BCI based password validation system

Project Title

BCI based password validation system

Project Area of Specialization NeuroTechProject Summary

An authentication system is the system that decides

whether to accept or reject the claiming identity of a person.

Biometric-based authentication utilizes the individuality in

human physiological and behavioral characteristics to authorize

a person. Brain-signal-based authentication system is relatively

new comparing to other types of biometric data. In this project, we

proposed a novel method that applies P300-based Brain

Computer Interface (BCI) technique to the authentication system.

The main concept for P300-BCI-based authentication is that the

Oddball paradigm eliciting P300 waves is secret to the attacker.

The  P300 classification model has 0.831 accuracy rate.

And the proposed authentication system has 0.325 False

Rejection Rate (FRR), 0.00 False Acceptation Rate (FAR) for

secret-unknown attack and 0.10 FAR for secret-known attack after traning.

This project will be showing that P300 wave has good potential as a

biometric for highly secured authentication system.

Project Objectives

The Objective are as follows:

To create a system that decides whether to accept or reject the claiming identity of a person.

To Authenticate identity by brain signals.

To make system safe from attackers.

Project Implementation Method

This proposed  project have following method:

1 )  P300-based BCI

P300 is a positive ERP   that occurs in the scalp-recorded

EEG after a stimulus that is delivered under a specific set of

circumstances. P300 latency may vary from 250 ms to 750 ms

from onset of the stimulus and it is strongest in parietal area of

human brain. The set of circumstances that elicit P300 ERP is

known as Oddball paradigm in which a subject is presented

with a series of 2-classes stimuli where the low-probability

target stimuli are mixed with high-probability non-target

stimuli. The low-probability target stimuli elicit a P300. The

most common use of P300-based BCI is P300 speller, where

the desired character works as the low-probability target stimuli

thus P300 waves can be detected and the user is able to type

words without using any kind of movement.

2 ) P300-BCI-Based Authentication System

The main idea for the proposed method, P300-BCI-based

authentication system, is that the Oddball paradigm is secret

only the client know. In other words, given the same sequence

of stimuli to both client and imposers, only client will be able

to distinguish the low-probability target stimuli from the high-

probability non-target stimuli. In this project, the stimuli are

pictures of person and the low-probability targets are pictures

of client’s known people. Knowing this setting, client’s P300

wave can be detected from watching a sequence of stimuli in

Oddball paradigm and used as data to authorize the system. In

contrast, the same sequence specifically made for the client

would be perceived as just pictures of random person to the

imposers and no P300 wave would elicit.

The system begins by having the user register to the

system in which user has to provide the username (notated as

IDX for the user X) and N target pictures . The

system then generates sequences containing pictures of random

person randomly mixed with target pictures due to Oddball

paradigm for users to perform P300 BCI. The user’s EEG

signal responded to each of stimuli is extracted, preprocessed

and used to train the two-class (non-P300 response and P300

response) classification model using supervised machine

learning technique. Finally, user’s target pictures, trained P300

classification model along with the registered ID are saved to

the system database. when a person uses authentication system. Given an identity-unknown person, Y,

with the claiming identity, IDX, the system will pull the client

X’s target pictures and P300 classification model from the

database to construct the P300-BCI authentication system. If Y

is indeed the client, P300 signals can be detected and the

system will accept Y as X, the true owner of identity IDX.

Benefits of the Project

The benifits of project are as follows :

Oddball paradigm eliciting P300 waves is secret to the attacker so less possiblity of being attacked.

The authentication accuracy for False Rejection Rate (FRR) and False Acception Rate (FAR) would be much better then other authentication systems.

P300 wave has good potential as a biometric for highly secured authentication system.

System utilizes the individuality in human physiological or behavioral characteristics in order to authorize a person. It provides a much more reliable user authentication than the password-based authentication system.

Brain signal is almost impossible to mimic since  it  is  unique to each individual person so It’s less likely to steal or force a person to authorize the system as the brain activity is sensitive to stress and mood of the person.

Technical Details of Final Deliverable

finally we will be going to design and implement brain computer interaction password validation system,

we proposed a biometric-based authentication

system that use P300 waves eliciting due to the target stimuli

in Oddball paradigm as the data.

P300 wave has good potential as a

biometric and P300-BCI-based authentication system is a

promising authentication system that, with some minor

improvement, could be used as an authentication system in the place that requires high security that includes bank account of a person etc.

Final Deliverable of the Project HW/SW integrated systemCore Industry OthersOther Industries IT , Security Core Technology NeuroTechOther TechnologiesSustainable Development Goals Sustainable Cities and CommunitiesRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	77112
EEG sensor	Equipment	1	67112	67112
publications	Miscellaneous	2	5000	10000