Japanese study finds IoT networks face fairness & throughput trade-offs

Researchers from Japanese universities have developed a detailed mathematical model evaluating how grant-free communication schemes perform under conditions of massive Internet of Things (IoT) connectivity.

The study, published in the journal Computer Communications, investigates how slotted ALOHA—a widely used grant-free access method—operates when networks host extremely high densities of IoT devices. The results reveal notable performance constraints and fairness issues that may affect the design of future 5G and 6G wireless systems supporting applications such as smart cities, connected vehicles, and remote healthcare.

Grant-free schemes examined

Massive Machine Type Communication (mMTC) is intended to connect extremely large numbers of IoT devices, with some projections suggesting up to one million devices per square kilometre may require sporadic but reliable data transmission. Grant-free communication methods, including slotted ALOHA, enable such devices to send data without first requesting access from base stations. This approach reduces the processing and power loads on both end devices and central infrastructure, but it poses a risk of data collision and congestion, especially in crowded environments.

The study was led by Professor Shigeo Shioda from the Graduate School of Informatics at Chiba University, with contributions from Mr. Yuki Ichimura, also of Chiba University, and Professor Takeshi Hirai from the Graduate School of Information Science and Technology at The University of Osaka. Their analytical model relies on stochastic geometry to represent both base stations and IoT devices as randomly distributed entities across the network area.

Performance and fairness trade-offs

Three different variants of the slotted ALOHA protocol were evaluated: the basic version, a version with interference cancellation known as Non-Orthogonal Multiple Access (NOMA), and a version with power control, where devices adjust their transmission strengths.

The researchers focused on two primary metrics: transmission success probability and base station throughput, defined as the rate at which a base station can successfully receive data. While the use of interference cancellation improved base station throughput by as much as 20 per cent in some configurations, it did not solve the so-called 'near-far problem'. This issue arises when devices located close to a base station have a much higher success rate than those situated further away.

Notably, the study observed that interference cancellation benefited devices at intermediate distances in particular, whereas those nearest or farthest from the base station saw less improvement. By contrast, introducing power control addressed the near-far problem and promoted more equitable transmission opportunities across the network but decreased the overall throughput performance.

"Our study reveals that ALOHA-based communications face an inherent trade-off between two conflicting objectives: fairness, in the sense that devices should have equal opportunities to communicate regardless of their distance from the base station, and throughput, the goal of enabling a single base station to receive data from as many devices as possible," explains Prof. Shioda. "In other words, it is fundamentally difficult to achieve both fairness and maximum throughput simultaneously."

This finding indicates that future designers of wireless communications for IoT environments may need to carefully balance optimal data flow and equitable access, rather than relying on a single approach to meet both goals.

Implications for future networks

The authors believe their findings will inform the future development of IoT systems where high device density and reliable data exchange are critical. Specific applications mentioned include vehicle-to-everything communications, in which information is shared among vehicles and road infrastructure, and remote healthcare involving wearable medical devices.

"We have shed light on the inherent limitations of IoT networks that will form the backbone of future IoT-driven societies. These limitations stem from the use of grant-free communication schemes and may potentially be overcome by adopting grant-based schemes. In future work, we intend to explore this possibility further," concludes Prof. Shioda.

This study is an extended version of earlier research recognised with the Best Paper Award at ACM MSWiM 2023, an international conference on wireless and mobile system simulations. The project was supported in part by funding from JSPS KAKENHI and the Support Center for Advanced Telecommunications Technology Research.

The work underscores that knowledge of the limitations and trade-offs inherent in current grant-free communication schemes will be important in shaping next-generation networks designed for both efficiency and fairness in IoT-driven societies.