Recently, mobility has gathered tremendous interests as users desire for consecutive connections and better quality of service (QoS). An accurate prediction of user mobility in mobile networks provides efficient resource and handover management, which can avoid unacceptable degradation of the perceived quality. Therefore, mobility prediction in wireless networks is of great importance and many works have been dedicated to this issue. In this paper, the necessity of mobility prediction, together with its intrinsic characteristics in terms of movement predictability, prediction outputs and performance metrics is discussed. Moreover, the learning perspective of solutions to mobility prediction has been studied. Specifically, an overview of state-of-the-art approaches is provided, including Markov chain (MC), hidden Markov model (HMM), artificial neural network (ANN), Bayesian network and data mining based on different kinds of knowledge. Lastly, this work also explores the open research challenges due to the advent of the fifth-generation mobile system (5G) and puts forward some potential trends in the near future.

In existing works of deep learning-based resource allocation, the scalability degrades heavily with the increase of network complexity, which is due to their limited learning ability of shallow neural networks and insufficient knowledge of network. Nowadays, to address the growth of cell density, cooperative beamforming in user-centric network (UCN) is emerged, where the additional degrees of freedom of multi-antenna and cell coordination aggravate the challenges. This paper proposes a deep residual learning framework, UcnBeamNet, to enhance the ability of approximating the iterative algorithm for sum rate maximization, where multi-branch subnets are connected in parallel to extract extra information. Specifically, a weighted minimum mean square error (WMMSE)-based algorithm is derived to determine the optimal clusters and beamforming matrices; then UcnBeamNet is trained to learn the input-output mapping and provide direct insight of UCN from association matrices in addition to plural inputs. Extensive experiments demonstrate UcnBeamNet still reaches 90.38% sum-rate relative to conventional algorithm even with a large network size, and achieves more than 50,000× speed up in computational efficiency.

Cooperative communication has used to be a hot topic and it has been studied extensively in the past 10 years, but in recent years, it becomes less likely to find substantial innovation in this field as before. In this paper, we propose a new hybrid decode-forward and amplify-forward with non-orthogonal multiple access (NOMA) (HDAF-NOMA) transmission scheme for a cellular system with multiple relays. To the best of our knowledge, this is the first work that attempts to integrate decode-forward (DF), amplify-forward, and NOMA into one strategy design to improve system performance. To verify the performance advantages, the proposed HDAF-NOMA scheme is compared with the other four traditional schemes in terms of channel capacity and average system throughput, and the optimal number of selected DF relays is also determined for the HDAF-NOMA scheme. Simulation results show that compared with the traditional schemes, the proposed HDAF-NOMA scheme can achieve larger sum channel capacity for the transmission of x ₁ and x ₂ , and it can also achieve larger average system throughput at high SNR region.

Wireless sensor nodes (SNs) are usually energy-limited, a low energy efficiency of SN may be caused by the occlusion of transmission link in the unmanned aerial vehicle (UAV)-enabled data collection. Reconfigurable intelligent surfaces (RISs) provide potential communication opportunities by reflecting and enhancing the signal, thus increasing SNs’ energy efficiency. This paper investigates energy-efficient data collection aided by N RIS elements, where a height-fixed and flight time-constrained UAV is considered as the collector, a tractable air-to-ground channel model is established considering the cascaded RIS channel and time-varying trajectory of UAV, and energy consumption of SNs is minimized. Specifically, coupling two intractable fourth-order moment variables, the air-to-ground channel model is processed into an easily computable expression through the Gamma function. Furthermore, based on the proposed two RIS deployment schemes, maximum energy consumption of SNs minimization problem is formulated as a non-convex mixed-integer nonlinear programming (NMNP) problem with optimization of UAV trajectory and SNs wake-up schedules. Additionally, through the slack technique and Taylor expansion, the NMNP problem is decoupled as linear programming and quadratically constrained quadratic programming, and an effective algorithm is designed to obtain the sub-optimal solution. Simulations prove the validity of our algorithm and RIS deployment insights are given for SNs energy saving.

Existing researches on UAV networks mostly adopt omni-directional antenna pattern, which results in overlapping between coverage areas and bringing severe co-channel interference due to air-to-ground dominant line-of-sight (LOS) channels. Considering directional antennas that concentrate transmission power to small cells, this letter analyzes system performance of UAV-assisted networks with conic directional antenna model by deriving network coverage probability via stochastic geometry, where UAVs and macro base stations (MBSs) are deployed according to independent Poisson point process. Specifically, realistic and mathematically tractable directional antenna pattern is adopted that radiation gain is modeled by a continuous function of the elevation angle $\varphi $ and the directivity factor $\omega $ . In addition, the tradeoff between cell association and interference isolation is exploited by adjusting $\omega $ and UAV LOS/NLOS probabilities. Theoretical results show that, system coverage probability could be improved by 41% in urban environment when $\omega =6.2$ and UAV density $\lambda _{u}=300/k{m^{2}}$ .

Exploiting multiple-input multiple-output (MIMO) technology in UAV networks will mine the potential of spatial multiplexing. However, considering UAVs 3D distribution and line-of-sight (LOS)/non-line-of-sight (NLOS) channel dynamics, multi-antenna transmission will exacerbate network irregularity and render interference management a complicated problem to solve. This paper proposes 3D coordination model for interference management via multi-cell beamforming and signal-level cooperation in multi-antenna UAV networks, and analyzes system performance by deriving semi-closed expression of coverage probability using stochastic geometry. Specifically, UAVs are deployed according to 3D Poisson point process (PPP) with maximum height limit $L$ , and user-centric cell group selection is considered according to specific signal threshold $\eta $ for interference coordination with the consideration of LOS/NLOS channel conditions. Zero-forcing beamforming and joint transmission is performed within the UAV cluster for interference mitigation with limited channel state information feedback taken account. In addition, considering the irregularity of UAV group, system performance is construed as three scenarios and coverage probability is derived by Gamma approximation with proposed novel coordinate system transformation. Numerical results match well with simulations and provide optimal deployment parameters for UAV deployments, where coverage probability can reach 92% when SIR threshold $T=0$ dB.