15 June 2021 in IEEE Transactions on Communications
This paper studies an unmanned aerial vehicle (UAV)-assisted wireless powered IoT network, where a rotary-wing UAV adopts fly-hover-communicate protocol to successively visit IoT devices in demand. During the hovering periods, the UAV works on full-duplex mode to simultaneously collect data from the target device and charge other devices within its coverage. Practical propulsion power consumption model and non-linear energy harvesting model are taken into account. We formulate a multi-objective optimization problem to jointly optimize three objectives: maximization of sum data rate, maximization of total harvested energy and minimization of UAV's energy consumption over a particular mission period. These three objectives are in conflict with each other partly and weight parameters are given to describe associated importance. Since IoT devices keep gathering information from the physical surrounding environment and their requirements to upload data change dynamically, online path planning of the UAV is required. In this paper, we apply deep reinforcement learning algorithm to achieve online decision. An extended deep deterministic policy gradient (DDPG) algorithm is proposed to learn control policies of UAV over multiple objectives. While training, the agent learns to produce optimal policies under given weights conditions on the basis of achieving timely data collection according to the requirement priority and avoiding devices' data overflow. The verification results show that the proposed MODDPG (multi-objective DDPG) algorithm achieves joint optimization of three objectives and optimal policies can be adjusted according to weight parameters among optimization objectives.
28 April 2020 in IEEE Transactions on Communications
24 August 2021 in IEEE Transactions on Wireless Communications
The intelligent reflecting surface (IRS)-assisted millimeter wave (mmWave) communication system has emerged as a promising technology for coverage extension and capacity enhancement. Prior works on IRS have mostly assumed perfect channel state information (CSI), which facilitates in deriving the upper-bound performance but is difficult to realize in practice due to passive elements of IRS without signal processing capabilities. In this paper, we propose a compressive channel estimation techniques for IRS-assisted mmWave multi-input and multi-output (MIMO) system. To reduce the training overhead, the inherent sparsity of mmWave channels is exploited. By utilizing the properties of Kronecker products, IRS-assisted mmWave channel is converted into a sparse signal recovery problem, which involves two competing cost function terms (measurement error and sparsity term). Existing sparse recovery algorithms solve the combined contradictory objectives function using a regularization parameter, which leads to a suboptimal solution. To address this concern, a hybrid multiobjective evolutionary paradigm is developed to solve the sparse recovery problem, which can overcome the difficulty in the choice of regularization parameter value. Simulation results show that under a wide range of simulation settings, the proposed method achieves competitive error performance compared to existing channel estimation methods.
27 February 2020 in IEEE Transactions on Antennas and Propagation
This article presents a novel dual-polarized millimeter-wave (mm-Wave) patch antenna with bandpass filtering response. The proposed antenna consists of a differential-fed cross-shaped driven patch and four stacked parasitic patches. The combination of the stacked patches and the driven patch can be equivalent to a bandstop filtering circuit for generating a radiation null at the upper band edge. Besides, four additional shorted patches are added beside the cross-shaped driven patch to introduce another radiation null at the lower band edge. Moreover, by embedding a cross-shaped strip between these four stacked patches, the third radiation null is generated to further suppress the upper stopband. As a result, a quasi-elliptic bandpass response is realized without requiring extra filtering circuit. For demonstration, a prototype was fabricated with standard PCB process and measured. The prototype operates in the 5G band (24.25-29.5 GHz) and it has an impedance bandwidth of 20%. The out-of-band gain drops over 15 dB at 23 and 32.5 GHz, respectively, which exhibits high selectivity. These merits make the proposed antenna a good element candidate for the 5G mm-Wave massive MIMO applications to reduce the requirements of the filters in the mm-Wave RF front ends.
22 October 2018 in IEEE Transactions on Antennas and Propagation
A wideband filtering patch antenna is investigated in this paper. The patch antenna is fed by an F-shaped probe, which consists of one upper and two lower horizontal arms, as well as a vertical arm that connects them. Owing to the novel excitation scheme, two probe modes along with a patch mode are simultaneously excited within the passband, producing a wide bandwidth of 21.3% with stable antenna gains and radiation patterns. Meanwhile, cross-coupling is constructed in the antenna, generating two symmetrical radiation nulls right at the two sides of the passband. Consequently, a compact filtering antenna with bandpass response and high selectivity is obtained, without utilizing any extra filtering circuit. This design is also extended to realize a reconfigurable filtering antenna. Two varactor diodes are embedded in the F-shaped probe to continuously tune the frequency of the operating band from 2.05 to 2.52 GHz. The antenna bandwidth can be flexibly tuned from 2.2% to 21.3%, and simultaneously good filtering performance is kept during the tuning of different states.
2 August 2018 in IEEE Transactions on Antennas and Propagation
This paper presents a broadband dual-polarized filtering dipole antenna for base-station application, which has a compact size of 50 $\text {mm} \times 50\,\,\text {mm} \times 31.8$ mm. The antenna consists of four parts: main radiator, feeding baluns, reflector, and two parasitic loops. Without using complex filtering circuits, the dual-polarized dipole antenna realizes satisfactory filtering performance and enhanced bandwidth by employing only two parasitic loops. Two specific radiation nulls are thus generated and individually controlled by the two parasitic loops. To further improve the upper stopband selectivity and bandwidth, a simple open-ended stub is added to the arms of the dipole. As a result, the bandwidth can be tuned from 7.4% to 47.6%, and the realized gain is decreased dramatically from 8.6 dBi at 2.7 GHz (in-band) to −10 dBi at 2.9 GHz (out-of-band). For demonstration, a broadband dual-polarized dipole antenna is implemented. Measured results show that the proposed antenna has more than 34 dB port isolation over 48.7% (1.66–2.73 GHz) impendence bandwidth (VSWR < 1.5). The measured in-band gain is about 8.15 dBi with stable 3 dB beamwidth 65.4°±2.4° in the horizontal plane, whereas the out-of-band radiation suppression is more than 17 dB.