The automated classification of heart sounds plays a significant role in the diagnosis of cardiovascular diseases (CVDs). With the recent introduction of medical big data and artificial intelligence technology, there has been an increased focus on the development of deep learning approaches for heart sound classification. However, despite significant achievements in this field, there are still limitations due to insufficient data, inefficient training, and the unavailability of effective models. With the aim of improving the accuracy of heart sounds classification, an in-depth systematic review and an analysis of existing deep learning methods were performed in the present study, with an emphasis on the convolutional neural network (CNN) and recurrent neural network (RNN) methods developed over the last five years. This paper also discusses the challenges and expected future trends in the application of deep learning to heart sounds classification with the objective of providing an essential reference for further study.

To leverage massive distributed data and computation resources in the Internet-of-Things (IoT) networks, federated learning (FL) is considered to be a promising technique with benefits of improved data privacy and communication efficiency. Meanwhile, cell-free massive multiple-input multiple-output (cell-free massive MIMO) is a promising technology to enable the IoT networks to support FL. By deploying the access points (APs) closer to the IoT devices, path loss attenuation can be reduced. However, the performance of FL is still constrained by the limited power resources of IoT devices. To address this issue, we design an energy-efficient FL scheme over cell-free IoT networks by formulating an optimization problem to minimize the total energy consumption of the IoT devices participating in the FL process. To solve the intractable problem in hand, by exploiting its unique structure, we decompose it into three sub-problems that facilitate the development of the proposed scheme. First, we derive the optimal central processing unit (CPU) operating frequency for IoT devices. Then, we design an optimal power allocation scheme to mitigate the straggler effect. Next, nonlinear programming method is adopted to obtain a sub-optimal solution for the reformulated problem. Finally, a three-stage algorithm is proposed for energy consumption minimization by considering these sub-problems. Simulation results demonstrate the close-to-optimal performance of the proposed algorithm for energy savings compared with three baseline algorithms and the capability to support large numbers of IoT device access by mitigating the straggler effect.

This paper analyzes the performance of a satellite-terrestrial free-space optical (FSO)/millimeter wave (MMW) radio frequency (RF) relaying system with amplify-and-forward (AF) and decode-and-forward (DF) relaying. The Málaga ( $\mathcal {M}$ ) distribution with pointing error impairments and the multi-cluster fluctuating two-ray (MFTR) fading model are introduced to characterize the FSO link for both heterodyne detection (HD) and indirect modulation/direct detection (IM/DD) and MMW link, respectively. Therefore, the precise closed-form expression of end-to-end signal-to-noise ratio (SNR) for the outage probability (OP), average bit error rate (BER), ergodic capacity (EC) and effective capacity (EFC) is derived. In addition, we present an asymptotical result analysis for the OP and average BER at high SNR in terms of simple functions. Finally, we employ Monte-Carlo (MC) simulation results to verify all our analytical results.

Cell-free massive multiple-input multiple-output (mMIMO) has recently been proposed to improve cell edge performance. However, most prior works consider perfect hardware impairments (HIs), which are difficult to be achieved in practical systems. This paper studies the impact of HI in an uplink cell-free mMIMO system with both multi-antenna access points (APs) and multi-antenna user terminals (UTs) under the Weichselberger channel model.Firstly, we study a two-layer decoding scheme with local minimum mean-squared error or maximum ratio combining at the AP side and with optimal large-scale fading decoding in the central processing unit. We derive novel closed-form SE expressions and prove that the effect of HI can be mitigated in the case of UTs with multiple antennas. However, the achievable SE is constrained by the pilot contamination and pilot overhead. To this end, the superimposed pilot (SP) transmission method is considered in this paper, where all the coherence intervals are used for both pilot and data symbols transmission. Finally, numerical results verify our derived expressions and reveal the relationship between HI and the number of antennas per UT for different pilot schemes. Note that the advantages of SP over regular pilots disappear when the hardware quality decreases with multi-antenna UTs.

This study proposes an uplink reconfigurable intelligent surface (RIS)-aided free-space optical (FSO) and radio frequency (RF) hybrid satellite-terrestrial network (HSTN). We consider the impact of imperfect channel state information (CSI) on the RIS system under fixed-gain amplify-and-forward (AF) and decode-and-forward (DF) relay conditions. We derive closed-form expressions for various performance metrics of the end-to-end signal-to-noise ratio (SNR). We model the FSO link using the Málaga ( $\mathcal {M}$ ) distribution with two detection schemes and use an RIS-aided Rayleigh fading model for the RF link, both of which consider the satellite-terrestrial loss. At high SNRs, we determine the relevant asymptotic expressions and validate the analytical results through Monte Carlo simulations. Additionally, we evaluate the impacts of key system parameters, such as RIS configuration parameters and imperfect CSI, on the end-to-end performance.