Research On Dynamic Resource Scheduling Algorithm In TWDM-PON

As emerging services such as big data,cloud computing,and artificial intelligence gradually gains popularity,it is difficult to meet the needs of users for various bandwidth services through traditional time division multiplexing technology.Time and wavelength division multiplexed passive optical network(TWDM-PON)is selected as the main development direction of next generation passive optical network stage 2because it can provide bigger bandwidth and richer services and can coexist with existing architectures,and has longer transmission distances and wider coverages.Resource management technology is a key technology of TWDM-PON,so efficient resource allocation algorithms is crucial for its performance improvement.Focusing on the resource allocation problem in TWDM-PON,this thesis proposes an Efficient and Adaptive-Dynamic Wavelength Bandwidth Allocation(EA-DWBA)algorithm.In this algorithm,the priority of the optical network unit(ONU)is determined based on the service level agreement.In order to ensure the users' quality of service the algorithm takes traffic differentiation into consideration.Mathematical models are respectively established for idle slot filling,bandwidth allocation and wavelength allocation.Considering the problem of idle slots between cycles in TWDM-PON owing to information interaction,an adaptive hybrid slot filling strategy is designed to ensure the maximum bandwidth utilization.The principle of high-priority service first filling can meet the delay requirements of high-priority service.On this basis,oriented to the dynamic changes of network traffic,the adaptive bandwidth allocation strategy is designed to ensure the efficient bandwidth allocation.At the same time,considering the problem that ONU may transmit its data on different wavelengths and its transmission delay may be large,we set migration condition based on the earliest idle wavelength algorithm to ensure efficient data transmission,which can reduce the delays of migrations.The simulation results show that the proposed algorithm in this thesis can effectively reduce the total network delay,and can also meet the user's quality of service requirements for different services while ensuring high bandwidth utilization.