Experimentation Using Short-Term Spectral Features for Secure Mobile Internet Voting Authentication

Surendra Thakur,Emmanuel Adetiba,Richard Millham,Oludayo O Olugbara

doi:10.1155/2015/564904

Abstract

We propose a secure mobile Internet voting architecture based on the Sensus reference architecture and report the experiments carried out using short-term spectral features for realizing the voice biometric based authentication module of the architecture being proposed. The short-term spectral features investigated are Mel-Frequency Cepstral Coefficients (MFCCs), Mel-Frequency Discrete Wavelet Coefficients (MFDWC), Linear Predictive Cepstral Coefficients (LPCC), and Spectral Histogram of Oriented Gradients (SHOGs). The MFCC, MFDWC, and LPCC usually have higher dimensions that oftentimes lead to high computational complexity of the pattern matching algorithms in automatic speaker recognition systems. In this study, higher dimensions of each of the short-term features were reduced to an 81-element feature vector per Speaker using Histogram of Oriented Gradients (HOG) algorithm while neural network ensemble was utilized as the pattern matching algorithm. Out of the four short-term spectral features investigated, the LPCC-HOG gave the best statistical results withRstatistic of 0.9127 and mean square error of 0.0407. These compact LPCC-HOG features are highly promising for implementing the authentication module of the secure mobile Internet voting architecture we are proposing in this paper.

Highlights

Election is the process by which voters in a political entity elect leaders among competing candidates by casting of votes either on a ballot paper or electronically in order to actualize desired changes in their society
An immediate problem with the first generation Digital Recording Electronic (DRE)/Electronic Voting Machine (EVM) was that the votes were captured and a black box result was produced by the lack of a paper trail
The spectral features we examined are MelFrequency Cepstral Coefficients (MFCCs), Mel-Frequency Discrete Wavelet Coefficients (MFDWC), Linear Predictive Cepstral Coefficients (LPCC), and Spectral Histogram of Oriented Gradients (SHOGs)

Summary

Introduction

Election is the process by which voters in a political entity elect leaders among competing candidates by casting of votes either on a ballot paper or electronically in order to actualize desired changes in their society. An immediate problem with the first generation DRE/EVM was that the votes were captured and a black box result was produced by the lack of a paper trail This led to Mercuri [10] proposing a Voter Verifiable Audit Trail (VVAT), which is a printed equivalent of the computer choice for voters. The ubiquity of the Internet and the global attainment of the mobile moment serve as a motivation to propose a secure mobile Internet voting architecture in this study This voting scheme being proposed will be a paradigm shift in e-voting and it can potentially mediate some of the highlighted challenges facing the traditional electoral processes and the conventional e-voting systems. The neural network ensemble was generally utilized as the pattern matching algorithm in the four experimental models that were set up to investigate the short-term spectral features in this study

Mobile Internet Voting

Secure Mobile Internet Voting Architecture

Theoretical Foundation for Speaker Recognition Authentication

Experimental Results

Speaker 7

Conclusion