Abstract:To balance the contradiction between the rapid growth of data service demands and the limited spectrum resources, wireless network virtualization (WNV) has been proposed as a promising technology by isolating and sharing wireless resources among different virtual networks in the future wireless networks. In this paper, a two-dimension-time-scale hierarchical resource allocation scheme is proposed in the multi-service wireless virtualized network, which consists of three 5G generic scenarios. The resource slicing problem is decomposed into two time scales including large time period for inter-slice resource pre-allocation and small time slot for intra-slice resource scheduling. In large time period, the inter-slice resource pre-allocation problem is formulated as a multi-objective optimization problem (MOOP) by modeling the packets arriving and serving process of each slice as a queueing system. While in small time slot, the resource block (RB) and power scheduling in each slice is formulated as a stochastic optimization problem considering dynamic traffic arrivals and time-varying channel conditions, which is aimed at optimizing the overall performance subject to various quality of service (QoS) requirements such as network stability, delay, reliability, transmission rate and power constraints. The stochastic optimization problem can be transformed into a delay-aware optimization problem by applying Lyapunov optimization technique, and be solved by the proposed algorithm consisting of a heuristic algorithm and a concave optimization algorithm. The simulation results show that the proposed schemes are close to the optimal solution with a lower complexity, which can also achieve a performance-delay tradeoff related to the control factor.

Wireless Network Slice Assignment With Incremental Random Vector Functional Link Network

Hierarchical Resource Allocation in Multi-Service Wireless Networks with Wireless Network Virtualization

Recurrent Neural Network Based Collaborative Filtering for QoS Prediction in IoV

Federated Spectrum Learning for Reconfigurable Intelligent Surfaces-Aided Wireless Edge Networks

Random vector functional link network with subspace-based local connections

Probabilistic-Forecasting-Based Admission Control for Network Slicing in Software-Defined Networks

Safe and Fast Reinforcement Learning for Network Slicing Resource Allocation

Deep-Learning-Based Wireless Resource Allocation With Application to Vehicular Networks

Elastic Adaptive Network Slicing Scheme Based on Multi-Priority Cooperative Prediction in Fi-Wi Access Network

Efficient Embedding VNFs in 5G Network Slicing: A Deep Reinforcement Learning Approach

Dynamic RAN Slicing for Service-Oriented Vehicular Networks via Constrained Learning

A Heuristically Assisted Deep Reinforcement Learning Approach for Network Slice Placement

Ai-Assisted Resource Advertising And Pricing To Realize Distributed Tenant-Driven Virtual Network Slicing In Inter-Dc Optical Networks

Random vector functional link network: Recent developments, applications, and future directions

Fast and Scalable Network Slicing by Integrating Deep Learning with Lagrangian Methods

Prediction-based Hybrid Slicing Framework for Service Level Agreement Guarantee in Mobility Scenarios: A Deep Learning Approach

Autonomous Resource Slicing for Virtualized Vehicular Networks With D2D Communications Based on Deep Reinforcement Learning

Wireless Channel Prediction using Artificial Intelligence with constrained Data Sets

Density Encoding Enables Resource-Efficient Randomly Connected Neural Networks

DRL-based Slice Placement Under Non-Stationary Conditions

Topology-Aware Prediction of Virtual Network Function Resource Requirements