論文

The term industrial edge computing is used to describe a distributed platform that integrates communication, computation, and storage resources for performing real-time applications that can be directly accessed from the cloud. A step toward the industrial Internet revolution, industrial edge computing is designed to facilitate agile connectivity, real-time control, and data optimization, while enabling intelligent applications, ensuring tight security, and protecting privacy. Industrial edge computing makes use of what is known as edge computing nodes (ECNs), which bridge the gap between the physical world and the digital world by acting as smart gateways for assets, services, and systems. The IEEE P2805 Standards are being developed for defining protocols for self-management, data acquisition, and machine learning through cloud-edge collaboration on ECNs.

The spread of Internet of Things (IoT) devices and high-speed communications, such as 5G, makes their services rich and diverse. Therefore, it is desirable to perform functions of rich services transparently and use edge computing environments flexibly at intermediate locations on the Internet, from the perspective of a network system. When this type of edge computing environment is achieved, IoT nodes as end devices of the Internet can fully utilize edge computing systems and cloud systems without any change, such as switching destination IP addresses between them, along with protocol maintenance for the switching. However, when the data transfer in the communication is encrypted, a decryption method is necessary at the edge, to realize these transparent edge services. In this study, a transparent common key-exchanging method with cloud service has been proposed as the destination node of a communication pair, to transparently decrypt a secure sockets layer-encrypted communication stream at the edge area. This enables end devices to be free from any changes and updates to communicate with the destination node.

Due to liberalization processes initiated on Japan's power market since 2000, a significant higher congestion on the power grid is expected entailing a higher vulnerability to shortages. With increasing aspirations on power supply security, an interaction with the demand side has become inevitable to reduce the congestion without the need for new generation or transmission capacities. Attaining efficiency close to the optimum in curtailing and rescheduling loads, the fast automatic processing of essential information affecting the performance of the power grid is crucial. This approach shows a solution for hybrid demand control integrating a vast number of small and medium sized enterprises for centralized load management based on locally collected environmental data. The hardware and software of this system has been prototyped and evaluated. Evaluation tests show the capability of this system to deliver the necessary infrastructure necessary for demand control. © 2006 IEEE.

In this paper, lossy compression based on Integer DCT (Discrete Cosine Transform) for haptic data is presented and implemented to an FPGA (Field Programmable Gate Array). The lossy data compression accomplishes to reduce data size of the haptic data. In the lossy data compression, DCT-based techniques are often used. The Integer DCT (IntDCT) achieves the operation of the DCT with the format of integer. The compatibility of the IntDCT with the FPGA is therefore high. The lossy compression method is applied to a bilateral control system and the efficacy is confirmed by experiments. © 2010 IEEE.

FPGA has been focused on as a controller of robot systems. Haptic robot system based on bilateral control system is one of the significant applications using the FPGA. In this paper, the bilateral control system including mode transformation is implemented on the FPGA. The bilateral control system is based on acceleration control system. The mode transformation is a transformation between robot mode space and operational mode space. Furthermore, multirate sampling method is introduced to consider hardware limitation. Some experimental results are shown in order to verify the validity of the implementation. © 2008 IEEE.

Bilateral teleoperation has been needed in many scenes such as space activities or undersea, work on a atomic power plant and minimally invasive surgery, etc. In bilateral teleoperation, transfer of keen tactile sense as if the operator directly touch the environment without using the robot is required. Wide frequency bandwidth of force sensing is required to achieve transfer of a keen tactile sensation. Bilateral controller based on acceleration control with disturbance observer is available to achieve transfer of a keen tactile sensation. The frequency bandwidth of force sensing is determined by cut-off frequency of low-pass filter used in disturbance observer. The cut-off frequency is in inverse proportion to a sampling period. Therefore, shortening of the sampling period improves performance of bilateral teleoperation. In this paper, bilateral teleoperation system by using FPGA(Field Programmable Gate Array) is introduced. FPGA is a LSI that user can design its internal logic. When motion controller is implemented on FPGA, it operates faster than that implemented in a PC(Personal computer) with RTOS(Real Time Operating System). The sampling period is shortened from 100/its to 10μs by using FPGA. © 2006 IEEE.

© 2016 IEEE. A typical service provided by a Smart City is a Home Energy Management System (HEMS), an automated system that uses sensor networks to collect usage of home appliances and devices such that the collected data are accessible by the local government. For example, Smart meters, installed within individual homes, can send measures of electricity consumption for analysis in the context of the global community. Such data can be used by a forecasting tool to avoid electrical outage and improve the overall management of electricity usage by the community. However, the data collected may be protected by individual privacy laws and it is possible that the network may be violated by intruders to gain access to the lifestyle of individuals within their home. For example, an intruder may be able to identify active appliances and the number of individuals present within the residence. There is a requirement to protect the data emanating from these sensors to preserve this individual privacy information. Thus, a data concierge service capable of analyzing individual HEMS data is required; such a service would improve the lifestyle by allowing access to HEMS data by a third-party analyst. Although such a service may not require raw data, one must evaluate the trade-off between the preservation of privacy information and the effectiveness of the impact of the data services on the entire community. In this paper, we implemented a prototype application in conjunction with two noise-based perturbations. The application recommends improvements to household electricity usage based on its measured electricity consumption collected by smart meters. The perturbations provide a method to support Privacy Preserving Data Mining (PPDM) techniques. Evaluation of the performance results of the prototype applications illustrates the influence of the PPDM in improving the accuracy of said application.

Worldwide the demand on a higher level of automatic control of power systems rises to comply with system objectives of reliability, efficiency, cost minimization as well as emission requirements. With the transition of the power systems towards a more volatile on-site generation based on renewable resources, a higher degree of flexibility in energy consumption for reduced system congestion is evident. This paper presents a real-time information system for active control of the demand side in Japanese power grids. The proposed system allows local data processing of environmental and consumption data to schedule and curtail loads of shops, small and medium size enterprises. The concept comprises specialized control hardware for end-use supervision, local and internet based server architecture and control methods for global energy management. © 2006 IEEE.

Data integration of different infrastructure is fundamental to smart communities. In this chapter, case studies of data-oriented smart community projects, such as energy reduction and efficient-use services, recommendation services, mobility services, and health services, are described.

KNIVES is proposed as Internet based demand and supply control system and controls power by using distributed and cooperative power control algorithm. However, if KNIVES aims to maintain indoor environmental amenity, installed locations and the numbers of environmental sensors are critical issues. Wireless sensor network is useful to measure environmental data of various points where a wired sensor is not available or cost of wired installation is expensive, and makes it easy to install or to relocate the sensors. In this paper, we implement ZigBee wireless sensor network and make operational test to discuss about the application of demand control method by using wireless sensor network. Furthermore, we propose some demand control application with ZigBee, customized environment-based control and head-count-based control, to make DSM more efficient. ©2008 IEEE.

KNIVES (Keio University Network oriented Intelligent and Versatile Energy saving System) is an IT based system, designed for Japanese power grids, that supports efficient control and balanced power supply. As KNIVES focuses on the integration of a vast numbers of end-users, it provides the necessary scalability, reliability, and dynamic adaptability. The latest generation of KNIVES system (V2) provides a new hardware concept for a local control that reflects environmental affinity. KNIVES terminals use Internet communication and supports ZigBee based environment sensors. The implementations of this system and practical field tests are described in this paper. ©2007 IEEE.

Structural Health Monitoring (SHM) is a key technology for determining the health condition of structures, such as buildings and bridges. SHM is widely used especially in gigantic structures due to the requirements of safety. However, conventional sensor network system for SHM did not take account of an accidental collapse. As a result, there is a possibility that a sensor network node, a network link, and also a center server cause failures in processing or storing of data gathered system. These failures may lose important data that contains useful information for a post-analyzing of the collapse. To solve this problem, a lightweight data duplication system for SHM that prevents the data failures or loss is proposed. Proposed system consists of three functions; node search, backup node selection and backup data transfer function. In backup node selection, round-trip delay and a size of free memory space of backup nodes are used.

© 2016 ACM. It will become a serious problem to increase power consumption of routers resulting from explosive increase of network traffics caused by IoT data, big data, and so on. Table lookups in packet processing are known as a bottleneck of the router from the points of both processing performance and power consumption. Packet Processing Cache (PPC), which accelerates the table lookups and reduces the power consumption of them by using cache mechanism, was proposed. However, it is difficult for PPC to obtain high cache hit rate because the size of PPC should be small, such as a L1 cache of processors, to get higher access speed. For this reason, an effective line replacement algorithm was considered in this study for reducing a cache miss without increasing the cache size. First, defects of applying typical line replacement algorithms to PPC were examined. Secondly, two algorithms, LRU Insertion Policy (LIP) and Elevator Cache (ELC), and improved algorithms of LIP and ELC called LIP1, LIP2, ELC1, and ELC2 were considered for improving the above defects. In simulation, it was shown Elevator Cache could reduce the cache miss by at most 17.4% compared with Least Recently Used, which applied to many cache systems.

In recent years, instead of wired communication, wireless communication technology has been utilized in many applications. Particularly, wireless sensor networks can be utilized as a communication medium of multiple mobile robots for environmental monitoring, security service, space exploration etc. In this paper, to accomplish reliable packet routing in mobile ad-hoc network, a novel QoS sensitive routing algorithm is discussed. Wireless network topology could be changing dynamically by robot movement, radio disturbance or fading in continuously shifting situation. Therefore, a packet routing algorithm should guarantee QoS by using some quality indicators of a wireless network. In proposed routing algorithm, routing table is constructed by using Received Signal Strength Indicator(RSSI) and the table information is periodically flooded in the network. Novel hash based routing entry management by using extended-MAC address can eliminate the overhead of message flooding. An analysis of the collision of hash values contributes to determine the length of the hash values, which is minimum required. Simulations are conducted to evaluate the effectiveness of proposed algorithm and to validate the theory in general wireless network routing algorithm.

The next-generation backbone routers need large-bandwidth communication capability because of increased future Voice over Internet Protocol (VoIP) services. Providing these or other future ubiquitous services will require new Network Processor (NP) designs, since these protocols or services are expected to require enormous increases in processing power in order to address their fine-grain communication pattern. These improvements of processing power will not be enough for future NP designs due to the requirements of cost reduction and lower power consumption. To meet these requirements, the cache mechanism has received increasing attention in NP design. Cache can reduce the load and time needed for table lookup by taking advantage of temporal localities in network traffic. For these reasons P-Gear, an advanced cache-based network processor, has been proposed by our laboratory. P-Gear, which provides a sophisticated cache architecture, can be used not only for table lookup but also for determining processing results, and achieves high-speed packet processing. In this paper, the cache architecture of P-Gear is shown and an extended cache mechanism called Multi-Context-Aware cache is proposed. Multi-Context-Aware cache can improve the speed of packet processing because it can control cache entry by analyzing the packet header and payload. To evaluate the validity of this proposed method, simulators of SIP (Session Initiation Protocol)-Aware cache, IKE (Internet Key Exchange)-Aware cache and FTP (File Transfer Protocol)-Aware cache were designed as subsets of Multi-Context-Aware cache. According to the simulation results, SIP-Aware, IKE-Aware and FTP-Aware cache can all make contributions to improving the cache hit rate by managing cache entry. In addition, they all can reduce processing unit requirements.

This paper introduces an implementation of a standard-based sensor network for the Internet of Things (IoT) applications adhering to the Institute of Electrical and Electronics Engineers (IEEE) P1451.0 and P1451.1.6 draft standards. The architecture of standards-based sensor networks consists of P1451.0 and P1451.1.6-based wide-area networks (WAN), P1451.0, and P1451.5.X-based wireless local area networks (WLANs), and P1451.0 and P1451.2-based wired local-area networks (LANs). A few case studies P1451.0 and P1451.1.6 WAN are provided with preliminary results to verify some network service specifications of the sensor network based on P1451.0 network services and P1451.1.6 MQTT interfaces in order to verify and improve draft specifications helping to achieve interoperability.

In a smart community, which is an initiative to create an efficient regional society by leveraging information and communications technology (ICT), various data exchanges and data processing are performed. Owing to the demand for network transparency in data processing, packet rewriting is required at network edges. In the packet rewriting process, packets need to be analyzed after obtaining the necessary permissions. A conventional method for packet analysis is using a 5-tuple. While this method can analyze packets by a service user, it cannot analyze packets by a service. Moreover, service provision needs to take place at the appropriate locations with respect to the privacy level, and service provision locations need to be flexible enough to change even subsequent to the commencement of services. To address this issue, there is a need to enable positional flexibility in determining the location for service provision between the data source and cloud. In this study, assuming that the permission to analyze packets had been obtained within the smart community, we implemented a framework that could analyze not only 5-tuple but also layer-7 (L7) information and rewrite packet contents based on the analysis results. The process performed at the edge first analyzed the header of the incoming packet to determine whether the packet was subject to L7 information analysis. If the packet was subject to analysis, the L7 information was analyzed to determine whether the packet was subject to rewriting. A packet subject to rewriting were rewritten and forwarded accordingly. For this series of packet process, we used the Data Plane Development Kit (DPDK), a tool that could speed up packet processing. The framework was applied to the anonymization process, and the delay due to the anonymization process was 0.195 ms. With this framework, it is possible to achieve the packet content rewriting process at any point between the IoT device and the cloud by modifying the location of the edge node.

The network infrastructure in a smart community uses the cloud and compute servers, and the edge and fog areas to perform processing in the optimal place. Nodes in the edge and fog areas have various processing capacities, and there must be load-balancing mechanisms to meet the requirements of applications. However, Internet of Things (IoT) devices are generally small, computationally constrained, and numerous, so reconfiguration and functionality updates for the network must be kept to a minimum. In addition, if nodes establish connections with other nodes to exchange their information, the number of connections will be large, and communication costs will be high. In this study, we use nodes called 'Distributor' that can monitor data flow through the network without changing the existing infrastructure. Piggyback technique is used to add information onto the packets flowing in the network. Distributors add the information to the packets and exchange information to each other without establishing connections with other nodes, and they perform load balancing based on the information. By changing and migrating the processing locations among Distributors, IoT devices can continue to communicate with other nodes even during load balancing. Experiments confirmed that load balancing could be performed to prepare for sudden load increases and that the impact of piggybacking on occupied bandwidth and processing delay was 0.177% and 0.06%, respectively. We also confirmed that sensor data can be sent without reconnecting communications, even during load balancing.

Smart communities use myriad IoT devices and the data they transmit to perform global optimization. In network-transparent service provision, data transmitted from IoT devices are analyzed using network devices such as switches and routers in the edge region to add value to the data without modifying the IoT devices. However, to establish a mechanism for providing such network transparency services while complying with the opt-in or opt-out of data provided by users, data distribution management is necessary to determine the stored location and path of the target data. In addition, because a smart community environment is assumed to be a heterogeneous computing-resource environment, it is necessary to envision a dynamic relocation function for transparent network services considering the usage of computing resources. In this study, we proposed a method for managing and migrating transparent network services deployed on edge nodes using containers and linked it to an existing information distribution management mechanism that enables data distribution management at the service level. In the experiments, we confirmed that dynamic placement was possible under a load. We also measured the response delay associated with the provision of a transparent network service and confirmed that the associated processing delay was approximately 10 ms.

Bilateral control systems are strongly required to apply to endoscopic surgeries. It is necessary that the system has enough degrees of freedom for applying various operation procedures including endoscopic surgery. When the degrees of freedom of the system are increased, the amount of control calculation is also increased, and it is hard to keep sampling periods short. The bilateral control systems, however, require comparatively shorter sampling periods, particularly the control system based on an acceleration control system. Hence, it is a difficult issue to increase the degrees of freedom of the bilateral control system. In this paper, the sampling period is kept short in the multi-degree of freedom system by using field programmable gate arrays as processors. The bilateral control system based on the acceleration control system is implemented in a robot system, which has 12 DOF, and some experimental results are shown, and the errors between the response of the master and slave robots are discussed. © 2009 IEEE.