A New Method to Prevent Malicious Code Attack

碩士 === 國立中興大學 === 資訊科學與工程學系所 === 97 === In the past few years, various information attacks have been increased frequently due to the fast progress of the information network technique. The attack, which has influenced to the whole system security, can be divided into the two major types: one is atta...

Full description

Bibliographic Details
Main Authors: Jhih-Han Wu, 吳致翰
Other Authors: De-Cheng Huang
Format: Others
Language:zh-TW
Online Access:http://ndltd.ncl.edu.tw/handle/97353917980986002439
Description
Summary:碩士 === 國立中興大學 === 資訊科學與工程學系所 === 97 === In the past few years, various information attacks have been increased frequently due to the fast progress of the information network technique. The attack, which has influenced to the whole system security, can be divided into the two major types: one is attack for the virus and another is attack for the memory buffer overflow. It is not only the memory buffer overflow attack has been popular used in the internet, but also this attack has been widely used in the steal control authority for a computer system. In this thesis, we first propose a secure-control mechanism which is based on the hardware architecture to prevent memory-buffer-overflow attacks. The method compels to lock the marked data, non-rewritable instructions or the related memory contents of the instructions by the hardware. Here, there are several theorems have been provided and proved the idea correctness. To demonstrate the theorems, we have fulfilled a series simulation to verify the effectiveness of our method by using C language. Meanwhile, we adopt Verilog HDL to demonstrate the feasibility of our idea. By using software and hardware to simulate experiments, we prove that the secure-control mechanism can defend the various memory buffer overflow attacks effectively.