(A) study on traffic analysis and scheduling schemes for efficient and secure communications in broadband networks

Abstract

Traffic control schemes such as traffic scheduling, analysis and congestion control are very useful technique to utilize system resources and provide QoS of each user in both wired and wireless communication networks. In this thesis, we propose an efficient traffic scheduling scheme for wireless OFDM systems and traffic analysis and control schemes for wired large scale networks. In wireless system, orthogonal frequency division multiplexing (OFDM) is regarded as a very promising digital modulation technique for achieving high rate transmission. However, the increasing number of wireless data users and the deployment of broadband wireless networks have brought about issues of fairness among users and system throughput. In this thesis, we propose an efficient scheduling algorithm to maximize system throughput while providing a level of fairness among users for non real-time data traffic in the downlink of a multiuser OFDM system. We establish a practical scheduling procedure to implement our scheme considering fairness among users and also formulate the resource allocation problem for rate, power, and subcarrier allocation as an integer program that maximizes system throughput. Next, we present a computationally efficient heuristic algorithm for a problem based on the Lagrangian relaxation procedure. Through the computing simulation, we show that the proposed scheme performs better than other schemes in terms of both system throughput and fairness among users. Another issue dealt with in this thesis is a defending scheme for large scale networks based on traffic analysis and congestion control. The distributed denial-of-service (DDoS) attack presents a very serious threat to the stability of the Internet. In a typical DDoS attack, a large number of compromised hosts are amassed to send useless packets to jam a victim, or its Internet connection, or both. Therefore, the defense from DDoS attacks or the identification of their sources has been among the hardest re...