The increase in living alone and old households due to aging of the society has led to a decline in the quality of life (QOL) of elderly people. How to maintain and improve the quality of life for elderly people is a pressing issue of society. We watch over the elderly from both aspects of health-related QOL and subjective QOL and work on creating a social system that realizes a safe and secure life for the elderly. In this research, we promote the health information database which holds health related information in the shared database system such as the cloud by the will of the person him/herself, and develop a method for mining various kinds of life log information and medical/care related information in a hybrid manner. In this paper, we propose a remote watching method inferring resident’s behavior related to the operation of electric appliances from the measurement of total load current of the household. We also attempt to perform hybrid data mining by combining subjective mood data and objective data with wearable sensors that can collect biological data. We aim to provide a system that brings a safe and secure life to the elderly by carrying out assessment and early detection of abnormal symptoms of them such as depression and dementia.

In this research, we propose a sensor system for monitoring a house and residents. Residents’ various behavioral patterns different from the signal which shows a vibration characteristic of a house are measured by the prototyped MEMS acceleration sensors installed in the house. And the degree of danger of the behavior of the resident is recognized from the vibration data and it is possible to detect abnormality and issue necessary warnings. A preliminary experiment was performed in standard housing (wooden and steel frame construction) using prototyped MEMS acceleration sensors and the highly precise servo type speedometer as a reference measuring instrument and some of the experimental results are discussed in this report.

In this study, we conducted long-distance radio propagation experiments at 1mW/145 MHz to realize low-power long-distance communication for wildlife research and disaster prevention telemetry [1]. We successfully realized long-distance communications from a weather balloon (elevation: ~14,000 m) with 1 mW radio power. Furthermore, we demonstrated that our method can achieve such long-distance communication at 10 BPS/1mW with battery cell operation in the SSB mode and that the SNR in the SSB bandwidth is better than -10 dB.

A life log records man’s activity information as digital data. After these life logs are recorded, they are used in order to look back upon an individual’s life. There are some problems in the method of life log acquisition. The system needs to install many sensors in a house. In order to install sensors, much initial investment is needed and the user’s burden is heavy. Moreover, since the mental resistance is also strong to installing sensors in a house, introduction of the system takes time and effort. For the solution of these problems, we propose a method of acquiring and analyzing life log from Android based home electric appliances. In this study to develop an experimental system, the life log information which could be taken with Android home electric appliances is acquired by the simulated environment with home electric appliances with minimal sensors, and the life log is used and very rough life pattern is estimated. The reliability of the acquired log information is evaluated by getting subjects to live for one week in this system.

We conducted long distance radio propagation experiments at 1-10mW/145MHz, to realize a low-power long-distance communication for wildlife research and disaster prevention telemetry. In this paper, we describe that we succeeded in long distance communications, from a ferryboat to the top of Mt. Asugiyama (elevation: 501m, distance: 15km) in Kure, Hiroshima, Japan, in the verification test of our method using 3W radio power. We found out that our method has sufficient capability to achieve such a long distance communication at 10BPS/10mW in battery cell operation on a marine, if we use SSB mode and the SNR in SSB bandwidth is better than -10dB.

In this paper, we describe the construction of a long distance communication sensor network node for Arduino. We develop a Mad-SS Shield prototype system, which succeeds in about 5-km transmission with a 1 mW output.

In this paper, our research objective is to develop a database virtualization technique so that data analysts or other users who apply data mining methods to their jobs can use all ubiquitous databases in the Internet as if they were recognized as a single database, thereby helping to reduce their workloads such as data collection from the Internet databases and data cleansing works. In this study, firstly we examine XML scheme advantages and propose a database virtualization method by which such ubiquitous databases as relational databases, object-oriented databases, and XML databases are usable, as if they all behaved as a single database. Next, we show the method of virtualization of ubiquitous databases can describe ubiquitous database schema in a unified fashion using the XML schema. Moreover, it consists of a high-level concept of distributed database management of the same type and of different types, and also of a location transparency feature. Finally, we propose a database incompatibility trouble-recovery technique for use in a virtualized ubiquitous database use environment.

In recent years, the research of sensor networks has advanced. They are anticipated for use in a wide variety of fields such as traceability systems of products, environmental morning, health care, etc. However, developers must develop applications to work in a sensor node and applications to make the sensor data visible. Therefore the knowledge that is necessary for development and the acquisition of the language will take time to develop. As described in this paper, we propose a development environment for sensor network application development. The environgment will be useful even by web programmers and a designers.

The demand for embedded software is growing, and the number of development steps tends to increase with shorter period for development. It becomes inevitable to re-use existing software resources. However, if the configuration management of the existing software for reuse is not sufficient, it becomes hard to predict which modules in a system would be affected by the reused software modules. This may result in generating new bugs in the system. In this paper, we propose a method to accomplish the effective reuse of software modules by managing dependency information among modules. The method analyzes mutual relations among existing modules and visually presents not only inter-module dependency but also a list of reusable modules based on the developer's demand. We evaluated the method by applying it to some actual developments.

In recent years, business modeling has attracted attention in the development of information systems for analyzing and designing the working procedures of the system under development. The paper concentrates upon the use of business modeling in the process of analysis from the point of view of early identification of procedural bottlenecks. It goes on to propose and evaluate a method of configuring information systems that links business modeling with UML. The proposed method builds a business process model from the requirement specifications and translates the model into UML. The proposed method was applied to the prototyping of a bus booking and control system, comparing the results obtained using the proposed method with those using only UML for the configuration. The proposed method proved effective in the early detection and removal of bottlenecks, in coping with changes in the specifications, and in accurate modeling.

The research and development results of agent technology are awaited as the software solution with autonomy and social ability, which can provide the way to realize social activities on the Internet. However, the current status of the agent research is that many researches are for a specific application domain and are lacking in generality, and it is hard to apply the framework of the existing research to other fields. Moreover, it is thought that practical usage of agent is in immature state that there are few practical application examples and the system development methodology is not sufficient. In this research, the cooperative problem solving system by the multi-agent is first shown as an application development framework by the agents. Then, we present usage of templates of the agent model and the task model as the development support technique. And the evaluation according to an application construction experiment in a planning problem or the field of network community support is discussed.

It is our purpose to develop a novel fin type of microrobot in pipe that has the characteristics of flexibility, driven by wireless, good response and safety in body. This paper proposes a novel fin type of model microrobot utilizing electromagnetic actuator as the servo actuator to realize moving motion for biomedical application. We also discuss the structure, motion mechanism and characteristic evaluation of a driving fin. The moving fin driven by a permanent magnet can be controlled by the frequency adjustment of the alternate magnetic field. The running speed of the microrobot has been measured in vertical direction and at turning point by changing the frequency of input current. Experimental results showed that the maximum running speed 40mm/s could be obtained in vertical direction. The developed microrobot has characteristics of compact volume, quick response, can correspond to the dirt adhering to the inner wall, or long distance movement, and running speed control can be easily performed by frequency adjustment. It will be very useful for industrial application and for microsurgery.

Advances in information technology are changing the social environment, and have brought about the so-called information-oriented network society. One of the major issues in the network society is how to bring people together, that is how to form the communities that share similar interests. The formation of communities brings a great deal of benefits to the members such as social collaboration on the Internet, recommendations from people who are interested in similar topics, etc. However, acquiring information about communities and selecting appropriate communities to join is a burden for most of the people and the burden restricts their lives in the network society. We propose a model of community formation with multi-agents with consideration of the cognitive balance model of the user. Personal agents can help the users by gathering and exchanging information, and recommending or assisting the users in making choices. The proposed model is evaluated through experiments.

Information-oriented network society is advancing far beyond our speculation based on the rapid innovations of information and network technologies. Large amount of people become suffering from so-called information barrier When we consider people with informational handicaps, a large gap is opening up between those who can use computers effectively to process information and those who cannot. People interact in the network society through network communities and one of the major issues in the network society is how to bring people together, that is how to form the communities that share similar interests. Active usage of information with the help of personal agents that interact with each other, gather information, and assist the user seems inevitable. Although, users interact through explicit communities, it is more natural and with fewer barriers to form communities as implicit communities using user's profiles and then to introduce users to explicit communities. In this paper, we propose a multi-agent supporting system for implicit community formation and discuss methods to encourage effective information exchanges between community members and to improve the quality of their lives.

In this paper, we propose a practical method for solving the delivery scheduling problem with multiple agents and discuss its implementation. The method is based on the distributed cooperative problem solving. In the delivery scheduling system, the covered region is partitioned into multiple subregions and each subregion is assigned a subproblem solving agent which solves the problem of scheduling of drivers and trucks in the subregion. Integrating those subproblem solving agents, an integration-and-evaluation agent solves the total problem. By employing the distributed cooperative problem solving framework for the delivery scheduling problem, we achieved an easy incorporation of various evaluation parameters in the process of scheduling, efficient use and management of scheduling knowledge of various levels, and reduction of computer processing by division of the problem into subproblems.