Abstract:Deploying long term evolution (LTE) system into unlicensed spectrum for offloading faces with two challenges, (a) The system can not serve the users during its non-channel access period, which makes the metrics of instantaneous performance in traditional LTE system be no longer applicable for unlicensed LTE (LTE-U) system. That is, the user satisfaction can not be fairly evaluated, (b) The scheduled users to transmit / receive data within the same frame must end their data transmissions / receptions simultaneously so as to prevent Wi-Fi nodes from sensing channel idle and then accessing the channel. Therefore, the users with insufficient data are mandated to add padding bits, which deteriorates the channel utilization. To address the abovementioned issues, it is necessary to adjust the system's channel occupancy duration (COD) dynamically and allocate subcarriers to the users properly according to user satisfaction and channel condition. In this paper, we first define the unified channel utilization and user satisfaction. Then, we model a two-dimensional optimization problem, which maximizes the weighted sum of channel utilization and user satisfaction by optimizing COD in time domain and subcarrier allocation in frequency domain. By introducing a control factor, the tradeoff between channel utilization and user satisfaction can be achieved. By using Lyapunov optimization theory we transfer the original optimization problem into a new problem, so that it can be solved by using the instant data arrival and channel state information (CSI) of the current frame instead of knowing the statistical knowledge of CSIs and data arrivals of all frames in advance. Simulation results reveal that as compared with the baselines, our proposed heuristic algorithm for solving the new problem can flexibly balance channel utilization and user satisfaction, and can meet different applications on demand.

Reinforcement Learning-Based Computing and Transmission Scheduling for LTE-U-Enabled IoT

Edge-Aided Computing and Transmission Scheduling for LTE-U-Enabled IoT

Multi-Agent Reinforcement Learning Based Unlicensed Resource Sharing for LTE-U Networks.

BiLSTM Based Reinforcement Learning for Resource Allocation and User Association in LTE-U Networks

Deep Reinforcement Learning Based Massive Access Management for Ultra-Reliable Low-Latency Communications

Rethinking Mobile Data Offloading in LTE and WiFi Coexisting Systems

Rethinking Mobile Data Offloading for LTE in Unlicensed Spectrum

Throughput Analysis and Energy Efficiency Optimization for Standalone LTE-U Networks with Randomly Delayed CSI

A Distributed User Association Method for LTE-U by Enabling Q-Learning in Minority Game.

Q-Learning Based Fair and Efficient Coexistence of LTE in Unlicensed Band

Optimal Channel Occupancy Duration and Subcarrier Allocation for LTE-U Systems

Joint Resource Allocation for LTE over Licensed and Unlicensed Spectrum.

Bidirectional Mobile Offloading in LTE-U and WiFi Coexistence Systems

Resource Management of LTE-U Systems for Channel Utilization and User Satisfaction

Deep Reinforcement Learning-Based Computation Offloading and Resource Allocation in IoT

Deep Reinforcement Learning for Task Offloading in Edge Computing Assisted Power IoT

Age-of-Information-Based Computation Offloading and Transmission Scheduling in Mobile-Edge-Computing-Enabled IoT Networks

An intelligent task offloading method based on multi-agent deep reinforcement learning in ultra-dense heterogeneous network with mobile edge computing

Deep Reinforcement Learning Based Task Offloading and Resource Allocation in Small Cell MEC

Online Task Offloading for 5G Small Cell Networks

Learning-Based URLLC-Aware Task Offloading for Internet of Health Things