Misao Kataoka

Bio: Misao Kataoka is an academic researcher from Nippon Telegraph and Telephone. The author has contributed to research in topics: Cloud computing & Software. The author has an hindex of 5, co-authored 21 publications receiving 84 citations.

Automatic GPU Offloading Technology for Open IoT Environment

Abstract: Recently, Internet of Things (IoT) technologies have been progressed. To overcome of the high cost of developing IoT services by vertically integrating devices and services, open IoT enables various IoT services to be developed by integrating horizontally separated devices and services. For open IoT, we have proposed Tacit Computing technology to discover the devices that have data users need on demand and use them dynamically. However, existing Tacit Computing does not consider performance. Therefore, in this paper, we propose an automatic graphics processing unit (GPU) offloading technology as a new elementary technology of Tacit Computing that uses genetic algorithm to extract appropriate offloading areas from parallelizable loop statements automatically. This can improve performance of IoT applications. We evaluate our proposed GPU offloading technology by applying it to five C/C++ applications of image processing, matrix manipulation, and so on to verify its effectiveness and find that it can process them more than ten times as quickly as only using central processing units within 1 h tuning time.

Device Identification Based on Communication Analysis for the Internet of Things

Abstract: As the Internet of Things (IoT) is rapidly expanding, a huge variety of devices is being connected to the Internet. Device management is becoming an important topic for IoT. Especially for using devices properly and securely, it is necessary to visualize what types of devices are in the network. However, most conventional device identification methods are not suitable for resource-constrained IoT devices. Therefore, we have developed a method of device identification that identifies the type and model of devices on the basis of general communication information. It determines the type and model of devices by calculating the similarity of features extracted from their network packets. The great merit of our proposed device identifier is that it can be applied to various IoT devices without special equipment. We conducted three experiments to evaluate its effectiveness. In the experiments, we focused on devices specialized for specific functions such as network cameras and factory-used devices because they are effective targets of our device identifier. In addition, we tried identifying models from devices of the same type. The first experiment revealed the relationship between the packet header information used for identification and the success of identification. The second experiment with 11 types of network cameras showed that the device identifier correctly identified nine of them. In addition, the third experiment in a simulated factory environment showed the device identifier correctly identified six types of factory-used devices. Thus, we have demonstrated the feasibility of the proposed device identifier in a real environment.

Proposal of Parallel Processing Area Extraction and Data Transfer Number Reduction for Automatic GPU Offloading of IoT Applications

Abstract: Recently, IoT (Internet of Things) technologies have been progressed. To overcome of the high cost of developing IoT services by vertically integrating devices and services, Open IoT enables various IoT services to be developed by integrating horizontally separated devices and services. For Open IoT, we have proposed Tacit Computing technology to discover the devices that have data users need on demand and use them dynamically and an automatic GPU (graphics processing unit) offloading technology as an elementary technology of Tacit Computing. However, it can improve limited applications because it only optimizes parallelizable loop statements extraction. Therefore, in this paper, to improve performances of more applications automatically, we propose an improved method with reduction of data transfer between CPU and GPU. This can improve performance of many IoT applications. We evaluate our proposed GPU offloading method by applying it to Darknet which is general large application for CPU and find that it can process it 3 times as quickly as only using CPUs within 10 h tuning time.

Distributed Search Architecture for Object Tracking in the Internet of Things

Abstract: Internet of Things (IoT) is rapidly expanding, which will enable many devices to be installed in various environments. However, current IoT services cannot maximally utilize devices because of their silo model. To solve this problem, we aim to realize Open IoT, in which services share devices. In this paper, we propose an architecture for an object tracking service, one of the main services of Open IoT. The architecture uses video data from shared devices, such as surveillance cameras or pedestrians’ smartphones. An important research task is to discover the most appropriate devices for a service out of a huge number of devices connected to the Internet. We named real-time data generated by devices “live data” and are trying to use these data to discover appropriate devices. However, it is difficult to collect and handle all live data in the cloud because of the network band limit. Therefore, we propose a distributed search architecture. Generally, distributed architecture uses network and computing resources less efficiently than cloud architecture. Our proposed architecture overcomes this by deploying a search function dynamically and copes with arbitrary searches. We developed a system that embodies our proposed architecture and evaluated its feasibility. An experiment simulating a moving object tracking service with network cameras is shown that the architecture reduces the communication bandwidth of the core network to 1000th or less of that when cloud computing is used. In addition, another experiment is shown that the architecture search speed is sufficient for a walking-person tracking service.

Autonomous device identification architecture for Internet of Things

Abstract: A wide variety of devices is being installed in various environments such as homes, factories, and streets with the rapid expansion of the Internet of Things (IoT). To properly and securely use IoT devices, the states of various devices that have different properties and protocols must be managed. However, it is difficult to understand the consistency of an individual device when its installation place or software changes, since many individual IoT devices do not have unique identifiers. In this paper, we propose an architecture that estimates individual devices by analyzing and combining multiple pieces of information that can be obtained from the device. This architecture estimates individual device identity based on the time change pattern of the feature amount extracted from a signal transmitted by a device. Results of a simulation revealed that this architecture could identify the individual devices to find the correct device.

A Survey of Healthcare Internet of Things (HIoT): A Clinical Perspective

Abstract: In combination with current sociological trends, the maturing development of Internet of Things devices is projected to revolutionize healthcare. A network of body-worn sensors, each with a unique ID, can collect health data, that is, orders-of-magnitude richer than what is available today from sporadic observations in clinical/hospital environments. When databased, analyzed, and compared against information from other individuals using data analytics, Healthcare Internet of Things data enables the personalization and modernization of care with radical improvements in outcomes and reductions in cost. In this article, we survey the existing and emerging technologies that can enable this vision for the future of healthcare, particularly, in the clinical practice of healthcare. Three main technology areas underlie the development of this field: 1) sensing, where there is an increased drive for miniaturization and power efficiency; 2) communications, where the enabling factors are ubiquitous connectivity, standardized protocols, and the wide availability of cloud infrastructure; and 3) data analytics and inference, where the availability of large amounts of data and computational resources is revolutionizing algorithms for individualizing inference and actions in health management. Throughout this article, we use a case study to concretely illustrate the impact of these trends. We conclude this article with a discussion of the emerging directions, open issues, and challenges.

A review on machine learning–based approaches for Internet traffic classification

Abstract: Traffic classification acquired the interest of the Internet community early on Different approaches have been proposed to classify Internet traffic to manage both security and Quality of Service (QoS) However, traditional classification approaches consisting of modifying the Transmission Control Protocol/Internet Protocol (TCP/IP) scheme have not been adopted due to their complex management In addition, port-based methods and deep packet inspection have limitations in dealing with new traffic characteristics (eg, dynamic port allocation, tunneling, encryption) Conversely, machine learning (ML) solutions effectively classify traffic down to the device type and specific user action Another research direction aims to anonymize Internet traffic and thwart classification to maintain user privacy Existing traffic surveys focus on classification and do not consider anonymization Here, we review the Internet traffic classification and obfuscation techniques, largely considering the ML-based solutions In addition, this paper presents a comprehensive review of various data representation methods, and the different objectives of Internet traffic classification Finally, we present the key findings, limitations, and recommendations for future research