Economy/Home · Economy

NVIDIA Expands 6G Open Source Ecosystem; Robot Task Time Cut 52.6%, Network Power Consumption Reduced 45.6%

NVIDIA is rapidly expanding its open-source software ecosystem supporting the development of 6G, the next-generation mobile communications technology. Research achievements are being unveiled one after another, including reducing robots' task processing times, improving wireless signal reception per

Wooil Shim
Staff Reporter
17 min read
NVIDIA Expands 6G Open Source Ecosystem; Robot Task Time Cut 52.6%, Network Power Consumption Reduced 45.6%
CBC News

NVIDIA is rapidly expanding its open-source software ecosystem supporting the development of 6G, the next-generation mobile communications technology. Research achievements are being unveiled one after another, including reducing robots' task processing times, improving wireless signal reception performance, and lowering communication network power consumption. As moves to integrate artificial intelligence (AI) into network operations gather pace, the technological convergence of semiconductors and communications infrastructure is drawing attention.

On the 6th, NVIDIA introduced on its official blog cases of technology development by major research institutions worldwide using its wireless communications research platform. What stands out in this announcement is the pace of the platform's adoption. According to NVIDIA, the global developer adoption of AI Aerial and Sionna grew tenfold over the past year. Cumulative downloads of its 5G and 6G open-source tools have exceeded 1 million, and more than 8,000 research and academic institutions worldwide use the platform.

About a year ago, NVIDIA released its Aerial wireless software as open source. Together with the existing Sionna, it expanded the development environment so that researchers can test wireless communication algorithms and compare results. This technology development is not aimed solely at communication speed. It is progressing toward integrating communication and computing functions so that future autonomous driving systems, industrial robots, and AI-based devices can respond quickly to changes in their surroundings.

■ Robot task times cut by up to 52.6%

A research case from King's College London in the UK demonstrates the potential of this change. The researchers built an experimental environment combining wireless communications and edge AI by connecting a camera-equipped robotic arm with the NVIDIA Jetson AGX Orin. The key is enabling the robot to perform necessary judgments while reducing the amount of information it transmits.

In the conventional approach, the robot transmits video or images it captures at regular intervals for analysis by an external system. In contrast, the goal-oriented semantic communication method applied by the researchers selectively transmits only the information needed to perform tasks, such as object positions and characteristics and relations with surrounding objects.

In experiments involving fruit packaging and block-stacking tasks, task completion time was reduced by up to 52.6% compared with the existing periodic image transmission method. The researchers reported that task success rates also improved. Notably, the system demonstrated the ability to recognize and respond to errors such as dropping objects or mistakes during tasks. In experiments applying digital twins, the researchers also tested a function in which the robot checks the expected path for potential collisions before performing actual movements. However, these results were measured under specific experimental conditions, and the same improvements do not necessarily appear for all robots or wireless communication environments.

■ AI receiver improves uplink throughput by up to 60%

The potential of AI was also confirmed in wireless signal reception technology. SoftBank of Japan integrated a neural network-based receiver with NVIDIA's Aerial CUDA-accelerated radio access network technology and tested it on commercial 5G smartphones. In wireless environments, when multiple signals overlap or interference occurs, data transmission errors increase and retransmissions become necessary, potentially reducing communication efficiency. SoftBank approached this problem by having AI learn the characteristics of received signals.

In controlled test environments including interference, the neural network receiver recorded higher uplink throughput than the conventional receiver in all 15 test conditions, with the largest improvement reaching about 60%.

Related research was also conducted at ETH Zurich in Switzerland. Using the NVIDIA Aerial testbed, the researchers tuned a neural network receiver with measurements collected in real wireless environments. As a result, the block error rate was reduced by more than one-third compared with some pre-trained models. The two studies show that AI can improve signal restoration capability in communication data processing. Such technologies could be considered for use in environments where stable data transmission is critical, such as disaster response and industrial sites.

■ Network power consumption reduced 45.6%

Research on improving the energy efficiency of communication networks is also drawing attention. Researchers at the University of Leeds in the UK developed a system that predicts communication demand in advance and controls base station power usage. The system analyzes expected network demand over the next 10 minutes and is designed to switch some base stations to power-saving mode within a range that does not affect communication services.

Before reducing the power usage of a specific base station, the system checks whether neighboring base stations can handle communication demand in the area. It then continues to monitor network conditions and re-secures communication capacity if needed. The researchers validated this operational method using a portable 5G system based on the NVIDIA DGX Spark and digital twin technology.

Experimental results showed that radio access network (RAN) power consumption decreased by 45.6% relative to maximum power usage while maintaining device connections. This is a case demonstrating that AI-driven optimization of network operations can contribute to improving energy efficiency. However, the reduction rate is a figure based on the study's test conditions and comparison baseline, and actual savings in commercial networks may vary depending on communication demand, facility configuration, and operating environment.

■ 6G simulation: 3 hours down to 10–20 minutes

Simulation times required for 6G development are also being reduced. The Centre Tecnològic de Telecomunicacions de Catalunya (CTTC) in Spain conducted research to reduce the computational burden of wireless network simulations using NVIDIA's GPU acceleration technology. To deploy wireless communication technologies in real environments, scenarios such as many devices connecting simultaneously or signal interference conditions must be examined in advance, but the required computational load increases sharply as network scale grows.

CTTC combined NVIDIA's GPU-accelerated wireless channel model with 5G-LENA, an open-source network simulator. In initial evaluations, simulation run times that previously took about 3 hours were reduced to the 10–20 minute range. NVIDIA explained that processing times were shortened while maintaining expected statistical characteristics and network performance, though results vary depending on the integration method. Faster simulations allow researchers to compare more design options and test diverse communication environments within the same amount of time. They can also be more useful in the process of identifying technical problems before actual equipment deployment.

The research cases disclosed this time show that NVIDIA's 6G technology strategy is expanding beyond GPU-based computing to wireless communication software, AI models, and network operations technology. Robot control, signal reception, base station power management, and network simulation are different fields, but they share the commonality of using AI to improve communication system efficiency.

In next-generation 6G networks, environments where not only people and smartphones but also robots, autonomous systems, and AI-based devices connect simultaneously are emerging as an important research challenge. In this process, communication networks may evolve into infrastructure that supports computing and environmental awareness functions in addition to data delivery. However, most of the results currently disclosed are at the stage of experiments and technical verification by research institutions, and they should be distinguished from evidence proving 6G commercialization schedules or actual adoption outcomes by telecom operators.

The significance of the cases NVIDIA disclosed lies in the fact that multiple research institutions are testing technologies needed for next-generation networks in a shared development environment. Attention is focusing on how the combination of AI semiconductors and wireless communication technology will change the design and operation of future communications infrastructure. [Some AI assistance was used in producing this article.]

Wooil Shim
Staff Reporter

CBC Globe publishes verified stories with editorial review, source checks, and tenant-specific publication standards.