The evolution of 5G networks from their initial deployment phase into extensive mission-critical infrastructure requires network performance to evolve from basic bandwidth expansion to intelligent adaptive management. The modern application ecosystem requires three essential performance characteristics which include minimal latency, dependable operations, and stable data transmission rates.
Networks require a complete transformation of their traffic management systems and resource allocation methods to achieve this change. Ameer Shohail leads the transition of 5G Core Applications Architecture as a Senior Member of IEEE and 5G Core Applications Architect. Through his innovative work of merging L4S with NaaS APIs, Shohail creates the foundation for future 5G networks that will operate as programmable congestion-aware systems optimized for low-latency applications.
The Changing Demands of 5G
The worldwide deployment of 5G networks transforms digital technology usage between businesses and their customers. The network faces an extreme test because of its need to provide ultra-low latency and stable performance under various unpredictable traffic conditions during this transformation.
The implementation of sub-millisecond response times for applications including immersive AR/VR experiences and real-time industrial automation and autonomous systems pushes network design to its maximum capacity. The traditional QoS systems which have managed telecommunications since the 1980s have become outdated because they fail to adapt to sudden traffic increases, resulting in network congestion and service breakdowns.
The performance gap between current 5G capabilities and required standards stands as a major obstacle for 5G network development, according to Shohail. The current network infrastructure needs to evolve beyond speed because it must develop intelligent capabilities. The practice of adding excessive network capacity no longer functions as the primary solution. The development of self-adjusting systems requires the creation of real-time optimization and adaptation mechanisms.
Moving Beyond Static QoS
Shohail proposes a system that merges L4S principles with NaaS API technology to transform network infrastructure into a programmable system. QoD APIs through Quality on Demand allow developers to ask for better QoS, but these requests remain limited to pre-defined policy templates which fail to adapt when network conditions change quickly.
The L4S system uses dual-queue technology with Explicit Congestion Notification (ECN) to separate critical delay-sensitive data from large bulk transfers. The system enables real-time applications to receive priority treatment while maintaining overall network performance. The integration of L4S with 5G Core components through NEF and PCC enables real-time traffic priority adjustments that replace traditional service-based static configurations.
This approach shifts networks from congestion-based reaction systems into architectures that are theoretically capable of anticipating and preventing congestion. The L4S API integration within Shohail’s framework illustrates how communication service providers could create adaptive performance delivery systems for diverse applications.
AI-Powered UPF: Predicting Congestion Before It Happens
The User Plane Function (UPF) serves as the core element of Shohail’s framework because it can be enhanced with AI analytics beyond traditional data forwarding functions in 5G Core networks. Under this vision, the UPF evolves into an intelligent predictive controller, drawing from real-time data such as buffer depth, traffic bursts, and ECN signaling patterns to anticipate congestion before it impacts performance.
Based on estimates informed by global mobility reports and industry forecasts, this model could achieve:
- Up to 80% reduction in latency compared to static QoS approaches.
- Nearly 80% better SLA compliance, supporting enterprise-grade reliability.
- Around 66% improvement in throughput stability during peak usage.
These values represent modeled expectations and market-aligned estimates, not live deployments, yet they underscore the potential of transforming the UPF into a proactive decision-making system. This kind of architecture would provide the consistency and predictability required for low-latency, mission-critical applications.
Unlocking New Monetization Pathways for Operators
Shohail emphasizes that the technical advancements in this model create business possibilities for communication service providers (CSPs). Operators could provide customized services through programmable NaaS API exposure to developers and enterprise clients, enabling service levels that align directly with application requirements.
Potential monetization pathways include:
- Premium service levels for AR/VR applications, cloud gaming, and telemedicine.
- Automatic QoS enhancements through application requests, granting priority to latency-sensitive sessions.
- Pay-per-use performance models supporting real-time industries such as logistics and manufacturing.
These capabilities position networks to move beyond rigid subscription pricing, introducing flexible revenue models tied to real-time demand—much like cloud platforms operate today.
Strategic Relevance for the 6G Era
The integration of L4S into programmable APIs represents a fundamental step toward developing self-optimizing architectures that will be essential for 6G networks. Shohail notes that future network infrastructure will need to be dynamic, adapting at speeds that match application requirements.
Three pillars define this transition:
- Technical scalability through ultra-low latency at reduced costs of over-provisioning.
- Commercial flexibility with revenue streams tied directly to application needs.
- Programmable AI-native foundations for intelligent, self-adjusting networks.
This approach unites operational and financial considerations, equipping operators to evolve from 5G to 6G with superior performance and business resilience.
A Vision for Intelligent, Programmable Networks
Shohail’s influence extends beyond network architecture design. He contributes to the IEEE INGR Testbed Technical Program Committee, where he advances federated testbed frameworks for 6G research. His work includes IEEE conference presentations and editorial board contributions, producing white papers on new network approaches and telecom policy structures.
Through his IEEE publication review activities, he helps shape the direction of wireless technology research, blending academic insight with real-world application. He continues to advocate for open, standards-based systems to ensure interoperability, scalability, and long-term growth for 5G and 6G ecosystems.
Building the Blueprint for Intelligent Networks
Shohail argues that the telecommunications industry must undergo a fundamental shift. Speed alone is no longer the benchmark; instead, networks must evolve into intelligent, predictive systems capable of optimizing performance in real time.
By integrating L4S into NaaS APIs and enhancing the UPF with AI-driven analytics, this framework delivers a theoretical roadmap for measurable gains in latency, congestion management, and throughput stability. While the figures are based on modeled scenarios and industry reports, they illustrate a path toward creating programmable, adaptive, and monetizable networks.
Conclusion
The telecommunications industry stands at a pivotal moment. Speed enhancement alone is insufficient—the true challenge is developing networks that can match the pace of application evolution.
Ameer Shohail’s work on L4S-enabled NaaS APIs and AI-based traffic management frameworks provides a visionary pathway for building low-latency, intelligent networks. His approach outlines how operators can transform infrastructure into adaptive systems that sustain digital innovation and prepare the foundation for the 6G era.
Disclaimer: The performance metrics referenced in this article are modeled estimates informed by global mobility reports and market research. They are intended to illustrate theoretical potential and are not based on live network deployments.
