Noncentralized Estimation of Degree Distribution

Abstract

In this chapter, a distributed network degree distribution estimation (DNDD) algorithm for estimating the network degree distribution and degree matrix of a WSN is introduced. Degree distribution and degree matrix are important metrics to characterize the structure of networks and are used in several applications [144-146]. For example. the knowledge of degree distribution and degree matrix can be used to infer the properties of distributed networks such as minimum/maximum node degree, k–connectivity, and edge connectivity [144, 145]. It is shown in [146] that different network topologies have different network degree distributions. For example, random graphs usually have Poisson degree distribution, whereas in real-world networks, the degree distribution usually follows power-law degree distribution [146]. Therefore the topology of the network such as random topology and tree topology can be inferred from the degree distribution. Note that part of the content in this chapter is presented in [147].