Adaptive multilevel modulation for grooming in elastic cloud optical networks

In order to mitigate the mismatch of granularities between fixed grid and client traffic, the elastic optical network (EON) was proposed by using orthogonal frequency division multiplexing. In EONs, the bandwidth variable transponder adjusts the number of bits per symbol, so that an optical signal generates by using just enough sub-carriers each with appropriate modulation level. Owing to the advantage of line-rate adaption above, the application of cloud computing has witnessed rapid growth in EONs. However, bandwidth variable transponders consume more power compared with ordinary ones, which will lead to a power-thirsty EON if no effective measure is taken. As a result, the green grooming was proposed for EONs. Unfortunately, the adaptive multilevel modulation was neglected in the current works focusing on green grooming. Thus, in this paper, we design a novel modulation adaptive grooming with guaranteeing transmission performances in green EONs. The distance-adaptive spectrum resource allocation is applied to the green grooming algorithm previously designed by us for EONs. The simulation results show that the adaptive multilevel modulation plays an important role on saving spectrum and power consumption for the green grooming in EONs, because the spectral bandwidth can be saved by increasing the number of bits per symbol to transmit the same data rate.