This thesis focuses on the performance of terrestrial communication systems that use channel assignment schemes to allocate base stations in a scenario that implements the coexistence of mixed terrestrial communication systems based on cognitive radio technology. Interaction and coexistence of different channel assignment schemes is investigated. Reinforcement learning is applied into multicast downlink transmission with power adjustment to develop the intelligence of cognitive radio. We focus on investigating channel assignment schemes that select channels based on optimizing the coverage area supported by a terrestrial network. Four channel assignment schemes are developed and compared individually followed by an interaction of mixed schemes. It was found that for mixed schemes, different combinations will affect performance, either delivering better coexistence or more interference. It is shown in this thesis that the dynamic channel assignment used in different situations can efficiently improve the performance of spectrum management. We investigate how channel assignment in multicast terrestrial communication systems with distributed channel occupancy detection can be improved using intelligence based on reinforcement learning and transmitter power adjustment. A weighting factor is used to determine the highest priority channels and help in controlling the performance of the system. It is shown how such schemes significantly reduce the number of reassignments and improve the dropping probability at the expense of increased blocking. It is found that using different minimum quality of service threshold percentages can partly control and improve performance in place of the more traditional SINR (Signal to Interference plus Noise Ratio) threshold levels. We also show how a power adjustment technique is developed, that significantly reduces the level of overlap between adjacent base stations and further reduces interference and transmitter power.
展开▼
