METHOD OF SYNTHESIS OF PROTECTED INFORMATION NETWORKS BASED ON THE INTRODUCTION OF CORRECTION LINES OF COMMUNICATION

Topicality. Modern distributed software systems operate in environments characterized by continuous evolution of hardware platforms, heterogeneous computing resources, and dynamic failure conditions. Ensuring functional stability and rapid self-recovery in such environments is a critical challenge, especially under potential software or hardware failures, increased workloads, and external disruptions. Objective: This research aims to develop a novel method for synthesizing the structure of heterogeneous information networks based on the principle of maximizing the functional stability index. The proposed approach leverages the introduction of corrective communication links to enable self-recovery mechanisms that account for the heterogeneity and complexity of modern distributed systems. Methods: A mathematical model of a hypernetwork was developed using two interdependent hypergraphs to represent network components and their interactions. Based on this model, three algorithms (AI, AII, and AIII) were designed to optimize network connectivity, ensure vertex redundancy, and dynamically reconfigure the network in response to failures. These algorithms facilitate real-time fault detection, localization, and restoration through resource redistribution. Results: Simulation results demonstrate that the proposed method significantly improves system resilience and reduces recovery time. The maximum optimization effect was observed at a relative value of 0.869698, where the difference in the functional stability index reached 0.467232, confirming the effectiveness of the structural optimization process. Conclusions: The integration of corrective communication links into network topology enables enhanced robustness and fault tolerance. The adaptive algorithms developed within the hypernetwork framework support autonomous reconfiguration and self-healing, making them suitable for next-generation intelligent and resilient distributed systems.