Eiji Oki

Knapsack based multicast traffic grooming for optical networks

Abstract This paper proposes a light-tree based heuristic algorithm, called 0/1 knapsack based multicast traffic grooming, in order to minimize the network cost by reducing the number of higher layer electronic and optical devices, such as transmitters, receivers, and splitters, and used wavelengths in the network. The proposed algorithm constructs light-trees or sub light-trees, which satisfy sub bandwidth demands of all multicast requests. We present a light-tree based integer linear programming (ILP) formulation to minimize the network cost. We solve the ILP problem for sample four-node and six-node networks and compare the ILP results with the proposed heuristic algorithm. We observe that the performance of the proposed algorithm is comparable to the ILP in terms of cost. When the introduced ILP is not tractable for large network, the proposed algorithm still able to find the results. Furthermore, we compare the proposed heuristic algorithm to existing heuristic algorithms for different backbone networks. Numerical results indicate that the proposed heuristic algorithm outperforms the conventional algorithms in terms of cost and resource utilization.

Self-adjustable domain adaptation in personalized ECG monitoring integrated with IR-UWB radar

Abstract To enhance electrocardiogram (ECG) monitoring systems in personalized detections, deep neural networks (DNNs) are applied to overcome individual differences by periodical retraining. As introduced previously [4] , DNNs relieve individual differences by fusing ECG with impulse radio ultra-wide band (IR-UWB) radar. However, such DNN-based ECG monitoring system tends to overfit into personal small datasets and is difficult to generalize to newly collected unlabeled data. This paper proposes a self-adjustable domain adaptation (SADA) strategy to prevent from overfitting and exploit unlabeled data. Firstly, this paper enlarges the database of ECG and radar data with actual records acquired from 28 testers and expanded by the data augmentation. Secondly, to utilize unlabeled data, SADA combines self organizing maps with the transfer learning in predicting labels. Thirdly, SADA integrates the one-class classification with domain adaptation algorithms to reduce overfitting. Based on our enlarged database and standard databases, a large dataset of 73200 records and a small one of 1849 records are built up to verify our proposal. Results show SADAs effectiveness in predicting labels and increments in the sensitivity of DNNs by 14.4% compared with existing domain adaptation algorithms.

Designing a Hadoop system based on computational resources and network delay for wide area networks

This paper proposes a Hadoop system that considers both slave server’s processing capacity and network delay for wide area networks to reduce the job processing time. The task allocation scheme in the proposed Hadoop system divides each individual job into multiple tasks using suitable splitting ratios and then allocates the tasks to different slaves according to the computational capability of each server and the availability of network resources. We incorporate software-defined networking to the proposed Hadoop system to manage path computation elements and network resources. The performance of proposed Hadoop system is experimentally evaluated with fourteen machines located in the different parts of the globe using a scale-out approach. A scale-out experiment using the proposed and conventional Hadoop systems is conducted by executing both single job and multiple jobs. The practical testbed and simulation results indicate that the proposed Hadoop system is effective compared to the conventional Hadoop system in terms of processing time.

A routing and wavelength assignment scheme considering full optical carrier replication in multi-carrier-distributed optical mesh networks with wavelength reuse

Abstract This paper proposes a routing and wavelength assignment (RWA) scheme that considers full optical carrier replication to minimize the number of required wavelengths for wavelength-reusable multi-carrier-distributed (WRMD) mesh networks. Unlike the conventional wavelength division multiplexing networks where each node contains multiple laser diodes, the WRMD networks use a light source, called the multi-carrier light source, to ease the difficulty of controlling many light source devices. The optical carrier replication at any node of optical carrier or any end node of lightpath, called full optical carrier replication, improves the performance of the conventional schemes, which do not consider full optical carrier replication. This paper first formulates the RWA problem considering full optical carrier replication as an integer linear programming problem that minimizes the number of required wavelengths to satisfy the given lightpath setup requests. A heuristic RWA scheme is then proposed to solve the RWA problem in practical times. Simulation results show that our proposed heuristic RWA scheme for the WRMD network achieves a near-optimum number of wavelengths. In addition, it is able to reduce the number of required wavelengths for lightpath establishment, compared to the conventional scheme.

Recent research progress on spectrum management approaches in software-defined elastic optical networks

Abstract The flex-grid technology or elastic optical network (EON) is accepted to be a propitious solution for the future transport network due to its unique properties, which are bandwidth aggregation that creates super wavelength channels, different data rate accommodation, bandwidth segmentation that creates sub-wavelength channels, and elastic variation of assigned bandwidth. The software-defined network (SDN) is incorporated with the emerging technology of EON for enhancing its performance by enabling dynamic provisioning and releasing of optical lightpaths. This reduces the required time for lightpath control, which leads to improve the network performances. This paper exploits the software-defined elastic optical network (SD-EON) by identifying its architectures, existing literature, resource management approaches, and future challenges. In addition, the performance of EONs related to the SDN technology is compared in terms of blocking probability. Numerical results indicate that the SD-EON accommodates 25% more admissible traffic than the EON without considering SDN, when the acceptable blocking probability is considered 0.01.

Application-triggered automatic distributed cloud/network resource coordination by optically networked inter/intra data center [invited]

Everything is being connected to the Cloud and Internet of Things, and network robots with big data analy- sis are creating important applications and services. The cloud network architecture is moving towards mega-cloud data centers (DCs) provided by companies such as Amazon and Google in combination with distributed small DCs or edge computers. While the traditional restrictions im- posed by distance and bandwidth are being overcome by the development of advanced optical interconnection, modern applications impose more complex performance and quality of service requirements in terms of processing power, response time, and data amount. The rise in cloud perfor- mance must be matched by improvements in network per- formance. Therefore, we propose an application-triggered cloud network architecture based on huge-bandwidth optical interconnections. This paper addresses edge/center cloud and edge/edge integration with the use of virtual ma- chine migration. In addition, to reduce energy consumption, an application-triggered intra-DC architecture is described. Using the proposed architectures and technologies can real- ize energy-efficient and high-performance cloud service.

Expected capacity guaranteed routing method based on failure probability of links

In a high-speed backbone network, the failure of a network link may cause large data losses, so it is necessary to reserve spare network resources for faster recovery. The conventional protection methods to reserve backup routes do not consider the failure probability of each network link, so the same amount of network resources for the backup route are needed regardless of the failure probability of network links. This leads a decrease in the number of connections that can be accepted into the network. This paper proposes a routing and capacity allocation method that guarantees the expected value of allocated capacity. We formulate a mixed integer liner programming model for the proposed method. We conduct simulations to study the advantage of the expected capacity guaranteed routing over the conventional routing method in terms of bandwidth blocking probability. The results show that the proposed method reduces the bandwidth blocking probability to about 1/3 as compared to that of the conventional path protection method.

Performance of elastic optical network with allowable spectrum conversion at intermediate switches

This paper evaluates the performances of elastic optical networks (EONs) in terms of request blocking probability with and without incorporation of spectrum conversions at intermediate switches. In EONs, dynamically setting up and tearing down of lightpath connections generate spectrum fragmentation problems. The spectrum fragmentation problem occurs when the available spectrum slots are isolated from each other as neither they are aligned along the routing path nor they are contiguous in the spectrum domain. The spectrum conversion at intermediate switches is performed for blocked requests without triggering any defragmentation process. Simulation results indicate that the incorporation of spectrum conversions significantly suppresses the request blocking probability and increases the traffic admissibility when the traffic load is low. In a congested network under heavy traffic load, the performance is comparable with and without incorporation of spectrum conversion.

Performance evaluation of first-last-exact fit spectrum allocation policy for elastic optical networks

This paper evaluates the performance of the first-last-exact fit spectrum allocation policy for elastic optical networks. The first-last-exact fit spectrum allocation policy is intended to increase the number of aligned available slots and avoid small contiguous available slots, and hence suppresses the blocking probability. This policy separates the allocation of disjoint and non-disjoint connections. Connections with disjoint paths are allocated using the first-exact fit allocation policy, whereas connections with non-disjoint paths are allocated using the last-exact fit allocation policy. The separation of the disjoint and non-disjoint connections provides a higher number of aligned available slots. Simulation results show that the first-last-exact fit spectrum allocation policy improves the traffic admissibility compared to the conventional first fit and first-last fit policies.

Estimating the effect of Wavelength Selective Switch latency on optical flow switching performance

Optical networks are well suited to support massive data exchanges between data centers. Elephant traffic flows can be routed over provisioned and dedicated lightpaths (optical flows) while other (mice) flows, which are routed by electronic switches, are unaffected. In some solutions, Wavelength-Selective Switches (WSSs) are employed in the optical nodes to individually route the lightpath towards its destination. WSSs take time to be switched and delay the lightpath setup time. In this paper, the authors compare three service policies for WSS devices aiming to reduce the lightpath setup and tear down times. The conventional service policy assumes that each setup (tear down) request is handled individually. Two other service policies assume that groups of setup (tear down) requests are handled together by the WSS. These policies are implemented in a discrete event simulator, which is used to estimate the end-to-end lightpath setup and tear down time across an arbitrary mesh network. Simulation results show that group service policies outperform the conventional policy at high loads. The grouping policies are useful to reduce the lightpath setup time especially in the presence of lightpaths that are frequently set up and have relatively short holding time (short duration of elephant-optical flows).