This paper investigates the downlink performance of a cell-free massive multiple-input multiple-output (CF-mMIMO)-aided simultaneous wireless information and power transfer Internet-of-Things (SWIPT-IoT) network. Considering a more practical scenario with non-ideal radio frequency responses and low-resolution analog-to-digital converters (ADCs) and digital-to-analog converters (DACs), closed-form expressions for the lower-bound achievable rate and harvested energy (HE) under conjugate beamforming (CB) and enhanced normalized CB (ECB) schemes are derived. By virtue of these closed-form results, we conduct comprehensive HE and rate performance analyses with respect to various system parameters. To further enhance the downlink rate performance, a novel hybrid precoding scheme that combines the benefits of both CB and ECB is proposed. Aside from the above, we also develop a particle swarm optimization framework in terms of ADC/DAC resolution to maximize the sum-HE and sum-rate. All theoretical analyses and the performance of the proposed optimization algorithm are validated via extensive simulations.

Video-based scoring using neural networks is a very important means for evaluating many sports, especially figure skating. Although many methods for evaluating action quality have been proposed, there is no uniform conclusion on the best feature extractor and clip length for the existing methods. Furthermore, during the feature aggregation stage, these methods cannot accurately locate the target information. To address these tasks, firstly, we systematically compare the effects of the figure skating model with three different feature extractors (C3D, I3D, R3D) and four different segment lengths (5, 8, 16, 32). Secondly, we propose a Multi-Scale Location Attention Module (MS-LAM) to capture the location information of athletes in different video frames. Finally, we present a novel Multi-scale Location Attentive Long Short-Term Memory (MLA-LSTM), which can efficiently learn local and global sequence information in each video. In addition, our proposed model has been validated on the Fis-V and MIT-Skate datasets. The experimental results show that I3D and 32 frames per second are the best feature extractor and clip length for video scoring tasks. In addition, our model outperforms the current state-of-the-art method hybrid dynAmic-statiC conText-aware attentION NETwork (ACTION-NET), especially on MIT-Skate (by 0.069 on Spearman’s rank correlation). In addition, it achieves average improvements of 0.059 on Fis-V compared with Multi-scale convolutional skip Self-attentive LSTM Module (MS-LSTM). It demonstrates the effectiveness of our models in learning to score figure skating videos.

This letter investigates the downlink simultaneous wireless information and power transfer in a reconfigurable intelligent surface (RIS)-aided Internet-of-Everything network relying on cell-free architecture. In our network, the access points beam the energy and information signals toward users by means of time-switching protocol. Based on the correlated Rician fading channels, we derive tight closed-form expressions for the downlink achievable harvested energy and rate under imperfect channel state information. Moreover, we propose a particle swarm optimization-based algorithm in terms of RIS phase shift to maximize the achievable harvested energy and rate. All theoretical analyses and the effectiveness of the proposed algorithm are verified via extensive simulations.