This paper proposes a telecom big data based user offloading self-optimisation (TBDUOS) scheme. Its aim is to assist operators effectively balancing the load distribution with achieving good service performance and customer management in heterogeneous relay cellular systems. To achieve these objectives, cell-level offloaded traffic analysis stage, optimal calculated minimise total blocking probability. In user-level portrait drawn K-MEANS algorithm employed manage users clustering heavily...

When regarding the inherent uncertainty of differentiated services requirements as well non-uniform spatial distribution capacity requests, it is essential to flexibility adjust resources satellite satisfy different conditions. How match system demand with efficient utilization beam a brand-new challenge. The convention hopping methods ignores intrinsic correlation between decisions, do not consider long-term reward, and only achieve optimal solution at current time. Therefore, complexity...

Mobile edge computing (MEC) enhanced satellite based internet of things (SAT-IoT) is an important complement for terrestrial networks IoT, especially the remote and depopulated areas. For MEC SAT-IoT with multiple satellites gateways, coupled user association, offloading decision, communication resource allocation should be jointly optimized to minimize latency energy cost. In this paper, optimization are formulated as a dynamic mixed-integer programming problem, which hard obtain optimal...

Dynamic channel allocation (DCA) is the key technology to efficiently utilize spectrum resources and decrease co-channel interference for multibeam satellite systems. Most works allocate on basis of beam traffic load or user terminal distribution current moment. These greedy-like algorithms neglect intrinsic temporal correlation among sequential decisions, resulting in underutilization. To solve this problem, a novel deep reinforcement learning (DRL)-based DCA (DRL-DCA) algorithm proposed....

Dynamic resource allocation (DRA) is the key technology to improve network performance in resource-limited multibeam satellite (MBS) systems. The aim find a policy that maximizes expected long-term utilization. Existing iterative metaheuristics DRA optimization algorithms are not practical due high computational complexity. To solve problem of unknown dynamics and prohibitive computation, deep reinforcement learning-based framework (DRLF) proposed for problems MBS A novel image-like tensor...

Power amplifier (PA) models, such as the neural network (NN) models and multilayer NN have problems with high complexity. In this article, we first propose a novel behavior model for wideband PAs, using real-valued time-delay convolutional (RVTDCNN). The input data of is sorted arranged graph composed in-phase quadrature ( I/Q ) components envelope-dependent terms current past signals. Then, created predesigned filter layer to extract basis functions required PA forward or reverse modeling....

Satellite assisted vehicle-to-vehicle (V2V) communication can provide services for vehicles in depopulated areas, and it be employed as an effective complementary component terrestrial vehicular networks. Since the available computing resource satellites are scarce, task offloading, allocation, which coupled with each other, critical issues satellite V2V communication. To tackle these problems, we formulate joint offloading decision, allocation problem a mixed-integer nonlinear programming...

Unmanned aerial vehicle (UAV) assisted emergency communication is an important technique for future B5G/6G scenario. The UAV usually considered as a mobile relay to forward information from the macro base station (MBS) users in area. In this paper, MBS power allocation, service zone selection, and user scheduling are jointly investigated enhance sum spectrum efficiency. We formulate allocation selection problem Markov Decision Process (MDP) delay ignored system (DIS) propose deep...

As a key enabler of emerging 6G applications, mobile edge computing (MEC) improves the system performance by coordinating computation resources between devices and edge. However, energy consumption carbon emission are boosted as device number increases. This paper constructs hierarchical network-assisted task offloading architecture for Green Internet Things (IoT). Then authors propose an energy-efficient scheduling based on traffic mapping with slack-space proportional pre-allocation...

With the rapid development of low Earth orbit (LEO) constellation satellite networks and on-board processing, in-network content caching becomes feasible, which can further improve system performance. In order to minimize user terminals access delay, a novel algorithm by optimizing placement in LEO is proposed this letter. The formulated problem proved be NP-complete (NPC) reducing it 2-disjoint set cover problem, known NPC problem. This letter considers interactions among distributed...

Centralized radio resource management method puts all of the computational burdens in an agent, which is unbearable with increasing data dimensionality. This letter focuses on how to schedule limited satellite-based resources efficiently enhance transmission efficiency and extend broadband coverage low complexity. We propose a cooperative multi-agent deep reinforcement learning (CMDRL) framework achieve strategy. The bandwidth allocation problem taken as example analyze proposed novel...

Information dissemination in mobile networks turns out to be a problem when the network is sparse. Mobile begin establish separate cluster attributable limited communication range of terminals. The multi-beam satellite systems can play significant role providing direct-to-user services and connecting separated clusters. This paper focuses on how efficiently schedule satellite-based radio resources enhance transmission efficiency meet requested traffic with low complexity. Taking inter-beam...

High peak-to-average power ratio (PAPR) in orthogonal frequency division multiplexing (OFDM) systems is one of the major drawbacks wireless transmitters. In this letter, a novel low complexity PAPR reduction method based on real-valued neural network (NN) proposed. This first builds module using NN to achieve PAPR. To reconstruct transmission signal, decompression introduced into receiving end recover signal order minimize bit error rate (BER) systems. The and can be jointly trained offline,...

This letter proposes a joint interference suppression scheme in heterogeneous networks (HetNets) with dense small cells (SCs) and users. Different from the majority of existing studies, we adopt co-tier intra-cell alignment (IA), while inter-cell cross-tier is suppressed by centralized power control macro base station (MBS). Specifically, problem modeled as Markov Decision Process (MDP) aim maximizing sum spectrum efficiency. Considering exponential growth output layer neurons faced general...

Benefit from the enhanced onboard processing capacities and high-speed satellite-terrestrial links, satellite edge computing has been regarded as a promising technique to facilitate execution of computation-intensive applications for communication networks (SCNs). By deploying servers in gateway stations, SCNs can achieve significant performance gains at expense extending dimensions complexity resource management. Therefore, this paper, we investigate joint management problem minimize...

Benefiting from the development of satellite on-board processing capability, orbital computing can be realized by deploying edge servers on satellites to reduce task latency. However, based geostationary Earth orbit (GEO) or low-Earth (LEO) alone hardly meet latency requirements Satellite-assisted Internet Things (SIoT) services. Moreover, uneven distribution tasks generated SIoT devices will also cause load unbalancing among different satellites. In this article, hybrid GEO–LEO networks is...

Driven by the demands of diverse artificial intelligence (AI)-enabled application, Mobile Edge Computing (MEC) is considered one key technologies for 6G edge intelligence. In this paper, we consider a serial task model and design quality service (QoS)-aware offloading via communication-computation resource coordination multi-user MEC systems, which can mitigate I/O interference brought reuse among virtual machines. Then construct system utility measuring QoS based on application latency user...

The rapid development of smart healthcare and the increasing number wearable devices in consumer electronics can facilitate health monitoring services with lower latency less energy consumption, provide more reliable computation communication for medical cares. In this paper, we consider next generation (next-gen) architecture introduce D2D a trustworthy system that utilizes collaborative computing among different wearables to reduce execution cost tasks improve efficiency services, which...

Beam hopping (BH) is the key technology to improve system throughput and decrease transmission delay in multibeam satellite systems. The objective of this study find a policy maximise expected long‐term resource utilisation. BH illumination plan (BHIP) optimisation problem aimed at minimising formulated modelled as partially observable Markov decision process. To tackle issue unknown dynamics prohibitive computation, an artificial intelligence method named deep reinforcement learning (DRL)...

Vehicular communications have recently attracted great interest due to their potential improve the intelligence of transportation system. When maintaining high reliability and low latency in vehicle-to-vehicle (V2V) links as well large capacity vehicle-to-infrastructure (V2I) links, it is essential flexibility allocate radio resource satisfy different requirements V2V communication. This paper proposes a new resources allocation system for based on proximal strategy optimization method. In...

The convolutional neural network (CNN) based power amplifier (PA) model has been proven to reduce the complexity significantly. However, due calculation mode of structure, application CNN-based predistortion still faces problem high computational complexity. In this letter, we use one lightweight CNN propose a modeling method predistorter with low for wideband PA. This first decomposes traditional two-dimensional convolution kernels into two kinds one-dimensional kernels, create predesigned...

Carrier aggregation (CA) is one of the promising techniques for further advancements third-generation (3G) long-term evolution (LTE) system, referred to as LTE-Advanced. When CA applied, a well-designed carrier scheduling (CS) scheme essential LTE-Advanced system. Joint user (JUS) and separated random (SRUS) are two straightforward CS schemes. JUS optimal in performance but with very high complexity, whereas SRUS contrary. Consequently, authors propose novel scheme, termed 'separated...

Multibeam satellite communication plays an important role in future integrated satellite-terrestrial systems. However, the large area covered by beams makes itself vulnerable to be wiretapped, especially for cases that eavesdroppers are located near legitimate receivers. In this paper, cooperative jamming from both and full-duplex receivers investigated where we aim minimize power consumed a multibeam system with secrecy outage probability constraints. To tackle problem, algorithm of joint...