Earticle

Multipath TCP(MPTCP)는 다중 인터페이스를 이용하여 데이터를 동시에 전송하는 전송계층의 프로토콜이다. MPTCP는 비슷한특성의 망을 사용하는 네트워크 환경에서는 기존 TCP에 비해 우월한 성능을 보이지만 이종 망을 사용하는 MPTCP의 다중경로중 특정 경로에서 버퍼블로트 현상이 일어나게 되면 급격한 지연시간의 증가로 인해 느린 경로로 전송 된 패킷이 MPTCP 수신버퍼에 도착하지 않게 되어 다른 경로로 수신한 패킷이 수신버퍼에서 대기하는 Head-of-Line(HoL) blocking 현상을 초래하여 기존 TCP 대비 더욱 악화된 성능을 보이기도 한다. 따라서 본 논문에서는 이러한 버퍼블로트 문제를 해결하기 위해 Active Queue Management(AQM) 기법 중 Adaptive Random Early Detection(ARED), Controlled Delay(CoDel), Proportional Integral Controller Enhanced(PIE)를 네트워크에 적용하여 이종 망에서 MPTCP 성능 개선을 목표로 실험을 진행하였다. 성능 향상을 살펴보기 위해Linux 기반의 유선 테스트베드를 구축하고 기존 드롭 테일과 AQM을 비교분석하여 AQM 기반의 TCP가 bufferbloat가 있는 다중경로 MPTCP의 성능을 향상시키는 것을 살펴보았다.

Multipath TCP (MPTCP) is a transport layer protocol that simultaneously transmits data using multiple interfaces. MPTCP is superior to existing TCP in network environment with homogeneous subflows, but it shows worse performance compared to existing TCP in network environment with bufferbloat. If a bufferbloat occurs in one of the MPTCP multipaths, the packet will not arrive at the MPTCP receive buffer due to a sudden increase in delay time, resulting in a HoL blocking phenomenon. It makes the receive window of the other path to be zero. In this paper, we apply Adaptive Random Early Detection (ARED), Controlled Delay (CoDel) and Proportional Integral Controller Enhanced (PIE) among the proposed Active Queue Management (AQM) to limit the delay of bufferbloat path. Experiments were conducted to improve the performance of MPTCP in heterogeneous networks. In order to carry out the experiment, we constructed a Linux-based testbed and compared the MPTCP performance with that of the existing droptail.

Enabling cooperation between vehicles form vehicular networks, which provide safety, traffic efficiency and infotainment. The most vital of these applications require reliability and low latency. Considering these requirements, this paper presents a multitier heterogeneous adaptive vehicular (MHAV) network. Comprising of transport operator or authority owned vehicles in high tier and all the other privately owned vehicles in low tier, integrating cellular network with dedicated short range communications. The proposed framework is implemented and evaluated in Glasgow city center model. Simulation results demonstrate that the proposed architecture outperforms previous multitier architectures in terms of latency while offloading traffic from cellular networks.

빅데이터 라는 개념이 등장하며 이전과 달리 대용량 데이터에 대해 가용할 수 있게끔 기술이 발전함에 따라 최근 추천시스템은 대량의 리소스 내에서 유저에게 적절한 개체를 발견하여 제공하는 용도로 이용되고 있으며 이 프로세스가 그 플랫폼내에서의 서비스 만족도 등 고객만족지표에 영향을 끼칠 만큼 온라인 서비스에서도 점차 그 역할의 중요도가 커져가고 있는 추세를 보이고 있다. 전통적인 추천시스템 방법론은 주로 유저-아이템 간의 상호 기록을 분석하여 효율적으로 예측하고자 하였으나 앞서 언급하였듯이 '유비쿼터스', '5G' 등의 기술들이 주변 곳곳에서 데이터를 수집하고 가용할 수 있게 되었으며 이로 인해 데이터에 대해 중요도가 커져가며 다소 지엽적인 시각에서만 바라보았던 데이터를 좀 더 다양한 측면에서도 생각하며 그 내에서 유용한 데이터를 추출하고자 하는 노력이 주를 이루고 있다. 그에 따라 추천시스템도 다양한 종류의 보조적인 성격을 가진 데이터도 이용하고자 하는 연구들이 진행되고 있긴 하나 그 내에서는 몇가지 문제가 존재 한다. 상대적으로 다양한 정보들 덕분에 유용하나 다른말로 표현해보자면 다소 복잡한 구조를 가지고 있다는 특징 때문에 그러한 피쳐들을 잘 가공하여 어떻게 모델에 접목시킬것인가에 대해 현재까지도 추천시스템뿐만 아니라 데이터마이닝 측면에서 도전적인 과제로 남아있으며 이걸 어떻게 효율적으로 해결할 것인가에 대해 새로운 접근법을 모색하고 있다. 그 중에서도 네트워크 임베딩이 데이터 내의 잠재적인 구조를 추출하여 좋은 성능을 보이며 '분류', '클러스터링', '추천' 등 많은 데이터 마이닝에서 활용하고 있다. 특히 추천 쪽에서는 'random walk' 그리고 'skip-gram'이 결합된 'Deepwalk'라는 방법론을 주로 이용하며 임베딩 된 구조 내에서 또 각 노드간의 연결에 가중치를 둔 'LINE' , 'SDNE' 등의 알고리즘도 사용하고있다. 허나 지금까지 언급한 방법들은 '동종 네트워크('homogeneous network)'에만 적용할 수 있다는 한계점을 지니고 있다. 본 연구에서는 그래프 구조 데이터를 활용하여 추천 시스템의 알고리즘의 정확도를 제고하기 위한 새로운 접근법을 제안한다. 구체적으로 본 연구에서는 그래프 상의 노드(node)의 임베딩(embedding)을 주변 이웃 노드로부터 생성하는 GCN(Graph Convolutional Networks)의 변형모델인 GraphSAGE(Sample and aggreGatE)에 기반하여 사용자-상품 네트워크 구조를 학습하고, 이를 기반으로 추천 상품을 예측하는 추천 모델을 제안한다. 일반적인 GraphSAGE 알고리즘은 네트워크를 구성하는 모든 노드가 동질하다고 전제하는 동종 네트워크(homogeneous network)를 기반으로 동작한다. 하지만, 추천 시스템에서의 노드는 크게 사용자와 상품이라는 서로 이질적인 두 종류의 노드로 구성되어 있다. 이에 본 연구에서는 이러한 이종 네트워크(heterogeneous network) 구조를 다룰 수 있도록 확장된 GraphSAGE 알고리즘인 HinSAGE(Heterogeneous SAGE) 알고리즘을 적용하여 추천 모델을 구축하고, 이를 전통적인 추천 알고리즘인 협업필터링(CF, Collaborative Filtering)과 비교하여 성능의 개선이 있는지 실증적으로 확인해 보고자 한다.

본 논문에서는 비동기식 이동통신망의 동일 주파수 대역에서 신형 원격 기지국(RRH)과 중계기간에 RF 신호가 상호 정합이 가능하도록 하는 정합 모듈 시스템 설계에 관한 것이다. 이 정합 시스템은 통신 중에 있는 모듈의 RF Gain을 제어하고 품질상태를 모니터링 한다. 또한 Network의 RF 환경변화에 따라 RF Gain을 적절히 조절하여 중계기의 수신 특성에 맞게 정합 할 수 있는 기능을 부여하였다. 본 연구의 결과로 신형 기지국과 기존 중계기간에 상호 인터페이스 품질문제를 개선하였다.

This paper concerns the design of a matching module system enabling the RF signal to be harmonized between the new base stations (RRH) and the repeater in the same frequency band of the asynchronous transfer network. This matching system controls the RF Gain Control of the module while monitoring the quality of the quality. Additionally, the RF environment has been adapted accordingly to adjust the RF Gain Control to match the receiver characteristics of the relay. As a result of this study, we improved the quality of the interface between the new base stations and existing relays.

UHD(Ultra High Definition) 방송은 초고선명 비디오와 다채널 오디오 재현으로 최근 관심이 높아지고 있다. 그런데 UHD 방송 서비 스를 제공하기 위해서는 지상파 현재 방송 시스템으로는 대역폭, 인코딩 등 해결해야할 기술적인 어려움이 있다. 이러한 문제를 해결 하기 위한 방안 중 하나로 MMT(MPEG Media Transport)가 제안되었다. MMT는 UHD와 같은 초고화질의 대용량 멀티미디어를 방송할 수 있는 표준으로, IP를 기반으로 구성되어 있기 때문에 다른 종류의 여러 망을 동시에 사용하여 전달할 수 있다는 특징이 있다. MMT를 기반으로 미디어 데이터를 계층적으로 분리하여 이기종망을 통해 전송하면 현재 방송 시스템으로도 UHD 미디어 서비스가 가능하다. 그러나 MMT를 이용하여 계층화된 UHD 미디어 데이터가 서비스되기 위해서는 이기종망을 통해 들어오는 데 이터들이 시그널링을 통해 관리되어야 하고, 수신기가 계층구조에 대한 정보를 빨리 전달받을 수 있어야 빠른 화면 구성 및 제공이 가능하기 때문에 이를 위한 정보가 시그널링 내에 존재해야 한다. 따라서 본 논문에서는 현재 MMT 내에 정의되지 않은 이기종망에 대한 정보를 적절하게 제공하는 시그널링 테이블 구조 설계를 제안한다.

UHD(Ultra High Definition) Broadcasting is the focus of attention recent days. However, current broadcasting system faces with difficulty of bandwidth, encoding and so on. To solve this problem, MMT was suggested as the one of solutions. MMT is designed to be based on IP network and it has characteristics that can deliver multimedia through many networks at the same time. UHD Media service can be available with the current broadcasting system that divides media data into hierarchical data based on MMT. To provide this service, information about heterogeneous network should be delivered and signalling should be given to perceive it. For UHD media data service, the information about data from heterogeneous networks should be transported for providing presentation information and service for the receiver models. The present MMT signalling has not much information about heterogeneous services with hierarchical media data. In this paper, we suggest the design of service signaling structure based on MMT for UHD broadcasting systems.

중소기업에 예상되는 통신 관련 이슈는 1) IT 시스템이 구축되어 있는 경우 연계가 쉽지 않고, 2) 이기종 시스템 에서 데이터 수집과 통합이 어렵고, 3) 다양한 Fieldbus 및 프로토콜이 존재하여 인터페이스가 쉽지 않다는 것이다. 보통 스마트공장 도입 전에 중소기업은 자동화가 구축되어 있는 경우가 많다. 중소기업의 자동화 대상 설비와 통신네트 워크에 노후 된 Sensor와 이기종 시스템 수준에 맞게 Sensing 등 통신기술을 함께 제공할 필요가 있다. PI를 활용한 사전준비 시점에 최신 네트워크 기술을 적용하기 전에 각 유형별 네트워크 인터페이스 개선방안을 고찰하고자 한다.

The communication issues expected for SMEs are that 1) IT systems are not easy to connect, 2) data collection and integration by heterogeneous systems are difficult, and 3) various fieldbuses and protocols make interfaces difficult. Usually, SMEs often have automation built before the introduction of smart factories. It is necessary to provide communication technology such as Sensing to meet the heterogeneous system level with the old aged sensors in the automation equipment and communication network of SMEs. We will consider how to improve the network interface before applying the latest network technology at the time of preparation using PI.

In Heterogeneous Network (HetNet), the small base station such as femtocells are deployed in indoor and multiple business complexes in order to extend the coverage and capacity. The deployment of co-channel (i.e. shared channel) for femtocells along with macrocell induces the severe interference issues. This is because of effects on SINR and thereby capacity become deteriorated. Therefore, this paper investigates the interference mitigation techniques and come out with a proposal to improve the Signal to Interference Noise Ratio (SINR) performance as well as total capacity in Heterogeneous Network (HetNet).The simulation results show that the achieved capacity is maximized than the existing technique.

Cell selection/reselection and handover (HO) are important mobility features and challenges of wireless cellular system. When network topology transforms from macro-only to heterogeneous network (HetNet) deployment, the imbalance of transmitted power and coverage area of different kinds of cells in HetNet introduce more challenges to those two important processes. In this paper, we investigate and elaborate some new algorithms and strategies to meet those challenges. On cell selection/reselection, there are two directions in associating more UEs with small cells, one is RSRP-based by introducing a empirical bias to expand small cells’ coverage area, the other is fairness-based to maximize the system throughput and usage of resources. As to more important and difficult HO process, the proposed methods include optimizing the parameters of event A3, handling time to trigger(TTT), optimizing HO process and utilizing inter-cell interference coordination algorithm(ICIC) etc, to maximize the usage of capability of deployed small cell. The presented new algorithms and strategies promise specific performance improvement with introducing receivable time and computing complexity.

We present a novel approach for the measurement of one-way delays and throughput between network nodes in heterogeneous environment over Wide Area Network (WAN). This research investigates performance evaluation of remote data transfers on heterogeneous services and technologies environment. We propose an enhanced equation to evaluate the performance of network traffic management via Little Law and Queuing theory to improve the evaluation algorithm. To get accuracy results on the performance of simulation model, we measure (verify and validate) data from lab experiment. We use network management tool to capture those data and Ping Plotter application (network analyzer) to generate traffic. As a result, this simulation model can provide a good approximation of the real traffic (one-way delays and throughput) observed in the real network environment (Wide Area Network, WAN). Through laboratory and field experiments, it shows that the model via simulation is capable of approximating the performance of remote data transfers over heterogeneous services and techniques within a minimum error range.

With the continuous improvement of the transmission rate and the quality of wireless communications services, the need for a wireless communication network structure to adjust. The traditional rules of cellular networks no longer meet the high-density, large number of people for high-speed data transmission. Heterogeneous network can solve the problem of network coverage and network-efficient transport, but because of the heterogeneous network dense arrangement of the network, so the introduction of a number of inter-cell interference, especially at the edge of the cell. To solve this problem, we use collaborative tools of game theory, to be distributed among the power cells to determine the coverage of each pico cell. But the need to optimize collaboration between cells, select the number and scope of the coordinated cell. Simulation results show that the proposed algorithm can effectively improve the transmission rate of the system, eliminating the interference between cells.

Prediction of links or relations between the objects in any network is no longer a new task these days; in fact it has become a high rated area of research and has attracted many researchers seeking their contribution to the mentioned area. Research has seen an exponential growth over the passing years, and the active researchers do not hesitate in linking with fellow researchers working in same domain irrespective of their geographic location. However this in turn has generated a very complex network of objects and links which are needed to be analyzed and dealt with. Prediction of co-authorship is the sub domain of link prediction and with the increasing complexity of co-authorship network the authors are treated as heterogeneous entity not as homogeneous ones. The rule is simple analyze the data preprocess it, train the classifier according to desired classification rules and then get the classified form of data. But irrelevant features always reflect various impacts and issues on generation of a classifier and consequently the impact is sustained to further classification results. Therefore, this paper proposes streaming classification algorithm combined with Correlation based Feature selection as a solution to the stated problem. The consistent and relevant features are selected with the help of feature selection algorithm and then these features are classified with the help of streaming classification algorithm- Very Fast Decision Tree (VFDT). VFDT is a streaming classification algorithm and it takes the dataset in the form of continuous stream as an input. Finally the effectiveness of the proposed algorithm can be seen in the experimental results.

This paper studied the power allocation and downlink interference issues in orthogonal frequency-division-multiple-access (OFDMA) system to mitigate the interference issues. In Orthogonal Frequency Division Multiplexing Access (OFDMA) frameworks, resource allotment to the subcarrier is crucial attributable to the deficient assets accessible at the base station. In OFDMA, subcarrier and power distribution are not separate, along with these lines this two portion are not self-overseeing. This paper examines the power allo-cation approach through exploring the HGBBDSA approach. The integration of GA with Biogeography Based Optimization algorithm benchmarked over the Particle Swarm Op-timization (PSO). The both of the algorithms cross usefulness is evaluated and looked at the exhibitions in Heterogeneous Network (HetNet). The reenactment results demonstrate that the HGBBDSA approach efficiently allocates the power than that of existing PSO approach.

In orthogonal frequency division multiplexing Access (OFDMA) systems resource allocation to the subcarrier is essential owing to the insufficient resources available at the base station. In OFDMA, subcarrier and power allocation are not separate; thereby this two allocation are not self-governing. This paper proposes the subcarrier allocation approach through investigating previous methods, and also the Enhanced Particle Swarm Optimization (PSO) algorithm is analyzed. The both of the algorithms cross functionality is analyzed and compared the performance of the subcarrier allocation of OFDMA systems in Heterogeneous Network (HetNet). The simulation results show that the PSO approach efficiently allocates the OFDMA subcarriers.

Event coreference resolution is an important step for business intelligence. In this paper, we present a method for event coreference resolution based on heterogeneous information network. We adopt holistic, hierarchical clustering algorithm to solve this problem. In terms of similarity measure, the proposed method utilizes the relationship between entities, events, documents and data source, calculates event comprehensive similarity through the event properties similarity. Experiment on business event sets, characters event sets and product event sets, the experiment results show that the algorithm of this paper can effectively complete event coreference resolution, and has good recall and precision.

The continuously increasing diffusion of application scenarios requiring pervasive data sensing has lead to the development of novel monitoring solutions leveraging on Heterogeneous Sensor Networks (HSNs). At the moment, only little attention has been dedicated to the design of multi-purposes applications devoted to process, filter and use data from HSNs and aimed at providing composite services to the final users. Most available networks are currently managed by custom stand-alone applications, where many low level details are handled at application level. Actually, this approach presents many drawbacks in terms of system flexibility and adaptability. In such a scenario, the main goal is to decouple application logic from low level implementation details including the HW/SW platform and the programming language. Moreover, the specific units of functionality should be organized into autonomous and connectable modules operating in a distributed computing environment. This paper presents a management framework for rapid and flexible deployment of HSN-based applications based on XMPP/Jabber, a well-known XML-based open-source Instant Messaging protocol. The proposed solution aims to speed up the integration of different software modules into a network of services. The main benefits of the proposed solution include an increase of code reusability, platform independency, system scalability and modularity. The advantages of the proposed approach are demonstrated by presenting three use-cases.

Future Heterogeneous Network (HetNet) will foster interoperability to achieve seamless multimedia services. To exploit the ubiquitous diversity across HetNet, Mobile Nodes (MNs) will prefer to perform Vertical Handover (VHO) for better guaranteed Quality of Experience (QoE) and optimal resource utilization. To perform VHO, the radio frequency part of the emerging wireless standards needs to accurately estimate the channel for network selection. Further in HetNet, MNs may perform VHO in group (in bus/ train) where HetNet utility increases abruptly due to simultaneous VHO requests and severely degrades the HetNet performance. Therefore, paper proposes a proactive group VHO model, whose network discovery is based on error vector magnitude (EVM) measurement which can be mobile or network controlled for optimal network selection (new attachment node). Proposed model is compared with probabilistic and non-cooperative selection schemes. Performance is evaluated in terms of link availability, packet loss rate, transmission delay, resource utilization to achieve overall optimal load balance.