Satoru Aikawa

With the recent expansion of electromagnetic (EM) waves, undesired EM waves generated by device can cause malfunctions of other electronic equipment and communication failure. EM wave absorbers are widely used as countermeasures against these problems. Therefore, EMC (Electromagnetic Compatibility) technology is reguired to improve the electromagnetic environment. In this study, a microwave metamaterial EM wave absorbers using a sheet with a periodic array of square metal patterns were designed and its reflection characteristics at vertical incidence were experimentally evaluated.

This contribution focuses on the accuracy and measurement cost of indoor Area Estimation using wireless LAN. The system estimates the area, such as a room or a shopping store, where the user is located. The user-measured AP information (UD), and the pre-measured AP information (DB) are used in the FP method. In the previous study, RSSI probability distributions were obtained from the measured DB and UD. The area with the largest overlap in probability distribution was selected as the estimation result. In this study, we propose Area Estimation by CNN using data measured while walking at various locations within an area. The method using CNN improved the estimation accuracy compared to conventional Area Estimation. Moreover, the method measuring AP information while walking reduced the measurement time.

This paper examines an indoor location estimation method using wireless LAN. The Fingerprint method is used as a location estimation technique. This method can estimate the current location based on the RSSI and MAC address obtained from the wireless LAN AP. However, this method is greatly affected by changes in the radio propagation environment, and the presence or absence of obstacles reduces the accuracy of location estimation. In this study, we propose a DB created by measuring the data of the different radio propagation environments. Environments changed by opening and closing doors were confirmed experimentally.

With the rapid development of electronic devices in recent years, undesired electromagnetic waves affecting other devices have become a problem. To solve this problem,electromagnetic shielding materials are frequently used. In this study, in order to design an electromagnetic shielding material for high-power devices, a magnetic field sources of different heights are placed inside a metal enclosure with an opening on the top surface, and the magnetic field leakage from the metal plate placed in the opening was investigated.

Recently, indoor navigation system is one of the important technologies. We are studying an indoor location estimation technique using wireless LAN employing a Finger Print scheme. A database (DB) and user data (UD) are measured from wireless LAN radio waves to estimate the location using the Finger Print. A CNN is utilized to compare the UD and DB. However, the CNN estimation accuracy may deteriorate due to the degradation problem caused by the increase in the number of middle layers and the gradient vanishing problem. ResNet is a solution to this problem. In this study, we have evaluated estimation accuracies of CNN and ResNet experimentally.

In this study, we attempted to interpolate the coordinates output by the fingerprinting method for indoor location estimation using wireless LAN. Using Wi-Fi-based RSSI fingerprinting, we can know our indoor location. The fingerprinting method requires the RSSI measurement by the operator in advance. That has a problem the user can only estimate the location at the pre-measured coordinates. It is known that this problem can be solved by using a particle filter. By using a particle filter, we can estimate not only the pre-measured coordinates but also the inter- coordinates among them. Not only that, we can improve the accuracy of the position estimation based on the temporal dependency of the user assuming a pedestrian. In this paper, the result of experimental validation shows the proposed method improves the accuracy. We evaluated and enhanced the accuracy of position estimation using MSE with a particle filter having more particles than before. In addition, we did the accuracy of position estimation when using CNN with a particle filter.

In this study, it is assumed that UAV (unmanned aerial vehicle) is used as a repeater. However, in that case, it is necessary to consider the following two problems. The first is that the movement of the transmitting station causes the radio waves from the repeater to interfere with the radio waves that arrive directly from the transmitting station. The second problem is that the direction of radio wave irradiation cannot be accurately determined due to the movement of the repeater. Therefore, in order to solve the above problems, the authors designed a small circular polarized wave antenna with wide directivity by FDTD calculations. After that, the radiation pattern of designed antenna is considered.

This contribution describe "area estimation" using Finger printing for indoor localization. A method for updating DBs with user-acquired AP information is necessary for the system. There is a problem of lowering the estimation accuracy due to the bias of the updated DBs. To solve this problem, we proposed a method to create a large distribution by mixing multiple databases.

The authors have been proposed the bandpass space filters consisting of the metal wire grid with negative permittivity and the dielectric materials. In this study, we designed the space filters using a conductive film sheet, instead of a metal wire grid.

Recently, location information has been emphasized with the spread of smartphones. The authors have been researching indoor location estimation with the goal of developing indoor navigation applications. In this research, we handle RSSI as time series data, and propose and verify the position estimation method using RNN.

In this study, the measurements of near magnetic field on the opening surface of metallic enclosure are investigated. First, the authors simulated the magnetic fields of a high power equipment enclosure including a power circuit. Next, the near magnetic fields are evaluated by comparing measured and calculated results.

First of all, as a research background, position estimation by GPS has become the mainstream in recent years, but it is known that position estimation accuracy deteriorates in places with poor radio wave environment such as indoors and underground facilities. Therefore, our laboratory is studying position estimation by the Finger Print method using a wireless LAN access point.

In this study, we designed the electromagnetic (EM) wave absorber using a square metallic pattern array sheet. Furthermore, we confirmed that the resonant frequency and the EM wave absorption can be controlled by changing the construction parameters of the metallic pattern array sheet.

Total knee Arthroscopy (TKA) is an operation which replaces the damaged knee with an artificial knee implant. There are some kinds of TKA procedures, and various kinds of prosthesis. Thus, it becomes a tough work for surgeons to select an appropriate procedure and prosthesis for individual patients. This study proposes a prediction method of post-operative implanted knee kinematics. It predicts the post-operative kinematics from only pre-operative kinematics using a machine learning method with clinical big data. In 46 TKA subjects, the method predicts the post-operative anterior-posterior translation with a correlation coefficient of 0.77 and a root-mean-squared error of 0.7mm.

To realize automated checker system of cafeteria, we have developed a meal menu recognition system. This system employs a RGB-D sensor to get both color image and depth image. From these images, first we extract and distinguish each dishes. By knowing what type of dish is used for the meal, we can reduce candidates for meal menus, it leads up to improvement of menu recognition accuracy. Some experimental result show that we can recognize dish type over 99% accuracy except for black dishes. Further, we can recognize a dish menu about 97% accuracy.

In order to improve the communication performance and to reduce the deterioration due to fast-fading channel variations, a continuous-switching type array antenna receiver is proposed. The receive antennas are placed in a linear array form, and by sequentially switching the elements in the opposite direction to the motion of the train, it is possible to fix the receiving point virtually. In this paper, as part of the results of a scale-model experiment of high-speed movement using the prototype antenna, we show that it is possible through optimum antenna selection and appropriate signal processing to reduce the fading rate and improve the performance of an OFDM system. A simulation based on the WiMAX physical layer parameters was made and data transmitted at RF in a 5 MHz bandwidth with symbol rate reduced to 1/10. The eight element receive antenna moves along at 3.6 km/h and the bit error rate with and without switching compared.