BASIS: A Biological Approach to System Information Security

With the increase of size, interconnectivity, and number of points of access, computer networks have become vulnerable to various forms of information attacks, especially to new, sophisticated ones. It should be pointed out that biological organisms are also complex and interconnected systems that have many points of access; these systems are vulnerable to sabotage by alien microorganisms. During evolution, biological organisms have developed very successful immune systems for detecting, identifying, and destroying most alien intruders. In this paper we intend to establish a connection between the basic principles that govern the immune system and potential uses of these principles in the implementation of information security systems (ISS) for computer networks. In order to be dependable, existing ISS require an enormous amount of data p rocessing that adversely affects the network performance. A modern ISS must include a number of semi-autonomous software agents designated to prevent particular kinds of threats and suppress specific types of attacks. The ability of such an ISS to provide the required level of information security without burdening the network resources could be achieved only by adopting the most advanced principles of interaction between particular agents. Such principles have been used in biological immune systems. These systems have distributed cells (agents) of various types that attack anything suspected to be alien. Cells interact by sharing information about the type and location of an intruder, utilize the feedback principle for engaging only “as many cells as necessary,” and are capable of learning about intruders that results in immunity to repeated attacks. When necessary functions of the ISS agents are established, the multi-agent system theory can be utilized as a mathematical apparatus to facilitate the formalization of the complex interaction between particular agents in the fashion that is observed in biological immune systems. The individual and collective behavior of particular immune cells need to be investigated to establish a synthetic immune system op erating in computer networks.