and evaluation

This paper presents a unique probing scheme and a rate adjustment algorithm that can be used for estimating the available bandwidth (ABW) of an end-to-end network path more accurately and non-intrusively. The proposed algorithm is based on the well-known concept of self-induced congestion and it features a unique probing train structure in which there is a region where packets are sampled more frequently than in other region. This high-density region enables our algorithm to find the turning point more accurately. When the dynamic ABW is outside of this region, we readjust the lower rate and upper rate of the packet stream to fit the dynamic ABW into that region. We appropriately adjust the range between the lower rate and the upper rate using spread factors, which enable us to keep the number of packets few and thus we measured the ABW non-intrusively. Simulation results show that our algorithm outperforms PathChirp, a state of the art measurement algorithm, estimating the ABW with greater accuracy and stability in presence of different cross traffic sources.

A deployable scheme of CMT-SCTP with off-the-shelf android smartphones

We present a deployable scheme that achieves seamless handover in the context of heterogeneous wireless access networks based on concurrent multipath transfer (CMT) using the Stream Control Transmission Protocol (SCTP) with off-the-shelf Android smartphones and protocol translation servers. Since the scheme is accomplished by neither embedding special modules nor relying on super-user (root) privileges on off-the-shelf Android smartphones, it has considerable potential for general acceptance. In our scheme, we modified the mechanism of the dynamic address reconfiguration (DAR) extension of SCTP to ensure seamless handover, especially for the scenario where the mobile users frequently join/leave heterogeneous Wi-Fi networks that provide Internet access via network address translation (NAT) devices. We confirmed the feasibility of the scheme through prototype implementation.

Implementation of a proxy-based CMT-SCTP scheme for Android smartphones

Since various types of wireless networks have been widely deployed, the Concurrent Multipath Transfer (CMT) scheme over multiple wireless networks has become vital to ensure that current mobile users can gain advantage in the context of high data transmission throughput and communication robustness. Aiming at deploying the CMT scheme, in this paper, an implementation design of the CMT scheme based on the Stream Control Transmission Protocol (SCTP) with Android smartphones and protocol translation servers is presented. This scheme is based on a proxy technique that transparently translates the TCP flows into CMT-SCTP flows (streams) to avoid any changes to the existing service applications and legacy servers. The proposed scheme was preliminarily implemented and evaluated through a real-world experiment in Japan, and was shown to have considerable potential to achieve the above advantages.

Implementation Design of Available Bandwidth Measurement Scheme: A Proxy based Approach

To widely deploy the performance measurement scheme on wireless access networks, in this paper, implementation design of an active measurement scheme with off-the-shelf Android smart-phones is presented and the data transformation server on the network paths is introduced as a proxy server. The presented scheme efficiently estimates the available bandwidth based on a proxy technique that conveys application data by piggybacking inside the probing packets, which results in a significantly fewer overheads of native active measurement. At the same time, since the scheme is accomplished without relying on superuser (root) privileges on off-the-shelf Android smart-phones and without any changes to the existing client applications and legacy servers, it has considerable potential for general acceptance. Through the preliminary experiments with the prototype implementation on the Internet, we confirmed the feasibility and the effectiveness of the proposed scheme where deployability is the chief concern.