NewMiniprojectdoc (1) 1
NewMiniprojectdoc (1) 1
Bachelor of Technology
in
Computer Science and Engineering
Submitted by
2021- 2025
CMR COLLEGE OF ENGINEERING & TECHNOLOGY
KANDLAKOYA, MEDCHAL ROAD, HYDERABAD – 501401
CERTIFICATE
This is to certify that the Mini Project I report entitled "Securing the Testing
Environment: The USB and Wi-Fi Management System" being submitted by
Gajulapalle Sree Lakshmi (21H51A0504), G. Pavani (21H51A0505), J. Akansh
(21H51A0506) in partial fulfillment for the award of Bachelor of Technology in
Computer Science and Engineering is a record of bonafide work carried out
his/her under my guidance and supervision.
The results embodies in this project report have not been submitted to any
other University or Institute for the award of any Degree.
With great pleasure we want to take this opportunity to express my heartfelt gratitude to all
the people who helped in making this project work a grand success.
We are grateful to, Dr. Siva Skandha Sanagala, Associate Professor and Head of
Department of Computer Science and Engineering for his valuable technical suggestions and
guidance during the execution of this projectwork.
We would like to thank Dr. Siva Skandha Sanagala, Head of the Department of Computer
Science and Engineering, CMR College of Engineering and Technology, who is the major driving
forces to complete my project work successfully.
We are very grateful to Dr. G. Devadas, Dean-Academics, CMR College of Engineering
and Technology, for his constant support and motivation in carrying out the project work
successfully.
We are highly indebted to Major Dr. V A Narayana, Principal, CMR College of
Engineering and Technology, for giving permission to carry out this project in a successful and
fruitful way.
We would like to thank the Teaching & Non- teaching staff of Department of Computer
Science and Engineering for their co-operation
We express our sincere thanks to Shri. Ch. Gopal Reddy, Secretary, CMR Group of
Institutions, for his continuous care.
Finally, We extend thanks to our parents who stood behind us at different stages of this
Project. We sincerely acknowledge and thank all those who gave support directly and indirectly in
completion of this project work.
REFERENCES 32
GitHub Link 33
List of Figures
FIGURE
NO. TITLE
1 Goals of ML Based approach 21
2 Graph 28
3 Essentials of proposed system 32
List of Tables
FIGURE
NO. TITLE PAGE NO.
ABSTRACT
Our mini-project aims to enhance the integrity of university and college practical and
other computer-based exams by implementing robust security measures in computer labs. We
are developing a solution that selectively disables USB ports (except for keyboards and mouse)
and temporarily suspends Wi-Fi connectivity during exams. This prevents students from using
external devices or accessing online resources, ensuring a fair testing environment. Our user-
friendly interface allows easy configuration and management of these restrictions, supporting
both Windows and Linux systems. By maintaining academic integrity, our project promotes
fairness and honesty in examinations
The selective disabling of USB ports prevents students from using external storage
devices, ensuring that they cannot access unauthorized materials or bring in pre-prepared
content to the exam environment. Temporarily suspending Wi-Fi connectivity is crucial in
preventing students from accessing online resources, Eliminating the risk of communication or
collaboration during the exam period. Admininstrators can easily navigate through the interface
to set specific parameters for USB port control, define the duration of Wi-Fi suspension, and
customize the security settings to the unique requirements of the exam.
CHAPTER 1
INTRODUCTION
CHAPTER 1
INTRODUCTION
Creating a secure and integrity-focused testing environment is crucial for fair examinations.
.
Motivated by the goal to prevent students from using external devices or accessing online
resources during exams, promoting a testing environment that upholds academic integrity and
contributed to fairness in examinations. Designing a solution to enhance the integrity in
examinations
Develop a secure exam integrity solution using Tkinter, Pyinstaller, and Python to prevent
cheating in university and college practical and computer-based exams.
Implement a user-friendly program for both Windows and Linux platforms that selectively
disables or connect to the internet during tests.
Enhance the overall integrity of exams an maintains fairness by leveraging technology to
prevent cheating in academic assessments.
• Cross-Platform Compatibility:
Supporting both Windows and Linux systems ensures the applicability of the solution in
diverse computer lab environments.
• User-Friendly Interface:
Development of a user-friendly interface for easy configuration and management of security
restrictions, facilitating efficient use by administrators and exam proctors.
• USB Port Control:
The project aims to selectively disable USB ports, allowing only essential peripherals, such as
keyboards and mice, to be connected during exams.
Project limitaions:
• Dependency on Physical Access:
The project primarily relies on controlling physical access to USB ports and
temporarily suspending Wi-Fi. Students with advanced technical knowledge might
find ways to circumvent these restrictions, particularly if they have physical access to
the computer before the exam.
• Dynamic Cheating Methods:
The project may face challenges in keeping pace with evolving cheating methods and
technologies. Continuous updates and improvements are essential to address emerging
threats effectively.
• Scalability Challenges:
While the project is designed to be adaptable, there may be challenges in scaling the
solution to accommodate larger institutions or exams with diverse requirements.
Ensuring seamless scalability requires careful consideration of various factors.
CHAPTER 2
BACKGROUND WORK
CHAPTER 2
BACKGROUND WORK
2.1 Machine Learning Based approach to exam cheating detection
2.1.1 Introduction
The paper introduces a unique approach using machine learning, combining recurrent
neural networks and outlier detection, to identify anomalous scores on final exams.
The proposed method involves analyzing student assessments, including quizzes and
midterms, using outlier detection after processing data with a recurrent neural network.
The algorithm aims to provide a robust tool for identifying potential cases of exam fraud,
contributing to global efforts to maintain academic integrity in remote learning
environments.
ML models may inherit biases from training data, leading to unfair detections, and lack
transparency, making it challenging to understand decision-making processes.
ML models may produce false positives, incorrectly flagging normal behavior, or false
negatives, failing to detect actual cheating due to the complexity of student behavior and diverse
learning environments.
Challenges:
The algorithm for detecting potential cases of cheating on the final exam. The inputs to the
algorithm are sequences of grades—quizzes, midterm exam, the final exam—of an entire
class, while the output is a collection of labels—one label per student—indicating whether
each student cheated or not. The proposed method consists of two parts: regression and
unsupervised outlier detection. First, a recurrent neural network model is trained to predict
the final exam scores based on the previous assessment scores. Then an outlier detection
model is applied to identify the instances where the difference between the actual and the
predicted final exam scores is abnormal. Since the input data is unlabeled, the proposed
method is an unsupervised algorithm.
Anomaly Detection:
Algorithm:
1. Train an LSTM-based regression model on the given dataset.
a. The input features (X) are the scores prior to the final exam, i.e. quizzes, the
midterm exam, project, and others.
2. Calculate the error between the scores predicted by the trained model and the actual
exam scores. Apply a KDE-based outlier detection method on the set of errors.
.
CMRCET B.Tech(CSE) Page No. 7
Securing the Testing Environment: The USB and Wi-Fi Management system
2.2 Proctorio
2.2.1. Introduction
Proctorio views each proctored exam, online or in-person, as a step closer to learners achieving
their academic and career goals. That's why this platform makes proctored assessments
accessible to anyone, at any time, anywhere. Securing exam integrity is essential to protecting
the reputation of the institution and the success of test takers.
Proctorio’s online proctoring, identity verification, plagiarism detection tools, and content
protection tools make up the digital tool box within your assessment platform to ensure your
online assessment’s integrity.
Merits:
• Proctorio encrypts collected exam recordings and doesn't require additional personally-
identifiable information beyond what is required by the assessment platform.
• This software is largely easy to use on any computer, quick to install, and
straightforward in install and set up.
Demerits:
• Proctorio can sometimes “bug out” with chrome and have issues starting and functioning
during tests. The issues are few, but can cause devastation for some students
Challenges:
• Privacy concerns:
The intrusive nature of monitoring, including webcam and microphone access, raises
significant privacy concerns among students.
• Technical Issues:
Users often encounter technical challenges with Proctorio, such as software compatibility
issues, installation difficulties, and occasional glitches during exams.
The application uses facial detection capability to capture students' IDs and images, record them
in Proctorio Gradebook, allowing instructors to review and flag anomalies. It provides a data
analytics module, which allows administrators to spot trends, create usage reports, filter activities
based on departments or courses and manage exam parameters or faculty controls. Proctorio
provides an application programming interface, which allows developers to integrate the system
with several third-party applications such as Blackboard, Canvas, Questionmark.
2.3. The Detection of Students' Abnormal Behavior in Online Exams Using Facial
Landmarks in Conjunction with the YOLOv5 Models
2.3.1 Introduction
The research focuses on enhancing exam integrity through proctoring remote online exams using
computer vision and artificial intelligence. The proposed system employs You Only Look Once
(YOLO) models and Dlib facial landmarks to identify abnormal student behavior, including facial
expressions, eye and hand movements, mouth opening, gaze direction, and mobile phone usage.
The "Behavioral Detection Model" achieved a mean Average Precision (mAP) of 0.87, while the
"Mouth Opening Detection Model" and "Person and Objects Detection Model" had accuracies of
0.95 and 0.96, respectively, demonstrating high detection accuracy. The study emphasizes the
effectiveness of using computer vision and deep learning models trained on a private dataset to
identify and prevent irregularities in student behavior during exams.
Merits:
• Enhanced Exam Security and Comprehensive Detection: The combination of facial
landmarks and YOLOv5 models provides a powerful tool for identifying abnormal
behaviors during online exams, contributing to a more secure testing environment.
• Customization with Private Dataset: Training the models on a private dataset ensures
that the system is tailored to the specific behaviors exhibited by the target student
population, improving accuracy and reducing the risk of false positives.
• High Detection Accuracy: Reported high accuracies, such as the mean Average
Precision (mAP) of 0.87 for the "Behavioral Detection Model" and accuracies of 0.95 and
0.96 for other models, indicate a reliable and effective detection system.
Demerits:
• Limited to Visible Behaviors: The system may not detect non-visible behaviors,
potentially allowing some forms of cheating or irregularities to go unnoticed.
Limitations in visible behaviors should be acknowledged.
Challenges:
• Adaptation to Diverse Behaviors: Ensuring the system can adapt to a diverse range of
behaviors and account for cultural differences in expressions and gestures poses a
challenge in creating a universally effective solution.
• Constant Model Updating: Staying ahead of new cheating methods and adapting the
system to evolving behaviors requires continuous model updates and improvements,
demanding ongoing resources and attention.
Preprocessing
Frames Behavioral
(640x640x3) Analyses Model
Report &
Persons & Statistical
Objects Detect Information
Model Generator
Mouth Open
Detect Model
The pool of questions includes multiple-choice options. During the acquisition of the dataset, the
six participants were required to engage in various deliberate cheating scenarios in the context of
online and closed-book exams throughout the videos; these motions included utilizing a phone,
moving a hand or an eye, and shifting the head left or right. Participants joined the trials while
being exposed to various test locations, camera settings, lighting conditions, etc. These changes
make it challenging to catch cheating incidents. The videos mimic a student completing an online
test in front of a webcam. The overall running time of all films is roughly six hours, with each
video lasting roughly fifteen minutes for each scenario.
CHAPTER 3
PROPOSED
SYSTEM
CHAPTER 3
PROPOSED SYSTEM
Fairness in Technological
assessment Adaptance
Securing the environment through USB and Wi-Fi controls contributes to a more controlled and
predictable testing environment.
1.Regulatory Compliance:
Educational institutions must adhere to regulations and policies governing exam security and
integrity. Implementing measures such as disabling USB ports and managing Wi-Fi access
ensures compliance with there regulatory requirements.
4. Fairness in Assessment:
By preventing technology-enabled cheating, institutions ensure a fair assessment process for all
students. Disabling USB ports and managing Wi-Fi access contribute to a level playing field,
where each student is evaluated based on their own knowledge and skills.
5. Technological Adaptation:
The implementation of measures like USB disabling and Wi-Fi access management reflects the
educational institutions' adaptation to the evolving technological landscape. It demonstrates a
proactive stance in addressing the challenges posed by technology in maintaining the fairness and
security of exams.
Testing Environments:
The Operating system and Testing server – An operating system (OS) is nothing but a program
that you load into the computer through a boot program. It would manage every application
program on your computer.
The boot is derived from bootstrap, a small strap or loop at the back of a leather boot that enables
you to pull the boot on.
The Booting of an OS works by loading a small program into the computer and then giving that
program control so that it, in turn, loads the entire operating system.
Testing Environment
System32 is the system folder of the Windows operating system. It is the nerve centre of the
Windows operating system, containing many files. Besides, System32 is the folder that holds
the main files of the operating system. It is usually created automatically when the operating
system is installed, and the relevant files are placed in the appropriate folders.
Function: The primary function of system 32 holds is saving critical files and folders intending
to run the computer well. It is very impactful to the Windows operating system.
In Windows it we can either use python gui interface , or windows power shell.
It would be very easier to call funtions of powershell script rather than writing the whole
source in python. In Windows the all drivers have to diasabled or else none of them , but
partially disabling is not possible. It would be possible only in linux.
3.3 Designing
Software Design Process:
Software Design is the process to transform the user requirements into some suitable
form, which helps the programmer in software coding and implementation.
The Software Design process can be divided into the following 3 levels of phases of design:
1. Interface Design
2. Architectural Design
1. Interface Design :
Command Interface:
In our project we are using command interface by invoking the cmd command admin
in windows. The command prompt can be opened with the command windows key + R.
Fig7 User Interface design
2. Architectural Design:
Start
Python Source
Code
GUI User
Interface Using
Tkinter
Network
USB
Interface
Drivers
End
System1
System 2
Examination System
System 3
System 4
The steps involved in implementing the idea of a establishing a secure exam environment involves:
• Interfacing the python code and GUI using tkinter.
Import
tkinter
Module
Initialize the
Window
manager
Define the
Label
Widget
Geometry
Management
using pack
This python code is saved in windows C > System 32 files. Now the command prompt is been
invoked.
To run it as an administrator. Now a new window admin gets opened in this we run our python
code.
• Network Interface and USB Driver Interface:
The Network and the usb drivers gets enabled and disabled as per the action.
GUI Components:
Labels:
Labels are used to display text or describe the purpose of other GUI elements.
Buttons:
Buttons trigger actions when clicked. They can be used to submit a form, open a dialog, or
perform various other functions.
Window - Rectangular section of the computer's display that shows the program currently
being used
Dialog box - A type of window that displays additional information and asks a user for input
Code:
import tkinter as tk
from tkinter import messagebox
class ExamIntegrity:
def __init__(self, root):
self.root = root
self.root.title("Exam Integrity Control Panel")
self.usb_frame = tk.Frame(root)
self.usb_frame.grid(row=0, column=0, padx=10, pady=10)
tk.Label(self.usb_frame, text="USB Controls").pack()
tk.Button(self.usb_frame, text="Enable USB", command=self.enable_usb).pack(pady=5)
tk.Button(self.usb_frame, text="Disable USB", command=self.disable_usb).pack()
self.wifi_frame = tk.Frame(root)
self.wifi_frame.grid(row=0, column=1, padx=10, pady=10)
tk.Label(self.wifi_frame, text="Wi-Fi Controls").pack()
tk.Button(self.wifi_frame, text="Enable Wi-Fi", command=self.enable_wifi).pack(pady=5)
tk.Button(self.wifi_frame, text="Disable Wi-Fi", command=self.disable_wifi).pack()
self.ethernet_frame = tk.Frame(root)
self.ethernet_frame.grid(row=0, column=2, padx=10, pady=10)
tk.Label(self.ethernet_frame, text="Ethernet Controls").pack()
tk.Button(self.ethernet_frame, text="Enable Ethernet",
command=self.enable_ethernet).pack(pady=5)
tk.Button(self.ethernet_frame, text="Disable Ethernet", command=self.disable_ethernet).pack()
def enable_usb(self):
messagebox.showinfo("USB Enabled", "USB ports are enabled.")
def disable_usb(self):
messagebox.showinfo("USB Disabled", "USB ports are disabled.")
def enable_ethernet(self):
result = subprocess.run('netsh interface set interface "ethernet" enable', check=False, shell=True,
capture_output=True, text=True)
if result.returncode == 0:
messagebox.showinfo("Ethernet Enabled", "Ethernet is enabled.")
else:
messagebox.showerror("Error", f"Failed to enable Ethernet:\n{result.stdout}\n{result.stderr}")
def disable_ethernet(self):
result = subprocess.run('netsh interface set interface "ethernet" disable', check=False, shell=True,
capture_output=True, text=True)
if result.returncode == 0:
messagebox.showinfo("Ethernet Disabled", "Ethernet is disabled.")
else:
messagebox.showerror("Error", f"Failed to disable Ethernet:\n{result.stdout}\n{result.stderr}")
def enable_wifi(self):
result = subprocess.run('netsh interface set interface "Wi-Fi" enable', check=False, shell=True,
capture_output=True, text=True)
if result.returncode == 0:
messagebox.showinfo("Wi-Fi Enabled", "Wi-Fi is enabled.")
else:
messagebox.showerror("Error", f"Failed to enable Wi-Fi")
def disable_wifi(self):
result = subprocess.run('netsh interface set interface "Wi-Fi" disable', check=False, shell=True,
capture_output=True, text=True)
if result.returncode == 0:
messagebox.showinfo("Wi-Fi Disabled", "Wi-Fi is disabled.")
else:
messagebox.showerror("Error", f"Failed to disable Wi-Fi")
if __name__ == "__main__":
root = tk.Tk()
app = ExamIntegrity(root)
root.mainloop()
Code :
Importing Libraries:
subprocess: Used to run system commands.
tkinter: GUI toolkit for creating the graphical user interface.
messagebox: Part of tkinter to display message boxes.
GUI Layout:
The GUI is divided into three frames, each controlling a specific aspect (USB, Wi-Fi, Ethernet).
Buttons within each frame trigger corresponding functions for enabling or disabling the respective
component.
CHAPTER 4
RESULTS AND
DISCUSSION
CHAPTER 4
RESULTS AND DISCUSSION
Administrator window:
Actions Invoked
Future Scope and Modifications / Extensions which can be done to our project:
The expand the functionality of this USB and Wi-Fi access management system, here are
some future modifications you could consider:
there are:
1.User Authentication:
Implement a user authentication system, so only authorized users can enable/disable USB
ports or Wi-Fi.
You might consider integrating this with existing authentication systems like LDAP or
Active Directory.
2.Security Enhancements:
Implement additional security measures, such as encryption, to protect communication and
prevent potential vulnerabilities.
3.Scalability:
Design the system to scale efficiently, accommodating growth in the number of users,
devices, or network complexity.
CHAPTER 5
CONCLUSION
CHAPTER 5
CONCLUSION
In conclusion, our project enhances exam integrity by restricting unauthorized access, ensuring
consistent and stable performance in various environments with a user-friendly interface.
Operates smoothly on Windows and Linux without disrupting normal computer use during
exams, ensuring secure and rule-abiding assessments.
Signifies substantial potential for academic institutions seeking fair and secure exam
environments.
In simple terms, our project is a powerful tool for schools and colleges that want to make sure
exams are fair, secure, and up-to-date with technology. It deals with the challenges of modern
exams and provides a practical solution that is easy to use and can be adjusted to fit different
needs. As schools adapt to digital exams, our project becomes a trustworthy and important tool
for keeping academic standards high and ensuring honesty in assessments.
REFERENCES
REFERENCES
https://www.researchgate.net/publication/368847353_An_Improved_Authentication_and_Monitoring_System_fo
r_E-Learning_Examination_Using_Supervised_Machine_Learning_Algorithms
https://www.researchgate.net/publication/349458625_A_Systematic_Review_of_Online_Exams_Solutions_in_E
-Learning_Techniques_Tools_and_Global_Adoption
(PDF) The Detection of Students' Abnormal Behavior in Online Exams Using Facial Landmarks in
Conjunction with the YOLOv5 Models (researchgate.net)
GitHub Link