Abstract: A method of building an ad-hoc network of a wireless relay node and an ad-hoc network system are disclosed. The method includes verifying state information representing a relative distance and angle to a neighboring node capable of receiving data when each of relay nodes transmits the data, determining an action representing a change over time of each relay node, determining, based on an amount of change in a network throughput determined according to the state information and an amount of energy consumption according to the action, a reward corresponding to the action, and building a network including a source node, a destination node, and a plurality of relay nodes by generating, based on a reward of each of the relay nodes, a policy that allows a cumulative reward to be maximized.

Abstract: According to one embodiment of the present invention, a group signature-based federated learning method by a group signature-based federated learning system including at least one client terminal and an administrator terminal provided as a group, and a server distributing a federated learning model to the at least one client terminal may comprise requesting, by the client terminal, a group signature to the administrator terminal; transmitting, by the administrator terminal, a group signature response to the client terminal in response to the request for the group signature; generating, by the client terminal, update information of signed federated learning using the group signature response and transmitting it to the server to request participation in federated learning update; transmitting, when the server receives the request for participation in the federated learning update, a latest group signature request to the administrator terminal and receiving a latest group signature response from the administrato

Abstract: A method for a deobfuscation apparatus that deobfuscates a malicious program obfuscated using an obfuscation technique, and the deobfuscation method comprises executing the malicious program to identify and extract memory information containing a trampoline code used in the obfuscation technique, executing the trampoline code based on the memory information to classify a type of obfuscation technique of the malicious program, and deobfuscating the malicious program according to the classified obfuscation technique and generating a deobfuscation program. According to the constitution, the techniques for obfuscating OEP and IAT can be deobfuscated.

Abstract: A method of detecting malicious behavior in a malicious behavior detection apparatus that detects malicious behavior by analyzing the code of a program, comprises generating a first control flow graph (CFG) by performing dynamic analysis on the program, generating a second CFG by extracting a code block that is likely to be executed by a conditional branch instruction included in the program, and comparing the first CFG and the second CFG to detect hidden code included in the program, and detecting a malicious behavior of the program by determining whether the detected hidden code is a malicious code. According to the constitutions, the time it takes to discover and respond to new types of malicious code can be shorten by extracting hidden code related to the malicious behavior and detecting malicious behavior.

Abstract: The present invention is to predict physical properties of a new composition, especially changes in physical properties at ultra-high temperatures, and has the effect of selecting compositions with desired physical properties without going through direct experiments by predicting changes in physical properties of new compositions, especially at ultra-high temperatures, predicting changes in physical properties due to changes in composition at absolute zero, and predicting changes in physical properties at ultra-high temperatures using the QHA (Quasi-harmonic approaching) technique and the Debye approximation for the stable phase of the composition.

Abstract: The present invention relates to an activity recognition apparatus and method in sports videos using CGAM.

Abstract: A high-efficiency cordless hair dryer according to an exemplary embodiment of the present invention may include: a main body housing having, at a rear side thereof, an air inlet port through which air is introduced, having, at a front side thereof, a discharge port through which the air is discharged, and having an internal space in which a heating element is mounted; and a power supply module configured to supply power and detachably coupled to the rear side of the main body housing at which the air inlet port is formed, in which the power supply module is detachably coupled to the main body housing so as to be coincident with a direction of a flow of air in the main body housing.

Abstract: The method for detecting a deepfake video according to one embodiment of the present invention comprises receiving a target video, which is a video containing a person speaking; analyzing the direction of the person's face in time series for each of an image and voice constituting the target video; and determining whether the target video is a deepfake video by comparing analysis results of the image and the voice.

Abstract: The present disclosure relates to an apparatus for screening cathode active material candidates for secondary batteries, and may include a database constructing unit configured to receive a data-set labeled with properties of a cathode active material structure for secondary batteries; a pre-processing unit configured to pre-process a part of the data-set to a learning data-set; a prediction model generating unit configured to generate a cathode active material prediction model for predicting performance indicators of target materials that may be arranged to fit a predetermined structure based on the learning data-set; and a candidates generating unit configured to generate cathode active material candidates for secondary batteries based on a result of the cathode active material prediction model.

Abstract: Provided is a method for providing information-centric networking (ICN) service based on a 5G network, and the method comprises transmitting, by a producer server, a content name prefix to all ICN routers constituting an ICN data network (DN), requesting, by an ICN gateway (GW) connected to the 5G network among the ICN routers, registration of the content name prefix from an ICN service/network controller, and generating, by the ICN service/network controller, ICN application configuration information including an internet protocol (IP) address of the ICN GW and the content name prefix, and registering the generated ICN application configuration information to an ICN application function (AF).

Abstract: Provided is a method for providing a routing path of a producer server in an information-centric networking system, and the method comprises measuring a routing cost for each of a plurality of routing schemes based on a transition cost between routing schemes, an update packet transfer cost, an interest packet transfer cost, and a data packet transfer cost at each predetermined period Tk, estimating a routing cost of period Tk based on an estimated routing cost and a measured routing cost in period Tk-1 for each of the routing schemes, selecting a routing scheme having a minimum estimated routing cost for each of the routing schemes, and providing a routing path between the producer server and the user equipment by applying the selected routing scheme.

Abstract: A method for detecting a mobile malicious application based on an implementation feature in a mobile malicious application detection apparatus based on an implementation feature and the method comprises decompiling a labeled application to remove preset information; extracting abstract syntax tree (AST) that is an implementation feature for each method; generating an AST node list; generating and vectorizing the generated AST node list as a learning dataset for deep learning; generating a classification model by learning a vectorized learning dataset; and outputting a classification result of a target application based on the classification model. This can reduce the false positive rate, extract many features from the obfuscated application, and detect malicious applications by classifying mobile applications as normal or malicious behaviors based on the behaviors performed by the application.

Abstract: A system for constructing a database for a structure and a property of a material based on big data includes an input module that receives chemical information about the material based on the big data, a controller that obtains information about the structure and the property of the material by modeling the chemical information about the material based on quantum mechanics and calculating elastic constants, a database constructing module that constructs a database for the information about the structure and the property, and an output module that outputs the information about the structure and the property in the database with respect to a model selected through the modeling.

Abstract: An apparatus for providing sodium superion conductor material information using a machine learning method is provided. The apparatus includes a classification unit for generating classification information by classifying sodium superion conductor material information by comparing an output value obtained by inputting material characteristic information into a sodium superion conductor material search model with a preset threshold value: a first verification unit for selecting information on a structure in an electrochemically stable state using Ewald summation from the generated classification information; and a second verification unit for selecting and providing final information on a structure having high ionic conductivity using AIMD simulation from the information on the structure in the electrochemically stable state.

Abstract: Provided is a unit test case-based security design flaw detection method performed in a security design flaw detection apparatus for detecting a security design flaw of a software system, and the method comprises collecting a unit test case for the software system from an external device and preprocessing the unit test case; generating a first test case by testing whether the software system violates a security policy using the preprocessed unit test case; generating a second test case that is a data set for testing a function of the software system based on the first test case; and detecting a vulnerability of the software system by executing the second test case.

Abstract: Provided are a content caching optimization system and method. The content caching optimization system in which content priority in an information-centric networking (ICN) environment is taken into consideration includes one or more producer terminals configured to generate and provide content, one or more user terminals configured to transmit content requests according to users and receive content according to the content requests, one or more mobilelmultiple access edge cotnputings (MECs) configured to predict the number of requests for each piece of content to be requested later on the basis of the content requests received from the user terminals, and a software-defined network (SDN) controller configured to calculate a content popularity using the number of requests for each piece of content predicted by the MECs and perform content caching optimization on the basis of the calculated content popularity and a preset content priority.

Abstract: The present invention provides a high-purity glucose-based compound ultra-highly substituted with a biopolymer, in which a surface of a glucose-based compound as a natural material is ultra-highly modified with a biopolymer to improve compatibility and dispersibility with biodegradable polymers, thereby significantly improving the mechanical properties of the biodegradable polymer composite.

Abstract: Provided is a method for detecting a synthetic voice based on a biological sound, the method comprises receiving an audio stream; extracting a biological feature vector corresponding to a meaningless voice from the audio stream; extracting a synthetic voice feature vector from the audio stream; combining the biological feature vector and the synthetic voice feature vector to generate a combined feature vector; and determining whether the audio stream is a synthetic voice based on the combined feature vector. Accordingly, it is possible to detect a synthetic voice at a lower computational cost than a conventional neural network that detects a synthetic voice by learning the correlation between frames.

Abstract: Proposed is an all-solid-state battery including a self-standing sheet layer positioned on a negative electrode current collector. The all-solid-state battery may include the negative electrode current collector, the self-standing sheet layer positioned on the negative electrode current collector, a solid electrolyte layer positioned on the sheet layer, a positive electrode active material layer positioned on the solid electrolyte layer, and a positive electrode current collector positioned on the positive electrode active material layer. The sheet layer contains carbon nanotubes.

Abstract: Provided is an eBPF-based hot patch engine device for protecting kernel vulnerabilities. The eBPF-based hot patch engine device comprises a container-aware code generating unit for generating a container-aware code for identifying a target container, to which a hot patch is attached; and a hot patch configuring unit for configuring an eBPF-based hot patch code for attaching a hot patch to the target container based on the container-aware code. Accordingly, it is possible to prevent attacks based on CVEs, which are known vulnerabilities for container systems, by hot patching kernel-related CVEs at runtime without rebooting and freezing.