Electronics and Telecommunications Research Institute

Yang Seo Choi

Ik Kyun Kim

Jong Soo Jang

Dae Won Kim

Jin Tae Oh

Jong Hyun Kim

Jung Tae Kim

Dae Sung Moon

Min Ho Han

A behavior-based malicious code detecting apparatus and method using multiple feature vectors is disclosed. A malicious code learning method may include collecting characteristic factor information when a training target process comprising a malicious code is executed, generating a feature vector for malicious code verification based on the collected characteristic factor information, learning the generated feature vector through a plurality of machine learning algorithms to generate a model of representing the malicious code and a model of representing a normal file, and storing the model of representing the malicious code and the model of representing the normal file generated through the learning.

Behavior-based malicious code detecting apparatus and method using multiple feature vectors

Disclosed are provided a method and a system for network connection chain traceback by using network flow data in order to trace an attack source site for cyber hacking attacks that goes by way of various sites without addition of new equipment of a network or modification a standard protocol when the cyber hacking attack occurs in the Internet and an internal network.

Method and system for network connection chain traceback using network flow data

There are provided an apparatus and method for detecting an executable code, capable of verifying reliability of an extracted signature by determining whether there is present an executable code in network data by using instruction pattern information related calling mechanism of function for distinguishing the executable code from a non-executable code, the method including: forming instructions by reverse assembling network data suspicious as an attack; comparing the respective formed instructions with instruction patterns according to calling mechanism of function; and determining whether there is present an executable code in the network data according to a result of the comparing.

Method and apparatus for detecting executable code

There are provided a network attack detection apparatus and method capable of determining even unknown network attack, the apparatus connected between two networks or connected by port mirroring of an Ethernet switch to real-time monitor all packets flowing through the networks. The apparatus decodes a payload portion of an inputted network packet into a machine code instruction, determines whether an executable code is included in the decoded machine code by analyzing relationship between instructions, and determines whether the packet is harmful based on statistics with respect to a possibility that an executable code exists in a service and a certain transaction of the service when the executable code is included.

Growing community of inventors

Daejeon, South Korea

Yang Seo Choi