Dynamic coding system using neural networks

Dissertant

Talia, Thurayya Abd al-Qadir

University

University of Technology

Faculty

-

Department

Computer Sciences Department

University Country

Iraq

Degree

Master

Degree Date

2002

English Abstract

It is noteworthy that cryptography and steganography is the only known practical means for in formation coding that is protecting information transmitted through a large communication network.

Various cryptographic and steganograpbic methods are explored, analyzed and attacked.

Hence, understanding the limitation of these methods provides for the construction of more robust method that can better survive manipulation and attacks.

Cryptography scrambles a message so it cannot be understood ; steganography hides the message so it cannot be seen.

As a new trend in coding systems, we try to develop a coding system by combining these two fields; that is encrypting the message and then hide it by a meaningful message using artificial neural network the artificial neural network proposed is an associative memory network called Bidirectional Associative Memory (BAM).

The reason is mainly due to its simplicity, high-speed computation capability, robust behavior, and its ability of associative recall.

A fundamental rule of security is that one must assume that the attacker knows the general method of coding used (i.e., coding method and the key used).

The amount of effort necessary to invent, test, and install a new method every time the old method is compromised has always made it impractical.

This is where the dynamic coding enters.

Dynamic coding consists of using a-new,-randomly-chosen session key for each new connection.

Instead of changing the method every few years (after it has broken), the key.

Can be changed as often as required (hence the name dynamic) Meanwhile, making attacker task more difficult.

Besides message confidentiality, any private communication requires a management protocol; that is the communicator need to authenticate one another and to exchange the session key, which represents a crucial step.

In this work, a new approach, called DYNAMIC CODING SYSTEM USING NEURAL NETWORK (DCSUNNW), has been suggested in this coding system there are 2-step process : first step achieve message security using bidirectional associative memory network.

The second step allows the sender and the receiver to authenticate each other and exchange the session key.

This step is called authentication ^d key exchange protocol.

Main Subjects

Mathematics

Topics

• Writing
• Encryption
• Symbolism
• Neural networks(Computer science)
• Data

• APA
• MLA
• AMA

Language

English

Data Type

Arab Theses

Record ID

BIM-305781