ABSTRACT
A network is a system that enables (i) communication and (ii) automation and control, between two or more network terminating points that may refer to either users or machines. A network consists of geographically dispersed components—links and nodes, of which links transport traffic while the nodes provide the distributed computational intelligence necessary in the network. The computational demand stems from the key networking functions of route determination, routing of traffic cells, switching the traffic cells in the switch fabric, ensuring the authenticity and security of each user traffic, and buffer management. In turn, the nodes of a network are composed of computationally intelligent modules corresponding to each of the networking functions. This chapter focuses on the issue of computational intelligence in buffer management, especially the use of algorithms, fuzzy sets, and fuzzy thresholds. Buffers are incorporated into a switch fabric to temporarily store the excess cells of a bursty traffic and achieve the dual functions of minimizing cell loss and smoothing traffic. For details on the nature of the computational demand for the remainder of the networking functions, the reader is referred to Reference 1. The use of sophisticated fuzzy set oriented computational techniques in networking is a recent phenomena and is growing rapidly. The remainder of this chapter is organized as follows. Section 9.1 presents a brief overview of the state of the current research in the use of fuzzy sets in networking. Section 9.2 focuses on an important networking function, buffer management, and presents the role of computationally intelligent algorithms and fuzzy thresholds in minimizing cell loss from buffer overflow, under bursty traffic. Finally, Section 9.3 develops a recommendation for further studies that are needed to maintain continued growth in this field.
