1 June 2021 in IEEE Transactions on Vehicular Technology
Generalized beam modulation (GBM) is a recently proposed spectrum efficiency (SE)-enhanced index modulation (IM) scheme for narrowband millimeter wave (mmWave) massive multiple-input multiple-output (mMIMO) systems with hybrid transceivers. Unlike most existing fully-digital IM techniques whose application to wideband channels is straightforward after adopting the orthogonal frequency division multiplexing (OFDM) technique, extending GBM from the narrowband to wideband channels is quite challenging. The reason is that the per-subcarrier process in hybrid OFDM systems is no longer independent across individuals as all subcarriers share a common analog beamformer. Hence, the designed wideband GBM (wGBM) has to be compatible with hybrid OFDM systems while maintaining the SE-enhanced merit of GBM. Towards this objective, a symbol-based instead of the prevalent subcarrier-based paradigm is applied to guide wGBM design. To avoid the high detection complexity in large-scale OFDM systems, we devise a low-complexity detector by transforming the complicated symbol (block) detection into the manageable subcarrier (sub-block) detection. Finally, a simple first-order Doppler compensator is carefully designed to enhance the robustness of wGBM against Doppler. Both theoretical analyses and numerical simulations have been carried out to demonstrate the superiority of wGBM over existing alternatives in terms of error performance and energy efficiency.
11 September 2023 in IEEE Transactions on Aerospace and Electronic Systems
Power line extraction is not only crucial for UAVs obstacle avoidance, but also a fundamental step for fault diagnosis of power lines. Therefore, achieving robust and accurate extraction of power lines in aerial images is essential to enable intelligent UAVs inspection. Unfortunately, power line extraction is an extremely challenging task, and all the current methods attempt to utilize a single model to solve the problem of power line extraction in complex and variable scenes. This results in insufficient generalization ability and suboptimal computational efficiency. In this work, we propose a power line scene classification network based on complexity assessment, named SceneNet, which can provide a solution for tackling power line extraction challenges. Firstly, we propose a human-machine hybrid reasoning model to obtain the ground truth of image complexity reasonably and build the first benchmark dataset that can be used for automatic classification research of power line scenes. Secondly, we propose an improved StyleGAN3 model and loop transfer learning strategy for data augmentation. Most importantly, the SceneNet comprises a multi-feature joint embedding module and a feature encoding-decoding module. On the one hand, it achieves the multi-level fusion of artificial features and high-dimensional semantic features. On the other hand, we use a self-attention mechanism to enable full use of the contextual association between each block of the fusion feature map. The SceneNet has successfully achieved the mapping and pattern recognition between the abstract concept and the concrete features. Experimental results demonstrate that the SceneNet is obviously superior to the existing 12 state-of-the-art models, and it provides guidance and delineation of applicable scenes for power line extraction methods.
28 November 2019 in Mobile Networks and Applications
21 December 2023 in IEEE Geoscience and Remote Sensing Letters
Robust power line detection from aerial images is an essential step for intelligent unmanned aerial vehicles (UAVs) inspection. However, compared to regular object detection tasks, the generalization ability and accuracy of current methods are seriously insufficient due to the variability of power line scenes and the low proportion of power lines in the image. In this letter, we propose an object-based Markov random field with discrete multi-neighborhood system (OMRF-DMNS) model, which can provide a solution for tackling power line detection challenges. Firstly, the proposed method constructs a discrete multi-neighborhood system (DMNS) for the non-contact nodes. This not only greatly improves the anti-noise ability of the model, but also lays the foundation for the transfer of context information among nodes under different receptive fields. Secondly, a multi-level logical model with multi-neighborhood joint reasoning (MLL-MNJR) is defined on the DMNS, which makes full use of the line features and location information to achieve semantic inference of nodes. Therefore, the OMRF-DMNS model can fully mine the multi-dimensional feature associations and context information among nodes, enabling it to effectively distinguish between power line segments and interference noises. Experimental results demonstrate that the proposed method significantly outperforms previous methods in different power line scenes.
24 July 2025 in IEEE Transactions on Communications
In this paper, we investigate the near-field wideband terahertz (THz) massive multi-input multi-output (MIMO) integrated sensing and communication (ISAC) systems. Specifically, we propose an energy-efficient serial true-time-delay (TTD)-based reconfigurable antenna architecture for the dual-function base station (DFBS). This architecture enables DFBS to switch between modular and compact MIMO configurations by dynamically controlling the activation of antenna subarrays, corresponding to the sensing and communication stages, respectively. During the sensing stage, a modular MIMO architecture is deployed by deactivating several subarrays and all TTDs. This configuration utilizes fewer antennas to achieve a larger array aperture with enhanced energy efficiency. Moreover, we take advantage of beam squint effects at both the main and grating lobes to extend sensing range and enable rapid user sensing. During the communication stage, a compact MIMO configuration is employed by activating all antennas, and the TTD network is activated to mitigate the near-field beam squint effect. Based on the channel state information obtained during the sensing stage, we formulate a joint optimization problem of the hybrid analog/digital beamforming and TTD network time delays to maximize the system sum rate. To solve it, we first design analog beamforming and time delays through a serial-delay approach, followed by an alternating iterative optimization for digital beamforming. In addition, our proposed scheme accounts for the finite resolution and limited delay range of TTDs. Simulation results demonstrate the effectiveness of our designed antenna architecture and optimization scheme.
26 October 2020 in IEEE Wireless Communications Letters
To better serve future vehicular network, we judiciously design an anti-Doppler index modulation termed as pre-compensation assisted generalized beam index modulation (PC-GBIM) for wideband mmWave massive multiple-input multiple-output (mMIMO) systems. The proposed PC-GBIM design comprises two key elements. First, we generalize the existing beam index modulation (BIM) from the hybrid narrowband to hybrid wideband systems and from activating single beam to activating multiple beams. Secondly, an effective first-order Doppler pre-compensator is introduced to assist index modulation, such that the originally complicated block-based maximum likelihood (ML) detection degenerates into the simple subcarrier-based ML detection. Both the theoretical analyses and numerical simulations have been made and demonstrate the superiority over existing alternatives in doubly-selective channels.