Congestion in different topologies of traffic networks

In the present paper, we consider three different types of networks (random, small-world, and scale-free) with dynamic weights and focus on how the characteristic parameters (degree distribution exponent, rewiring probability, and clustering coefficient) affect the degree of congestion and the efficiency. Experiment simulation shows that the scale-free and small-world networks are more prone to suffering from congestion than random ones at low traffic flows, but the scale-free network is more sensitive than the small-world one. Compared with other two topologies, the scale-free network, while its congestion factor rises slowly, can support much more volume of traffic as the traffic flow increases. Results also indicate that for the same value of congestion factor, there may be a different efficiency, which shows that only congestion or efficiency alone cannot evaluate the performance of networks effectively.