Abstract: An optical transmission unit (3) transmits an optical signal having a light intensity set as a Low level component of a pulse. An optical partial reflector (6) is provided on a path through which transmission light is transmitted from a circulator (5) to the atmosphere, and reflects the optical signal. A detection unit (11) performs coherent detection on reception light using, as local light, a signal in a Low level section in the optical signal reflected by the optical partial reflector (6).

Abstract: Provided is a sensor network capable of acquiring, from a data center, learning data composed of a pair of sensor data (input) and a classification label (output) necessary for adding/changing a classification label for updating a classifier while reducing the volume of communication between a sensor and the center. One aspect of the present invention relates to a sensor network, including one or more sensor nodes and a data center. The sensor node includes: an encoding unit for encoding sensor data by an encoder part of an autoencoder; and a transmission unit for transmitting the encoded data. The data center includes: a reception unit for receiving data encoded from sensor data by the encoder part of the autoencoder; a decoding unit for decoding the encoded data by a decoder part of the autoencoder; and a storage unit for storing the decoded data therein.

Abstract: A gas wiping nozzle manufactured from parts divided along the slit length direction and maintains a gap in the width direction over the length direction in high temperature atmospheres and a method for manufacturing a hot-dip metal strip. In a gas wiping nozzle, a first and a second nozzle member are each divided along the length direction X of a slit into a plurality of nozzle members. The dimension of a divided face of the first nozzle member is 1.5T1 or more in a section of the first nozzle member where T1 is the thickness of the first nozzle member in the width direction Z of the slit, and the dimension of a divided face of the second nozzle member is 1.5T2 or more in a section of the second nozzle member where T2 is the thickness of the second nozzle member in the width direction Z of the slit.

Abstract: Provided is a sensor network capable of acquiring, from a data center, learning data composed of a pair of sensor data (input) and a classification label (output) necessary for adding/changing a classification label for updating a classifier while reducing the volume of communication between a sensor and the center. One aspect of the present invention relates to a sensor network, including one or more sensor nodes and a data center. The sensor node includes: an encoding unit for encoding sensor data by an encoder part of an autoencoder; and a transmission unit for transmitting the encoded data. The data center includes: a reception unit for receiving data encoded from sensor data by the encoder part of the autoencoder; a decoding unit for decoding the encoded data by a decoder part of the autoencoder; and a storage unit for storing the decoded data therein.

Abstract: A small electronic apparatus employs a first case for storing a display panel and a second case for storing a keyboard which are foldably coupled to each other. The apparatus also includes an inclination mechanism for retaining the first case at an inclined plane with respect to the second case to see the display panel clearly. The first case is provided with a memory card insertion case, a metal plate, and a membrane of the keyboard. A plurality of openings are formed in an upper portion of the card insertion case. Openings are respectively positioned at upper portions of electronic parts disposed in the case to reduce the thickness of the apparatus. The metal plate is disposed on the upper portion of the card insertion case for reinforcing it. With a thin construction, the apparatus keeps the balance on a desk, even if the first case is inclined to an outward plane of the second case.