Fuzzy Controlled Dynamic QoT Routing and Wavelength Assignment in Transparent Optical Networks

In a transparent optical network, a transmitted signal remains in the optical domain over the entire route (lightpath) assigned to it between its source and destination nodes. In this paper, a novel Fuzzy dynamic routing and wavelength assignment technique is proposed for a wavelength division multiplexing (WDM) optical network, to achieve the best quality of network transmission and network service. This paper presents a novel Fuzzy logic controlled quality of transmission aware dynamic routing and wavelength assignment (FCD-QoT-RWA) algorithm, where the optimum path is chosen by a Fuzzy Rule Based System (FRBS) with hop constrained paths and at a specified traffic load. The optical signal traverses a number of cross connect switches (XCS’s), fiber segments, and optical amplifiers such as erbium doped fiber amplifier (EDFA). Thus, the signal quality degrades as it encounters crosstalk at the XCS’s and also picks up amplified spontaneous emission (ASE) noise at the EDFA’s. These impairments continue to degrade the signal quality as it progresses towards its destination, and the received optical signal to noise ratio (OSNR) at the destination node becomes unacceptably low. The proposed algorithm incorporates quality of transmission aware linear impairments such as polarized mode dispersion (PMD), amplified spontaneous emissions noise (ASE), switch crosstalk (SC), demultiplexer crosstalk (DC) for computation of OSNR for each of the path generated for the call request. The proposed FRBS takes OSNR and two quality of service aware parameters such as hop count (HC) and free channel availability (FCA) of each path as input parameters. The defuzzified value determines the optimal path for establishment of lightpath. In this study, the performance metrics of the hop-constrained lightpaths evaluated at different number of system wavelengths at a specified load. This paper attempts to improve the mean blocking probability, the mean execution time and mean FCA using fuzzy system.