Beam tracking for unmanned aerial vehicle (UAV) communications is an enabling technology that maintains the connection between UAV and ground base stations. In this paper, we propose an efficient training-based beam tracking scheme for UAV system. To track the beam, we need to answer how frequently the training should be performed. The proposed scheme uses joint beam training and angular velocity estimation, based on which the beam coherence time can be obtained. Accordingly, the training frequency can be adjusted based on the variation of beam coherence time to save the training overhead. Then, we study the maximization of flying range and speed of UAV. To this end, we optimize the beamwidth by choosing different codebook. For flying range maximization, we first propose an efficient algorithm that can obtain the optimal beamwidth and the corresponding maximal flying range at a given speed of UAV. To gain more insight, we then show that in most practical scenarios, the maximal flying range is achieved with the narrowest beamwidth. Furthermore, we show that when the flying range is small, broad beam should be used to support high speed of UAV, and vice versa.

Wireless power transfer (WPT) via dedicated radio frequency (RF) transmission is an appealing technology to provide cost-effective energy supply to low-power devices in the future era of Internet of Things. To achieve efficient power delivery over moderate distance, WPT usually relies on highly directional power transmission from the energy transmitter (ET) to the energy receiver (ER). To this end, the ET needs to be equipped with a large number of antennas and employ adaptive energy beamforming to flexibly control the energy focusing directions to the ER based on the multipath channels between them. However, this renders the conventional fully digital beamforming with one dedicated RF chain for each transmit antenna too costly, in terms of both hardware implementation and energy consumption. To overcome this issue, we study in this paper, a new WPT system based on the hybrid analog or digital beamforming technique, where the number of RF chains is in general significantly less than that of transmit antennas. We first show that for a general point-to-point multiple-input multiple-output WPT system over frequency-selective channels, hybrid beamforming is able to achieve the optimal performance as the fully digital beamforming, as long as the number of RF chains at the ET is no less than twice the number of sub-bands used or twice the number of channel paths. Furthermore, for the special cases of line-of-sight channel or multiple-input single-output WPT, the required number of RF chains can be further reduced to equal the number of channel paths only. Finally, for the scenarios when the given number of RF chains is insufficient to achieve the fully digital beamforming performance, we propose efficient algorithms for the hybrid beamforming design to maximize its efficiency. Numerical results are provided to validate our analytical results and demonstrate the effectiveness of the proposed hybrid beamforming design.

Millimeter wave (mm-wave) communication using lens antenna arrays is a promising technique for realizing cost-effective large multiple-input multiple-output (MIMO) systems with only limited radio frequency chains. This paper studies a multi-user mm-wave single-sided lens MIMO system, where the base station (BS) is equipped with a full-dimensional lens antenna array and each mobile station employs the conventional antenna arrays. By exploiting the angle-dependent energy focusing property of lens antenna array and the multi-path sparsity of mm-wave channels, we propose a low-complexity single-carrier (SC)-based path-division multiple access (PDMA) scheme for the general wide-band frequency-selective channels. To this end, a new technique called path delay compensation is proposed at the BS to transform the multi-user frequency-selective MIMO channels to parallel frequency-flat small-size MIMO channels. In addition, we propose an efficient channel estimation scheme tailored for the SC-based PDMA, which requires negligible training overhead in practical mm-wave systems and yet leads to comparable performance as that with perfect channel state information. Numerical results show that the proposed design achieves comparable performance as the state-of-the-art benchmark systems in terms of spectrum efficiency, but with significantly reduced hardware/power consumption cost and signal processing complexity.

Millimeter wave (mmWave) communication with advanced lens antenna arrays is a promising technology for achieving cost-effective 5G wireless systems. This paper studies a multi-user mmWave single-sided lens multiple-input multiple-output (MIMO) system in the uplink communication, where the base station (BS) is equipped with a full-dimensional (FD) lens antenna array with both elevation and azimuth angle resolution capabilities, and each mobile station (MS) has the conventional uniform planar array (UPA). With limited radio frequency (RF) chains at the BS and one single RF chain at each MS, we propose a low-complexity path division multiple access (PDMA) scheme to enable virtually interference-free multiuser communications, by exploiting the angle-dependent energy focusing property of the lens antenna array at the BS as well as the multi-path sparsity of mmWave channels. Besides, a new technique called path delay compensation is proposed at the BS to effectively transform the frequency-selective MIMO channel to parallel frequency-flat small-size MIMO channels, for each of which the low-complexity single-carrier (SC) transmission can be applied. Numerical results show significant sum-rate gain with the proposed design over benchmark systems.

In this paper, an efficient channel estimation scheme is proposed for the point-to-point lens-antenna-array enabled millimeter wave (mmWave) multiple-input multiple-output (MIMO) system with limited number of radio frequency (RF) chains at both source and destination. By exploiting the energy focusing property of lens antenna arrays at both the transmitter and receiver, as well as the multi-path sparsity of mmWave channels, the proposed scheme employs two-way training to achieve energy-based antenna selections at both the source and destination, thus significantly reducing the number of RF chains required. The proposed scheme has low complexity for implementation and yet does not require any prior knowledge of the channel. It is shown by simulations that compared to the traditional channel training method for MIMO systems without lens antenna, the effective data rate achieved by the lens array enabled mmWave MIMO system can be significantly improved, thanks to the energy focusing capability of lens antenna arrays and the matching efficient channel estimation scheme proposed.

This paper studies a device-to-device (D2D) LAN underlaying a cellular uplink, where multiple D2D users (DUs) intend to communicate with a D2D receiver. Two interference alignment (IA) based D2D communication schemes are proposed to manage the interference between two networks. The first scheme is referred to as “interference-free” IA scheme. In this scheme, the interference signals from DUs are designed to be aligned in the orthogonal space of cellular links at the evolved NodeB (eNB), so that the links of cellular users (CUs) are completely free from interference. The second scheme is referred to as “interferencelimiting” IA scheme. In this scheme, the DUs' signals are allowed to occupy some links of CUs, but the peak interference power on each of the “interfered” links is controlled under a certain threshold γ. The “interference-limiting” IA scheme is most efficient when there are a large number of DUs as the scheme can take advantage of multiuser diversity. Performance analysis shows that based on the proposed schemes, the interference generated on the cellular links is eliminated or well controlled, while the quality of service of the D2D LAN can also be guaranteed.

This work studies achievable degrees of freedom (DoF) of two-cell two-hop interfering broadcast channels. In the two cells, base stations (BS) equipped with Mi and M2antennas intend to communicate with M1and M2single-antenna users, respectively. The messages sent from each BS are first received by two shared full-duplex multiantenna relays simultaneously in the first hop and then are forwarded to the intended users from the relays in the second hop. We study the achievable DoF of the network with various channel state information (CSI) available at the BSs and the users while the relays have CSI of the two-hop channels. We first consider the scenario when the BSs have the CSI of the first hop and the users have the CSI of the second hop, respectively. A linear scheme with precoding at the BSs and interference neutralization at the relays is designed to achieve the cut-set DoF of M1+ M2if both relays have M1+ M2- 1 antennas. We then consider the scenario when the BSs do not have any CSI but the users have the full CSI of both hops. A relay beamforming scheme is proposed to achieve the DoF of M1+ M2when both relays have M1+ M2- 1 antennas.

This paper studies multiuser diversity of uplink MIMO cognitive radio network and proposes a two-stage opportunistic user scheduling scheme. In the first stage, a cognitive beamforming design is proposed to ensure the interference caused by secondary signals is canceled or minimized on the spatial dimensions occupied by primary MIMO system. Then, some secondary users that cause minimal interference leakage at primary system are pre-selected as candidate users. In the second stage, some candidate users that produce maximum sum secondary rate are further selected for uplink scheduling. The proposed scheme enables the secondary link to take advantage of multiuser diversity while ensuring that the interference on primary link is within a certain threshold. Analytical results show that the sum rate of secondary uplink scales as Nslog logK for K secondary users and Nsantennas on secondary receiver for very large K.

This paper studies the achievable degrees of freedom (DoF) of 3-user MIMO interference channels based on linear spatial beamforming scheme. Previous work mainly focus on the scenario where each user must have the same DoF and the achievable sum DoF is 3⌊Doutover 3⌋, where Douterdenotes the outer bound of sum DoF. In this paper, we investigate the scenarios where three users are allowed to have different DoF. Three interference alignment based signal transmission strategies are proposed and the achievable DoF is derived. We show that the achievable DoF of the proposed interference alignment scheme achieves the DoF of ⌊Dout⌋ in some cases, which is equal to or higher than 3⌊Doutover 3⌋.

It is known that with conventional spatial interference alignment, i.e., linear interference alignment without symbol extension, the achievable Degrees of freedom (DoF) of MIMO X channels is no more than ⌊Douter⌋, where Douterdenotes the outer bound of DoF of MIMO X Channels. In this paper, we propose a spatial interference alignment and zero-forcing framework in combination with asymmetric complex signaling, which can achieve the DoF of 1/2⌊2Douter⌋ for arbitrary number of antennas on each terminal. Note that 1/2⌊2Douter⌋ is equal to or larger than ⌊Douter⌋. The result shows that the technique of asymmetric complex signaling, which was originally proposed for single-antenna systems, can be useful for MIMO channels, in particular when symbol extensions are not allowed.

While the outer-bound of sum degrees of freedom (DoF) of 3-user interference channel is known, the entire DoF region is still unknown in terms of both outer-bound and achievable region. In this paper, we first give the outer-bound of DoF region. Then, we present a linear beamforming scheme based on interference alignment chain whose achievable DoF region is the same as the outer bound, with the consideration of integer DoF only.

Degrees of freedom (DoF) of heterogeneous two-cell interfering uplink network is investigated in this paper. In each cell, there is one base station (BS) equipped with M antennas and arbitrary number of users equipped with different number of antennas (smaller than M). Suppose k and l active users are selected to communicate with the two BSs, respectively, it is shown that the DoF of k + l is achievable. A distributed opportunistic interference alignment scheme is then proposed which can obtain the optimal achievable DoF of the network. Numerical result shows that the scheme outperforms other existing schemes in terms of achievable sum rate.

Degree of freedom of MIMO interference channels with constant channel coefficients is studied in this paper. An asymmetric interference alignment and cancelation scheme is proposed for the 3-user MIMO interference channels with constant channel coefficients. With M antennas at each transmitter and N antennas at each receiver, the total degree of freedom (M+2N)/2 can be achieved by the proposed scheme for N <; M ≤ 2N.