The Thesis is devoted to the resource management problem in next generation networks. The aim of this research is to develop and evaluate network bandwidth allocation mechanisms, which allow dynamically adaptive distribution of bandwidth resources correspondingly to traffic class QoS under uncertain network conditions. We propose fuzzy logic based methodology of decision making on bandwidth allocation for virtualization-enabled network infrastructures. Its performance was evaluated and improved within simulation experiments realized by using CPN Tools and Mamdani fuzzy inference system for traffic QoS control. We describe traffic classification and anomaly detection algorithms by using fuzzy clustering technique applied to transformed traffic data. We show the effectiveness of fuzzy transforms in traffic data preprocessing and dimensionality reduction.