The Approach for Simulation of the Noise-Related Characteristics for Radio-Over-Fiber Transmission Systems

Recently, significant attention has been paid to utilizing fiber optical access networks for the highly-demanded mobile X-haul (X-front or back). Integrating new technology means changing the architecture, evaluating the coexistence, and defining new standards and recommendations for the existing fiber optical access network. Consequently, the reliable and effective mathematical modeling of the physical layer becomes very important for evaluating the various options for developing the access network. Addressing this, we focus on improving the modeling of the converged fiber optical transmission systems aimed at defining the limits and requirements that can be utilized in the design and standardization of the transmission systems. In the scope of current research, we offer the solution for the problem of the diversity of the parameters of the elements of the converged fiber optical transmission system to be simulated. Respectively, we propose the approach for the simulation of the thermal noise and received optical signal power for the orthogonal-frequency-division multiplexed, multi-frequency-band, radio-over-fiber transmission system. The approach aims to adapt the simulation results for different power budgets, signal quality thresholds, and optical receiver sensitivities without additional simulation. The adoption is ensured by the recalculations performed in Matlab software. The signal quality is evaluated in terms of error-vector-magnitude (EVM). The analysis of the number of simulated bits is performed to ensure the efficiency and precision of the simulation. The high-ranking fiber optical simulation software (VPI Design Suite) provides the EVM versus thermal noise curve. Further, it is expanded to received optical signal power versus thermal noise and EVM versus received optical signal power curves via mathematical calculations implemented in program-development software (Matlab). The approach's feasibility is explicitly proved by comparing calculated and simulated results, with the difference not exceeding 0.03% of EVM.