Abstract: A string actuator-based exoskeleton robot includes: a driving force conversion unit including a fixed frame and a rotation pulley and mountable on a side portion of the waist of a human body; a driving unit including a driving motor providing a driving force and a pair of strings provided in double rows to have one side connected to the driving motor and the other side connected to the rotation pulley, and assisting the rotation of the rotation pulley by twisting or untwisting the pair of strings with or from each other by a rotational driving force of the driving motor to vary a length of the pair of strings; and a support unit formed to extend from the rotation pulley to at least one of a front surface or a rear surface of the femoral region of the human body to support the femoral region of the human body.

Abstract: An optical phased array (OPA) radiator includes a plurality of unit radiators configured to serve as optical waveguides, each of the unit radiators being made of a silicon material and each having a predetermined length, where the unit radiators are disposed in parallel; a cladding portion configured to cover the plurality of unit radiators; and a plurality of electrodes arranged in parallel with the plurality of unit radiators on the cladding portion, where the plurality of electrodes are arranged so as not to overlap the plurality of unit radiators in a vertical direction. A beam radiated through the unit radiators using a phased array can be efficiently vertically steered.

Abstract: Provided is an ultrasensitive spectrometer. The ultrasensitive spectrometer includes a scattering medium, a light-receiving sensor disposed at a rear end of the scattering medium, a storage unit storing previously predicted speckle pattern information according to wavelengths with respect to the scattering medium, and a signal processing unit configured to process a sensing signal generated by the light-receiving sensor. The scattering medium scatters incident analyzing light to form a random speckle pattern, and the light-receiving sensor senses the speckle pattern formed by the scattering medium. The signal processing unit restores spectrum information of the analyzing light from sensing information sensed of the speckle pattern formed by the scattering medium.

Abstract: An electronic package molding device includes a planar heating element configured to mold an electronic package by approaching one surface of an electronic package to melt a molding agent disposed on the one surface of the electronic package and apply pressure to bring the epoxy molding compound into close contact with the electronic package and a gas pressure regulator configured to adjust gas pressure between the electronic package and the planar heating element.

Abstract: The present disclosure relates to a computer system for maximizing wireless energy harvesting using an intelligent reflection surface and a method thereof. A multi-path channel including a direct channel through which power is directed from an ET apparatus toward an EU apparatus and a reflection channel through which the power is directed from the ET apparatus toward the EU apparatus after being reflected by an IRS apparatus may be implemented. The computer system may be designed to maximize a gain of the multi-path channel in order to maximize the harvested power.

Abstract: A golf swing practice system is provided. The golf swing practice system includes one or more electromyography (EMG) sensors attached to a specific muscle point on a body of a user and a golf swing analysis device that is connected to the EMG sensor to enable data communication, wherein the golf swing analysis device may include a sensor signal collector configured to receive an EMG signal of the user from the EMG sensor, a data comparison analyzer configured to compare and analyze EMG data of the user and a professional golfer to be compared, and a user interface configured to receive information about a muscle that the user needs to train and display an analysis result of the data comparison analyzer.

Abstract: Provided is a 3D rendering accelerator based on a DNN trained using a weight of the DNN using a plurality of 2D photos obtained by imaging the same object from several directions and then configured to perform 3D rendering using the same, the 3D rendering accelerator including a VPC configured to create an image plane for a 3D rendering target from a position and a direction of an observer, divide the image plane into a plurality of tile units, and then perform brain imitation visual recognition on the divided tile-unit images to determine to reduce a DNN inference range, an HNE including a plurality of NEs having different operational efficiencies and configured to accelerate DNN inference by dividing and allocating tasks, and a DNNA core configured to generate selection information for allocating each task to one of the plurality of NEs based on a sparsity ratio.

Abstract: An object detecting device includes a feature extracting circuit configured to extract first feature data from an input image; a feature transforming circuit configured to transform the first feature data into transformed feature data according to a transformation function; and a decoder circuit configured to decode the transformed feature data into a region map indicating a detected object.

Abstract: An autonomous driving device includes an execution network configured to determine a target speed of a driving vehicle according to state information history including a plurality of state information for road environment. The plurality of state information are generated at a plurality of times. The execution network includes a spatial attention network configured to receive the state information history and to generate feature data reflecting spatial importance based on the state information history; and a temporal attention network configured to determine the target speed of the driving vehicle by applying temporal importance to an output of the spatial attention network.

Abstract: The present disclosure relates to a method and device for the tomography of a Rydberg atom graph using a location-adjustable ancilla qubit. The method and device may be configured to prepare at least one ancilla qubit around a Rydberg atom graph of interest having at least one atom, measure interactions between the Rydberg atom graph and the ancilla qubit while adjusting a location of the ancilla qubit around the Rydberg atom graph, and reconstruct a quantum state of the Rydberg atom graph based on the measured interactions.

Abstract: A protocol state fuzzing method for security of a control plane of a distributed software-defined network is provided. The protocol state fuzzing method includes receiving input alphabets being abstract symbols of a protocol message in an ambusher of a distributed network operating system (NOS), converting the input alphabets into the protocol message, and sending the protocol message to a cluster, monitoring, by the cluster, intercommunication between instances in the distributed NOS, and selecting a set of sequences executable in the cluster and searching a cluster log for an output by executing the sequence to generate an attack result.

Abstract: Various example embodiments provide a computer system for hybrid of epoch- and pointer-based memory reclamation, and a method thereof. Various example embodiments are configured to implement a pointer and epoch-based memory reclamation scheme by hybridizing a pointer-based memory reclamation scheme and an epoch-based memory reclamation scheme, and perform a memory reclamation of a concurrent data structure, based on the pointer and epoch-based memory reclamation scheme.

Abstract: An edge detecting device includes a feature extracting circuit configured to extract first and second feature data from an input image; a prototype generating circuit configured to generate prototype data using the first feature data and an input label, the prototype data including foreground and background information of an object; a region detecting circuit configured to generate a segmentation mask by detecting a region of an object using the first feature data and the prototype data; and an edge extracting circuit configured to generate an edge map by combining the segmentation mask and the second feature data.

Abstract: A calibration circuit includes a selection unit, a power detector, an analog-to-digital converter and a calibrator. The selection unit is connected to the plurality of channels, selects two channels from among the plurality of channels, provides a test signal to the selected two channels, and receives a test result signal from the selected two channels. The power detector detects power of the selected two channels. The analog-to-digital converter performs an A/D conversion on an output of the power detector. The calibrator calibrates the phases and the gains between the plurality of channels based on an output of the analog-to-digital converter. One of the plurality of channels is set as a reference channel, phases and gains of remaining channels other than the reference channel among the plurality of channels are sequentially optimized based on the reference channel, and a phase and a gain of the reference channel is optimized.

Abstract: The present disclosure relates to an anti-reflective nanostructure and a method of manufacturing the same, and more particularly, to an anti-reflective nanostructure having a refractive index close to a quintic refractive index profile and exhibiting an anti-reflection effect in which a reflectance is close to almost 0. The anti-reflective nanostructure according to an embodiment of the present invention includes: a base part having a top surface; and a plurality of nanostructures arranged in a first direction on the top surface and each having a shape in which an upper portion has a thin and sharp shape, a lower portion has a width greater than that of the upper portion, and the width gradually increases in a direction from the upper portion to the lower portion.

Abstract: Various embodiments provide carbon dots emitting light having multiple colors and a method of fabricating the same. According to various embodiments, the carbon dots include at least two structural characteristics. At least two emission levels indicative of different emission wavelengths, respectively, may be introduced into the carbon dots based on the structural characteristics. A ratio of emission colors of the carbon dots may be adjusted when the carbon dots are synthesized or after the carbon dots are synthesized.

Abstract: A device includes a threshold voltage distribution estimation network configured to generate an estimated distribution using a feature distribution and read trial information, a set of feature distributions generated from a plurality of threshold voltage distributions for a plurality of pages of a memory device, and a read reference voltage estimation network configured to generate a read reference voltage from the estimated distribution. The read trial information includes a read trial vector and an output value, the output value being generated by applying the read trial vector to a threshold voltage distribution for a page to be read among the plurality of threshold voltage distributions.

Abstract: A signed bit slice generator includes a divider configured to divide input data, which is 2's complement data having N (where N is a natural number)-bit precision, and divide remaining bits excluding a sign bit of the input data into a predetermined number of bit slices, a sign bit adder configured to add a sign bit to each of the bit slices, a sign value setter configured to set a sign bit of an MSB slice among the bit slices to a sign value of the input data and to set sign bits of the remaining bit slices to positive sign values, and a sparse data compressor configured to perform sparse data compression on each of the signed bit slices, thereby generating a predetermined number of signed bit slices having the same number of bits where each bit slice includes a sign bit.

Abstract: For non-destructively measuring the yield strength of a metallic member using magnetic incremental permeability, a quasi-static excitation magnetic field is applied to two points of the metallic member, while a small alternating magnetic field generated by a transmitting coil is applied to the metal member. Intensity of the quasi-static excitation magnetic field applied to the metallic member is measured using a Hall sensor. The magnetic field induced by the magnetized metallic member is detected using a sensing coil. Using signals from the Hall sensor and the sensing coil, a reversible permeability (MIP) of the metallic member is obtained. A grain size of the metallic member is obtained from the relationship between the reversible permeability and the grain size of the metallic member, and the yield strength of the metallic member is calculated using the grain size. The grain size and yield strength of the metallic member can be measured non-destructively.

Abstract: A neural network data quantizing method includes: obtaining local quantization data by firstly quantizing, based on a local maximum value for each output channel of a current layer of a neural network, global recovery data obtained by recovering output data of an operation of the current layer based on a global maximum value corresponding to a previous layer of the neural network; storing the local quantization data in a memory to perform an operation of a next layer of the neural network; obtaining global quantization data by secondarily quantizing, based on a global maximum value corresponding to the current layer, local recovery data obtained by recovering the local quantization data based on the local maximum value for each output channel of the current layer; and providing the global quantization data as input data for the operation of the next layer.