The design of a high-gain and low-profile H -plane horn antenna embedded into a large metallic platform is proposed. The antenna is composed of three parts: coaxial-to-ridged transition, wideband H -plane antenna, and dielectric lens. The coaxial-to-ridged transition makes an efficient conversion from TEM modes in a coaxial probe to the fundamental TE 10 mode in a ridged horn, and by choosing the ratio between horn aperture and length and suppressing the TE higher mode, the optimal design of the H -plane horn antenna can be obtained. Then, a tapered dielectric lens is located in the end-fire direction, which converts the TE wave into the leaky-wave mode, enhancing the peak gain. A prototype of the proposed antenna is finally fabricated and tested. The measured results are in good agreement with the simulated ones, which show that the proposed antenna has a very wide bandwidth from 2.5 to 20 GHz for the voltage standing wave ratio (VSWR) < 2.5 and exhibits a low thickness of only 8 mm (0.066λ L , λ L is the free-space wavelength at the lowest operating frequency). Good radiation pattern and high gain can be achieved over a wide frequency band.INDEX TERMS End-fire antenna, horn antenna, tapered dielectric lens, wideband, high gain.

No abstract

Feature extraction is a key part of the electronic tongue system. Almost all of the existing features extraction methods are "hand-crafted", which are difficult in features selection and poor in stability. The lack of automatic, efficient and accurate features extraction methods has limited the application and development of electronic tongue systems. In this work, a convolutional neural network-based auto features extraction strategy (CNN-AFE) in an electronic tongue (e-tongue) system for tea classification was proposed. First, the sensor response of the e-tongue was converted to time-frequency maps by short-time Fourier transform (STFT). Second, features were extracted by convolutional neural network (CNN) with time-frequency maps as input. Finally, the features extraction and classification results were carried out under a general shallow CNN architecture. To evaluate the performance of the proposed strategy, experiments were held on a tea database containing 5100 samples for five kinds of tea. Compared with other features extraction methods including features of raw response, peak-inflection point, discrete cosine transform (DCT), discrete wavelet transform (DWT) and singular value decomposition (SVD), the proposed model showed superior performance. Nearly 99.9% classification accuracy was obtained and the proposed method is an approximate end-to-end features extraction and pattern recognition model, which reduces manual operation and improves efficiency.

In this paper, a metasurfing (MS) concept is demonstrated and applied in the design of the low profile and wideband endfire antenna on metallic surface environments. The MS comprises an array of varying patch printed on homogeneous host medium, and fed by a surface wave launcher (SWL). Each row patch is designed according to the operating wavelength and by altering the surface reactance of the MS the surface-wave mode can be manipulated into the free-space wave mode. Meanwhile, two row of rectangle patches with same size are located on between the surface wave launcher (SWL) and non-uniform MS, which is regarded as an impedance modulation to obtain a good impedance matching. The VSWR of the proposed antenna is below 2.5 from 3.8 GHz to 16.7 GHz in the measured results, which are in good agreement with the simulated results and the thickness of proposed antenna is only 5 mm (0.065λ L , λ L is the free-space wavelength at the lowest operating frequency). Moreover, a stable end-fire beam and low side lobe level (SLL) is obtained in a wide frequency band, and the group delay and the time-domain result also are shown to prove the good wideband transmission. INDEX TERMS End-fire antenna, non-uniform metasurface, low profile, wideband antenna.

No abstract