4
Huawei and China Unicom Beijing recently announced the commercial deployment of what they describe as the world’s largest 5G-A 100 MHz GigaUplink network
In sum – what to know:
Mobile AI – Huawei says real-time AI applications are shifting network priorities toward stronger uplink performance rather than traditional download-centric optimization.
100 MHz network – China Unicom and Huawei have launched what they describe as the world’s largest commercial 5G-Advanced 100 MHz GigaUplink network across Beijing, covering key areas with more than 10,000 base stations.
Future evolution – Huawei says continued innovation in multi-band pooling, uplink enhancement, and multi-dimensional coordination will be needed as 5G-A networks continue to evolve.
Chinese vendor Huawei believes the rapid emergence of Mobile AI applications will shift operators’ investment priorities toward stronger uplink performance, as artificial intelligence increasingly depends on transmitting large volumes of data from devices to the cloud in real time.
The comments were made by a Huawei source after the vendor and China Unicom Beijing announced the commercial deployment of what they describe as the world’s largest 5G-Advanced (5G-A) 100 MHz GigaUplink network. The deployment provides continuous 100 MHz network coverage across key areas of Beijing using more than 10,000 base stations.
While previous generations of mobile networks largely focused on improving download speeds, Huawei argues that AI-powered services are creating fundamentally different traffic patterns.
“In the mobile AI era, strong uplink capabilities are becoming an essential foundation for real-time interaction,” a Huawei source told RCR Wireless News. “This trend is also expected to drive operators’ demand for further investment in uplink network infrastructure,” the source added.
The deployment is designed to support AI-enabled applications that continuously generate data at the network edge, including AI glasses, AI smartphones, smart cameras, and other intelligent devices that rely on cloud-based processing.
During a live media drive test across a 34-kilometer route in Beijing, the companies said the network achieved a peak uplink speed of 1 Gbps and an average uplink throughput of 397 Mbps. China Unicom also reported that the network maintained a 100 MHz uplink effective ratio of 83%, while cell-edge weak points accounted for only 0.1% of the route.
According to China Unicom, these capabilities supported services including AI glasses with millisecond-level transmission, AI phone interactions with intelligent agents, and real-time cloud upload for AI-powered image analysis.
Huawei said the deployment represents the world’s largest commercial 5G-Advanced network capable of delivering 100 Mbps uplink performance. The network combines coordinated deployment across the 3.5 GHz and 2.1 GHz spectrum bands.
Asked what made deployment at this scale possible, the Huawei source pointed to the coordinated use of those spectrum bands while noting that further innovation remains necessary. “Looking ahead, continued innovation will be required in areas such as multi-band pooling, uplink enhancement, and multi-dimensional coordination,” the source said.
Huawei also believes enhanced uplink capabilities will help operators generate new commercial opportunities as AI services mature. During MWC Shanghai, the company demonstrated AI-focused consumer, enterprise and home solutions, including pure voice calling and intelligent home assistants. According to Huawei, these offerings are intended to help operators create new sources of commercial value as Mobile AI services expand.
Although the Beijing deployment demonstrates the architecture at large scale, Huawei stopped short of discussing international deployment plans, noting only that overseas implementations will comply with local regulations and spectrum requirements in each market.
The deployment reflects a broader evolution of 5G-Advanced as vendors and operators increasingly position uplink performance alongside download capacity as networks adapt to AI-driven traffic patterns. Rather than focusing solely on delivering content to users, operators are beginning to prepare networks for an environment where billions of connected devices continuously upload data for real-time AI processing.
