Wi-Fi Method and Apparatus of Intrusion Detection forWi-Fi NetworkBased on Weight-Selected Neural Networks

Presented are a method and a system of intrusion detection for Wi-Fi based on weight selection for neural network. The method proposed in the present invention comprises the steps of: selecting strong points using a decision tree and artificial neural network based on a weight according to data stan...

Full description

Saved in:
Bibliographic Details
Main Authors MUHAMAD ERZA AMINANTO, TANUWIDJAJA HARRY, KWANGJO KIM, CHOI RAKYONG
Format Patent
LanguageEnglish
Korean
Published 26.07.2019
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:Presented are a method and a system of intrusion detection for Wi-Fi based on weight selection for neural network. The method proposed in the present invention comprises the steps of: selecting strong points using a decision tree and artificial neural network based on a weight according to data standardization and threshold adjustment; classifying the strong points selected by using the decision tree and artificial neural network using an artificial neural network classification unit; and calculating IDS matrix for the classified strong points and testing an actual network using the classified strong points. 신경망에 대한 가중치 선택 기반 Wi-Fi 침입 탐지 방법 및 시스템이 제시된다. 본 발명에서 제안하는 신경망에 대한 가중치 선택 기반 Wi-Fi 침입 탐지 방법은 데이터 표준화 및 임계값 조정에 따른 가중치에 기초하고, 결정 트리 및 인공신경망을 이용하여 특장점을 선택하는 단계, 결정 트리 및 인공신경망을 이용하여 선택된 특장점을 인공신경망 분류부를 이용하여 분류하는 단계 및 분류된 특장점에 대하여 IDS 행렬을 계산하고, 상기 분류된 특장점을 이용하여 실제 네트워크로 테스트 하는 단계를 포함한다.
Bibliography:Application Number: KR20180006369