I was planning to attend Mobile World Congress (MWC) in Barcelona this year. A busy stretch of analyst days made that impossible. But Huawei released a wave of announcements — some before the show opened, some on opening day — and I found them worth covering for the CIOs and CTOs I talk to globally, many of whom operate in markets where Huawei is an active infrastructure partner.

Full disclosure on my personal context: I have used Huawei phones and watches and genuinely appreciated the technology. I have no comment on the political and regulatory decisions that have restricted Huawei's market access in the United States, Canada, the United Kingdom, Australia, and parts of Europe. Covering only vendors who operate in North American markets is a disservice to a global audience. Huawei is building real infrastructure across Asia-Pacific, the Middle East, Africa, and Latin America, and the CIOs and CTOs in those regions deserve the same analytical attention.

Before getting into the announcements, it helps to understand how the layers of a mobile network relate to artificial intelligence (AI) workloads — because several of these announcements only make sense in that context.

How the Network Layers Connect to AI: A Simple Map

Think of mobile network infrastructure as four floors of a building.

The ground floor is the radio layer — the cell towers and antennas that send and receive wireless signals. This is the part most people think of when they think of 5G.

The second floor is the transport layer — the wires, cables, and microwave links that carry traffic from towers back toward the centre of the network. This is what most MWC announcements are actually about, and the part that most frequently becomes a bottleneck.

The third floor is the core network — the intelligence centre that manages connections, routes traffic, handles identity, and enforces policies across the entire system.

The fourth floor is where services live — the applications, assistants, and experiences that run on top of everything below.

AI enters at every floor. At the radio layer, AI optimises signal quality and manages interference. At the transport layer, AI detects congestion and reroutes traffic automatically. At the core, AI manages identity for millions of connected devices and matches network resources to what each application actually needs. At the service layer, AI powers the experiences users interact with directly.

Huawei's MWC 2026 announcements cover all four floors. Here is how they break down.

Announcement 1: Faster, Smarter Backhaul for Business-Critical Applications

Floor 2 — Transport Layer  ·  February 28, 2026

On February 28, Huawei announced upgrades to its 5G-A (5G-Advanced) Mobile Backhaul (MBH) architecture — the network layer that carries traffic from cell towers to the core. Huawei identified three problems with current infrastructure.

First, the physical connections from towers to the network are too slow to carry what modern 5G generates. Second, more than half of access-layer devices still operate as simple switches rather than routers, forcing traffic between nearby towers on unnecessarily long detours and adding delay. Third, current networks deliver data on a best-effort basis with no guarantees — fine for casual browsing, unworkable for machines and automated systems.

The upgraded architecture addresses all three with faster physical links, smarter routing at the edge, and automated traffic optimisation across the full network path. Data usage across deployments has increased by more than 20 percent from releasing traffic that congestion was previously blocking.

Enterprise use cases: Factory automation and connected robotics where a signal delay can mean a machine collision. Autonomous vehicles that need real-time data to make split-second decisions. Commercial drones managing delivery routes and airspace coordination. Port and logistics operations using automated cranes and vehicle guidance systems. All of these depend on the network making and keeping a delivery promise — not just trying its best.

Announcement 2: New Spectrum Band Products Bridging 5G-A and 6G

Floor 1 — Radio Layer  ·  March 1, 2026

Huawei announced a full product range for the U6GHz spectrum band — a frequency range above existing 5G spectrum that offers wider channels for significantly more data capacity. More than 20 countries are in the process of allocating or have already allocated this band for mobile communications, including China, the UAE, Brazil, and several in Europe. Consumer devices supporting this band are expected to reach the market in 2026.

Huawei's products cover outdoor towers, indoor small cells, and the microwave transmission links between sites. The headline capacity figure — up to 100 Gbps (Gigabits per second) in optimal conditions — is a laboratory ceiling; real-world deployments will produce figures well below it. The practical significance is that this band handles a tenfold increase in network traffic as AI-connected devices multiply, and it is the stepping stone toward 6G architecture.

Enterprise use cases: Stadiums, airports, convention centres, and dense urban areas where thousands of devices connect simultaneously. Private 5G campus networks for manufacturing, logistics, and healthcare where consistent indoor and outdoor coverage is required. Extended Reality (XR) collaboration tools for remote workers and field engineers that require sustained high-bandwidth connections.

Announcement 3: Smarter Network Transport with Lower Running Costs

Floor 2 — Transport Layer  ·  March 1, 2026

Huawei announced upgrades to its broader mobile transport solutions with three specific figures worth noting. A 30 percent reduction in Total Cost of Ownership (TCO — the full cost of owning and running technology over its lifetime), with energy savings estimated at 3 million kilowatt-hours per 1,000 devices over three years. Automated congestion detection paired with traffic rerouting that improves data usage by more than 20 percent. Self-diagnosing fibre fault detection that reduces manual site visits and improves fault resolution speed by 20 percent.

Enterprise use cases: Operators and large enterprises running private networks benefit directly from the lower energy bill and reduced maintenance labour. For organisations that have deployed edge computing at remote sites — energy facilities, mines, agricultural operations — automated fault detection reduces the cost of keeping distant infrastructure running reliably.

Announcement 4: AI Built Into the Network Core

Floor 3 — Core Network  ·  March 1, 2026

Huawei announced the Agentic Core solution — its approach to making the network core capable of supporting AI workloads directly rather than just transporting them. Three capabilities are described.

The first is identity and session management for AI agents. As physical AI devices — robots, autonomous vehicles, connected medical equipment — multiply into the billions, the network needs to register, authenticate, and manage connections between them the same way it currently manages connections between people. Without this, intelligent devices cannot reliably find or communicate with each other.

The second is intent-driven network management. Instead of human engineers writing fixed rules for every scenario, AI agents within the network understand the needs of different organisations and dynamically match network resources accordingly. An AI robot needing 100 Mbps (Megabits per second) bandwidth and 20 milliseconds of latency gets exactly that allocated automatically.

The third is new monetisable services carriers can layer on top — personalised AI assistants, immersive communications, and on-demand computing power delivered over the network rather than from a local device.

Enterprise use cases: Manufacturers deploying fleets of connected robots that need guaranteed connectivity to a central control system. Logistics operators managing thousands of autonomous guided vehicles across a warehouse. Healthcare providers connecting remote diagnostic equipment that must communicate with hospital systems in real time. Smart city operators managing traffic, utilities, and public safety infrastructure from a centralised platform. In each case, the network core is no longer a passive pipe — it actively manages AI workloads.

Announcement 5: AI-Native Framework for Network Operations

Floor 3 — Core Network  ·  March 1, 2026

Developed over a year with industry organisations and leading global carriers, Huawei announced what it calls the first AI-Native framework for intelligent network operations. AI-Native means AI is built into the foundation of how the system works, not added on top of an existing architecture.

The framework has three principles. Focus on outcomes that were previously impossible to achieve with traditional methods. Use digital twins — virtual replicas of the physical network used for simulation, testing, and prediction — and specialised AI models to drive decisions and close the loop between action and result. Redesign operational workflows so that human engineers and AI systems work together, each doing what they do best.

Enterprise use cases: For large enterprises running private 5G networks, this means faster fault resolution, automated capacity planning, and reduced dependence on specialist network engineers for routine operations. For CIOs evaluating managed connectivity services from carriers, this framework is the underlying reason a carrier can credibly offer Service Level Agreements (SLAs — contractual guarantees of network performance) that they could not enforce with manual operations alone.

Announcement 6: AI-Powered Services for Everyday Users

Floor 4 — Service Layer  ·  March 1, 2026

Huawei outlined its vision for carriers moving from selling connectivity to delivering AI-powered services directly to consumers. Several examples cited are already in production.

AI noise cancellation on voice calls reduces background noise from 80 decibels to 40 decibels — the difference between standing next to a busy road and sitting in a library. The system distinguishes between environmental noise (crowd noise, wind, traffic) and human voice in real time. AI-powered home entertainment assistants identify individual household members by voice and recommend personalised content based on viewing habits. On mobile internet, AI identifies individual users' quality of experience in seconds and adjusts network priority accordingly.

Looking ahead, Huawei describes three directions for consumer services as networks evolve toward 6G. Immersive communication using AI and Augmented Reality (AR) devices to create the sense of someone being physically present rather than on a screen. Computing-network synergy, where processing power is delivered over the network so that lightweight, affordable devices can run demanding applications without expensive local hardware. Network-native personal AI assistants that understand intent and proactively act — not a chatbot responding to commands, but a system that anticipates what the user needs.

Consumer use cases: Mobile users in high-noise environments — commuters, travellers, outdoor workers — getting clear call quality without any change in behaviour. Households getting personalised entertainment recommendations without separate accounts or manual settings. Students and workers in emerging markets accessing AI-powered applications on affordable devices because the computing power comes from the network, not the handset.

Analyst Perspective

Reading these six announcements as a set, Huawei is building toward one position: the infrastructure partner for carriers that want to move from selling bandwidth to delivering guaranteed, AI-powered experience. Nokia and Ericsson are pursuing the same destination through their own approaches. The direction is consistent across the industry.

Where Huawei's situation differs is in its deployment base. Carriers across Asia-Pacific and Africa are already running 5G-A in production at scale — not in pilot. The South Africa revenue result and the China demonstration deployments are live. That gives Huawei commercial evidence that Western-market counterparts cannot yet match in volume.

Sources

Huawei press release, February 28, 2026 — 5G-A MBH Architecture

Huawei press release, March 1, 2026 — U6GHz Portfolio

Huawei press release, March 1, 2026 — 5G-A Mobile Transport

Huawei press release, March 1, 2026 — Agentic Core Solution

Huawei press release, March 1, 2026 — MBB Service Intelligence

Huawei press release, March 1, 2026 — AI-Native Framework

Disclaimer: This blog reflects my personal views only. Artificial intelligence tools may have been used for research support. This content does not represent the views of my employer, Info-Tech Research Group.

Strategic Engagement