Modeling Flow of Smart City Network: Review and Analysis

Diverse applications of IoT technology generate the expected data traffic that is generated by different networks. Diverse traffic requires the most suitable mathematical apparatus to create an adequate model. Some traffic models are more correctly described based on fractal analysis methods. Application flow models that approximate various distributors with both “light tails” (Gauss, Poisson distribution) and “heavy tails” (Pareto, Weibul distribution, lognormal distribution). Self-similar traffic models are widely used to describe traffic in packet-switched networks. The degree of self-similarity of traffic can be determined by various methods. Traffic samples generated by IoT devices are considered: traffic model for a wireless sensor network (WSN) based on a Gaussian distribution; ON / OFF WSN traffic model for tracking randomly and linearly moving targets; IoT traffic models with distributions over a limited time interval; self-similar model of aggregated traffic on the WSN gateway.