Spectrum allocation method in elastic optical network based on service priority awareness

Abstract

The invention discloses a spectrum allocation method in elastic optical network based on service priority awareness. The method comprises: first, ranking the arriving services according to their requirement on the bandwidth; according to the sizes of the idle frequency blocks in the network, selectively abandoning the services with high requirement on the bandwidth; ranking the priorities of the remaining services in a descending order according to their requirement on the bandwidth wherein the services requiring the bandwidth more enjoy higher priority and are allocated in priority. According to the invention, with the constant arrival of and the increase in the services of the network, when the sizes of the remaining frequency spectrum pieces become small, the services requiring the bandwidth less are more likely to be filled into the remaining idle frequency blocks, therefore, increasing the overall utilization efficiency of the frequency spectrum resource. The method of the invention can effectively increase the service completion degree of the network, reduces the service clogging rate and reduces the discretization of frequency spectrum pieces in the network.

Description

technical field [0001] The invention relates to a spectrum allocation scheme based on service priority perception in an elastic optical network (EON), and belongs to the technical fields of elastic optical network and resource allocation. Background technique [0002] A traditional wavelength division multiplexing (WDM, Wavelength Division Multiplexing) network adopts a fixed spectrum division mechanism, which has disadvantages such as low resource utilization rate and poor flexibility. Elastic Optical Network (EON, Elastic Optical Network) adopts Orthogonal Frequency Division Multiplexing (OFDM, Orthogonal Frequency Division Multiplexing) technology to further divide network spectrum resources into narrow spectrum units, namely Frequency Slot (FS, Frequency Slot), and can Appropriate adjacent spectrum units are allocated according to the actual number of business needs, and dynamic and effective spectrum resource allocation is performed accordingly, thereby improving resour...

AI Technical Summary

Problems solved by technology

Eiji Oki and others proposed a block-based First-Last Fit spectrum allocation scheme, which divides the connections with intersecting links into different color blocks and labels the color blocks, and the connections of odd-numbered color blocks start from the beginning of the spectrum Start allocation, and the even number is allocated from the end. The advantage is that the business blocking rate is low, and the disadvantage is that the block process is too complicated when the routing table is complex [E.Oki and B.C.Chatterjee, "Performance evaluation of partition scheme with first-lastfit spectrum allocation for elastic optical networks,"in 2016 18thInternational Conference on Transparent Optical Networks(ICTON),2016,pp.1-4.]

Hideki Tode et al. proposed another scheme, dividing the spectrum on each link in the network along the spectrum axis into a priority area and a general area with increasing fixed lengths, which has the advantages of low traffic blocking rate and fragmentation in the network The disadvantage is that due to the imbalance of dynamic service bandwidth requirements, spectrum resources in priority areas may be wasted, reducing resource utilization [H.Tode and Y.Hirota,"Routing,spectrum and core assignment for space division multiplexing elastic optical networks,"in Telecommunications Network Strategy and Planning Symposium (Networks), 2014 16th International, 2014, pp.1-7.]

Liu Huanlin and others proposed a spectrum-aware-based service segmentation-combination elastic optical network resource allocation strategy. This solution reduces the service blocking rate by dividing the service into multiple services with small bandwidth requirements. The disadvantage is that the service segmentation Too many guard bands will be generated in some cases, resulting in a waste of spectrum resources [Liu Huanlin, Xu Yifan, Chen Yong, "A Resource Allocation Strategy for Flexible Optical Network Based on Spectrum Sensing-Based Service Segmentation and Merging," Journal of Electronics and Information Technology, pp.892- 898, 2016.]

Images