Machine learning FSO-SAC-OCDMA code recognition under different weather conditions

Abstract Nowadays, transmitting and receiving data with high speed a level of security are the main demands. So, new model spectral amplitude coding optical code division multiple access (SAC-OCDMA) is suggested in this paper, based on free space (FSO) communication system using coherent sources. Three different codes: enhanced double weight (EDW), modified (MDW) multi-diagonal (MD) codes assigned to our proposed direct detection (DD) technique. Furthermore, simulated under weather conditions including clear air (CA), light mist (LM), very fog, fog (LF). The performance evaluated through its bit error rate (BER), Q-factor, received power, signal noise ratio (SNR). Moreover, classification information by three SAC-OCDMA (EDW, MDW, MD) still challenging. two machine learning (ML) algorithms used, study, classify codes. Detecting which at receiver end important order reduce detection. Two algorithms: K-Nearest Neighbor (KNN) Support Vector Machine (SVM) adopted used for transmission four conditions. ML input dataset consists obtained simulation results, BER, SNR. Each feature be normalized before ML. results show that FSO achieved longest propagation range CA conditions, 7 km, while shortest LF, 0.98 km. This due attenuation causes degradation. uses EDW gives best result MD worst performance. Also, highest power when 5.5 has ? 21.58 dBm, 22.04 dBm 23.8 MDW codes, respectively. As between coming from both algorithms, KNN SVM, achieve promising most cases. They showed more than 97% accuracy CA, LM, LF