Grey model and polynomial regression for identifying malicious nodes in MANETs

Author

Silva, Anderson A. A. ; Pontes, Elvis ; Fen Zhou ; Kofuji, Sergio Takeo

Author_Institution

Univ. Paulista, Sao Paulo, Brazil

fYear

2014

fDate

8-12 Dec. 2014

Firstpage

162

Lastpage

168

Abstract

Nodes positioning is an essential issue for diverse applications in Mobile Ad Hoc Networks (MANETs). However, besides misbehaving nodes that could cause power depletion, MANETs are also susceptible to cyber-attacks, which can make the network unstable and/or unavailable. Therefore, considering the gaps aforementioned, the goal of this paper is to propose a model for identifying malicious/misbehaving nodes by: (1) the use of two forecasting methods (Grey Model and Polynomial Regression); (2) variability analysis; and (3) simulation of fake node positions. The obtained results allow concluding our model has high rate of accuracy for detecting malicious/misbehaving nodes.

Keywords

forecasting theory; grey systems; mobile ad hoc networks; polynomials; regression analysis; MANET; cyber-attack; fake node position simulation; forecasting method; grey model; malicious-misbehaving node identification; mobile ad hoc network; polynomial regression; power depletion; variability analysis; Ad hoc networks; Forecasting; Mathematical model; Mobile computing; Polynomials; Predictive models; Time series analysis; Grey Theory GM(1, 1); MANET; Polynomial Regression; malicious node identification; misbehaving node detection; prediction model;

fLanguage

English

Publisher

ieee

Conference_Titel

Global Communications Conference (GLOBECOM), 2014 IEEE

Conference_Location

Austin, TX

Type

conf

DOI

10.1109/GLOCOM.2014.7036801

Filename

7036801