Date of Award

3-10-2021

Publication Type

Master Thesis

Degree Name

M.A.Sc.

Department

Electrical and Computer Engineering

First Advisor

Roozbeh R.R. Razavi-Far

Second Advisor

Mehrdad M.S. Saif

Keywords

Imputation, Missing tolerance, V2X communication

Rights

info:eu-repo/semantics/openAccess

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Abstract

Misbehavior detection is a critical task in vehicular ad hoc networks. However, state-of-the-art data-driven techniques for misbehavior detection are usually conducted through complete V2X communication data collected from simulated experiments. This thesis evaluates the main strategies for the treatment of missing values to find out the best match for misbehavior detection with incomplete V2X communication data. This thesis proposes three novel methods for imputing and tolerating missing data. The first two are novel imputation methods that are based on cooperative clustering and collaborative clustering. The latter is a missing-tolerant method that is an ensemble learning based on the random subspace selection and Dempster-Shafer fusion. The effectiveness of the proposed techniques is evaluated in the ground truth vehicular reference misbehavior data. Moreover, a multi-factor amputation framework has been developed to induce missingness over V2X communication data with different missing ratios, mechanisms, and distributions. This framework provides a comprehensive benchmark resembling missingness over V2X communication data. The proposed methods are compared with some missing-tolerant and imputation methods. The attained results over benchmark data are analyzed and indicated the winner treatments in each aspect.

Available for download on Wednesday, March 16, 2022

Share

COinS