Priority Random Access and Power Control for NOMA-ALOHA in Heterogeneous mMTC

Technical Summary: Priority Random Access and Power Control for NOMA-ALOHA in Heterogeneous mMTC

Problem Statement
The paper addresses the challenges of supporting massive machine-type communications (mMTC) in heterogeneous networks, specifically focusing on devices with diverse Quality of Service (QoS) requirements, such as delay-sensitive and delay-tolerant Machine-Type Devices (MTDs). While Non-Orthogonal Multiple Access (NOMA) assisted ALOHA (NOMA-ALOHA) is a promising grant-free solution, existing schemes face three primary limitations in practical scenarios:

1. Non-Ideal Conditions: Most existing models assume perfect Channel State Information (CSI) and perfect Successive Interference Cancellation (SIC). In reality, imperfect CSI and imperfect SIC introduce residual interference, which can severely degrade signal decoding performance and system stability.
2. Lack of Differentiated QoS: Conventional NOMA-ALOHA schemes do not adequately support priority-based access. They fail to provide distinct access advantages for delay-sensitive MTDs over delay-tolerant ones, especially under dynamic traffic loads where the proportion of high-priority devices fluctuates.
3. Energy Efficiency and Overload: Existing user barring algorithms (UBA) used to mitigate user overload often rely on random number generation without considering channel quality. This can lead to MTDs with poor channel conditions being allowed to transmit, requiring excessively high transmit power (due to channel inversion-based power control), thereby reducing energy efficiency.

Methodology
The authors propose a comprehensive framework involving a new channel model, two priority access strategies, and an enhanced power control algorithm.

1. Received Power Level Model under Non-Ideal Conditions:
   The paper first derives a received power level model that explicitly incorporates imperfect CSI (modeled as Gaussian noise) and imperfect SIC (characterized by a SIC efficiency factor $\delta$). This model determines the required received power levels ($r_q$) to achieve a target Signal-to-Interference-plus-Noise Ratio (SINR) $\Gamma$. The study establishes a critical threshold for SIC efficiency ($\delta^*$); if the actual efficiency falls below this threshold, the power levels become infeasible (negative), preventing stable system operation.

2. Priority Random Access (PRA-NA) Schemes:
   To support heterogeneous MTDs, the authors divide the available power levels into two groups: a "high-power group" (more reliable, fewer collisions) and a "low-power group."

   • Fixed PRA-NA (FPRA-NA): Designed for scenarios with stable traffic proportions. The number of power levels allocated to high-priority and low-priority MTDs ($Q_1$ and $Q_2$) remains fixed.
   • Adaptive PRA-NA (APRA-NA): Designed for dynamic traffic scenarios. The allocation of power levels ($Q_1$ and $Q_2$) is dynamically adjusted based on the real-time proportion of high-priority MTDs ($\mu$) and priority weights, ensuring that high-priority devices receive more access resources when their traffic load increases.
   • Throughput Analysis: The paper derives upper-bound throughput expressions for both schemes, assuming that collisions in lower power levels do not affect decoding in higher power levels (a simplifying assumption to derive tractable bounds).

3. Enhanced User Barring Algorithm (EUBA):
   To address energy efficiency and overload, the authors propose EUBA, which combines a conventional User Barring Algorithm (UBA) with a channel quality-based check.

   • Mechanism: An MTD first undergoes conventional UBA. If it passes, it transmits. If it fails, it undergoes a secondary check against a channel quality threshold ($H^*$). Only MTDs with channel quality above this threshold are permitted to transmit.
   • Optimization: The algorithm calculates an optimal channel quality threshold and a trade-off parameter ($\theta$) to balance between location-dependent fairness (preventing distant devices from being completely barred) and energy efficiency (preventing poor-channel devices from transmitting at high power).

Key Contributions

• Robust Modeling: The development of a received power level model for NOMA-ALOHA that accounts for imperfect CSI and imperfect SIC, enabling reliable system design under non-ideal conditions.
• Priority Strategies: The design of FPRA-NA and APRA-NA schemes that provide differentiated access priorities. The APRA-NA scheme specifically addresses the challenge of fluctuating traffic loads by adaptively adjusting resource allocation.
• Energy-Efficient Control: The proposal of the EUBA, which integrates channel quality thresholds into the barring process. This reduces the average transmit power by barring MTDs with poor channel conditions while maintaining system fairness through the trade-off parameter $\theta$.
• Performance Validation: Extensive mathematical analysis and simulation results demonstrating the effectiveness of the proposed schemes in terms of throughput, access delay, and energy efficiency.

Results
Simulation results, utilizing 3GPP Beta traffic models and parameters reflecting practical constraints (e.g., $\sigma^2_e = 10^{-6}$, $\delta = 0.1 \times \delta^*$), show:

• Throughput: The proposed PRA-NA schemes achieve throughput comparable to conventional NOMA-ALOHA but with superior performance for high-priority MTDs. The APRA-NA scheme maintains performance close to ideal benchmarks even when the proportion of high-priority MTDs is high.
• Access Delay: The proposed schemes significantly reduce the average access delay for high-priority MTDs compared to conventional NOMA-ALOHA. The APRA-NA scheme is particularly effective in dynamic traffic conditions, maintaining low delay for high-priority devices even as their population increases.
• Energy Efficiency: The EUBA effectively reduces the average transmit power. By restricting access for MTDs with poor channel quality, the system avoids the high power costs associated with channel inversion. The trade-off parameter $\theta$ allows for a tunable balance between power reduction and location fairness.
• Fairness: The EUBA mitigates the location-dependent unfairness inherent in pure channel-quality-based barring. When $\theta$ is appropriately set, MTDs far from the base station are still granted transmission opportunities, albeit with reduced frequency compared to near-field devices.

Significance and Claims
The paper claims that its proposed framework offers a significant improvement over conventional NOMA-ALOHA in heterogeneous mMTC environments. By explicitly modeling non-ideal channel conditions, the work ensures that priority access strategies are robust and feasible in real-world deployments. The introduction of adaptive resource allocation (APRA-NA) addresses the dynamic nature of mMTC traffic, ensuring that delay-sensitive applications receive the necessary QoS guarantees. Furthermore, the EUBA provides a practical solution to the energy efficiency problem, demonstrating that user overload can be managed without forcing low-power MTDs to transmit at unsustainable power levels. The authors conclude that their approach significantly improves throughput, access delay, and energy efficiency, making it a viable candidate for 6G mMTC systems.