This paper proposes a priority-aware EP-ALOHA framework with predictive radio resource allocation for heterogeneous machine-to-machine (M2M) devices in 5G massive machine-type communication (mMTC) networks. The proposed framework extends conventional EP-ALOHA by introducing M2M priority classes, priority-dependent delay constraints, and Exploration Phase resource block (RB) allocation. The RB-allocation problem is formulated as an integer-constrained optimization problem, where the objective is to improve effective radio channel utilization while satisfying delay constraints for different priority classes. A Genetic Algorithm-based optimization procedure is used to generate optimization-derived RB-allocation targets under different traffic and system parameter settings. These targets are then used to train and evaluate predictive RB-allocation models, including Random Forest, Neural Network, Gradient Boosting, and Linear Regression. The simulation results show that the proposed priority-aware EP-ALOHA method achieves a higher successful access probability than the considered baseline schemes within the feasible operating region. For predictive RB allocation, the Neural Network achieved the best test-set performance, with MSE = 25.5002, RMSE = 5.0498 RBs, MAE = 3.2629 RBs, and R² = 0.9810. A separate computational evaluation showed that Random Forest inference reduced the mean allocation-decision time from 213.54 ms for GA-based optimization to 15.20 ms, corresponding to a 14.05-fold speed-up on the evaluated platform. In addition, M2M device activity probability forecasting is evaluated using Bayesian estimation, LSTM, moving average, and exponential smoothing. LSTM achieves the lowest forecasting error, while exponential smoothing provides a close and computationally simpler alternative. The results indicate that the proposed framework can support proactive and priority-aware resource management for heterogeneous M2M traffic, while the learning-based components are used as approximation and forecasting tools rather than as universally superior solutions.

Reliable vehicular safety broadcasting requires channel access that limits collisions without imposing excessive delay on urgent messages. However, fixed contention windows and single-indicator adaptation do not fully represent the combined effects of contender population, channel occupancy, and repeated backoff interruption. This study proposes Multi-Indicator Responsive Adaptation integrated with Enhanced Distributed Channel Access (MIRA-EDCA), a deterministic contention-control mechanism that separates potential contention through a normalized contender indicator, realized channel occupancy through the Channel Busy Ratio (CBR), and node-level backoff interruption through the Backoff Freeze Ratio (BFR), and combines them through a bounded contention-pressure function. The measure drives class-specific contention-window updates through discrete levels and a one-level-per-epoch transition rule while preserving priority differentiation between event-driven and periodic safety traffic. MIRA-EDCA was evaluated in a controlled slot-level IEEE 802.11p simulation over 30 density–traffic operating points formed by ten vehicle-population levels and three event-load levels, with 30 Monte Carlo realizations per operating point. Relative to Standard EDCA, MIRA-EDCA reduced collision probability by 31.92% while increasing conditional mean event-message MAC access delay by 0.66%; for vehicle populations of 100 or more, collision reduction reached 33.40% with a delay increase below 1%. Under the implemented common-environment proxy comparisons, MIRA-EDCA exhibited an 88.6% lower conditional mean event-message MAC access delay than the Dynamic Contention Window common-environment proxy (DYCW-2023-CE) and a 31.7% lower conditional mean MAC access delay than MA-DRL-2026-CE-PS. The corresponding transmitter-side SBR difference relative to DYCW-2023-CE was +0.899 percentage points. These comparisons refer only to the proxy controllers implemented under the matched assumptions of this study and not to the complete source protocols. These results support the proposed MAC-layer contention-control logic within the evaluated IEEE 802.11p common-contention-domain abstraction; they characterize collision–access-delay behavior at the modeled MAC layer and do not constitute receiver-side or end-to-end reliability validation.

The expansion of Internet-of-Things (IoT) deployments in 5G New Radio (NR) networks has made periodic measurement reporting a growing burden for low-mobility devices, which benefit little from frequent updates yet must report as often as highly mobile ones. At present, User Equipment (UE) transmits MeasurementReport messages at a fixed ReportInterval—typically 240 ms—regardless of mobility. This continuous transmission needlessly depletes UE battery energy and consumes critical uplink signaling capacity. This paper proposes a five-level adaptive ReportInterval selection scheme driven by the statistical properties of Reference Signal Received Power (RSRP) and Signal-to-Interference-plus-Noise Ratio (SINR). A low-mobility criterion combines four statistical conditions—the variance and gradient of both RSRP and SINR—through a logical AND, while a two-stage hysteresis mechanism (a 3 dB margin and a 2 s holding timer) suppresses unnecessary level transitions. The scheme is slice-agnostic: By relying on observed signal statistics rather than network-slice labels, it serves low-mobility mMTC and stationary eMBB devices while leaving URLLC and high-mobility UEs at their standard configuration. In Monte Carlo simulations over the 3GPP TR 38.901 Urban Micro (UMi) channel model (200 UEs, 300 s, 100 iterations), the algorithm attains a classification accuracy of 91.32% and a sensitivity of 98.77%. Based on the DRX energy model, it yields an estimated 10.87% reduction in average UE power (from 28.15 to 25.09 mW) together with a 51.09% reduction in the network-wide MeasurementReport count. The hysteresis mechanism cuts level transitions by a factor of 31.33 (from 6852.7 to 218.7 per iteration), substantially lowering RRC reconfiguration signaling. Operating at O(n) complexity on the gNodeB and using only conventional MeasConfig signaling, the scheme requires no protocol additions or UE-side modifications, making it directly deployable on existing 3GPP Release 17 infrastructure as a gNB-side software update.