Compensation control study based on impairment-aware for Wavelength Switched Optical Networks (WSON)

In an optical network, the optical signal transmitted along the lightpath may need to travel through a number of cross connect switches (OXCs), optical amplifiers, and fiber segments. While the signal propagates toward its destination, the optical components would continuously degrade the signal quality by inducing impairments. When the signal degradation is so severe that the received bit-error rate (BER) becomes unacceptably high, the lightpath would not be able to provide good service quality to a connection request. Such a lightpath, which has poor signal quality due to transmission impairments in the physical layer, should not be used for connection provisioning in the network layer. This paper presents an adaptive PID controller based on the power compensation of BP neural network to restrict the influence of the impairment power for a networked control system (NCS) with the presence of controller time-delay and power compensation at amplifiers' node firstly. Control algorithms continuously adjust their channel powers in response to dynamic information from the network links. And the controller could achieve the on-line adaptive power compensation without changing the parameters of PID controller. The results of simulation show that the proposed controller could adjust better channel power at the transmitter sites and achieve channel optical signal-to-noise ratio (OSNR) optimization with controller's time-delay.

► Control strategy is introduced continuously adjust channel powers in response to dynamic information from the network links, in order that the lightpath could able to provide good service quality to a connection request.
► And we improved the Bp model to better achieve the on-line adaptive power compensation without changing the parameters of PID controller.
► The controller could adjust better channel power at the transmitter sites and achieve channel optical signal-to-noise ratio (OSNR) optimization.
► Future work will focus on this efficiency study.