As IEEE 802.11bn (Wi-Fi 8) approaches completion, attention across the Wi-Fi ecosystem is naturally shifting toward what comes next. Naturally, that’s Wi-Fi 9. However, the next mainstream generation of Wi-Fi is not being shaped by a single breakthrough technology, but by a convergence of pressures that challenge long-standing assumptions in WLAN design. Across industry and academia, there is growing alignment that future progress will be defined less by peak data rates and more by reliability, predictability, and intelligent operation under real-world conditions.
A central theme emerging across contributions is the need for deterministic behavior. Many next-generation use cases—industrial automation, robotics, immersive entertainment, haptics, and emerging AI-driven applications—are fundamentally sensitive to worst-case latency rather than average performance. While recent Wi-Fi generations have made meaningful improvements in latency reduction, they remain largely best-effort systems. For these new scenarios, bounded latency, controlled jitter, and predictable scheduling are becoming baseline requirements. This represents a conceptual shift: Wi-Fi is increasingly expected to behave more like a scheduled system rather than a contention-driven one.
Closely tied to determinism is the broader requirement for end-to-end reliability. Mobility, dense deployments, external interference, and heterogeneous device capabilities all contribute to performance variability that is difficult to manage with existing mechanisms. Across sectors—from location-based entertainment venues to factory floors and medical environments—there is a clear demand for explicit reliability guarantees with defined failure probabilities. Importantly, this is not just a PHY-layer problem; it requires coordinated evolution across MAC behavior, scheduling, feedback mechanisms, and system-level control.
At the same time, capacity pressure is still a first-order concern, primarily driven by AI-era workloads. Edge AI, embodied intelligence, immersive media, and high-rate access networks are pushing WLANs toward limits that were not anticipated when downlink-dominant consumer traffic models prevailed. Several contributions highlight the risk of WLANs becoming bottlenecks relative to fiber, PON, and cellular backhaul, particularly as uplink traffic grows in importance. Indeed, the IEEE 802.11bq group specifically targets the use of 60 GHz spectrum to provide substantially greater datarate for both the donwlink and uplike. While reliability dominates many discussions, sufficient throughput headroom remains essential to make deterministic guarantees meaningful at scale. Indeed, Wi-Fi 9 could incorporate mmWave or potentially LiFi (IEEE 802.11br) as baseline to achieve this greater capacity and even address some of the requirements for deterministic behaviour?
Another cross-cutting insight is the growing diversity of deployment models. Traditional fixed access points are now joined by mobile access points, in-vehicle hotspots, personal tethering scenarios, and dense peer-to-peer clusters. These environments exhibit strong temporal and spatial correlations, but also face tight power, thermal, and coexistence constraints. Treating all of them as extensions of legacy enterprise WLAN assumptions is increasingly challenging.
Finally, there is broad recognition that backward compatibility, while critical for market success, constrains innovation. Several proposals explore the idea of greenfield or legacy-decoupled operation modes to enable fundamentally different scheduling and coordination paradigms. To this extent, both mmWave (IEEE 802.11bq) and LiFi (IEEE 802.11br) both provide the greenfield for such designs and deployments. While such approaches may not define the core of the next standard, they reflect a growing desire to restore design freedom that has narrowed over successive generations.
Taken together, the emerging vision for Wi-Fi 9 is not about a single feature or metric. It is about evolving Wi-Fi into a more predictable, reliable, and adaptable wireless system—capable of supporting AI-driven, interactive, and mission-critical applications while remaining practical, scalable, and power-efficient.
