The Laplacian Spectrum of Complex Networks

摘要: The set of all eigenvalues a characteristic matrix graph, also referred to as the spectrum, is well-known topology retrieval method. In this paper, we study spectrum Laplacian an observable part Internet graph at IPlevel, extracted from traceroute measurements performed via RIPE NCC and PlanetLab. order investigate factors influencing observed graphs, following complex network models: random Erdős-Renyi, smallworld Watts Strogatz scale-free derived Havel-Hakimi powerlaw degree sequence. Along with these models, corresponding Minimum Spanning Tree (MST). Extensive simulations show that spectra models differ substantially graphs. However, MST in Erdős-Renyi uniformly distributed link weights does bear resemblance it. Furthermore, discuss extensive topological characteristics real-world graphs well models.