Review of Error Analysis of FSO OFDM System

FSO communication provides range without a license, greater bandwidth, and a safe channel. Although PIN diodes are commonly employed for identification, avalanche photodiodes are preferable in order to identify high-speed FSO signals in various functions. APD's dispersion of noise is input-dependent Gaussian noise rather than input-independent Gaussian noise. Companding methods are commonly used to lower the peak-to-average-power ratio (PAPR) in OFDM systems. We examine analysis of error using on-off keying for different kinds of detection. This research provides a machine learning and non- machine learning approach and evaluates its execution against rigid and flexible judgments. We derive soft values for IDGN and IIGN and analyze optimal and suboptimal detection thresholds. The proposed unique ML technique outperforms the existing alternatives.