Qualcomm and Nokia unveil groundbreaking interoperability tech for AI in wireless systems

Qualcomm Technologies and Nokia Bell Labs have jointly developed a prototype technology that could overcome an important challenge of interoperability in wireless systems that prevents cooperation between multi-vendor artificial intelligence (AI) systems in devices and networks.

The breakthrough will pave the way for future wireless systems to harness AI for enhanced capacities, improved performance, and potentially lower energy consumption.

Peter Vetter, President of Bell Labs Core Research, Nokia, said, "The potential for AI is substantial in communications. A data-driven air interface can learn about its environment to improve its performance and efficiency. We are already seeing AI implemented in 5G and 5G-Advanced, however, it will achieve its fullest potential in the 6G era."

The technology has been successfully integrated into an AI proof-of-concept on a Nokia Bell Labs test network using a Snapdragon 5G modem-RF system. This proof-of-concept will be showcased at the Mobile World Congress 2024 in Barcelona.

The new approach utilizes sequential learning, wherein the network and device share data about the connection relevant to AI training, but do not have to share the actual AI model. By employing independent encoders and decoders, the technology facilitates the training of network and device AI models separately. These encoders and decoders act as translators, enabling the seamless exchange of knowledge between distinct AI systems, irrespective of the vendor.

The technique can also enable wireless handset or devices from multiple vendors with unique encoders to interwork with the same base station with its common decoder in the network, enhancing training and scalability for future real-world deployments.

Commenting on this development, Tingfang Ji, VP of Engineering at Qualcomm Technologies, said, "AI has revolutionized many aspects of our life and society, and Qualcomm is leading the edge AI technology. For AI to bring new capabilities, efficiency and performance to multi-vendor wireless systems, the ecosystem needs to work together to enable wireless AI on a global scale. Collaborating with Nokia on AI interoperability is an encouraging first step toward defining a scalable global standard."