Koichiro Yamashita

-Possibility of ESL-: a software centric system design for multicore SoC in the upstream phase

The embedded systems for which both hardware and software are rapidly advancing and expanding, there is a growing need to be able to comprehensively and quantitatively estimate system performance at an early stage in the design process, especially multi-core based SoC. But it can be difficult to estimate system performance of actual target by employing only simple estimation methods.

Design of stable wireless sensor network for slope monitoring

This paper presents implementation report for slope monitoring on Wireless Sensor Network (WSN) system by use of IEEE802.15.4 (ZigBee). The purpose of this research is the construction of WSN system which can be sensing and monitoring the deformation of the ground caused by heavy rain in mountain areas with radio communication devices and sensor devices. For the stable data acquisition, some electrical and environmental problems should be solved. In this paper, the techniques for dynamical transition of the communication mode depending on battery capacity, a method of protecting wireless nodes from the lighting, and a optimized design of antenna for WSN are proposed. The experimental results which the proposed system installed at actual test field show that the dynamic wireless network configuration. Moreover, the status of field can be predicted by the obtained data from the improved system.

Relay Node Position Optimization in Complex Environment

Environment has an essential effect on the selection of relay node positions. Real application environment of Wireless Sensor Networks usually is complex and heterogeneous where wireless channel has different characteristics at different locations. In network deployment, the environment also constrains relay node locations. There are areas unfeasible for node deployment. This paper studies the relay node placement problem in such complex environment. Our target of relay node position optimization is to minimize the relay node number and improve network performance by optimizing route paths of sensor nodes. Heuristic operations for genetic algorithms are proposed to optimize relay node positions. Simulation results show that the relay nodes number of the proposed algorithm is less than that of 2-approximation Steiner tree algorithm and the route paths of selected relay nodes have almost the best cost among route paths of all feasible relay nodes.

Architecture Design for the Environmental Monitoring System over the Winter Season

One of the applications as a source of big data, there is a sensor network for--the environmental monitoring that is designed to detect the deterioration of the infrastructure, erosion control and so on. The specific targets are bridges, buildings, slopes and embankments due to the natural disasters or aging. Basic requirement of this monitoring system is to collect data over a long period of time from a large number of nodes that installed in a wide area. However, in order to apply a wireless sensor network (WSN), using wireless communication and energy harvesting, there are not many cases in the actual monitoring system design. Because of the system must satisfy various conditions; measurement location and time specified by the civil engineering; communication quality and topology obtained from the network technology; the electrical engineering to solve the balance of weather environment and power consumption that depends on the above-mentioned conditions. We propose the whole WSN design methodology especially for the electrical architecture that is affected by the network behavior and the environmental disturbance. It is characterized by determining recursively mutual trade-off of a wireless simulation and a power architecture simulation of the node devices. Furthermore, the system allows the redundancy of the design. In addition, we deployed the actual slope monitoring WSN that is designed by the proposed method to the snow-covered area. A conventional similar monitoring WSN, with 7 Ah Li-battery, it worked only 129 days in a mild climate area. On the other hand, our proposed system, deployed in the heavy snow area has been working more than 6 months (still working) with 3.2 Ah batteries. Finally, it made a contribution to the civil engineering succeeded in the real time observation of the groundwater level displacement at the time of melting snow in the spring season.

Environmental Data Recovery using Polynomial Regression for Large-scale Wireless Sensor Networks

In the near feature, large-scale wireless sensor networks will play an important role in our lives by monitoring our environment with large numbers of sensors. However, data loss owing to data collision between the sensor nodes and electromagnetic noise need to be addressed. As the interval of aggregate data is not fixed, digital signal processing is not possible and noise degrades the data accuracy. To overcome these problems, we have researched an environmental data recovery technique using polynomial regression based on the correlations among environmental data. The reliability of the recovered data is discussed in the time, space and frequency domains. The relation between the accuracy of the recovered characteristics and the polynomial regression order is clarified. The effects of noise, data loss and number of sensor nodes are quantified. Clearly, polynomial regression offers the advantage of low-pass filtering and enhances the signal-to-noise ratio of the environmental data. Furthermore, the polynomial regression can recover arbitrary environmental characteristics.

A Deployment Algorithm for Multi-hop Wireless Networks

Multi-hop wireless networks have a wide range of applications. In the practice of real applications, network deployment is a challenging task. Because deployment environment is complex, it is difficult to select suitable positions for node deployment, and the performance of deployed network cannot be guaranteed. This paper proposes a network deployment algorithm which aims to deploy a real network which has the same performance with the designed network. A deployment quality is defined to indicate whether a position is suitable for node deployment. When a position is unsuitable to deploy nodes, the algorithm gives information of a better deployment position. A simulation method based on network simulators is used to verify the proposed algorithm. According to our simulation results, the deployed network generated by the proposed algorithm can keep the performance of designed network including disconnected node number, connectivity, packet delivery ratio, and transmission delay.