• Journal of Information Processing Systems
• /
• 제12권1호
• /
• pp.57-72
• /
• 2016

To provide effective communication in the wireless mesh network (WMN), several algorithms have been proposed. Since the possibilities of numerous failures always exist during communication, resiliency has been proven to be an important aspect for WMN to recover from these failures. In general, resiliency is the diligence of the reliability and availability in network. Several types of resiliency based routing algorithms have been proposed (i.e., Resilient Multicast, ROMER, etc.). Resilient Multicast establishes a two-node disjoint path and ROMER uses a credit-based approach to provide resiliency in the network. However, these proposed approaches have some disadvantages in terms of network throughput and network congestion. Previously, the buffer based routing (BBR) approach has been proposed to overcome these disadvantages. We proved earlier that BBR is more efficient in regards to w.r.t throughput, network performance, and reliability. In this paper, we consider the node/link failure issues and analogous performance of BBR. For these items we have proposed a resilient packet transmission (RPT) algorithm as a remedy for BBR during these types of failures. We also share the comparative performance analysis of previous approaches as compared to our proposed approach. Network throughput, network congestion, and resiliency against node/link failure are particular performance metrics that are examined over different sized WMNs.

• 한국통신학회논문지
• /
• 제32권5A호
• /
• pp.441-450
• /
• 2007

본 논문에서는 위성망 자원할당 및 분석 툴(SNRAT : Satellite Network Resource Allocation & Analysis Tool)을 활용하여 위성통신체계에 대한 링크성능 분석결과를 제시하였으며, 모의위성망을 구성하여 성능분석 툴로 예측한 링크 성능을 검증하였다. 통신위성은 지상장비와 달리 정지궤도에 발사하여 궤도 내 운용을 시작하면, 더 이상 성능의 수정이나 수리가 불가능하고, 운용 가능한 수명이 제한되어 있기 때문에 적기에 배치하여 운용하는 것이 매우 중요하다. 따라서, 위성 발사 전 위성망 자원할당 및 분석 툴을 이용하여 통신품질, 지원용량 및 항재밍 성능 등을 예측하고, 모의위성망을 구축하여 지상단말과 망제어기와 같은 지상장비 및 중계기 규격의 적합성을 증명하였으며, 여러 망 운용 시나리오에 대한 시험을 통해 최적의 운용방안을 도출하였다.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년도 제37회 하계학술대회 논문집 D
• /
• pp.1868-1870
• /
• 2006

Neural network is used in many fields of control systems, but input-output patterns of a control system are not easy to be obtained and by using as single feedback neural network controller. And also it is difficult to get a satisfied performance when the changes of rapid load and disturbance are applied. To resolve those problems, this paper proposes a new algorithm which is the neural network controller. The new algorithm uses the neural network instead of activation function to control object at the output node. Therefore, control object is composed of neural network controller unifying activation function, and it supplies the error back propagation path to calculate the error at the output node. As a result, the input-output pattern problem of the controller which is resigned by the simple structure of neural network is solved, and real-time learning can be possible in general back propagation algorithm. Application of the new algorithm of neural network controller gives excellent performance for initial and tracking response and it shows the robust performance for rapid load change and disturbance. The proposed control algorithm is implemented on a high speed DSP, TMS320C32, for the speed of 3-phase induction motor. Enhanced performance is shown in the test of the speed control.

• 전자공학회논문지B
• /
• 제31B권9호
• /
• pp.159-165
• /
• 1994

In this paper, we propose a multi-layer associative neural network structure suitable for hardware implementaion with the function of performance refinement and improved robutst capability. Unlike other methods which reduce network complexity by putting restrictions on synaptic weithts, we are imposing a requirement of hidden layer neurons for the function. The proposed network has synaptic weights obtainted by Hebbian rule between adjacent layer's memory patterns such as Kosko's BAM. This network can be extended to arbitary multi-layer network trainable with Genetic algorithm for getting hidden layer memory patterns starting with initial random binary patterns. Learning is done to minimize newly defined network error. The newly defined error is composed of the errors at input, hidden, and output layers. After learning, we have bidirectional recall process for performance improvement of the network with one-shot recall. Experimental results carried out on pattern recognition problems demonstrate its performace according to the parameter which represets relative significance of the hidden layer error over the sum of input and output layer errors, show that the proposed model has much better performance than that of Kosko's bidirectional associative memory (BAM), and show the performance increment due to the bidirectionality in recall process.

• Nuclear Engineering and Technology
• /
• 제54권2호
• /
• pp.770-780
• /
• 2022

In 1993 the University of Extremadura initiated the design, construction and management of the Radiological Alert Network of Extremadura (RAREx). The goal was to acquire reliable near-real-time information on the environmental radiological status in the surroundings of the Almaraz Nuclear Power Plant by measuring, mainly, the ambient dose equivalent. However, the phased development of this network has been carried out from two points of view. Firstly, there has been an increase in the number of stations comprising the network. Secondly, there has been an increase in the number of monitored parameters. As a consequence of the growth of RAREx network, large data volumes are daily generated. To face this big data paradigm, software applications have been developed and implemented in order to maintain the indispensable real-time and efficient performance of the alert network. In this paper, the description of the current status of RAREx network after 30 years of design and performance is showed. Also, the performance of the graphing software for daily assessment of the registered parameters and the automatic on real time warning notification system, which aid with the decision making process and analysis of values of possible radiological and non-radiological alterations, is briefly described in this paper.

• Smart Structures and Systems
• /
• 제26권2호
• /
• pp.175-184
• /
• 2020

Conventional Monte Carlo simulation-based methods for seismic risk assessment of water networks often require excessive computational time costs due to the hydraulic analysis. In this study, an Artificial Neural Network-based surrogate model was proposed to efficiently evaluate the flow-based system reliability of water distribution networks. The surrogate model was constructed with appropriate training parameters through trial-and-error procedures. Furthermore, a deep neural network with hidden layers and neurons was composed for the high-dimensional network. For network training, the input of the neural network was defined as the damage states of the k-dimensional network facilities, and the output was defined as the network system performance. To generate training data, random sampling was performed between earthquake magnitudes of 5.0 and 7.5, and hydraulic analyses were conducted to evaluate network performance. For a hydraulic simulation, EPANET-based MATLAB code was developed, and a pressure-driven analysis approach was adopted to represent an unsteady-state network. To demonstrate the constructed surrogate model, the actual water distribution network of A-city, South Korea, was adopted, and the network map was reconstructed from the geographic information system data. The surrogate model was able to predict network performance within a 3% relative error at trained epicenters in drastically reduced time. In addition, the accuracy of the surrogate model was estimated to within 3% relative error (5% for network performance lower than 0.2) at different epicenters to verify the robustness of the epicenter location. Therefore, it is concluded that ANN-based surrogate model can be utilized as an alternative model for efficient seismic risk assessment to within 5% of relative error.

• 한국정보통신설비학회:학술대회논문집
• /
• 한국정보통신설비학회 2007년도 학술대회
• /
• pp.69-72
• /
• 2007

In this paper, we analyze the performance of the network management system with intelligent mobile agent system. The proposed system dynamically selects appropriate its destinations. Thus, the system has an advantage of flexible network management in mobile network environments as well as dynamic change of traffic. Comparing its delay and throughput performance with the conventional SNMP based network management system, we find that the proposed mobile agent system performs better efficiency than the conventional one.

• ETRI Journal
• /
• 제28권4호
• /
• pp.513-516
• /
• 2006

Experimental demonstration of bipolar spectral encoding code-division multiple-access with modified pseudorandom noise codes is presented. Bipolar spectral encoding is achieved with an erbium-doped fiber amplifier amplified spontaneous emission source and arrayed waveguide gratings. The bit-error rate performance of 1.25 Gbps signal transmission over 80 km single mode fiber is measured in a multiple-user environment.

• 한국통신학회논문지
• /
• 제21권2호
• /
• pp.447-460
• /
• 1996

Network optimization and design procedures often separate quality of service (QOS) performance measures from reliability issues. This paper considers channel allocation and flow assignment (routing) in a network subject to link failures. Fault-tolerant channel allocation and flow assingments are determined which minimize network cost while maintaining QOS performance requirements. this approach is shown to yield significant network cost reductions compared to previous heuristic methods used in the design of packet switched network with unreliable links.

• 한국산업정보학회:학술대회논문집
• /
• 한국산업정보학회 2003년도 춘계학술대회
• /
• pp.52-56
• /
• 2003

In this paper, We evaluate the learning ability of non-monotonic DBM(Deterministic Boltzmann Machine) network through numerical simulations. The simulation results show that the proposed system has higher performance than monotonic DBM network model. Non-monotonic DBM network also show an interesting result that network itself adjusts the number of hidden layer neurons. DBM network can be realized with fewer components than other neural network models. These results enhance the utilization of non-monotonic neurons in the large scale integration of neuro-chips.