A Survey on Biologically Inspired Algorithms for Computer Networking

Abstract

Biologically Inspired Algorithms (BIAs), processes that mimic how organisms solve problems, offer a number of attributes well suited to addressing challenges presented by future computer networking scenarios. Such future networks will require more scalable, adaptive and robust designs to address the dynamic changes and potential failures caused by high heterogeneity and large scale networks. A variety of biological algorithms demonstrate characteristics desirable to network design, and significant effort has been placed on analyzing and developing the corresponding BIAs and applying them to computer networking applications. This paper provides a comprehensive survey of BIAs for the computer networking field, in which different BIAs are organized and explored based on their: (1) biological source; (2) mathematical model; (3) major application; (4) advantages to corresponding "classic" approach; (5) limitations and border conditions; and (6) potential directions for future applications. The paper also compares performance amongst each type of BIA, and compares BIAs that are inspired by different biological sources but are applicable to similar networking applications. The paper concludes by offering a framework for understanding the application of BIAs to problems in the computer networking space.