萬代 雅希

Now is the age of neuromorphic computing that creates brain circuits. The analog and digital circuit theory changes because the values of the basic conductance elements can be made variable by learning. The computer structure changes to in-memory computing technology in cooperation with a von Neumann architecture. In this paper, we propose a gyrator neuron (GN) that enables analog computer operations by nodal equation. The GN is constructed based on memristor elements. The GN executes learning by back propagation processing and association by forward propagation processing.

In this paper, we propose a tile quality selection method in tile-based video streaming. The proposed method suppresses the spatial quality variation within viewport caused by the change of the viewport region due to the movement of the user’s head. In the proposed method, client checks whether the difference of quality level between viewport and around viewport is large. When assigning quality, if the difference of quality level between the viewport and around viewport is large, the difference is reduced. As a result of simulation, when the segment length is long, quality variation can be suppressed without significantly reducing the perceived bitrate. Especially, the quality variation within viewport can be greatly suppressed.

In Adaptive Bit Rate (ABR) utilized in MPEG-DASH (MPEG Dynamic Adaptive Streaming over HTTP), there have been proposed three representative algorithms, rate-based, buffer-based and hybrid-based ones. In the information centric networks, which have been widely studied as a network architecture for efficient content distribution including video streaming, routers are generally equipped with cache storage. When some segments of a required content is occasionally downloaded from cache storage in a router, content download from a router might be faster than segment retrieval from the original content server. This inaccurate throughput estimation might lead to inappropriate bitrate selection of ABR algorithm running on the user's application and significant degradation of QoE (Quality of Experience) performance. In this paper, from the viewpoint of QoE, we evaluate the influence of in-network caches on three types of ABR algorithm in video streaming.

In this paper, we propose a quality-level control method based on quality of experience (QoE) characteristics for HTTP adaptive streaming (HAS). The proposed method works as an adaptive bitrate controller on the HAS client. The proposed method consists of two operations: buffer-Aware control and QoE-Aware control. We implement the proposed method on an actual dynamic adaptive streaming over HTTP (DASH) program and evaluate the QoE performance of the proposed method via both objective and subjective evaluations. The results show that the proposed method effectively improves both objective and subjective QoE performances by preventing stalling events and quality-level switchings that have a negative influence on subjective QoE performance.

Resource Pooling concept has been proposed as fair sharing of total network resources among all users sharing a network so that a whole network is treated as a single pooled resource. MPTCP (Multipath TCP) has been proposed as one of the most promising congestion controls which realize the Resource Pooling in the situation that each user can utilize multiple paths. In CCN (Content-Centric Networking), several multipath/multisource congestion control methods have been proposed. A fundamental difference between CCN congestion control and MPTCP where single source is assumed, is multisource and the Resource Pooling in multipath/multisource situation is quite a new research issue. In this paper, we discuss congestion control for CCN from the viewpoint of the Resource Pooling and show that our previously proposed congestion control satisfies the Resource Pooling concept. Our simulation results also show that our proposed congestion control achieves macroscopic fairness among users, which means the Resource Pooling concept is satisfied.

In the paper, we propose a new congestion control method based on hop-by-hop rate control in named data networking (NDN). The proposed method suppresses the excessive reduction of the Interest sending rate due to continuous negative acknowledgement (NACK) packets caused by the propagation delay. The proposed method limits the rate reduction to once per congestion. The point of the proposed method is to keep the Interest sending rate for NACK packets unchanged after the rate reduction for congestion until the rate reduction becomes effective. Performance evaluation is conducted through computer simulations in order to demonstrate that the proposed method improves the throughput performance as compared to the conventional method.

This paper first investigates how a network operates when multiple receivers download content simultaneously in content-centric networking (CCN) when the receivers' downloading speeds differ. The results indicate that the performance of the download completion time of a faster user degrades excessively due to a decrease in the cache-hit rate in the router. Based on the investigation, this paper proposes a novel in-network caching method for simultaneous download from multiple receivers in CCNs. The proposed method keeps cached data packets in a router until slower receivers download the data, in order to prevent slower users from directly downloading data from the content provider. We conduct computer simulations and confirm the effectiveness of the proposed method. We show that the proposed method can improve the download completion time performance in the situation where multiple receivers download content at different speeds in CCN.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.25(2017) (online)DOI http://dx.doi.org/10.2197/ipsjjip.25.164------------------------------This paper first investigates how a network operates when multiple receivers download content simultaneously in content-centric networking (CCN) when the receivers' downloading speeds differ. The results indicate that the performance of the download completion time of a faster user degrades excessively due to a decrease in the cache-hit rate in the router. Based on the investigation, this paper proposes a novel in-network caching method for simultaneous download from multiple receivers in CCNs. The proposed method keeps cached data packets in a router until slower receivers download the data, in order to prevent slower users from directly downloading data from the content provider. We conduct computer simulations and confirm the effectiveness of the proposed method. We show that the proposed method can improve the download completion time performance in the situation where multiple receivers download content at different speeds in CCN.------------------------------This is a preprint of an article intended for publication Journal ofInformation Processing(JIP). This preprint should not be cited. Thisarticle should be cited as: Journal of Information Processing Vol.25(2017) (online)DOI http://dx.doi.org/10.2197/ipsjjip.25.164------------------------------

One of the major challenges for the deployment of underwater acoustic sensor networks is the development of a medium access control (MAC) protocol catering for the harsh underwater environment. In particular, an underwater MAC protocol should provide high end-to-end throughput, low channel access delay, and fair share of the scarce network bandwidth. In this paper, a cross-layer MAC protocol is proposed. It interacts with a price-based rate allocation scheme at the network layer. To accurately reflect the clique constraint of the wireless medium, the clique-based price is generalized to act as the congestion signal, which controls the end-to-end rates of multihop flows. The MAC protocol then schedules contention-free packet transmissions of single-hop subflows in each maximum clique. Both the MAC protocol and rate allocation algorithm are simple and direct, thus owning low computational complexity. Through analysis and simulation, we show that the proposed MAC protocol enables multihop flows to acquire the max-min fair share of the network bandwidth from the end-to-end perspective.

Out-of-order packet problem incurs severe degradation of throughput performance of Multipath TCP (MPTCP) with heterogeneous communication links. In this paper, we propose an extended version of MPTCP to solve the out-of-order packet problem. The proposed method defines two types of ACK: Regular ACK and Extended ACK. When a receiver receives a data packet, the receiver sends a Regular ACK to the original path through which the data is transmitted, and Extended ACKs to all the other paths. If the Extended ACK arrives earlier than the Regular ACK, the proposed method solves the out-of-order packet problem. We evaluate the performance of the proposed method via NS-3 computer simulations, and show that the proposed method can improve the throughput performance when RTT of paths is heterogeneous.

In Wireless Sensor Networks (WSNs), saving energy is of critical importance and has attracted much research. Data aggregation is one of the methods to save energy in WSNs by reducing the amount of redundant data transmission. However, data aggregation causes transmission delay and reduces data accuracy, and may not be suitable for some applications where quick transmission or data accuracy is important. Our first contribution in this regard is giving the formulations of energy efficiency, transmission delay and data accuracy from the viewpoint of full data aggregation as well as non-aggregation.Secondly, as a way to overcome the problems, we propose a random partial aggregation, and compare it with non-aggregation as well as full aggregation to verify its effectiveness. Finally,we define and measure a trade-off index as a useful trade-off criterion among energy consumption, transmission delay, and data accuracy.

The recent advent of cognitive radio has made the opportunistic spectrum access scheme (OSA) a suitable approach to deal with the spectrum inefficiency issues. This paper concerns itself with the temporal aspect of the OSA. Over the years, a plethora of works have discussed the optimal access methods for the OSA environment. However, most of them have not considered the possible combination of a faster transmission rate and the assistance of a relay node in temporal spectrum sharing. In this paper, we affirm that the use of a faster transmission rate with the assistance of a relay (or relays) can be greatly beneficial in the temporal OSA environment. We propose a rate-relay adjustment scheme to dynamically exploit the nature of rate-relay. The scheme is composed of a receiver-based auto rate increase (R-ARI) scheme for rate adjustment, a relay selection scheme, and a rate limit scheme. The scheme takes into account the rate-relay adjustment and the PU activity factors. As demonstrated by simulation results, the use of a proper rate-relay adjustment scheme can lead to a gain in the secondary throughput and a lower primary collision percentage, as compared to the one hop optimal transmission rate scheme.

In this paper, we propose a dynamic Quality of Experience (QoE)-aware live streaming algorithm by utilizing client clustering and similarity metrics. Encoding scheme of the proposed system finds a sufficient bitrate of the video and creates quality levels adapted to the distribution of clients' bandwidth to improve overall average of user's QoE. We show that the proposed system can improve structural similarity method (SSIM) performance when the client's bandwidth is distributed in low or middle bandwidth, and decreases the number of clients who stop play back.

In wireless sensor networks, energy depletion of bottleneck nodes which have more data packets to relay than others, dominates the network lifetime referred to as the funnel effect problem. To overcome this problem, multiple sink methods have been proposed where sensor nodes send observed data packets toward several sinks to distribute traffic load of bottleneck nodes. If both of the topology and the traffic pattern are symmetric, bottleneck nodes are located near sinks. However, in a general sensor network with an asymmetric topology and/or an asymmetric traffic pattern, bottleneck nodes may exist any place in the network. In this paper, we propose DCAM (Dispersive Cast of packets to Avoid bottleneck nodes for Multiple sink sensor network), which is a load balancing method to improve lifetime of a sensor network with an asymmetric topology and an asymmetric traffic pattern. DCAM first finds bottleneck nodes, and then balances the load on the bottleneck nodes. Selected nodes send data packets to several sinks dispersively according to some criteria. The criteria classify DCAM into three variations: DCAM with probability (DCAM-P), DCAM with moving boarder (DCAM-MB), and DCAM with round-robin (DCAM-RR). This paper gives details of the DCAM methods, and thereafter evaluates them with asymmetric topologies and asymmetric traffic patterns. To deal with these dynamic asymmetry, the topology is modeled by a grid network with virtual holes that are defined as vacant places of nodes in the network. Asymmetry of traffic pattern is modeled by defining a hot area where nodes have heavier data traffic than the others. The evaluations are conducted as changing hot-area traffic patterns as well as fixing hot-area patterns. The results show that DCAM improves network lifetime up to 1.87 times longer than the conventional schemes, (i.e., nearest sink transmissions and optimal dispersive cast of packet). We also discuss DCAM on several aspects such as overhead, energy consumption, and applications.

Multi-view video is attracting extensive interest due to its innovative viewing experience. It allows user to change viewpoints by playing different video sequences. However, the traffic of multi-view video is much larger than that of conventional media, which would greatly decrease the quality of service, i.e., more transmission delay. In this paper, we analyze the successive motion switching model and find that only the frames in a triangle window are possible to be displayed at a subsequent period of time. A live-encoding scheme called UMS (User dependent Multi-view video Streaming) is proposed to reduce transmission bitrate. Instead of all the frames, UMS transmits only these specific frames by the periodic feedback. We also propose a prediction structure as a substitute of the prediction structure of MVC. Performance evaluation proves that UMS reduces the transmission bitrate by 53.64% compared to MVC with periodic feedback. Copyright © 2013 Binary Information Press.

Energy harvesting from ambient energy sources including solar and vibration has been studied as a candidate for powering next generation wireless sensor networks. However, energy harvesting is unstable to supply a sensor node with energy, and a node cannot know whether its neighbouring nodes have enough energy to receive a data packet. In this paper, we propose two data collection protocols for energy harvesting wireless sensor networks called the Probabilistic ReTransmission protocol (PRT) and PRT with Collision Consideration (PRT-CC). The idea is to derive the appropriate number of times to retransmit a packet based on the reception probability and the active intervals computed by the receivers themselves while, in PRT-CC, each node computes the reception probability with packet collision consideration. The performance evaluation shows that the proposed protocols are able to achieve higher delivery ratio than the previous work, namely, Geographic Routing with Duplicate Detection (GR-DD) and GR-DD with Retransmission.

In this paper, we propose and implement a cross layer protocol for ad hoc networks using directional antennas. In the proposed protocol called RSSI-based MAC and routing protocol using directional antennas (RMRP), RSSI is used for computing the direction of the receiver and also used for controlling backoff time. Moreover, the backoff time is weighted according to number of hops from a source node. In addition, simple routing functions are introduced in the proposed RMRP. We implement the proposed RMRP on a testbed with the electronically steerable passive array radiator (ESPAR) antenna and IEEE 802.15.4. From some experimental results, we confirm some throughput improvement and show the effectiveness of the proposed RMRP. Especially, the proposed RMRP can achieve about 2.1 times higher throughput than a conventional random backoff protocol in a multi-hop communication scenario. © 2012 Information Processing Society of Japan.

一般にキャンパスネットワークにおける幹線から支線への拡張手段として，スイッチングハブの多段接続が用いられることが多い．このような接続形態では，ハブ間の接続ポート部に通信データが集中し，輻輳が発生しやすいという問題点がある．しかし，輻輳の発生時にはTCPウインドウ制御など通信の抑制による方法がほとんどであり，ネットワークのスループットが低下することは避けられない．一方で通信効率を向上させるために複数のパケットを排他的論理和で多重化するネットワークコーディングが注目されている．本研究では，スイッチングハブの多段接続による潜在的なボトルネック問題をコンシューマ向けの市販実機によって明らかにするとともに，ネットワークの輻輳制御におけるフレーム処理方法にネットワークコーディングを活用し，スループットの向上と輻輳の緩和を同時に可能とする輻輳制御方式を提案する．実験環境として提案方式をアプリケーション層に実装し，提案方式の優位性を立証する．さらにキャンパスネットワークにおける実際の通信環境を分析し，提案方式が有効に機能する環境であることを明らかにした．In general, LANs are enhanced by cascade connection of switching hubs. In a LAN with the cascade connection, the communication at the port of connected points, congestion is subject to occur at the point. For the problem TCP's window control has been already proposed for a congestion control in full duplex Ethernet. However, throughput degrades due to controlling the transmission rate in these methods. On the other hand, the network coding is attracted attention as a technology that improves the bandwidth efficiency. In this paper, we clarify a potential problem by the cascade connection of switching hubs, and we propose a congestion control method that enables the throughput improvement and the congestion reduction at the same time. We implement the proposal method on application layer to prove the effectiveness of the proposal method. Furthermore, we analysis real communication data in a campus network and revealed that a proposal method functioned in this environment.

In this paper, we propose a multi-lobe directional transmission for network coding in multi-rate ad hoc networks to improve throughput performance of XOR-based WNC. The proposed scheme uses directional antennas that can make antenna beam forms with multiple main lobes. In the proposed scheme, a node transmits uncoded packets with a beam form of one or multiple main lobes for realizing high-rate unicast transmissions. In addition a node transmits XORed packets and native packets with a beam form of multiple main lobes with directional antenna to realize high-rate multicast transmissions. We evaluated and investigated the performance of the proposed scheme in terms of throughput via intensive computer simulations. The simulated results show that the proposed scheme can improve throughput performance of XOR-based WNC in some simple topologies such as 5 nodes straight line topology and X topology. In the fundamental evaluation, we have confirmed that the proposed scheme can achieve higher throughput than the conventional XOR-based WNC without using directional antennas in the same topologies. The proposed scheme is also applicable for other smart/directional antennas.

This paper discusses a buffering strategy for a delay-tolerant multimedia sensor network (DTMSN), whose typical application is video surveillance. In DTMSN, a sensor node observes events around it and stores the data in its own buffer memory. All the data is collected to the sink. Sensor nodes have restrictions on buffer memory as well as battery capacity. The entire data size is much larger than a single node’s memory size. Thus, developing a strategy for buffering satisfying these restrictions is a critical issue for DTMSN. In this paper, we propose a novel buffering scheme for DTMSN called cooperative buffering (CB). In the proposed CB, the sensor node which has a large amount of data cooperates with its neighbor nodes to buffer the data in a distributed manner. CB uses mobile sinks. The cooperatively buffered data are transmitted directly to the mobile sink when it arrives. After proposing CB, this paper discusses extension for easy collection of the sink, extension for multi source nodes, and some sink mobility strategies of sink mobility. It evaluates the power consumption performance of CB via theoretical formulation and computer simulation. As a result, we show from the results that the proposed CB can handle multimedia data while operating at low-power.

近年,指向性アンテナの使用を仮定した指向性MAC(Medium Access Control)プロトコルが提案されている.従来の指向性MACプロトコルでは,理想的なアンテナビームパターンが仮定されていた.しかし,実際のアンテナを用いた場合,サイドローブ,バックローブ等のマイナーローブの存在により,理想アンテナでは考慮されていなかった問題が発生することが考えられる.本論文では,マイナーローブが存在するアンテナビームパターンを用いて既存のMACプロトコルを評価し,問題点を定量的に明らかにする.問題点として,NAV(Network Allocation Vector)が適切に設定されないために発生する四つの問題を明らかにし,それらの問題について考察する.続いて,指向性アンテナのマイナーローブを考慮した指向性MACプロトコルを提案する.提案方式では,指向性巡回受信,NAVリクエストフレーム送信により,周辺端末との衝突を回避する.シミュレーション評価により,理想的なアンテナビームパターンを仮定して設計された従来のMACプロトコルと比較し,提案方式のスループット性能が向上することを示す.

Although the use of directional antennas in ad hoc networks is expected to provide significant improvements, directional MAC protocols inherently introduce new kinds of problems. Deafness is one of the major problems caused when a transmitter repeatedly attempts to communicate with its intended receiver, but it fails because the receiver has its beam pointed away from the transmitter. This paper proposes Receiver-Initiated Directional MAC (RI-DMAC) to handle the deafness problem. RI-DMAC is a combination of sender- and receiver-initiated operations. The sender-initiated mode is the default mode and the receiver-initiated mode is triggered when the transmitter experiences deafness. Each node maintains a polling table and polls a deafness node using the Ready to Receive (RTR) frame after the completion of every dialogue. Simulation results show that RI-DMAC outperforms other directional MAC protocols (e.g. Basic DMAC and Circular RTS MAC) in terms of throughput, fairness, overhead and packet drop ratio.

センサネットワークにおいて,観測データの発生が地理的に偏る場合には,センサノードの電力消費に差が生じ,シンク付近のノードが他のノードに比べて早期にバッテリ枯渇を起こす.これが原因となり,長いネットワーク寿命を達成できない.この問題に対し,本論文では,複数のシンクを用い,各ノードが複数のシンクを届け先とするバケット分配送信方式(DispersiveCast方式)を提案する.DispersiveCast方式は,各シンクヘのホップ数をもとに各ノードが分散的にバケット送信割合を決定するB-DOP(Basic DispresiveCast Of Packet)送信方式と,複数のシンクに到達するトラヒックを均衡化させるようにシンクが集中的にバケット送信割合を決定するO-DOP(Optima1 DispersiveCast Of Packet)送信方式からなる.シミュレーションによって,両送信方式の性能を評価し,DispersiveCast方式がネットワーク寿命の延長を可能とすることを示す.

In this paper, a routing protocol referred to as Directed Diffusion with Stepwise Interest Retransmission (DD/SIR) for wireless sensor networks is proposed to mitigate power consumption considering node mobility. In DD/SIR, a sink retransmits interest. The propagation areas of the interest are narrowed stepwisely. In addition, according to the number of hops between the sink and sensor nodes, the data transmission timing is controlled sequentially. By both theoretical analysis and computer simulation, we evaluate the performance of DD/SIR. We show that DD/SIR can mitigate control overhead and realize low power operation without degrading data reachability to the sink. Especially, at a small number of data sending nodes, DD/SIR is more effective than the conventional routing.

ワイヤレスマルチホップネットワークでは,端末のアクセス数の増加やホップ数の増加に伴い,著しくスループットが低下する問題点がある.我々は,この問題を解決するために,IEEE802.11をベースとしたAsymmetric Wireless Multi-hop Access Network(以下AWiMA Net)を提案している.AWiMA Netではアクセスポイント(以下AP)と携帯端末(以下端末)の通信を上りリンク配信区間と下りリンク配信区間に分離するノンコンテンションデータ転送方式によってデータパケットのふくそうを抑止する.更に下りリンク配信区間においてAPの信号出力を高めAPから端末へは1ホップでデータ送信を行いAPのサービス領域の拡大とスループット向上を行っている.本論文ではAWiMA Netに対するシミュレーションによって得られたスループットの特性評価を行った上で,ネットワークのスループット向上のための下りデータのバースト転送による高速化を提案する.シミュレーションにより提案方式を評価した結果,端末からAPへのデータ配信の要求間隔と,上りリンク配信区間と下りリンク配信区間で構成するサイクルタイム長の設定を適切に行うことで,提案方式によるスループットが3倍から4倍向上することを確認した.

In this paper, we propose a novel energy-efficient route-discovery scheme with transmission power control (TPC) for ad hoc networks. The proposed scheme is very simple and improves energy efficiency without any information about neighbor nodes. In the proposed scheme, when a node receives a route request (RREQ), the node calculates the routing-level backoff time as being inversely proportional to the received power of the RREQ. After the route discovery, source and intermediate nodes transmit packets by the power-controlled medium access control (MAC) protocol. In addition, we propose an extended version of the proposed scheme for discrete power control devices. Simulation results demonstrate the proposed schemes can discover more energy efficient routes than the conventional schemes.

本論文では,データ送信における電力効率を高める目的で送信電力と送信レートを適応制御するアドホックネットワークメディアアクセス制御(MAC)プロトコルを提案する.提案プロトコルでは,あらかじめ送信電力と送信レートのすべての組合せに対して,電力消費の効率を算出した送信電力レート制御テーブルを作成する.提案プロトコルでは制御フレームの交換によって得られた受信電力を用いて,送信電力レート制御テーブルを参照して,二つの送信シーケンスを使い分ける.送受信間の直接送信よりも他のノードが中継を行った方が電力効率が高い場合,リレーシーケンスを適用する.計算機シミュレーションによって提案プロトコルを評価し,提案方式の電力効率が向上できることを示す.

In this paper, to realize the low delay and high throughput route discovery in multi-rate ad hoc networks, we propose a novel on demand routing using signal strength, called signal strength aware routing (SSR). SSR is based on the on-demand routing with the route request (RREQ) and route reply (RREP) procedure. In SSR, a node measures the signal strength of a received RREQ, and calculate the appropriate data transmission rate. Nodes also calculate the standby time for the RREQ forwarding proportionally to the medium time at the data transmission rate. A RREQ through higher data rate links arrives at the destination earlier, and the destination can select a low delay and high throughput route easily. We evaluate the performance of SSR in terms of delay, throughput and route discovery delay by means of QualNet network simulator. As a result, we show that SSR can discover the lower delay and higher throughput route than the conventional shortest hop routing without increasing control overhead in multi-rate ad hoc networks.