Design And Simulated Implementation Of Energy-Saving Routing Algorithm In Multi-Granularity Transport Network

With the energy shortages and the greenhouse effect becoming more and more serious, the energy saving has become the focus issue in all fields. In recent years, as the improvements of the technology and the service request, the size and the cost of network equipments keep increasing, coming with the problem of energy excessive consumption, so that the energy saving in the network causes the attention of the whole world. At the same time, with the development of the optical communication, the multi-granularity transport network is becoming the backbone of network, so studying the energy saving in multi-granularity transport network is an effective method to reduce the energy consumption.In order to improve the drawbacks of the energy-saving algorithm presented, this thesis designs an energy-saving routing algorithm in multi-granularity transport network. First of all, design the reasonable structures of node and link to build the network model, considering the energy saving both of the IP layer and the optical transport layer. Second, design a reasonable and efficient auxiliary graph to make the routing be solved integrated. At last, design the unicast energy-saving routing algorithm and multicast energy-saving routing algorithm separately. The implementation of unicast energy-saving routing algorithm facing certain traffic matrix is based on the biogeography-based optimization algorithm. And implement unicast energy-saving routing algorithm facing uncertain traffic matrix using differential evolution with double populations algorithm based on intelligent water drops. To ensure the implementation of energy-saving routing algorithm facing one-to-many multicast requests adopt the intelligent single particle optimizer. And guarantee the implementation of energy-saving routing algorithm facing many-to-many multicast requests by the heuristic algorithm based on estimation of distribution.The energy-saving routing algorithms in multi-granularity transport network presented in this thesis are simulated implemented, and are tested the performance based on topologies with different scale. According to the result of performance evaluation, it is concluded that the four energy-saving routing algorithms facing requests of different type, can reduce the energy consumption obviously, under the premise of meeting the QoS requirements.