Electronics and Telecommunications Research Institute

Korea Atomic Energy Research Institute

Seong Mo Park

Nak Woong Eum

Sung Eun Kim

Tae Wook Kang

Kwang Il Oh

Jae-Jin Lee

In Gi Lim

Hyung-Il Park

Byounggun Choi

Mi Jeong Park

Moo Kyoung Chung

Sung Weon Kang

Jae Jin Lee

Kyung Hwan Park

Joo Hyun Lee

Han Jin Cho

Jun Young Lee

Suk Ho Lee

Jin Jong Cha

Jung Bum Kim

Jin Ho Han

Kyung Su Kim

Kyoung Seon Shin

Sang Heon Lee

Seung Hyun Cho

Jong Dae Kim

Kyung Soo Kim

Jin Joo Kim

Mi Young Lee

Hee-Bum Jung

Ju-Yeob Kim

Young Hwan Bae

Young Su Kwon

Jin Kyu Kim

Seung Chul Kim

Kyung-Hwan Park

Sang Mu Choi

Ig Kyun Kim

Young Rang Uhm

June Young Chang

Jong Bum Kim

Kwang-Jae Son

Jintae Hong

Provided is a spike neural network circuit including a synapse configured to generate an operation signal based on an input spike signal and a weight, and a neuron configured to generate an output spike signal using a comparator configured to compare a voltage of a membrane signal generated based on the operation signal with a voltage of a threshold signal, wherein the comparator includes a bias circuit configured to conditionally supply a bias current of the comparator depending on the membrane signal.

Spike neural network circuit including comparator operated by conditional bias current

Provided is a capsule endoscope. The capsule endoscope includes: an imaging device configured to perform imaging on a digestive tract in vivo to generate an image; an artificial neural network configured to determine whether there is a lesion area in the image; and a transmitter configured to transmit the image based on a determination result of the artificial neural network.

Capsule endoscope for determining lesion area and receiving device

Provided is a radioisotope battery. A radioisotope battery according to exemplary embodiments may include: a substrate; a shield layer disposed on the substrate and including a first material; a source layer embedded in the shield layer and including a second material which is a radioisotope of the first material; a PN junction layer on the shield layer and the source layer; and a window layer between the PN junction layer and the source layer.

Radioisotope battery

Provided is a spike neural network circuit. The spike neural network circuit includes an axon configured to generate an input spike signal, a synapse including a first transistor for outputting a current according to a weight and a second transistor connected to the first transistor and outputting the current according to an input spike signal, a neuron configured to compare a value according to the current output from the synapse with a reference value and generate an output spike signal based on a comparison result, and a radiation source attached to a substrate on which the synapse is formed, configured to output radiation particles to the synapse, and configured to increase magnitudes of threshold voltages of the first and second transistors of the synapse.

Spike neural network circuit including radiation source

A random number generating device includes a particle detector, a pulse generator, a clock counter, and a random number converter. The particle detector detects particles emitted from a radioactive isotope. The pulse generator generates pulses corresponding to the particles. The clock counter counts the number of clock cycles during time intervals between the pulses and generates a plurality of count values. The random number converter adjusts a clock frequency, based on a minimum value and a maximum value of the plurality of count values and converts a target count value generated depending on the adjusted clock frequency into a random number.

Random number generating device and operating method of the same

The exemplary embodiments of the present invention provide a quantum random number generation apparatus according to an exemplary embodiment of the present invention including: a space-division semiconductor detector including a plurality of cells, each individually absorbing a plurality of emission particles emitted from a radioactive isotope; and a signal processor that generates a random number based on an absorption event at which the plurality of emission particles are absorbed into the plurality of cells, and thus new type of random number conversion method that combines a spatial randomness and existing temporal randomness of the emission particle can be provided, there is no restriction generated due to the dead time, the random number generation rate can be remarkably increased, and it is possible to generate of a pure random number at high speed, which is required by a computer, a network processor, or an IoT device.

Apparatus and method for generating quantum random number

Provided is an artificial intelligence system. The system includes a first sensor configured to generate a first sensing signal during a sensing time, a second sensor disposed adjacent to the first sensor and configured to generate a second sensing signal during the sensing time, a pre-processing unit configured to select valid data according to a magnitude of a differential signal generated based on a difference between the first sensing signal and the second sensing signal, and an artificial intelligence module configured to analyze the valid data to generate result data.

Artificial intelligent system including pre-processing unit for selecting valid data

Provided are a random number generating device and a method of operating the same. The random number generating device includes a source detector, a pulse generator, a counter, and a verification circuit. The source detector detects particles emitted from a source to generate a detection signal. The pulse generator generates pulses corresponding to the detected particles, based on the detection signal. The counter measures time intervals among the pulses and generates binary count values respectively corresponding to the time intervals. The verification circuit determines an output of the binary count values, based on the number of 0 values and the number of 1 values included in the binary count values.

Provided is a neuromorphic arithmetic device. The neuromorphic arithmetic device may include a synapse circuit, a metal line having an inherent capacitance component, an oscillator, a comparator, and a capacitance calibrator. The synapse circuit may be configured to perform a multiplication operation on a PWM signal and a weight to generate a current. The metal line may include a metal line capacitor in which a charge of the current is stored. The oscillator generates a plurality of pulses on the basis of the charge stored in the metal line capacitor. The comparator may compare a frequency of the plurality of pulses and a target frequency, and may generate a control signal on the basis of a result of the comparison. The capacitance calibrator may adjust a capacitance value of the metal line capacitor on the basis of the control signal.

Neuromorphic arithmetic device

Provided is an operating method of a wearable device, the method including setting a threshold voltage when there is not a physical contact of an outsider, measuring an envelope voltage while performing a data communication, determining whether to be the physical contact with the outsider by comparing the threshold voltage with the envelope voltage, and stopping the data communication, when there is the physical contact with the outsider.

Wearable device and data security method thereof

The present disclosure relates to a capsule endoscope transmitter configured to transmit frames including control frames and data frames to a capsule endoscope receiver. The capsule endoscope transmitter includes a preamble generator configured to generate preambles for synchronizing and identifying the control frames used to select a reception electrode pair that receives the frames, and a line sync generator configured to generate a line sync for synchronizing the data frames and identifying a code value of each of the data frames.

Capsule endoscope transmitter and capsule endoscope receiver configured to perform human body communication and human body communication method using the same

The present disclosure relates to a neuromorphic arithmetic device. The neuromorphic arithmetic device may include first and second synapse circuits, a charging/discharging circuit, a comparator, and a counter. The first synapse circuit may generate a first current by performing a first multiplication operation on a first PWM signal and a first weight, and the second synapse circuit may generate a second current by performing a second multiplication operation on a second PWM signal and a second weight. The charging/discharging circuit may store charges induced by the first current and the second current in a charging period, and may discharge the charges in a discharging period. The comparator may compare a voltage level of the charges discharged in the discharging period and a level of a reference voltage. The counter may count output pulses of an oscillator on the basis of a result of the comparison by the comparator.

Provided is an image compression device including an object extracting unit configured to perform convolution neural network (CNN) training and identify an object from an image received externally, a parameter adjusting unit configured to adjust a quantization parameter of a region in which the identified object is included in the image on the basis of the identified object, and an image compression unit configured to compress the image on the basis of the adjusted quantization parameter.

Image compression device

Disclosed is a motion estimation apparatus and method capable of reduce a memory bandwidth by accessing reference image data occupying the largest part of the memory bandwidth based on a result of motion estimation skip determination. The motion estimation apparatus includes a motion estimation skip determination unit, a sum absolute of absolute difference (SAD) processing unit and a multiplexer (MUX) unit. The motion estimation skip determination unit determines the presence of motion estimation skip by reading current image data and skip reference image data, and outputs a motion vector in a skip mode. The SAD processing unit calculates an SAD value by reading the current image data and reference image data and outputs a motion vector in a normal mode. The MUX unit outputs the motion vector in the skip mode or the motion vector in the normal mode as a final motion vector based on a skip flag signal.

Motion estimation apparatus and method

Growing community of inventors

Daejeon, South Korea

Seong Mo Park