Abstract
The Internet of Things (IoT) is large-scale by nature, manifested by the massive number of connected devices and their vast spatial existence. Cellular networks are natural candidates to provide first-mile access for the resulting data tsunami, yet they face significant scalability problems in the uplink. This paper develops a traffic-aware spatiotemporal mathematical model for IoT devices using stochastic geometry and queueing theory. By accounting for traffic requirements, transmission strategies, and mutual interference, the model characterizes the cellular network's capacity to accommodate IoT traffic. We assess three transmission strategies—transmission persistency, backoff, and power-ramping—offering insights into the stability regions and effective scenarios for each.
Introduction
The Internet of Things (IoT) is fundamentally changing the demands on cellular networks. While traditional networks were optimized for high-bandwidth downlink traffic, IoT is characterized by a "data tsunami" of small, sporadic uplink packets from a massive number of spatially distributed sensors and machines.
This creates a significant scalability challenge. Existing mathematical tools often fall short: stochastic geometry is effective at modeling spatial interference but usually assumes a "backlogged" state where every device always has data to send. Conversely, queueing theory excels at modeling traffic timing but often uses oversimplified collision models that ignore the complex physics of wireless interference. This research bridges that gap to find the actual limits of how many IoT devices a cellular base station can support before the system becomes unstable.
Methodology
The developed spatiotemporal model operates on two distinct levels to capture the complexity of massive IoT deployments:
- Spatial Modeling: Independent Poisson Point Processes (PPPs) are used to model the random locations of both base stations and IoT devices. The model assumes a Rayleigh fading environment and full path-loss inversion power control.
- Temporal Modeling: Each IoT device is modeled as a Geo/PH/1 discrete-time Markov chain (DTMC). "Geo" represents the geometric arrival of packets, and "PH" (Phase-type) represents the complex service process, accounting for retransmissions and protocol states.
The research evaluates three specific transmission strategies by adjusting the DTMC structure:
- Baseline (Persistent) Transmission: Devices attempt transmission as soon as a packet is in the queue.
- Power-ramping: Devices increase transmit power in successive retransmission attempts to improve success probability.
- Backoff: Devices wait for a random period after a failed transmission to reduce immediate congestion.
A key contribution is the iterative algorithm designed to solve the "causality problem." Since a device's success probability depends on aggregate interference—which in turn depends on the queue status of all other devices—the authors use a fixed-point iteration to find the steady-state behavior of the network.
Key Insights and Results
The analysis identifies a clear "Stability Cliff": a maximum spatial traffic intensity beyond which device queues grow indefinitely, leading to a packet loss probability of 100%.
Strategy Performance
- Power-Ramping: Highly effective at extending the stability region when the SINR (Signal-to-Interference-plus-Noise Ratio) threshold is high, as it helps "flush" packets out of the queue faster.
- Backoff: While essentially unnecessary when the network is stable, backoff becomes mandatory in congested scenarios to prevent total network collapse.
- Baseline Transmission: Surprisingly, this is more effective than power-ramping in high-density, low-SINR scenarios because it avoids unnecessary interference spikes.
Analytical Accuracy
The analytical model showed a near-perfect match with system-level Monte Carlo simulations (conducted over a 100 km² area with varying device intensities). A significant finding was that under full channel inversion power control, the temporal correlation of interference is negligible, allowing for a tractable model without sacrificing accuracy.
Conclusion and Future Work
This framework provides a rigorous method for mapping the stability region of IoT-enabled cellular networks. While the current model relies on full path-loss inversion, future work will explore fractional power control, multi-antenna (MIMO) systems, and non-orthogonal multiple access (NOMA) to better reflect the constraints of power-limited IoT devices in complex urban environments.
Summary updated