A new study by Juniper Research, the foremost experts in telecommunications, has found the global number of cellular IoT devices will increase from 3.4 billion in 2024 to 6.5 billion by 2028. However, the study predicts that this 90% growth in connections will require deploying new services that enable efficient IoT device management and security automation.
The study identifies intelligent infrastructure management solutions as a crucial tool in managing the exponential increase in cellular data. These solutions empower IoT users to automate device configuration, security processes, and real-time connectivity. The research firm anticipates significant global cellular IoT data growth, reaching 46 petabytes in 2028, up from 21 petabytes this year. This projection underscores the need for further investment in IoT automation services, including federated learning.
Federated Learning to Enable Operators to Minimise Security Risks
Currently, most machine learning models are trained using data sources stored in a single location, which malicious entities can exploit. The study proposes a shift to federated learning models, a subset of machine learning. This approach leverages a decentralized data strategy, significantly reducing the risk of data fraud over IoT networks.
Federated machine learning limits the exposure of sensitive IoT data, thus reducing the threat of data breaches. Research author Alex Webb remarked: “As the number of cellular IoT connections grows, it is imperative that both platforms and operators ensure data is secure in transition and on the device. A failure to do so will dissuade IoT users in industries with sensitive data from using a cellular IoT-based approach to connectivity.”