Abstract
In this paper, we propose cumulative distribution function (CDF)-based multiuser scheduling (CMS) for downlink non-orthogonal multiple access (NOMA) to accommodate a massive number of IoT devices over limited radio resources and statistically analyze the ergodic capacity of devices. With the proposed CMS algorithm, multiple devices having relatively better channel gain to exploit multiuser diversity are selected while meeting QoS requirements by controlling the amount of resources. We first provide the user selection probability in terms of the weight of each device, which determines the amount of time resources. Based on some selected results, it is shown that equal weights ensure fair resource allocation. Further, for fair resource allocation scenario, the distribution of signal-to-interference-plus-noise ratio (SINR) when a specific device is selected is derived. With this derived statistical result of SINR, we analyze the ergodic capacity of the specific device. Some selected results show that the proposed CMS algorithm allocates the time resource more fairly than proportional fair scheduling. Further, the average throughput and amount of resources can be controlled by adjusting the weights of each device, so different QoS requirements can be guaranteed.
Introduction
As we move toward a world of "massive IoT," 5G networks face a daunting challenge: connecting tens of billions of low-cost, low-power devices over limited radio resources. Traditional Orthogonal Multiple Access (OMA) schemes, like TDMA or OFDMA, struggle because they assign unique "slots" to each user, which quickly run out.
Non-Orthogonal Multiple Access (NOMA) offers a solution by allowing multiple users to share the same time and frequency resources simultaneously, distinguishing them in the power domain. However, NOMA introduces a new scheduling challenge: deciding which users should share a resource to maximize throughput without leaving cell-edge users (those with poor signals) in the dark. Existing methods like Proportional Fair (PF) scheduling are computationally complex and often fail to guarantee specific Quality of Service (QoS) levels or fair resource distribution for every device in a massive network.
Proposed CMS Framework
The authors propose the CMS (CDF-based Multiuser Scheduling) algorithm to address these scheduling complexities. The approach leverages the statistical properties of user channels rather than raw signal strength alone.
The CMS Algorithm Breakdown
- CDF-Based Selection: Instead of looking at raw channel gains (which favors users close to the base station), the algorithm looks at the Cumulative Distribution Function (CDF) of each user's channel gain.
- The Ranking Metric: Users are ranked based on the value $F_i(\alpha_i)^{1/w_i}$, where $F_i$ is the CDF, $\alpha_i$ is the instantaneous channel gain, and $w_i$ is an adjustable weight. The top $2M$ users are selected to be paired across $M$ sub-bands.
- Weight Tuning: The weight $w_i$ acts as a control mechanism for the time fraction a user is selected. If all weights are equal, every user has an equal probability of being scheduled, ensuring "resource fairness."
- Statistical Analysis: The framework provides a rigorous mathematical derivation of the Signal-to-Interference-plus-noise Ratio (SINR) distribution and ergodic capacity (the long-term average data rate) for users under this scheduling regime.
Performance and Fairness
A major insight of this research is the distinction between resource fairness and throughput fairness. While Proportional Fair (PF) scheduling is effective at balancing data rates, it often gives cell-edge users more time resources to compensate for their poor signals. CMS with equal weights ensures every user gets the same amount of time, which is often more predictable for network planning.
Even in a crowded IoT environment, CMS successfully exploits "multiuser diversity." By picking users who are currently experiencing their own "best" channel conditions relative to their own statistics, the system improves overall efficiency. Furthermore, by adjusting the weight $w_i$, a network provider can guarantee that a high-priority industrial sensor gets more "airtime" than a low-priority smart meter, regardless of their physical distance from the base station.
Key Results
- Perfect Fairness: The authors proved mathematically that when weights are equal, the probability of any user being selected is exactly $2M/K$ (where $M$ is the number of sub-bands and $K$ is the total users).
- Superior SINR: CMS provides significantly higher SINR for cell-edge users compared to random scheduling because it waits for a "relatively good" channel state for that specific user before transmitting.
- Scalability: The sum rate of the system under CMS increases as the number of users grows, demonstrating excellent scalability for massive IoT deployments.
- Validation: The derived analytical formulas for ergodic capacity matched simulation results perfectly, providing a reliable tool for engineers to predict network performance without exhaustive simulations.
Limitations and Future Work
While promising, the study assumes a maximum of two users per sub-band due to the practical complexity of Successive Interference Cancellation (SIC). Future work could explore higher-order NOMA (3+ users per slot). Additionally, while the paper shows that weights can control QoS, it does not provide an automated algorithm to dynamically calculate the optimal weights needed to meet specific bit-rate targets. Finally, the analysis assumes perfect SIC; in real-world hardware, imperfect SIC would likely degrade the throughput results.
Summary updated