Research Of Attack Effect Evaluation Techniques In Softswitch

Soft switching was developed for the demands of separating control and carry in telecommunications industry, which is a set of multiple logical function entities, and provides call control, connection and part of business function. The essence is to software of hardware, and makes business truly independent of network. It becomes core equipment for call control and business providing in voice, data, and video business in next generation telecommunication network. The core technique in soft switching is a flexible and distributed software system which is independent of bottom hardware and operating system.Soft switching network has many advantages compared with traditional mobile techniques. However, it also brings several security problems due to the use of IP network for carrying. In order to guarantee the security of the soft switching platform, we need to perform a lot of testing before using it. These very important tests include attacking the platform and test the network performance under the attacks. Therefore, evaluating the performance of soft switching network under attacks is a method for testing the security and effectiveness of the network.In this thesis, we compare ongoing research works for evaluating network performance under attacks and introduce general evaluation techniques used in the area of network performance evaluation under attacks, which includes evaluation indicator criteria and evaluation methods. Then we analyze the techniques for evaluating soft switching networks under attacks based on the existing techniques for evaluating networks under attacks.First of all, we analyze potential security problems in soft switching networks from the perspective of its four layers and protocols. Against these security problems, we establish the evaluation indicator critiera for soft switching networks, which are proposed specifically for soft switching networks and meet the communication characteristics in soft switching networks. Basing on the indicator critiera and taking the characteristics of security problems in soft switching networks, we propose a three layer indicator architecture, namely attribute level, evaluation indicator layer and indicator collection layer top to bottom. Using the indicator architecture and combining specific security mechanisms, we propose an evaluation indicator system for soft switching network under attacks.We introduce two evaluation models for evaluating soft switching networks under attacks. After a brief description of the evaluating process, we illustrate the normalization of evaluation indicators and the method to select the weights. Finally, we apply fuzzy set analysis and expandable evaluation method on the established evaluation indicator architecture to build the two evaluation models for evaluating soft switching networks under attacks. These two models, which are based on different algorithms, both are able to calculate the effective level of each attack. But each model also has their own particularity. Moreover, the weights for the evaluation indicators also affect the evaluation results.At last, we perform experiment to get evaluation results using different weights under the two models, analyze their characteristics, and compare the evaluation results. We can see from the experiments that the weights of the evaluation indicators play a important role. When the same sets of weights are used, the two models provide a fundamentally consistent evaluation result, which verifies the effectiveness of the models.