Advanced service orchestration for intent-based networks: architectures, security, and algorithms

The deployment of 5G and the evolution toward 6G are transforming communication networks into dynamic, programmable ecosystems where conventional performance-driven management is no longer sufficient. Future networks must simultaneously satisfy technical requirements and broader value-oriented objectives formalized as Key Value Indicators (KVIs) alongside traditional Key Performance Indicators (KPIs). Against this background, three challenges remain open: securely translating high-level human intents into executable orchestration actions that jointly meet KPIs and KVIs; endowing Digital Twins (DTs) with the algorithmic intelligence to turn network state into context-aware decisions; and extending orchestration beyond centralized assumptions toward secure, autonomous control in decentralized environments. In response, this thesis advances service orchestration for Intent-Based Networks (IBNs) along three interconnected dimensions. First, it proposes a secure intent translation framework and a bi-objective optimization approach for joint KPI and KVI satisfaction. Second, it develops a DT-assisted algorithmic intelligence perspective grounded in Multi-Criteria Decision Making (MCDM) and game-theoretic methods. Third, it explores how agentic AI paradigms can enable autonomous, trust-aware orchestration at the edge of future Internet-of-Everything (IoE) environments.