WELCOME

TO
DEPARTMENT OF MASTER OF COMPUTER APPLICATIONS

.
ANNAMACHARYA P.G. COLLEGE OF COMPUTER STUDIES
NEW BOYANAPALLI, RAJAMPET - 516126.
ENCRYPTION AND DECRYPTION
ALGORITHM BASED ON NEURAL
NETWORK
Encryption and decryption algorithm based
on neural network

PRESENTED BY
UNDER THE GUIDANCE OF
D.AKBAR RASOOL,
ROLL NO:215N1F0003,
V. NIRMALA,
MCA-IV SEMESTER,
ASSITANT PROFESSOR,
APGCCS .RAJAMPET.
DEPT of MCA,
CONTENTS
Abstract  ER diagram
Introduction  UML diagram
 Testing
Existingsystem  Test cases
Disadvantages  Implementation
Proposed system  Sample screens
Advantages
 conclusion
Modules
System Architecture
Hardware requirements
Software requirements
Non functional requirements
ABSTRACT
A Neural Network is a machine which is designed to work like brain. It has the
capability to execute complex calculations easily. Cryptography is the interchange of
data between the users without seep of data to others.

Many public key cryptography are there which are based on numerical theory but it has
the limitation of availability of large analytical power, trigonometric and time utilization
during creation of key. To overcome these limitations, I am using self associative neural
network concept of soft computing in coordination with encryption technique to send
information securely on communication network.
Introduction
This project aims at removing the necessity for the coding to follow a general rule by
employing a neural network for cryptography the cipher text. Thus introducing the
randomness in secret writing creating it most harder to decrypt. I’ve conjointly
introduced the thought of as well as lies with in the data transmitted to misguide any
listener who manages to decipher the ciphertext.

Cryptography may be a science that involves two techniques secret writing and
decipherment and it essentially permits to send sensitive and confidential information
over the unsecure network. The root of the term "cryptography“, is came for the Greek
word “kryptos.” meaning of this word is hidden or secret.
Scope
The scope of an encryption and decryption algorithm based on neural network
involves designing and implementing a system that utilizes neural networks for
securing and processing sensitive data. This project entails researching existing
encryption algorithms, exploring the potential of neural networks in cryptography,
and designing a novel ANN-based algorithm.
Purpose
The purpose of an encryption and decryption algorithm based on neural network is to
provide a secure and efficient method for protecting sensitive data during
transmission or storage. By leveraging the capabilities of neural networks, the
algorithm aims to enhance the confidentiality and integrity of the data being
encrypted.
Existing System
The existing system for encryption and decryption algorithms based on neural network may
include various traditional encryption methods such as symmetric-key algorithms (e.g., AES) or
asymmetric-key algorithms (e.g., RSA). These conventional encryption schemes often rely on
mathematical operations and cryptographic techniques rather than neural networks.
Dis-Advantages
 Less accuracy
 low Efficiency
 limited adaptability
Proposed System
In this project I am using neural network to encrypt and decrypt and this neural
network will be trained with keys and plain text. While training neural network
application calculate weight between keys and neural network and this weight will be
consider as encrypted data. This encrypted data can be send to any receiver and then
receiver will perform below steps to decrypt text.
Advantages
 High accuracy
 High efficiency
 adaptability
Software Requirement
Analysis and Specification
Modules
 User
1. Generate key
2.build neural network model
3.neural network encryption
4.neural network decryption
 System
System Architecture
Hardware Requirements
 Processor : i3 or higher
 Hard Disk : 40 GB
 RAM : 4 GB (min)
Software Requirements
 Operating system : Windows XP/7/8
 Coding Language : Python
 Frame work : Tkinter
 Backend : TensorFlow
NON FUNCTIONAL REQUIREMENTS
 Usability

 Reliability

 Security

 Performance

 Maintainability
SDLC Model
System Design
ER Diagram
Data Tables
Table Name: User
Column Name Data Type(Size) Constraint

USER ID INT(10) PRIMARY KEY

NAME VARCHAR(20) NOT NULL

ADDRESS VARCHAR2(30) NOT NULL

Table Name: System

Column Name Data Type(Size) Constraint

SYSTEM ID INT(10) PRIMARY KEY

ENCRYPTED DATA VARCHAR(30) NOT NULL

ENCRYPTED KEY INT(30) NOT NULL

DECRYPTED DATA VARCHAR(30) NOT NULL

DECRYPTED KEY INT(30) NOT NULL

UML Diagrams
Use case diagram for overall system
Class diagram for overall system
Sequence diagram for overall system
Activity diagram for overall system
Deployment diagram
Testing

 Testing is a process, which reveals errors in the program. It is the

major quality measure during software development. During
software development, during testing, the program is executed with
a set of test cases and the output of the program for the test cases is
evaluated to determine if the program is performing as it. expected
to perform.
TESTING METHODOLOGIES

Types:
 White box testing
 Black box testing

Levels of testing:
 Unit Testing
 Integration Testing
 System Testing
 Acceptance Testing
Test Cases
Test Test CaseTest Case Test Steps Test Test
Case Name Description Case Priority
Id Step Expected Actual Status

01 Start the Host the If it doesn'tWe cannot run theThe application hostsHigh High
Application application Start application. success.
and test if it
starts making
sure the required
software is
available

02 Home Page Check the If it doesn’tWe cannot access theThe application isHigh High
Deployment load. application. running
environment for successfully
properly .
loading the
application.
03 User Mode Verify the working of If it doesn’t We cannot The application displays theHigh High
The application Respond use the Freestyle
in freestyle Freestyle Page
mode mode.

04 Data Input Verify if the If it fails We The High High

application to take the cannot application
takes input input or proceed updates the
and updates store in further input to application
The
Database
Implementation

 Implementation is the process of assuring the information

system which is operational and then allowing user take its
operation for its operations for use and evaluation.
Home Page
Message page
Key generation page
Neural network model page
Data encryption page
Data decryption page
Conclusions
 The concept of using neural networks in the field of cryptography is
growing at a rapid pace. Various neuro- crypto algorithms planned by
researchers are offered in literature. But most of them are limited to
the key generation and cryptanalysis. In the research work auto
associative memory network is utilized to encrypt the plain text into
the form which is totally independent from the previous one. The
formula is pretty easy to implement and has quicker coding and
decipherment speed. The algorithm is following the symmetric key
system which makes it vulnerable to leakage of key. To overcome this,
solely sure parties ought to be concerned in communication or a sure
third party will be used as associate authority to forestall the key run.
THANK YOU…!