This paper investigates the performance of secure image transmission over free-space optical (FSO) communication systems employing image encryption and decryption techniques. The analysis considers different atmospheric turbulence conditions, including weak turbulence and moderate-to-strong turbulence, which are modeled using Log-normal and Gamma-Gamma statistical distributions, respectively. The effects of atmospheric attenuation, transmission distance, and turbulence-induced fading on the quality and reliability of image transmission are systematically evaluated. Prior to transmission, the input image is encrypted using an encryption algorithm designed to provide high randomness and security, achieving an entropy close to the ideal value of 8, while the corresponding encrypted image exhibits a PSNR of approximately 30 dB. At the receiver, the encrypted image is recovered through the corresponding decryption process and compared with the original image to assess the reconstruction quality. The overall performance of the proposed secure FSO communication system is evaluated using bit error rate (BER), peak signal-to-noise ratio (PSNR), entropy, and mean square error (MSE). The results provide insights into the combined effects of atmospheric turbulence and attenuation on the security and image reconstruction performance of FSO-based optical wireless communication systems.
Currently, analyses of the propagation performance of vector vortex beams (VVBs) underwater primarily focus on the impact of oceanic turbulence phase screens, lacking consideration of the optical intensity attenuation caused by absorption and scattering effects. Therefore, this study proposes a semi-classical hybrid simulation framework. By integrating statistical sampling techniques with wave optics evolution theory, this framework utilizes Monte Carlo path tracing to simulate the power loss induced by absorption and scattering. Simultaneously, it characterizes the wavefront distortion caused by oceanic turbulence through cumulative phase screens, thereby facilitating a comprehensive propagation analysis of VVBs. The results demonstrate that, when taking optical intensity attenuation into account, the complex oceanic environment–including water quality and turbulence strength–significantly affects the propagation characteristics of VVBs. These findings offer valuable insights for optimizing high-bit-rate communication links between autonomous underwater vehicles and enhancing the stability of data transmission in deep-sea sensor networks operating under complex and dynamic water conditions.
The advancement of contemporary optical communication networks relies on the creation of efficient and flexible photonic technologies. This paper presents novel physics-informed neural network (PINN) approaches for the development and enhancement of 3D circular-shaped photonic crystal add/drop filters (PCADFs), which are essential for frequency control. While existing 3D designs are often technically costly and imprecise, existing 2D photonic crystal designs typically suffer from limited wavelength selection, higher power losses, and reduced adaptability to dynamic networking needs. To address these challenges, the proposed PCADF system integrates PINNs to optimize structural parameters by incorporating Maxwell's equations and other physical principles influencing light transmission into an intelligence system. By leveraging PINN, which can model complex wave interactions within 3D circular photonic crystals (PCs), this research reduces computational costs and enhances the precision of wavelength-specific responses. The approach employs adaptive materials, waveguide design optimization, and simulated thermal methods to minimize crosstalk and improve the add/drop filter's efficiency. The primary objectives are to enable robust and precise wavelength selection, reduce signal distortion, and increase energy efficiency in optical networks. Modeling results reveal significant improvements, including enhanced wavelength isolation, lower insertion losses, and greater adaptability to varying operating conditions. This study underscores the potential of PINNs to merge computational science and artificial intelligence, setting a new standard in photonic device design. It offers an innovative pathway for next-generation optical communication technologies for novel applications in optical computing, high-speed data transmission, and advanced networked communications.
Optical communications exhibit very low bit error rates (BER) and high-quality factors (Q-factor) with distance limitation. Our aim is to optimize output optical power using proposed Linear Regression-based Lagrange Optimization (LRLO) technique in Fiber Optic Communication (FOC) and Free Space Optics (FSO), employing external modulation to enhance performance over extended distances. In LRLO method, Lagrange optimization is performed using the predicted equations from linear regression. We have developed an optical signal processing algorithm using proposed LRLO technique whose performance is independent of data rate. Improvement in optimized system is achieved in terms of BER and Q-factor. These improvements are achieved for the long-haul distance of 14,000 km for FOC system and 40 km for FSO system under thick fog environment. The various environment considered for FSO system are clear air with drizzle, very clear air, very light mist with weak rain, light mist with average rain, very light fog with strong rain, light fog with storm, moderate fog, and thick fog conditions. This optimization technique deployed in this work is applicable to Wavelength Division Multiplexing (WDM) networks and achieves a minimum BER and a maximum Q-factor.
To address the security issue of physical layer authentication in wireless channels, we present a novel scheme that relies on measuring the bit error rate (BER) of free-space optical (FSO) channels. Our scheme consists of four components: transmitter, channel, receiver, and BER authentication. We separate the transmission and authentication channels and exploit the short-term channel correlation to measure the BER variations of both parties. We then set a dynamic threshold for the BER to determine the validity of the receiver. We validate our scheme using a simulated quadrature amplitude modulation (QAM) optical system. The simulation results demonstrate that our scheme can accurately identify an illegitimate transmitter when the link length difference between the legitimate and illegitimate transmitters exceeds 0.5 km. The scheme is also robust to different weather conditions, achieving 100
