Joint Assignment of Spatial Granularity, Routing, Modulation, and Spectrum in SDM-EONS: Minimizing the Network CAPEX Considering Spectrum, WSS, and Laser Resources
Mingcong Yang,Qian Wu,Yongbing Zhang
DOI: https://doi.org/10.1109/jlt.2018.2861750
IF: 4.7
2018-01-01
Journal of Lightwave Technology
Abstract:In space division multiplexing (SDM) systems, the spatial dimensions can be divided into one or several group(s). Spatial dimensions in the same group can be jointly switched based on joint switching (J-Sw)/fractional joint switching (FrJ-Sw), resulting in less wavelength selective switch (WSS) usage. Additionally, optical carriers on the same frequency of these spatial dimensions in the same group can share a single laser source. Therefore, a coarser spatial (switching) granularity can reduce the usage of both WSSs and lasers. However, compared to the case of a finer spatial granularity, this scenario leads to more guard bands (GBs), i.e., more spectrum resources, being required. Consequently, spatial granularity is a key factor of minimizing the cost in SDM network design. In this paper, we present a problem called routing, modulation, spatial granularity, and spectrum assignment (RMGSA). Our objective is to minimize the network CAPEX, i.e., minimize the total cost related to the network resources containing the spectrum, WSS, and laser. We first present a joint ILP-RMGSA formulation to solve this problem. Next, we present a decomposition method that divides RMGSA into two substituent subproblems, i.e., first, routing, modulation, and spatial granularity and, second, spectrum assignment, and then solve them sequentially. Because of the limitation of ILP formulation in larger scale problem instances, a rerouting-based heuristic algorithm is proposed. We examine the performances of the proposed approaches via simulation experiments and find that the best spatial granularity is related to the network resources to which network operators attached more importance. Moreover, as another contribution of this paper, we analyze the impact of GB width and the size of the connection requests on the decision regarding the best spatial granularity.