AI-DRIVEN APPLICATION INTENT FOR SEAMLESS END-TO-END GUARANTEED QUALITY OF SERVICE PACKET FORWARDING
Abstract
A next-generation approach to network Quality of Service (QoS) is proposed herein that enables direct, in-band communication of application intent using an innovative extension header, such as an Internet Protocol version 6 (IPv6) extension header. Using the proposed innovation, applications can embed concise, high-level QoS requirement strings or "intents" (e.g., 'ultra-low latency', 'AI/ML Training', etc.) into packet headers, allowing ingress edge devices to interpret these intents using a local policy database. Advanced matching engines, potentially leveraging machine learning, can enable flexible and semantically rich interpretation of these requirements/intents, surpassing traditional rigid classifications.
Publication Date
2026-01-07
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Rajamanickam, Jaganbabu; Sankaranarayanan, Madhan; Jensen, Paul; Mali, Sumant; and Deevi, Krishna, "AI-DRIVEN APPLICATION INTENT FOR SEAMLESS END-TO-END GUARANTEED QUALITY OF SERVICE PACKET FORWARDING", Technical Disclosure Commons, (January 07, 2026)
https://www.tdcommons.org/dpubs_series/9140