Rate-Distortion-Perception Optimized Neural Speech Transmission System for High-Fidelity Semantic Communications

We consider the problem of learned speech transmission. Existing methods have exploited joint source-channel coding (JSCC) to encode speech directly to transmitted symbols to improve the robustness over noisy channels. However, the fundamental limit of these methods is the failure of identification of content diversity across speech frames, leading to inefficient transmission. In this paper, we propose a novel neural speech transmission framework named NST. It can be optimized for superior rate-distortion-perception (RDP) performance toward the goal of high-fidelity semantic communication. Particularly, a learned entropy model assesses latent speech features to quantify the semantic content complexity, which facilitates the adaptive transmission rate allocation. NST enables a seamless integration of the source content with channel state information through variable-length joint source-channel coding, which maximizes the coding gain. Furthermore, we present a streaming variant of NST, which adopts causal coding based on sliding windows. Experimental results verify that NST outperforms existing speech transmission methods including separation-based and JSCC solutions in terms of RDP performance. Streaming NST achieves low-latency transmission with a slight quality degradation, which is tailored for real-time speech communication.

TRCCBP: Transformer Network for Radar-Based Contactless Continuous Blood Pressure Monitoring

Contactless continuous blood pressure (BP) monitoring is of great significance for daily healthcare. Radar-based continuous monitoring methods typically extract time-domain features manually such as pulse transit time (PTT) to calculate the BP. However, breathing and slight body movements usually distort the features extracted from pulse-wave signals, especially in long-term continuous monitoring, and manually extracted features may have limited performance for BP estimation. This article proposes a Transformer network for Radar-based Contactless Continuous Blood Pressure monitoring (TRCCBP). A heartbeat signal-guided single-beat pulse wave extraction method is designed to obtain pure pulse-wave signals. A transformer network-based blood pressure estimation network is proposed to estimate BP, which utilizes convolutional layers with different scales, a gated recurrent unit (GRU) to capture time-dependence in continuous radar signal and multi-head attention modules to capture deep temporal domain characteristics. A radar signal dataset captured in an indoor environment containing 31 persons and a real medical situation containing five persons is set up to evaluate the performance of TRCCBP. Compared with the state-of-the-art method, the average accuracy of diastolic blood pressure (DBP) and systolic blood pressure (SBP) is 4.49 mmHg and 4.73 mmHg, improved by 12.36 mmHg and 8.80 mmHg, respectively. The proposed TRCCBP source codes and radar signal dataset have been made open-source online for further research.

Anisotropic to Isotropic Transition in Monolayer Group-IV Tellurides

Monolayer group-IV tellurides with phosphorene-derived structures are attracting increasing research interest because of their unique properties. Here, we systematically studied the quasiparticle electronic and optical properties of two-dimensional group-IV tellurides (SiTe, GeTe, SnTe, PbTe) using the GW and Bethe-Salpeter equation method. The calculations revealed that all group-IV tellurides are indirect bandgap semiconductors except for monolayer PbTe with a direct gap of 1.742 eV, while all of them are predicted to have prominent carrier transport ability. We further found that the excitonic effect has a significant impact on the optical properties for monolayer group-IV tellurides, and the predicted exciton binding energy is up to 0.598 eV for SiTe. Interestingly, the physical properties of monolayer group-IV tellurides were subject to an increasingly isotropic trend: from SiTe to PbTe, the differences of the calculated quasiparticle band gap, optical gap, and further exciton binding energy along different directions tended to decrease. We demonstrated that these anisotropic electronic and optical properties originate from the structural anisotropy, which in turn is the result of Coulomb repulsion between non-bonding electron pairs. Our theoretical results provide a deeper understanding of the anisotropic properties of group-IV telluride monolayers.

Dynamics in the Sakaguchi-Kuramoto model with bimodal frequency distribution

In this work, we study the Sakaguchi-Kuramoto model with natural frequency following a bimodal distribution. By using Ott-Antonsen ansatz, we reduce the globally coupled phase oscillators to low dimensional coupled ordinary differential equations. For symmetrical bimodal frequency distribution, we analyze the stabilities of the incoherent state and different partial synchronous states. Different types of bifurcations are identified and the effect of the phase lag on the dynamics is investigated. For asymmetrical bimodal frequency distribution, we observe the revival of the incoherent state, and then the conditions for the revival are specified.

Efficient Polarization Beam Splitter Based on All-Dielectric Metasurface in Visible Region

In this paper, we present an all-dielectric gradient metasurface, composed of periodic arrangement of differently sized cross-shaped silicon nanoblocks resting on the fused silica substrate, to realize the function of polarization split in visible region. The cross-shaped silicon block arrays can induce two opposite transmission phase gradients along the x-direction for the linear x-polarization and y-polarization. By properly designing, the metasurface can separate the linearly polarized light into x- and y-polarized ones, which propagate at the same angle along the left and right sides of the normal incidence in the x-z plane. Particularly, when a beam with the polarization angle of 45.0° is incident on the proposed device, the x- and y-polarized transmitted ones possess nearly equal intensity within the wavelength range from 579 to 584 nm. We expect the proposed polarization beam splitter can play an important role for future free-space optical devices.

Optically Active Plasmonic Metasurfaces based on the Hybridization of In-Plane Coupling and Out-of-Plane Coupling

Plasmonic metasurfaces have attracted much attention in recent years owing to many promising prospects of applications such as polarization switching, local electric field enhancement (FE), near-perfect absorption, sensing, slow-light devices, and nanoantennas. However, many problems in these applications, like only gigahertz switching speeds of electro-optical switches, low-quality factor (Q) of plasmonic resonances, and relatively low figure of merit (FOM) of sensing, severely limit the further development of plasmonic metasurface. Besides, working as nanoantennas, it is also challenging to realize both local electric FE exceeding 100 and near-perfect absorption above 99%. Here, using finite element method and finite difference time domain methods respectively, we firstly report a novel optically tunable plasmonic metasurface based on the hybridization of in-plane near-field coupling and out-of-plane near-field coupling, which provides a good solution to these serious and urgent problems. A physical phenomenon of electromagnetically induced transparency is obtained by the destructive interference between two plasmon modes. At the same time, ultrasharp perfect absorption peaks with ultra-high Q-factor (221.43) is achieved around 1550 nm, which can lead to an ultra-high FOM (214.29) in sensing application. Particularly, by using indium-doped CdO, this metasurface is also firstly demonstrated to be a femtosecond optical reflective polarizer in near-infrared region, possessing an ultra-high polarization extinction ratio. Meanwhile, operating as nanoantennas, this metasurface achieves simultaneously strong local electric FE(|E_loc|/|E₀| > 100) and a near-perfect absorption above 99.9% for the first time, which will benefit a wide range of applications including photocatalytic water splitting and surface-enhanced infrared absorption.