**4. Simulation results**

In this section, we compare our simulation results with the recommended algorithm in (Chang & Hu, 2010), flooding-based approach and the algorithm proposed in (Marzolla et al., 2005; Marzolla et al., 2007) (MMO). In the first simulation tests, the total number of nodes which are visited during in the resource discovery and in updating are compared with algorithm (Chang & Hu, 2010) for 400 queries. In the second experiment, the nodes which queries send during resource discovery for the flooding-based approach, MMO algorithm and resource discovery tree algorithm, are shown and compared with our method.

As we know, for calculating the traffic which a method causes, we can calculate the number of links which are occupied for resource discovery or updating in the method. If the number of links visited for resource discovery or updating in a method is lower, the method has lower traffic and would be more efficient.

In the next simulation tests, the traffic in our method which is caused by the increased number of tree nodes in resource discovery and updating is indicated and the results are obtained for different nodes with 300 and 1000 queries. In the last simulation, we observe that the traffic caused in our method would be lower than other methods. This test is performed supposing there are 300 queries for different nodes, and the flooding-based, MMO, resource discovery tree and our methods are compared.
