The framework aims to mitigate these risks by integrating multiple modeling and optimization techniques into a unified control plane. Conventional routing approaches often rely on memoryless channel models, non-differentiable soft-computing heuristics, or centralized optimization that may not adequately capture blockage persistence and tail-latency risks. FARABI integrates a three-state semi-Markov blockage model to track blockage age and estimate residual interruption duration.

This model is combined with multi-class Markov-Modulated Poisson Process traffic modeling to account for bursty heterogeneous traffic patterns. The framework incorporates queueing, link reliability, route sustainability, and traffic-class requirements into an online hierarchical primal-dual optimization process. Central to the architecture is a seven-dimensional differentiable fuzzy inference system that works alongside Conditional Value-at-Risk-based risk-aware routing optimization.

This combination allows the system to manage complex routing decisions under stochastic conditions. The simulation-based evaluation was conducted under dynamic blockage, mobility, and bursty traffic conditions to assess the framework's performance. The results indicate that FARABI can improve the robustness of SDN-based 6G IoV routing against these specific stochastic challenges.

Computational efficiency is a key aspect of the FARABI framework's design. The control-plane execution time increases from 0.5 ms to 1.85 ms as the decision space grows from 450 to 18,000 variables. This performance corresponds to up to 42.08 times lower computational time than a centralized interior-point solver.

Under increasing obstacle density, the framework sustains an effective service rate from 4.62 Gbps down to 1.88 Gbps. However, the system faces limitations under extreme blockage scenarios. The semi-Markov model is used to estimate these residual interruption durations within the routing optimization process.

Latency performance was measured under severe blockage dynamics to evaluate the framework's responsiveness. In these conditions, 58.4% of packets are delivered within 1.6 ms. A higher percentage, 94.1%, is delivered within the 3.2-ms tail-latency target. The framework manages different traffic classes with varying quality of service requirements across burstiness levels from 1.5 to 5.

Safety packet-loss rates remain within the range of 0.002% to 0.098%. Cooperative-perception packet-loss rates range from 0.045% to 0.98%. Infotainment packet-loss rates are higher, ranging from 0.42% to 10.15%.

The study, published in Scientific Reports in October 2026, provides a detailed analysis of the FARABI framework's capabilities through simulation. The research highlights the integration of fuzzy logic, semi-Markov models, and risk-aware optimization in 6G networking. The results suggest that such frameworks can improve the reliability of 6G services while maintaining bounded computational and QoS performance.

The ability to handle heavy-tailed physical blockage is a significant focus of the framework's design.

Subject of Research: Fuzzy-assisted routing framework for 6G Internet of Vehicles

Article Title: FARABI: fuzzy-assisted routing for adaptive blockage-resilient 6G IoV

Article References: Sedaghat, S. (2026). FARABI: fuzzy-assisted routing for adaptive blockage-resilient 6G IoV. Scientific Reports. https://doi.org/10.1038/s41598-026-77473-x

Image Credits: AI Generated

DOI: 10.1038/s41598-026-77473-x

Keywords: 6G, Internet of Vehicles, Routing, Fuzzy Logic, Millimeter-wave, Terahertz, SDN, Blockage, Latency, Network Resilience, FARABI, fuzzy-assisted