Network performance & Quality of service in data networks involving spectrum utilization techniques
This research has developed technique to improve the quality of service in wireless data networks that employ spectrum utilization techniques based on Cognitive Radio. Most multiple dimension implementations focus on maximizing the Successful Communication Probability SCP in order to improve the wireless network utilization. However this usually has a negative impact on the Quality of Service, since increasing the SCP leads to increasing signal interference and Packet Loss, and thus network performance deterioration. The Multiple Dimension Cognitive Radio technique is a new technique, proposed in this thesis, that improves the Cognitive Radio Networks (CRN) efficiency by giving opportunity to secondary users (Unlicensed users) to use several dimension such as time, frequency, modulation, coding, and antenna directionality to increase their opportunity in finding spectrum hole. In order to draw a balance between improving the networking utilization and keeping the network performance at an acceptable level, this thesis proposes a new model of multiple dimension CR which provides a compromise between maximizing the SCP and network throughput from one side and keeping the QoS within the accepted thresholds from the other side. This is important so as to avoid network performance degradation which may result from the high user density in single wireless domain as a result of maximizing the SCP. In this research, a full Cognitive Radio model has been implemented in the OPNET simulator by developing modified nodes with the appropriate coding which include basic functionality. The Purpose of this model is to simulate the CR environment and study the network performance after applying the controlled multi dimension technique presented here. The proposed technique observes the channel throughput on TCP (Transmission Control Protocol) level, also QoS KPIs (Key Performance Index) like Packet Loss and Bit Error rate, during the operation of the CR multi dimension technique and alerts the system when the throughput degrades below a certain level. The proposed technique has interactive cautious nature which keeps monitoring the network performance and once find evident on network performance deterioration it takes corrective action, terminates low priority connections and releases over utilized channels, in order to keep the performance accepted.
- PhD