A novel MILP formulation for planning GMPLS transport networks with conversion and regeneration capabilities

Abstract

With the ever-increasing traffic in WDM transport networks, the development of GMPLS (or multi-granular) transport networks becomes essential to avoid the cost explosion of OXCs. This paper addresses the optimal planning problem of the GMPLS transport network by (1) considering the whole flow hierarchy defined in GMPLS, (2) allowing the optical signal conversion at all granularity levels, and (3) imposing the optical reach constraint on the length of all-optical paths. We call this planning problem routing and multi-granular paths assignment (RMGPA). The objective of the problem is to minimize the total weighted port count. The problem is formulated as a mixed integer linear programming (MILP) model. Due to the computational complexity of the problem, the MILP optimization is carried out on small-sized planning problems. The solutions to the MILP model are used as valuable quality references for previously developed sub-optimum methods, which we have proposed in a previous work to solve large-sized planning problems in a reasonable amount of time.