This paper proposes an adaptive stabilization control mechanism by using deep reinforcement learning (DRL) for hovering drones that have to execute a surveillance task for a long time. For long-endurance flights, we design and implement a buoyancy-aided autonomous aerial vehicle (AAV) that can use buoyancy lift to decrease the weight and increase the battery capacity so that the flight time can be significantly extended. However, the balloons of the buoyancy-aided AAV can cause “an inverted pendulum effect” and an instability issue on the drone attitude because the increased surface is easily affected by the gusty wind. We propose a buoyancy-aided adaptive stabilization control (BAASC) method with the DRL to stabilize the attitude and extend the flight time of the quadrotor-based buoyancy-aided AAV. This proposed model can immediately control the speeds of all rotors to balance the attitude based on the current state of the drone. Therefore, the degree of swing can be stabilized, and the inverted pendulum effect can be eliminated. The experimental results reveal that the designed buoyancy-aided AAV with the proposed BAASC scheme can effectively stabilize the attitude to extend the flight time by 112.8% compared with a nonbuoyancy-aided AAV under a gusty wind disturbance.

Reconfigurable intelligent surface (RIS) is a promising technology for energy efficient wireless communications and has drawn significant attention recently. Combining unmanned aerial vehicle with RIS (UAV-RIS) can provide on-demand deployment services in dynamic scenarios. However, reaping the benefits of UAV-RIS will be limited by the energy of the battery-powered UAV. To enhance the endurance of UAV-RISs, we develop a novel energy harvesting scheme for simultaneous wireless information and power transfer (SWIPT), resource allocation, and energy harvest from impinging radio-frequency (RF) signals. Different from the exist works, the proposed scheme creatively splits the passive reflected arrays on geometric space for transporting information and harvesting energy simultaneously. Furthermore, a deep deterministic policy gradient (DDPG) scheme is designed to continuously allocate UAV-RIS’s resources on both the time and space domains to maximize the total harvested energy, while guaranteeing the communication quality for each user. As shown by our simulation results, the proposed UAV-RIS SWIPT system improves performance significantly over the benchmark.

A mobile small-cell integrates the concept of mobile relay stations with fixed small-cells, and can be deployed in the public transportation to improve the service quality of users inside the vehicles. However, three-level interference among macrocells, fixed small-cells, and mobile small- cells is crucial to the improvement of throughput performance. In this paper, we investigate the impacts of resource partition schemes on the throughput and link reliability for the heterogeneous mobile small-cell network. We compare the system throughput and link reliability for the mobile small-cell system with orthogonal and non-orthogonal resource partition schemes. Simulation results show that the non-orthogonal resource partition scheme can improve 62% more average cell throughput for the mobile small-cell system than the orthogonal resource partition scheme when the network deploys seven mobile small-cells at the same time.

In this paper, we present a big-data self organizing network (Bi-SON) framework aiming to optimize energy efficiency of ultra-dense small cells. Although small cell can enhance the capacity of cellular mobile networks, ultra-dense small cells suffer from severe interference and poor energy efficiency. The self organizing network (SON) can automatically manage and optimize the system performance. However, current SON-enable mechanisms mostly focus on indoor femtocells. Our proposed Bi-SON suggests a data flow framework from data collection, analysis and optimization to reconfiguration. We adopt the statistics analysis approach to determine the optimal system parameters to improve the energy efficiency of a huge number of outdoor small cells. The Bi-SON mechanism periodically collects the management data of small cells, e.g. transmission power, reference signal receiving power and the number of users per cell. We find that simple sorting and filtering data analysis from huge number of small cells can already effectively find the almost optimal solution. Our simulation results show that Bi-SON can improve throughput and energy efficiency by 50% and 135% respectively, compared to the scheme without energy saving approach.

In this paper, we propose a low-complexity distributed channel-oriented channel allocation scheme (COCAS) to improve the link reliability of users and the spectrum efficiency of femtocells for multi-antenna orthogonal frequency-division multiple access (OFDMA)-based femtocell systems. By two steps of comparisons on channel gains of resource blocks and antennas, the COCAS can jointly allocate suitable antennas and resource blocks for transmission. We also investigate the impacts of channel allocation and multiple antennas on link reliability of users and spectrum efficiency of femtocells. Simulation results show that the proposed COCAS can achieve higher spectrum efficiency for femtocells than the random channel allocation scheme with a given number of antennas equipped on the femtocell base station, under the link reliability requirement.

The device-to-device (D2D) communications can improve the spectrum efficiency with low power and effectively offload the traffic from the macro-eNB. However, the macro/femto/D2D heterogeneous networks face a complicated three-tier interference issue. The completely centralized resource management is not applicable in such three-tier heterogeneous networks. Therefore, we propose a network-assisted device-decided (NADD) scheme to jointly select the channels and adjust the transmission power for D2D devices. In the NADD method, the macro-eNB broadcasts resource allocation instruction to D2D devices. According to the resource allocation instruction and the device's communication quality, D2D devices autonomously and dynamically select the suitable channels and adjust the transmission power. Simulation results show that NADD scheme can significantly improve the total throughput and ensure the link reliability of macrocell users, comparing to both the fixed power scheme and the network-decided (ND) power control scheme.

In this paper, we propose the group-based time control mechanism to improve the network overload and delay performance in the femtocell-based machine type communications (MTC) networks. The MTC network may be congested if numerous MTC devices concurrently deliver messages to the MTC server. Femtocells can solve the radio access network (RAN) congestion, however the core network (CN) congestion becomes more difficult to handle. The proposed group-based time control method can spread the traffic load of MTC devices over the time, and thereby mitigate the RAN overload and CN overload simultaneously. Numerical results show that the proposed approach can significantly improve the network congestion and message delay compared to the existing methods.

In this paper, we propose a low-complexity distributed stable subchannel allocation algorithm combined with a low-cost reconfigurable switched multi-beam directional antenna to improve the capacity and the fairness performance for the orthogonal frequency division multiple access (OFDMA)-based femtocell systems. With the proposed deferred acceptance algorithm, the stable subchannel allocation scheme can achieve a better tradeoff between capacity and fairness. Numerical results show that the proposed stable subchannel allocation scheme with the switched multi-beam antenna can significantly improve femtocell capacity compared to the traditional user-oriented subchannel allocation scheme and ensure the fairness as compared to the traditional channel-oriented subchannel allocation scheme.

Femtocells are low-power, low-cost and user-deployed home base stations that can improve poor indoor coverage and enhance system capacity. However, as femtocells are deployed extensively, the two-tier interference will become serious, including the interference from macrocells and that from femtocells. Thanks to the low side lobe, the E-plane horns based reconfigurable antenna can be used to reduce the interference and enhance the capacity. In this paper, we investigate the effects of applying E-plane horns based reconfigurable directional antenna on link reliability and capacity of the orthogonal frequency-division multiple access (OFDMA) based femtocell systems. We compare our newly designed E-plane horns based reconfigurable antenna,and the sector antenna provided in the IEEE 802.16 m, as well the omni-directional antenna. It is shown that E-plane horns based reconfigurable antenna system has higher femtocell capacity than the IEEE 802.16 m sector antenna system due to the advantages of low side lobe and high main-lobe gain.