Utility-Driven End-to-End Network Slicing for Diverse IoT Users in MEC: A Multi-Agent Deep Reinforcement Learning Approach

Sensors, Год журнала: 2024, Номер 24(17), С. 5558 - 5558

Опубликована: Авг. 28, 2024

Mobile Edge Computing (MEC) is crucial for reducing latency by bringing computational resources closer to the network edge, thereby enhancing quality of services (QoS). However, broad deployment cloudlets poses challenges in efficient slicing, particularly when traffic distribution uneven. Therefore, these include managing diverse resource requirements across widely distributed cloudlets, minimizing conflicts and delays, maintaining service amid fluctuating request rates. Addressing this requires intelligent strategies predict types (common or urgent), assess needs, allocate efficiently. Emerging technologies like edge computing 5G with slicing can handle delay-sensitive IoT requests rapidly, but a robust mechanism real-time utility optimization remains necessary. To address challenges, we designed an end-to-end approach that predicts common urgent user through T distribution. We formulated our problem as multi-agent Markov decision process (MDP) introduced soft actor-critic (MAgSAC) algorithm. This algorithm prevents wastage scarce intelligently activating deactivating virtual function (VNF) instances, balancing allocation process. Our aims optimize overall utility, trade-offs between revenue, energy consumption costs, latency. evaluated method, MAgSAC, simulations, comparing it following six benchmark schemes: MAA3C, SACT, DDPG, S2Vec, Random, Greedy. The results demonstrate approach, optimizes 30%, minimizes costs 12.4%, reduces execution time 21.7% compared closest related named MAA3C.

Язык: Английский

---

AI*LoRa: Enabling Efficient Long-Range Communication with Machine Learning at the Edge

Опубликована: Окт. 1, 2024

Язык: Английский

---

AI-driven Adaptive Data Rate for LoRaWAN Location-Based Services

IEEE Access, Год журнала: 2024, Номер 12, С. 168349 - 168359

Опубликована: Янв. 1, 2024

Язык: Английский

---

Localized Adaptive Channel and Power Selection With TinyML (LACPSA) in Dense IEEE 802.11 WLANs

Опубликована: Ноя. 14, 2024

Язык: Английский

---

A Contextual Aware Enhanced LoRaWAN Adaptive Data Rate for mobile IoT applications

Computer Communications, Год журнала: 2024, Номер unknown, С. 108042 - 108042

Опубликована: Дек. 1, 2024

Язык: Английский

---