Master of Engineering Science
Electrical and Computer Engineering
Dr. Xianbin Wang
Applications and services operating over Internet protocol (IP) networks often suffer from high latency and packet loss rates. These problems are attributed to data congestion resulting from the lack of network resources available to support the demand. The usage of IP networks is not only increasing, but very dynamic as well. In order to alleviate the above-mentioned problems and to maintain a reasonable Quality of Service (QoS) for the end users, two novel adaptive compression techniques are proposed to reduce packets’ payload size. The proposed schemes exploit lossless compression algorithms to perform the compression process on the packets’ payloads and thus decrease the overall net- work congestion. The first adaptive compression scheme utilizes two key network performance indicators as design metrics. These metrics include the varying round-trip time (RTT) and the number of dropped packets. The second compression scheme uses other network information such as the incoming packet rate, intermediate nodes processing rate, average packet waiting time within a queue of an intermediate node, and time required to perform the compression process. The performances of the proposed algorithms are evaluated through Network Simulator 3 (NS3). The simulation results show an improvement in network conditions, such as the number of dropped packets, network latency, and throughput.
Shamieh, Fuad, "Advanced Compression and Latency Reduction Techniques Over Data Networks" (2015). Electronic Thesis and Dissertation Repository. 2844.