5G was not just an incremental evolution over 4G; it was designed to address a broad set of new use cases (eMBB, URLLC, mMTC) and introduce flexible, service-oriented architecture concepts (e.g., slicing, QoS flows). Below are some of the key factors that influenced why the 5G RAN protocol stack looks the way it does—and how user scenarios and requirements helped shape it.
- High throughput and large bandwidths (sub-6 GHz and mmWave) are needed to deliver very fast data rates (multi-Gbps).
- This drives the need for massive MIMO, advanced beamforming, and efficient scheduling in lower layers.
- Stringent latency (as low as 1 ms) and very high reliability targets (99.999%).
- Implies optimized MAC scheduling, short Transmission Time Intervals (TTIs), robust link adaptation, and a streamlined stack to minimize overhead/delay (e.g., fast HARQ, fewer retransmissions).
- Potentially millions of IoT devices, each sending small amounts of data sporadically.
- Demands power-efficient designs, lightweight signaling, and the ability to handle massive access attempts.
These three service categories (sometimes expanded to include V2X, Industrial IoT, etc.) imposed very different performance requirements—leading the standard to be modular and flexible so it can adapt to each service type.
- The 3GPP tradition (from 2G/3G/4G) is to keep the radio stack split across PHY, MAC, RLC, PDCP, and RRC.
- Each layer has well-defined responsibilities (e.g., MAC for scheduling, RLC for segmentation/ARQ, PDCP for ciphering/reordering), ensuring interoperability and manageable complexity.
- SDAP (Service Data Adaptation Protocol) is a new layer in 5G, bridging the gap between 5G Core QoS flows (with QoS Flow Identifiers, QFIs) and the radio bearers.
- This meets the need for end-to-end quality-of-service control (eMBB vs. URLLC vs. custom slices), each with different performance guarantees.
- A continuation of the LTE concept but more clearly demarcated.
- Facilitates simpler, high-throughput user-plane processing vs. flexible, service-rich control-plane signaling (mobility, slicing config, security, etc.).
- The design ensures that new radio features (e.g., dynamic numerologies, advanced beam management) can be plugged into the PHY/MAC layers without breaking the entire stack.
- 3GPP wanted to future-proof the RAN so new features (AI-driven scheduling, further advanced waveforms) can be added over time.
- 5G is built around the idea of multiple logical networks (slices) on a single physical infrastructure, each slice targeting different service needs (e.g., a slice for automotive URLLC, another for eMBB streaming).
- This requires the RAN to differentiate traffic flows at the radio bearer level, which is what SDAP helps accomplish.
- In 4G/LTE, QoS was mostly handled via EPS bearers. 5G needed a more granular approach because slices might have drastically different latency or throughput requirements.
- Hence, mapping from 5G Core QoS flows to Data Radio Bearers was introduced to be more dynamic and service-driven.
- With potentially hundreds of antenna elements, the lower layers (PHY/MAC) had to be designed to handle advanced beam management procedures.
- RRC and MAC were enhanced to manage beam-based measurements, handovers, and resource coordination.
- 5G can use different subcarrier spacings (15, 30, 60, 120 kHz, etc.) within the same carrier to serve different use cases.
- The PHY layer thus needed to be more flexible and parameterizable than in LTE.
- Operators often wanted to co-deploy LTE and 5G in Non-Standalone (NSA) modes initially.
- The new stack had to be somewhat cohesive with existing LTE infrastructure (particularly in the control plane, e.g., for dual connectivity with an LTE eNB).
- Mobile Network Operators (MNOs) demanded support for new revenue streams (industrial IoT, slices for enterprises, etc.) while ensuring a reasonable upgrade path from 4G.
- Equipment Vendors needed a standard that could be implemented using new hardware (massive MIMO arrays) without discarding prior investments in software-defined base stations.
- Device Manufacturers needed a stack that could scale from small IoT modules to high-end smartphones, with consistent layering and security frameworks.
- Regulatory Bodies influenced aspects like band allocation, power limits, and EMF exposure, which in turn shaped the PHY/MAC design requirements (beamforming constraints, dynamic power control, etc.).
user scenarios and requirements (treated as services, e.g., eMBB, URLLC, mMTC) played a major role in how 3GPP defined the 5G RAN stack. However, several other factors came into play:
- Evolving technology (massive MIMO, flexible numerology).
- Standardization legacy (layered approach from LTE, continuity in naming).
- Operator business models (network slicing, QoS-based services).
- Regulatory and ecosystem constraints (spectrum, coexistence with LTE, device diversity).
The 5G stack therefore emerged from a consensus among these diverse requirements and technologies, aiming to be:
- Modular (with each layer having clear responsibilities).
- Service-centric (QoS flows, slicing).
- Future-proof (accommodating new features, frequency bands, and deployment scenarios).
In essence, while user (service) scenarios were a primary driver, they intersected with equally critical demands for flexibility, advanced radio features, backward compatibility, and scalability—all of which culminated in the 5G stack we see today.