FORMAL MODEL OF ADAPTIVE SELECTION OF CRYPTOGRAPHIC PARAMETERS FOR CHANNEL PROTECTION IN CORPORATE COMPUTER NETWORKS BASED ON DYNAMIC TRUST ASSESSMENT

The paper proposes a formal model for the adaptive selection of cryptographic parameters for protecting communication channels in corporate computer networks based on dynamic trust assessment and integrated risk. The relevance of the study stems from the fact that common practices of static configuration of encryption algorithms, modes of operation, and cryptographic strength parameters do not account for changes in access context and the behavior of interacting entities, which leads either to excessive computational overhead or to the emergence of vulnerability windows during threat escalation. The scientific novelty lies in interpreting the cryptographic profile as a controllable dynamic state of the security system, where trust acts as a direct control parameter of the cryptographic configuration rather than merely a factor in access decision-making. A protected channel is formalized as a state tuple combining the subject, resource, context, trust level, risk, and cryptographic profile, while adaptive parameter selection is described by a mapping that establishes a correspondence between (resource criticality, context) and a set of cryptographic characteristics (algorithm, mode, strength parameter, session lifetime). An optimization formulation for profile selection is developed that accounts for the trade-off between cryptographic strength and operational costs, along with an event-driven mechanism for updating the cryptographic state (Rekey/Upgrade/Revoke) in response to trust degradation, risk increase, or critical security events. Scenario analysis (normal operation, contextual/behavioral anomaly, critical event) demonstrates the model’s ability to coherently enhance strength and reduce cryptographic session lifetimes in high-risk situations, thereby reducing the potential attack window while maintaining acceptable performance under low-risk conditions. The obtained results provide a theoretical foundation for deploying adaptive cryptographic profiles in TLS/VPN and Zero Trust–oriented corporate environments.