Abstract
In this paper, we develop a spatiotemporal model to analysis of cellular user in underlay D2D communication by using stochastic geometry and queuing theory. Firstly, by exploring stochastic geometry to model the user locations, we derive the probability that the SINR of cellular user in a predefined interval, which constrains the corresponding transmission rate of cellular user. Secondly, in contrast to the previous studies with full traffic models, we employ queueing theory to evaluate the performance parameters of dynamic traffic model and formulate the cellular user transmission mechanism as a M/G/1 queuing model. In the derivation, Embedded Markov chain is introduced to depict the stationary distribution of cellular user queue status. Thirdly, the expressions of performance metrics in terms of mean queue length, mean throughput, mean delay and mean dropping probability are obtained, respectively. Simulation results show the validity and rationality of the theoretical analysis under different channel conditions.
Introduction
As we move toward 5G and beyond, Device-to-Device (D2D) communication—where nearby devices communicate directly rather than through a base station—has become essential for improving spectral efficiency. In the "underlay" mode, D2D users share the same frequency bands as cellular users (CUs), which introduces significant interference challenges.
Traditional performance analyses often assume a "saturation" state where users always have an infinite backlog of data. However, real-world data arrives in bursts, and device buffers are finite. This research addresses the gap by introducing a spatiotemporal model that combines stochastic geometry with queuing theory to capture the realities of random data arrivals, finite buffers, and adaptive transmission rates in dynamic traffic environments.
System Model and Methodology
The researchers employed a multi-layered mathematical approach to capture both the spatial distribution of devices and the temporal nature of data traffic.
Stochastic Geometry
Base stations and users are modeled as a Homogeneous Poisson Point Process (PPP). This spatial modeling allows for the calculation of the probability distribution of the Signal-to-Interference-plus-Noise Ratio (SINR) based on random spatial interference from both intra-cell and inter-cell D2D pairs.
Queuing Theory (M/G/1)
To handle dynamic traffic, the cellular user’s buffer is modeled as an M/G/1 queue (Poisson arrival, General service time). The service time is considered "General" because the transmission rate is not fixed; it adaptively changes based on the fluctuating SINR. An embedded Markov chain is used to find the stationary distribution of the queue status, enabling the derivation of exact formulas for throughput and delay.
Adaptive Transmission Strategy
The model assumes the network utilizes adaptive modulation (e.g., MQAM). The transmission rate is partitioned into intervals based on the SINR; if the signal quality is high, the rate increases, whereas high interference causes the rate to drop or stop entirely.
Key Insights and Results
The Reality of Finite Buffers
By modeling a finite buffer ($L_{max}$), the study accurately predicts the "dropping probability"—the point at which a user's buffer overflows. This is a critical metric missed by infinite-buffer models, which often lead to unrealistic delay predictions. For instance, the proposed model predicted a stable delay of ~2.4ms under certain conditions, while an infinite-buffer model erroneously predicted 22ms.
Saturation and Interference
There is a clear "knee" in the performance curves: throughput rises linearly with the packet arrival rate until it hits the channel's maximum capacity. Beyond this saturation point, additional data only increases delay and dropping probability. Furthermore, the study shows that interference is a direct byproduct of D2D pair density; as density increases, the effective service rate of the cellular user's queue slows down.
Impact of Network Parameters
- Path-Loss ($\alpha$): A higher path-loss exponent limits the reach of interference from distant D2D users, causing the queue to reach saturation faster.
- Base Station Density: In dense urban models, cellular users experience higher interference from the concentration of D2D pairs, leading to dropping probabilities as high as 30% even at moderate arrival rates.
Conclusion and Future Work
The proposed spatiotemporal model provides a more tractable and realistic framework for analyzing wireless requirements than traditional saturation models. Future work may expand this analysis by incorporating imperfect Channel State Information (CSI) and random D2D link lengths to further generalize the findings. Additionally, a dual-analysis tracking the queuing performance of both cellular and D2D users simultaneously would provide a more holistic view of network health.
Summary updated