-
公开(公告)号：JP2003263427A
公开(公告)日：2003-09-19
申请号：JP2002062625
申请日：2002-03-07
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： YAMAMOTO HIROSHI , KIKUI GENICHIRO
Abstract: PROBLEM TO BE SOLVED: To provide a method of dividing a sentence by learning without a teacher, without using heuristics. SOLUTION: This method of generating a word division model using a training sentence without word division comprises a first step of generating a network of dividable candidate words from all the given training sentences using a given dictionary entry, a second step of generating such a model as to minimize entropy, to the network of the candidate words generated in the first step, and a third step of smoothing the transition probability value which is the probability value for predicting the following word from a known word or a known word pair. COPYRIGHT: (C)2003,JPO
-
公开(公告)号：JP2003258811A
公开(公告)日：2003-09-12
申请号：JP2002055471
申请日：2002-03-01
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： UEDA TETSUO , SOMPRAKASH BANDYOPADHYAY , HASUIKE KAZUO
IPC: H04L12/28 , H04L12/733 , H04W16/28 , H04W40/02 , H04W40/12 , H04L12/56 , H04Q7/22 , H04Q7/24 , H04Q7/26 , H04Q7/30
Abstract: PROBLEM TO BE SOLVED: To transmit message data even when the opposite station is off-line and to provide a service when a self-station intends to be off-line in a radio network such as an ad hoc radio network. SOLUTION: When its own station intends to be off-line, the self-station selects a proxy radio station to be a proxy for the self-station from an MGLS table on the basis of a prescribed criteria (S142), and transmits to the proxy radio station a proxy request signal for requesting the proxy radio station to be a proxy for the self-station (S145). When the self-station receives a proxy request signal to the self-station, the self-station inserts proxy information showing that the self-station is a proxy radio station for a particular different ratio station in the MGLS table to update the MGLS table, and also updates the MGLS table of the different radio station by transmitting a prescribed packet signal with the proxy information included in transmitting the packet signal to the different radio station. COPYRIGHT: (C)2003,JPO
-
公开(公告)号：JP2003258532A
公开(公告)日：2003-09-12
申请号：JP2002051057
申请日：2002-02-27
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： IIGUSA KIYOUICHI , OHIRA TAKASHI
Abstract: PROBLEM TO BE SOLVED: To provide a control method of an ESPAR (Electronically Steerable Passive Array Radiator) antenna that can obtain an excellent converged value capable of obtaining an excellent evaluation function value with the smaller number of repetition times than that of a conventional control method when the major beam of an antenna system is directed to a desired wave and the null pattern is directed to an interference wave. SOLUTION: An adaptive control type controller 20 sequentially perturbs the reactance of each of variable reactance elements 12-1 to 12-6 by a prescribed difference width, calculates a prescribed evaluation function value for each reactance value, and controls to decrease the difference width ΔX and a step width μ depending on an SINR (signal to interference noise power ratio) calculated from the evaluation function value by using a prescribed reduction function indicative of the difference width ΔX when the reactance value of the variable reactance elements 12-1 to 12-6 is repetitively calculated and set in order to direct the major beam of the array antenna system 100 to the direction of the desired wave and the null to the direction of the interference wave so that the evaluation function value is maximized by using the maximum grade method on the basis of the calculated evaluation function value. COPYRIGHT: (C)2003,JPO
-
公开(公告)号：JP2003248496A
公开(公告)日：2003-09-05
申请号：JP2002047047
申请日：2002-02-22
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： NAKAJIMA HIDEJI , YAMAMOTO HIROSHI , WATANABE TARO
Abstract: PROBLEM TO BE SOLVED: To provide a language model adaptation method capable of creating a language model adaptable to a new task by using a mono-lingual corpus in the new task described in a language other than the language of the language model to be made adaptive. SOLUTION: The language model adaptation method for creating the language model of a first language adaptable to the new task when the task of a voice translation device is extended, is provided with a first step to create a second mono-lingual corpus in the new task described in the first language by translating the first mono-lingual corpus of the new task described in a second language other than the first language into the first language using a machine translation device, and a second step to make the language model adaptive based on the second mono-lingual corpus in the new task created in the first step. COPYRIGHT: (C)2003,JPO
-
公开(公告)号：JP2003233612A
公开(公告)日：2003-08-22
申请号：JP2002033072
申请日：2002-02-08
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： KURABAYASHI NORIYUKI , YAMAZAKI TATSUYA , YUASA TAICHI , HASUIKE KAZUO
Abstract: PROBLEM TO BE SOLVED: To develop communication of a high degree of freedom by enabling a user to acquire a variety of information from another user computer. SOLUTION: A communication system 10 includes a server 12. In a chat room provided by the server 12, a first to a fourth user of user computers 18 to 24 connected to the server 12 via the Internet 16 can chat. If, for example, the first user is silent on a subject, the user computer 18 requests information support from the server 12. The server 12 responsively mediates the information support request to the other user computers 20 to 24. Any one of the user computers 20 to 24 capable of information support provides information interesting the user computer 18. COPYRIGHT: (C)2003,JPO
-
Patent Agency Ranking
公开(公告)号：JP2003188630A
公开(公告)日：2003-07-04
申请号：JP2001385391
申请日：2001-12-19
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： HIRATA AKIHITO , TEI TAKASHI , OHIRA TAKASHI
Abstract: PROBLEM TO BE SOLVED: To perform adaptive control so that convergence time is significantly reduced, a main beam is directed to a wanted wave while null is directed to an interference wave with less calculation amount. SOLUTION: An adaptive control type controller 20 halves the range of reactance values which variable reactance elements 12 can take. The correlation factor between a reception signal and a learning sequence signal is calculated when a central value of each range after halving is set. The reactance value of each variable reactance element 12 corresponding to the larger correlation coefficient between two correlation coefficient corresponding to the central values of the ranges after halving is selected as an initial value. Then the range belonging to the selected reactance value is halved, and the correlation coefficient is calculated when the central values of the ranges after halving are set. The reactance value of the variable reactance element 12 corresponding to the larger correlation coefficient between the two correlation coefficient corresponding to the central values of the ranges after halving is selected. The latter process is repeated for adaptive control. COPYRIGHT: (C)2003,JPO
-
公开(公告)号：JP2003178303A
公开(公告)日：2003-06-27
申请号：JP2001376000
申请日：2001-12-10
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： UCHIUMI AKIRA , TETSUYA SHINJI
Abstract: PROBLEM TO BE SOLVED: To provide an object recognition device recognizing an object on the basis of a characteristic of appearance of the object itself and not by extracting the target object as a moving area or a particular color area. SOLUTION: A discrimination axis is determined by extracting geometrical structures of a background image and a person image picked up by a camera as a difference of picture element distributions and constructing a statistical model of appearance of the person image. The model and an inputted image are compared and a score is calculated on the basis of the discrimination axis of the model and a distance value vector of the inputted image. A recognition process is efficiently carried out by a small number of models by judging that a person has been detected when the score exceeds a threshold. COPYRIGHT: (C)2003,JPO
-
公开(公告)号：JP2003152745A
公开(公告)日：2003-05-23
申请号：JP2001346509
申请日：2001-11-12
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： MAESHIRO TETSUYA , SHIMOHARA KATSUNORI
Abstract: PROBLEM TO BE SOLVED: To provide a data transmission system that can reduce the area of a circuit board on which a transmission line is formed and enhance the data transfer speed, and to provide a data transmitter and a data receiver. SOLUTION: When a transmission request transmission section 11 transmits a transmission request to a receiver 2, a transmission request reception section 21 receives the transmission request via a data line D1, and a transfer clock generating section 24 transmits a transfer clock to a data transmission section 12 via a control line C1. In this case, the data transmission section 12 transmits transmission data in n-bits to a data reception section 22 via data lines D1 to Dn synchronously with the received transfer clock, and the data reception section 22 receives the n-bit received data sent from the data transmission section 12 synchronously with a delayed transfer clock outputted from a transfer clock delay section 23.
-
公开(公告)号：JP2003135863A
公开(公告)日：2003-05-13
申请号：JP2001342956
申请日：2001-11-08
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： MASE KENJI , SAITO MASASHI
Abstract: PROBLEM TO BE SOLVED: To achieve communications through the transmission of nonverbal information. SOLUTION: This music communication system 100 includes a network 102, and music communication devices 10a and 10b of the same constitution are connected together via the network 102. The music communication device 10a includes a stuffed animal 12 and a plurality of sensors for detecting the expression, that is, the behavior or voice, of a user A, are provided on different portions of the stuffed animal 12. For example, the roles of the plurality of sensors are varied (set) by the expression of the user A and sounds matching the sensor inputs are outputted for the roles determined. The music communication device 10a sets the roles of the sensors or outputs sounds according to the interacting condition of the music communication device 10b and the sensor inputs (state information). That is, the user A and user B communicate with each other by sending and receiving the state information through the music communication devices 10a and 10b.
-
公开(公告)号：JP2003121896A
公开(公告)日：2003-04-23
申请号：JP2001317895
申请日：2001-10-16
Applicant: ATR ADVANCED TELECOMM RES INST
Inventor： ANDO TARO , HOSODA MAKOTO , NISHIMURA KOTA , AIDA TADAHITO
IPC: G02F2/02
Abstract: PROBLEM TO BE SOLVED: To provide a wavelength conversion element capable of converting incident light of any wavelength into light of a shorter wavelength, and to provide a wavelength conversion device using the same. SOLUTION: The wavelength conversion element includes a unit element 32 and a unit element 33. The unit elements 32, 33 include barrier layers 41, 43 and well layers 42, 44. The barrier layers 41, 43 are formed of AlAs, and the well layers 42, 44 are formed of GaAs. An electron 51 excited at a sub- energy level 11 in the well layer 42 of the unit element 32 reaches a sub-energy level 15 in the well layer 44 via a sub-energy level 21 in the barrier layer 43. Moreover, a positive hole 52 generated in the well layer 42 of the unit element 33 reaches the well layer 44 of the unit element 32 via the barrier layer 41. Then, the electron 51 at the sub-energy level 15 is re-coupled with the positive hole at a sub-energy level 17, and thereby the exit light is emitted.
-
-
-
-
-
-
-
-
-
Search Results
127
records
(took 0.022 seconds)
