Abstract Human matting refers to extracting human parts from natural images with high quality, including human detail information such as hair, glasses, hats, etc. This technology plays an essential role in image synthesis and visual effects in the film industry. When the green screen is not available, the existing human matting methods need the help of additional inputs (such as trimap, background image, etc.), or the model with high computational cost and complex network structure, which brings great difficulties to the application of human matting in practice. To alleviate such problems, we use a segmentation network as the foundation and use multiple branches to achieve human segmentation, contour detail extraction, and information fusion. We also propose a foreground probability map module, which uses the feature maps in the segmentation network to pre-estimate the foreground probabilities of each pixel and obtain Semantic Guided Matting Net. Under the condition that only a single image is needed as the input, the human matting task can be realized by making full use of the semantic information in the image. We validate our method on the P3M-10k dataset. Compared with the benchmark, our method has made significant improvements in various evaluation indicators.

Abstract This paper investigates the power allocation in a massive multiple‐input multiple‐output (MIMO) downlink system, where the base station (BS) simultaneously transmits information and energy to information terminals and energy terminals without resorting to any wireless energy harvesting (EH) protocol, respectively. First, with respect to this system, the closed‐form expressions of achievable rate and harvested energy are, respectively, derived for Rayleigh fading channels. Then, from these expressions, three optimization problems are formulated based on quality‐of‐service (QoS) requirements, with the purposes of maximizing achievable sum rate, harvested sum energy and joint QoS requirements, respectively. The optimization problem of maximizing achievable sum rate is transformed into a sequence of geometric programmings (GPs), which can be solved efficiently with standard convex optimization tools. The optimization problem of maximizing harvested sum energy is proved to be solvable as a linear program. The optimization problem of maximizing joint QoS requirements is shown to be a multi‐objective optimization problem (MOOP) and solved by a one‐dimensional search based on GP. Finally, numerical results manifest that the proposed power allocation algorithms can provide good system performances of achievable sum rate and harvested sum energy, and jointly guarantee QoS fairness in terms of achievable rate and harvested energy.

Abstract This paper considers a relay‐assisted massive multiple‐input multiple‐output (MIMO) downlink system in which a base station (BS) communicates with all terminals via a relay node enabled simultaneous wireless information and power transfer (SWIPT). Both time‐switching (TS) and energy‐splitting (PS) protocols are respectively considered at the relay node for coordinating energy harvesting and information decoding. The quality‐of‐service (QoS) fairness problem among terminals is investigated in the considered systems. In detail, the power allocation coefficients and energy harvesting time duration for TS based SWIPT, and the power allocation coefficients and PS ratio for PS based SWIPT are jointly optimised, aiming at seeking to maximise the minimum achievable rate over all terminals. As the formulated problems are non‐convex and non‐linear, and hard to solve directly, suboptimal heuristic algorithms are proposed for solving them. Additionally, the convergence and complexity of the proposed algorithms are also analysed. Finally, the performance of the proposed algorithms is demonstrated by extensive computer simulations. Results manifest that the proposed algorithms can provide uniformly great QoS for every terminal by resource allocation.

Abstract The paper investigates simultaneous wireless information and power transfer (SWIPT) in massive distributed antenna systems, where the remote radio heads (RRH) equipped with a large number of antennas are arbitrarily distributed over the coverage area and each terminal operates via a power‐splitting (PS) device. Utilising the harvested energy, each terminal transmits its pilot signals for channel estimation. A closed‐form lower bound expression on the downlink ergodic capacity is derived for each terminal and then downlink ergodic sum capacity maximisation problem with respect to RRH‐terminal association, channel estimation duration, PS ratios and power allocation (PA) is designed. The designed problem is proved to be a mixed‐inter nonlinear programming with combinatorial variables, which is generally hard to solve due to its NP‐hardness. To this end, a hierarchical iterative algorithm is developed by utilising decomposition technique. Moreover, the convergence and computational complexity of this algorithm are analysed as well. Numerical results manifest the feasibility and effectiveness of this algorithm, and also show that it can achieve the maximum downlink ergodic sum capacity for massive distributed antenna systems with SWIPT protocol.

Abstract Both cooperative non‐orthogonal multiple access (NOMA) and physical layer security (PLS) are promising techniques for future communication systems. This paper investigates the PLS performance of a cooperative NOMA system where multiple sources transmit a superposed message to two users by utilizing either amplify‐and‐forward (AF) or decode‐and‐forward (DF) relays in both non‐colluding and colluding eavesdropper scenarios. A source‐relay selection scheme is proposed for this system aiming at improving the secrecy performance. Additionally, the closed‐form expression of the secrecy outage probability (SOP) is derived for the proposed source‐relay selection schemes. It is demonstrated that in the proposed schemes, the theoretical results match well with the simulation results. In addition, the proposed schemes outperform the conventional OMA system in terms of SOP.

In this study, the authors investigate the joint design problem of interference alignment (IA) and power splitting (PS) in simultaneous wireless information and power transfer (SWIPT) networks. The existing joint design method is generally divided into two steps. First, IA is performed, and then the PS ratio is optimised. To obtain IA and SWIPT, a new joint design multi‐objective alternating optimisation (MOAO) algorithm is proposed. Different from the existing joint design method, the proposed MOAO algorithm not only maximises the sum harvested power after PS, but also maximises the sum signal‐to‐intereference‐plus‐noise ratio (SINR) by jointly designing the transmitting precoding matrices, the receiving interference suppression matrices, and the PS ratio. For this multi‐objective optimisation problem, a weighted optimisation method is proposed, which compromises between maximising the sum harvested power and maximising the sum SINR. Simulation validates that the sum achievable rate and sum harvested power of the proposed MOAO algorithm are significantly superior to those of the current scheme under the reasonable weighting coefficients.

In this study, the authors propose a simultaneous wireless information and power transfer scheme based on the power‐splitting (PS) protocol for a massive multiple‐input multiple‐output (MIMO) network. On the downlink, the base station transmits the information and energy simultaneously to each terminal and each terminal divides the received signals by its PS device between information decoder and energy harvester (EH). On the uplink, a portion of the energy harvested by EH is drawn for uplink pilot transmission in the coming frame and then the other is used for uplink data transmission in the current frame. Based on this scheme, the closed‐form lower bound expressions are obtained for the achievable downlink rate and uplink rate of each terminal and then are utilised to formulate an optimisation problem that is to maximise achievable sum rate of the whole network while ensuring the minimum required uplink and downlink rates of each terminal. However, as the original optimisation problem is non‐convex and non‐linear and very hard to solve directly, an algorithm is proposed by utilising successive approximation method based on geometric program. Numerical results verify the correctness and feasibility of the proposed algorithm.

In this paper, in light of the imaging characteristics of missile-borne MMW spotlight SAR for ground target, some main problems including the imaging resolution and choice of pulse repetition frequency are discussed. Then a new scheme of motion compensation which consists of range alignment, phase adjustment and compensation for the movement with non-constant velocity is presented. Image reconstruction algorithm by using superresolution spectrum estimation is given based on linear spectrum extrapolation technique. Finally the simulation results prove its effectivity.