Abstract: A method by which a terminal transmits an SRS can comprise the steps of: receiving, from a base station, first information indicating symbols to which the SRS is transmitted among a plurality of symbols, when the terminal is configured to perform antenna switching for an SRS transmission on the plurality of symbols; and transmitting the SRS on the indicated symbols, wherein the SRS is transmitted through antenna ports respectively corresponding to the indicated symbols, and the antenna ports respectively corresponding to the indicated symbols can be different from each other.

Abstract: A semiconductor package includes a first semiconductor chip including a first semiconductor layer, a first through-electrode passing through the first semiconductor layer in a vertical direction, and a first bonding pad connected to the first through-electrode, and a second semiconductor chip including a second semiconductor layer on the first semiconductor chip, a wiring structure between the second semiconductor layer and the first semiconductor chip, a wiring pad connected to the wiring structure below the wiring structure, and a second bonding pad connected to the wiring pad below the wiring pad and in contact with the first bonding pad, wherein the second bonding pad includes a protrusion protruding toward the wiring pad.

Abstract: Provided is a memory-based reinforcement learning method and system capable of storing optional information in streaming data. A question-answering (QA) method using memory-based reinforcement learning method includes receiving, in an episodic memory reader (EMR), streaming data about an input context that is input from a user; analyzing, in the EMR, the received streaming data and storing preset semantic information used for QA in an external memory; and, in response to an input of a question front the user, determining, in a pretrained QA model, an answer to the input question based on semantic information stored in the external memory.

Abstract: The present disclosure relates to a key input device applicable to a smart mat and a smart mat to which the same is applied. More specifically, the present disclosure relates to a key input device including a first layer in which a plurality of row contacts are formed, a second layer in which a plurality of column contacts are formed, and a processor configured to detect that at least one of a plurality of key switches formed by the plurality of row contacts and the plurality of column contacts is pressed.

Abstract: The present disclosure relates to an interactive fitness system and a control method thereof for detecting an action of a user moving on a smart mat while wearing a wearable terminal, and providing feedback based on the detected result. More specifically, the present disclosure relates to a technology for transmitting and receiving data with at least one of a smart mat and a wearable terminal, and determining whether or not a user is performing training correctly based on the transmitted and received data.

Abstract: Provided is a memory-based reinforcement learning method and system capable of storing optional information in streaming data. A question-answering (QA) method using memory-based reinforcement learning method includes receiving, in an episodic memory reader (EMR), streaming data about an input context that is input from a user; analyzing, in the EMR, the received streaming data and storing preset semantic information used for QA in an external memory; and, in response to an input of a question from the user, determining, in a pretrained QA model, an answer to the input question based on sematic information stored in the external memory.

Abstract: The present disclosure provides a graphene-based touch sensor device using triboelectric effect, the device comprising: a substrate; a first electrode layer disposed on the substrate; a graphene channel layer disposed on the substrate, wherein the graphene channel layer is flush with the first electrode layer, and is spaced from the first electrode layer; a gate dielectric layer in partial contact with the electrode layer and the graphene channel layer respectively; source and drain electrodes formed on both opposing ends of the graphene channel layer respectively; and a triboelectric layer formed on the first electrode layer, wherein the triboelectric layer generates a triboelectric potential via contact of an external friction material therewith, wherein the contact of the external friction material is detected based on a current change in the graphene channel layer due to the triboelectric potential applied thereto.

Abstract: There is provided a microphone device comprising: a substrate having top and bottom portions, wherein the substrate has a cavity defined therein, wherein the cavity is open at the bottom portion of the substrate; a two-dimensional piezoelectric layer disposed on the top portion of the substrate, wherein the two-dimensional piezoelectric layer blocks a top of the cavity; and first and second electrode layers respectively arranged on both lateral end portions of the two-dimensional piezoelectric layer, wherein the first and second spaced electrode layers are spaced apart and electrically insulated from each other, wherein an electric potential energy is generated between the first and second electrode layers via piezoelectricity of the two-dimensional piezoelectric layer when the two-dimensional piezoelectric layer vibrates by sound energy applied thereto, wherein the piezoelectricity is generated in a parallel direction to a plane of the two-dimensional piezoelectric layer.

Abstract: The present disclosure provides a graphene-based touch sensor device using triboelectric effect, the device comprising: a substrate; a first electrode layer disposed on the substrate; a graphene channel layer disposed on the substrate, wherein the graphene channel layer is flush with the first electrode layer, and is spaced from the first electrode layer; a gate dielectric layer in partial contact with the electrode layer and the graphene channel layer respectively; source and drain electrodes formed on both opposing ends of the graphene channel layer respectively; and a triboelectric layer formed on the first electrode layer, wherein the triboelectric layer generates a triboelectric potential via contact of an external friction material therewith, wherein the contact of the external friction material is detected based on a current change in the graphene channel layer due to the triboelectric potential applied thereto.