This paper presents the results on developing an ensemble machine learning model to combine commercial gas sensors for accurate concentration detection. Commercial gas sensors have the low-cost advantage and become key components of IoT devices in atmospheric condition monitoring. However, their native coarse resolution and poor selectivity limit their performance. Thus, we adopted recurrent neural network (RNN) models to extract the time-series concentration data characteristics and improve the detection accuracy. Firstly, four types of RNN models, LSTM and GRU, Bi-LSTM, and Bi-GRU, were optimized to define the best-performance single weak models for CO, O3, and NO2 gases, respectively. Next, ensemble models which integrate multiple single weak models with a dynamic model were defined and trained. The testing results show that the ensemble models perform better than the single weak models. Further, a retraining procedure was proposed to make the ensemble model more flexible to adapt to environmental conditions. The significantly improved determination coefficients show that the retraining helps the ensemble models maintain long-term stable sensing performance in an atmospheric environment. The result can serve as an essential reference for the applications of IoT devices with commercial gas sensors in environment condition monitoring.
In this letter, we present results of a study on the directional enhanced radiation generated by a point resonant cavity operating at the band-gap frequency of a two-dimensional finite-size phononic crystal. We find that the directive radiation for operating frequency within the band gap comes from the direction-dependent transmittance property of the square-lattice phononic crystal. The directive-radiation characteristic can be capitalized on enhancing the emission from the phononic crystals by coupling the source with point-defect modes, and the enhanced directive radiation patterns with main lobe width less than are demonstrated.
Surface acoustic wave (SAW) devices are widely used in commercial products as filters and resonators. SAW devices are also applied as passive, wireless radio frequency identification (RFID) tags and sensors, which can be used in harsh environments and consume no batteries. In this study, we designed and fabricated an SAW-based RFID tag with the added function of a high-resolution temperature sensor. A coupling of modes model was adopted to design 433MHz SAW-based tags/sensors. An improved signal processing method was used to increase the resolution of time-domain signals, enabling the slight change of delay time caused by temperature variation to be detected. Subsequently, the SAW tags/sensors were fabricated on 128° Y-cut lithium niobate and used to detect temperature shifts. The results revealed that high-resolution delay-time SAW devices are feasible for measuring temperatures precisely and can be applied to other SAW-based sensors.
Phononic crystals (PnC) perform band gaps and promise applications such as filters and reflectors. Traditional PnC for surface acoustic waves (SAW) consisting of deep periodic holes in half space. To apply PnCs in SAW devices, the feasibility of fabrication was considered and layered PnCs were proposed. In this paper, a PnC comprised of cylindrical holes in a SiO 2 layer covering a 128°rotated Y-cut LiNbO 3 substrate was analyzed numerically and used to design SAW resonators. Acoustic dispersion of SAWs in the PnC was calculated by finite element method and the band gap range was indicated. The band gaps were manipulated by the layer thickness and cylinder size. Then the layered PnC was applied as gratings to reflect SAWs. Resonant cavities for SAW were further designed, and a 942.875 MHz SAW resonator with Q-factor of 433 was obtained. The result showed feasibility of applying PnC in acoustic electronics devices via MEMS fabrication.
Phononic crystal (PnC) in half-space allows anisotropic propagation and band gaps for surface acoustic wave (SAW) and promise applications such as filters and reflectors. For the feasible MEMS fabrication, a PnC comprised of cylindrical holes in a SiO 2 layer covering a 128°rotated Y-cut LiNbO 3 substrate was applied and the acoustic modes were analyzed numerically. The band gap for SAW was indicated and the PnC was used as reflective gratings to construct SAW resonators. By tuning the cavity length, the resonant frequency was controlled. Then a ladder-type bandpass filter at 933-947 MHz was defined by two phononic SAW resonators. The result showed feasibility of applying PnC in acoustic electronics devices.