In this paper we demonstrate continuous noisy speech recognition using connectionist temporal classification (CTC) model on limited Chinese vocabulary using electroencephalography (EEG) features with no speech signal as input and we further demonstrate single CTC model based continuous noisy speech recognition on limited joint English and Chinese vocabulary using EEG features with no speech signal as input.

This paper proposed a plasmonic resonator system, consisting of a metal-insulator-metal structure and two stubs, and a Fano resonance arose in its transmittance, which resulted from the coupling between the two stubs. On the basis of the proposed structure, a circle and a ring cavity are separately added above the stubs to create different coupled plasmonic structures, providing triple and quadruple Fano resonances, respectively. Additionally, by adjusting the geometric parameters of the system, multiple Fano Resonances obtained can be tuned. The proposed structure can be served as a high efficient refractive index sensor, yielding a sensitivity of 2000 nm/RIU and figure of merit (FOM) of 4.05 × 10 4 and performing better than most of the similar structures. It is believed that the proposed structure may support substantial applications for on-chip sensors, slow light and nonlinear devices in highly integrated photonic circuits.

No abstract

Ultra-high quality (Q) factor resonances derived from the bound states in the continuum (BICs) have drawn much attention in optics and photonics. Especially in meta-surfaces, they can enable ultrasensitive sensors, spectral filtering, and lasers because of their enhanced light-matter interactions and rare superiority of scalability. In this paper, we propose a permittivity-asymmetric all-dielectric meta-surface, comprising high-index cuboid tetramer clusters with symmetric structural parameters and configuring periodically on a glass substrate. Simulation results offer dual-band quasi-BICs with high Q values of 4447 and 11391, respectively. Multipolar decomposition in cartesian and electromagnetic distributions are engaged to analyze the physical mechanism of dual quasi-BIC modes, which reveals that they are both governed by magnetic quadrupole (MQ) and in-plane toroidal dipole (TD). The polarization-insensitive and scalable characteristics are also investigated. Additionally, we appraise the sensing performances of the proposed structure. As an example, our work supports an uncommon route to design dual-band polarization-insensitive TD quasi-BICs resonators and facilitates their applications in optic and photonics, such as low-threshold lasers and sensing.

This research combines a next-generation wireless network and a quadrotor unmanned aerial vehicle (UAV) to create a real-time wireless network quadrotor UAV flight control system. Three major problems will occur when the system operates, such as external disturbances, internal actuator failures and wireless network association failures (random delay and packet loss). We propose a hierarchical distributed comprehensive robust adaptive fault-tolerant control algorithm based on robust fault-tolerant theory to improve the performance of this system. The simulation and flight experimental test results show that when this system is analyzed with respect to external disturbances, internal actuator failures and wireless network association failures, the designed controller stability is asymptotically stable, the performance of this system is very good and the system provides a strategy for establishing a ground-to-air self-organizing wireless network.