Password Security Via Neural Networks


Password security and protection are one of the important research topics in modem computer systems. Providing privacy, authenticity, integrity and limited access to data, encryption methods are proposed for password security schemes. This paper proposes the use of neural network accessing the system, the system needs other information extracted from the user's password along with the password itself, these information is passed to two different neural networks to examine the authenticity of the user, and then decide whether the user is a legal user or an intruder. The extracted information can be summarized by the time period between each two successive characters in the password and the strength of strike of the user when each character is typed at login time. As a result is a powerful security scheme for password protection and the user has no wary about password being theft because the related password information can not be theft.