IoT-U: Cellular Internet-of-Things Networks over Unlicensed Spectrum

Hongliang Zhang, Boya Di, Kaigui Bian, Lingyang Song

Abstract

In this paper, we consider an uplink cellular Internet-of-Things (IoT) network, where a cellular user (CU) can serve as the mobile data aggregator for a cluster of IoT devices. To be specific, the IoT devices can either transmit the sensory data to the base station (BS) directly by cellular communications, or first aggregate the data to a CU through Machine-to-Machine (M2M) communications before the CU uploads the aggregated data to the BS. To support massive connections, the IoT devices can leverage the unlicensed spectrum for M2M communications, referred to as IoT Unlicensed (IoT-U). Aiming to maximize the number of scheduled IoT devices and meanwhile associate each IoT devices with the right CU or BS with the minimum transmit power, we first introduce a single-stage formulation that captures these objectives simultaneously. To tackle the NP-hard problem efficiently, we decouple the problem into two subproblems, which are solved by successive linear programming and convex optimization techniques, respectively. Simulation results show that the proposed IoT-U scheme can support more IoT devices than that only using the licensed spectrum.

Introduction

As we move toward a world with billions of connected devices, traditional licensed cellular bands (like those used for NB-IoT) are becoming increasingly congested. This "massive connectivity" requirement creates two primary challenges: spectrum scarcity and interference management. When many devices transmit simultaneously, they interfere with each other, significantly degrading network performance. To address this, the proposed IoT-U framework expands cellular IoT networks into the unlicensed 5GHz spectrum (typically used by Wi-Fi) to offload traffic and enhance capacity.

System Architecture

The IoT-U system models an uplink network where IoT devices choose between two primary modes of operation. In Cellular Mode, devices communicate directly with the Base Station (BS). In Aggregation Mode, devices relay their data through nearby Cellular Users (CUs) acting as mobile aggregators via Machine-to-Machine (M2M) communication. This hierarchical structure allows the network to leverage the proximity of CUs to reduce the transmit power required by low-power IoT sensors.

System model of the IoT-U network featuring cellular users (CUs) as data aggregators and coexistence with Wi-Fi users (WUs).

Coexistence Mechanism and Resource Allocation

A critical challenge in using unlicensed bands is ensuring fair coexistence with existing Wi-Fi users. The authors implement a duty-cycle-based protocol, where the IoT-U network operates during specific "ON" periods and remains silent during "OFF" periods to allow Wi-Fi traffic.

Duty-cycle-based scheduling for fair coexistence between IoT-U and Wi-Fi networks.

To manage this complex environment, the authors developed the IASPA (Iterative Association, Scheduling, and Power Allocation) algorithm. This algorithm solves a Mixed-Integer Non-Linear Programming (MINLP) problem by decoupling it into two manageable subproblems:

  1. Association and Scheduling: Determining which devices connect to which aggregator or the BS using successive linear programming.
  2. Power Allocation: Optimizing transmit power using Successive Convex Approximation (SCA) to transform non-convex constraints into convex ones.

Key Insights and Results

Numerical simulations, modeling a 500m cell radius with 300 IoT devices, revealed several key findings:

  • Capacity Gains: By moving M2M traffic to the unlicensed spectrum, the system reduces interference on licensed bands. This "offloading gain" allows the BS to support nearly double the devices compared to traditional licensed-only cellular networks.
  • Traffic-Dependent Association: Devices with high data requirements typically associate directly with the BS to avoid relay bottlenecks, while low-traffic sensors are better served by CU aggregators.
  • Efficiency and Convergence: The IASPA algorithm proved computationally efficient, typically converging within 7 to 11 iterations. Furthermore, the IoT-U scheme demonstrated superior energy efficiency per scheduled device compared to baseline optimization methods.

Conclusion and Future Work

While IoT-U significantly improves connectivity, the current model relies on centralized control at the Base Station, which may introduce signaling overhead in massive deployments. Additionally, the static duty cycle for Wi-Fi coexistence could be improved by dynamic mechanisms like Listen-Before-Talk (LBT) to better adapt to real-time Wi-Fi traffic density. Future research will focus on these decentralized and dynamic enhancements to ensure global regulatory compliance and scalability.

Summary updated

Abstract

In this paper, we consider an uplink cellular Internet-of-Things (IoT) network, where a cellular user (CU) can serve as the mobile data aggregator for a cluster of IoT devices. To be specific, the IoT devices can either transmit the sensory data to the base station (BS) directly by cellular communications, or first aggregate the data to a CU through Machine-to-Machine (M2M) communications before the CU uploads the aggregated data to the BS. To support massive connections, the IoT devices can leverage the unlicensed spectrum for M2M communications, referred to as IoT Unlicensed (IoT-U). Aiming to maximize the number of scheduled IoT devices and meanwhile associate each IoT devices with the right CU or BS with the minimum transmit power, we first introduce a single-stage formulation that captures these objectives simultaneously. To tackle the NP-hard problem efficiently, we decouple the problem into two subproblems, which are solved by successive linear programming and convex optimization techniques, respectively. Simulation results show that the proposed IoT-U scheme can support more IoT devices than that only using the licensed spectrum.
BibTeX
@article{zhang2019iot,
  author = {Hongliang Zhang and Boya Di and Kaigui Bian and Lingyang Song},
  title = {{IoT-U: Cellular Internet-of-Things Networks over Unlicensed Spectrum}},
  journal = {IEEE Trans. Wireless Communications},
  year = {2019},
  url = {https://openreview.net/forum?id=bhFyxfDhwc},
}
Figure