The fifth-generation (5G) mobile communication system introduces the use of high frequency bands with very wide bandwidth for realizing super high bit rate more than 20 Gbps. To use such high frequency bands, Massive MIMO technologies by hybrid beamforming is a promising technology in terms of power consumption and implementation cost. Experimental trials of hybrid beamforming in 28 GHz band were already conducted in an indoor anechoic chamber and in an outdoor light-of-sight (LOS) environment. However, as MIMO takes advantage of multipath, it is important to evaluate the system performance in multipath-rich environment. It is also important to evaluate the system performance in non LOS (NLOS) environment, because high straightness of high SHF band can degrade its performance. To clarify these actual impacts, in multipath-rich room with some obstacles, we conducted two indoor downlink experimental trials: SU-MIMO and 8-user MU-MIMO at 28 GHz frequency band with 500 MHz bandwidth. The results of SU-MIMO experiments show more than 20 Gbps user throughput was achieved in the multipath-rich environment. We also confirmed that strong reflector improves the throughput when a user is in the NLOS environment. MU-MIMO experiments show the hybrid beamforming can improve both the system throughput and user fairness compared to analog beamforming.

This paper describes a multimodal database which consists of image data of human gestures and corresponding speech data for the research on multimodal interaction systems.The purpose of this database is to provide an underlying foundation for research and development of multimodal interactive systems.Our primary concern in selecting utterances and gestures for inclusion in the database was to ascertain the kinds of expressions and gestures that artificial systems could produce and recognize.Total 25 kinds of gestures and speech were repeated four times for the recording of each subject.The speech and gestures for a total of 48 subjects were recorded, converted into files and in the first version, the files for 12 subjects were recorded on CD-ROMs.

Describes the development of a mouth-opening and -closing training robot WY-3 (Waseda Yamanashi-3) that has been designed and developed by combining oral-maxillofacial surgery and robotics. Mouth-opening training is useful for the rehabilitation of patients who have problems opening and closing their mouths. Conventional mouth-opening apparatus (such as wooden screws, bite blocks and clothespins type apparatus) only increase mouth-opening distance and do not have actuators, sensors, and/or control systems. Moreover, during therapy the quantitative force data is unknown in spite of its importance for the standardization of mouth-opening and -closing training. The mouth-opening and -closing training robot WY-3 consists of mechanical, actuation, sensor, and control systems. As a result of therapy using WY-3, the mouth-opening distance increased from 36 to 41 mm.

In this paper, we describe a multimodal interface prototype system based on Dynamical Dialogue Model. This system not only integrates information of speech and gestures, but also controls the response timing in order to realize a smooth interaction between user and computer. Our approach consists of human-human dialogue analysis, and computational modeling of dialogue.

This paper describes the quantification of masticatory efficiency by using a mastication robot that was developed as a mechanical simulator of human masticatory jaw motion on the basis of dental robotics. The masticatory efficiency is evaluated on the basis of human mastication and the quantification of masticatory efficiency is studied experimentally. In this paper, the authors use the sifting test to quantify efficiency. The results of chewing experiments with robot by comparing the differences in chewing motion between clenching and grinding showed that the masticatory efficiency of the latter was higher than that of the former.

Presents a pattern analysis scheme of autoradiogram images obtained with 2-D (two-dimensional) gel electrophoresis based on the RLGS (restriction landmark genome scanning) method for the purpose of high-speed genome scanning. By this means several thousands of spots are observed as signals; each spot corresponds to a genetic locus and its intensity reflects the copy number of the restriction landmark. The essential problems of image processing for automated analysis are to detect the location of each spot and to measure its intensity. Some kinds of operator including the gradient and ring operators are used for solving the problem. Then the matching and comparison of the RLGS patterns are also important to analyze the differential state. This is solved by using the Delaunay net and relative neighbourhood graphs. Some experimental results indicate that the authors' method is effective enough to construct an automated analysis system of RLGS patterns and genetic mapping.

We proposed a new expression of speech feature called orientation patterns which keeps its ability of detection higher in averaging of time domain. Because of this, we achieved to reduce number of frames in reference and input pattern in DP matching algorithm, then the calculation load were reduced. We constructed long term speech retrieval system by using this new expression. This system has RIFCDP as base matching algorithm which was already proposed. RIFCDP is an algorithm for spotting similar intervals between arbitrary reference pattern and arbitrary input pattern sequence synchronously with input frames.