A Novel Remote Desktop Protocol Data Security Approach Using Hybrid Cryptographic Algorithm and Unsupervised Neural Network

An empirical investigation uses AES, RSA, and a character replacement approach, along with the unsupervised neural network RBM, to enhance the security of Remote Desktop Protocol communication. It was discovered that Remote Access Trojans, such as PlugX, are used to establish control through remote access to compromised systems. Unauthorized access is obtained by an intruder to a system, resulting in the theft of sensitive data and the exploitation of a vulnerable workstation for malicious activities using a Remote Access Trojan (RAT). The proposed hybrid model is combined with RBM to ensure data transaction security between nodes. In addition, a comparative classification algorithm is integrated to further enhance the model. This study evaluates key size, data size, execution time, and memory usage of the algorithm. Cryptography is essentially a resource-intensive approach to computing security. Competitive and comparative analyses are being conducted to investigate the performance measures of the proposed model. In most cases, it was found that resource utilization was correlated with the key lengths of the AES and RSA algorithms. Due to the use of multiple categorical attacks on the remote desktop protocol, researchers have limited focus on secure data transaction of the remote desktop protocol. Therefore, the study limits the analysis of advanced security technologies that are used to gain a deeper understanding of the threats. This suggests that proposed algorithm is more reliable compared to AES and RSA when considering both security and efficacy. The research introduces a cutting-edge hybrid cryptographic model, enhanced with an unsupervised neural network, to outsmart potential intruders and protect sensitive data. By employing a Restricted Boltzmann Machine (RBM) for smart, autonomous threat recognition, the study assesses performance metrics crucial for real-world application and achieves a remarkable 98.75% accuracy, setting a new standard in RDP security.