Search

Exploiting simulation modeling of graphene synaptic field‐effect transistors is extremely important for helping researchers to construct carbon‐based neuromorphic computing systems. Here, lateral‐ionic‐gated graphene synaptic FETs with different gate lengths are fabricated, and they are modeled by using basic physic models combined with the ions migration‐diffusion model and graphene material model. The feasibility and accuracy of the proposed modeling are validated by showing an excellent agreement between simulations and experimental results. The slicing technique of the modeling is proposed to analyze the influence of ionic concentration and diffusion coefficient on the ions movement to reveal their working mechanism. The effect of key parameters about gate length, ionic concentration, and diffusion coefficient on synaptic behavior such as short‐term plasticity, and long‐term plasticity is simulated and discussed. In addition, three kinds of spike‐timing‐dependent plasticity are obtained by the device modeling. This research opens up promising avenues for the development of artificial synapse modeling and paths to new opportunities for the construction of carbon‐based neuromorphic networks. [ABSTRACT FROM AUTHOR]

In this paper, a subset-optimized eight-dimensional trellis-coded quadrature amplitude modulation (SO-8DTCM-16QAM) format for higher-order constellations in high-speed optical communications is proposed. This scheme increases the number of subsets of base 2D constellation divisions. On this basis, it is further optimized by using 2D subsets for Cartesian product combinations to obtain 4D subsets and eliminate the combinations with small Euclidean distances. Finally, the 4D subsets are utilized to construct interrelated 8D subsets for trellis coding modulation and signal transmission. The proposed scheme can effectively reduce the decoding complexity and outperforms the conventional scheme at a high signal-to-noise ratio (SNR). Simulation verification of the proposed scheme is carried out, and the results show that the SO-8DTCM-16QAM achieves signal-to-noise ratio (SNR) gains of 1.60 dB, 1.56 dB, 1.51 dB, and 1.33 dB, respectively, compared with the conventional 8D-16QAM signals when BTB and 5/20/30 km optical signal transmission are performed. The SO-8DTCM-16QAM also achieves an SNR gain of 1.86 dB, 1.75 dB, and 1.22 dB at a net transmission rate of 14/21/28 GBaud. In addition, the SO-8DTCM-16/32/64QAM achieves an SNR gain of 1.27 dB, 0.80 dB, and 1.24 dB, respectively, when compared with the unoptimized 8DTCM-16/32/64QAM. Meanwhile, the proposed eight-subset SO-8DTCM-QAM scheme reduces the complexity of the decoding computation in the subset selection part and the constellation point selection part by 93.75% and 50%, respectively, compared with the unoptimized eight-subset and four-subset 8DTCM-QAM schemes. It can be seen that the proposed scheme simultaneously optimizes the transmission performance and complexity of high-speed optical communication systems and has practical application value. [ABSTRACT FROM AUTHOR]

X-ray fluorescence (XRF) spectroscopy is a non-destructive differential measurement technique widely utilized in elemental analysis. However, due to its inherent high non-linearity and noise issues, it is challenging for XRF spectral analysis to achieve high levels of accuracy. In response to these challenges, this paper proposes a method for XRF spectral analysis that integrates an adaptive genetic algorithm with a backpropagation neural network, enhanced by an attention mechanism, termed as the AGA-BP-Attention method. By leveraging the robust feature extraction capabilities of the neural network and the ability of the attention mechanism to focus on significant features, spectral features are extracted for elemental identification. The adaptive genetic algorithm is subsequently employed to optimize the parameters of the BP neural network, such as weights and thresholds, which enhances the model's accuracy and stability. The experimental results demonstrate that, compared to traditional BP neural networks, the AGA-BP-Attention network can more effectively address the non-linearity and noise issues of XRF spectral signals. In XRF spectral analysis of air pollutant samples, it achieved superior prediction accuracy, effectively suppressing the impact of background noise on spectral element recognition. [ABSTRACT FROM AUTHOR]

IntroductionIn craniofacial surgery, magnetic navigation systems can effectively extend the doctor's limited visual range, improve their surgical precision, shorten the operation time, and reduce the incidence of surgical complications. Owing to the ease of magnetic navigation, the accuracy of the magnetic navigation system is affected by various equipment in the operating room. Therefore, its large-scale application is lacking because the navigation accuracy requirement can be extremely high during craniofacial surgery. Therefore, the accuracy of magnetic navigation systems is crucial. Various surgical instruments have been evaluated to effectively reduce the interference of magnetic navigation systems with surgical instruments. In craniofacial surgery, magnetic navigation systems can effectively extend the doctor's limited visual range, improve their surgical precision, shorten the operation time, and reduce the incidence of surgical complications. Owing to the ease of magnetic navigation, the accuracy of the magnetic navigation system is affected by various equipment in the operating room. Therefore, its large-scale application is lacking because the navigation accuracy requirement can be extremely high during craniofacial surgery. Therefore, the accuracy of magnetic navigation systems is crucial. Various surgical instruments have been evaluated to effectively reduce the interference of magnetic navigation systems with surgical instruments. In the surgical environment, the use of surgical instruments during mandibular surgery was simulated by selecting several conventional surgical instruments to record errors in the magnetic navigation system. The fluctuation values of the magnetic navigation errors were subsequently estimated and changes in its accuracy measured. MATLAB was used to calculate and analyze the fluctuations of the magnetic navigation errors. As results, the high-frequency electrosurgical system caused the greatest interference with the magnetic navigation system during surgery while powered on, with a maximum fluctuation error value of 1.8120 mm, and the maximum fluctuation error values of the stitch scissors, teeth forceps, and a needle holder were 1.3662, 1.3781, and 0.3912 mm, respectively. The closer the instrument is to the magnetic field generator or navigation target, the greater its impact. In conclusion, stitch scissors, teeth forceps, a needle holder, and the high-frequency electrosurgical system all affect magnetic navigation system accuracy. Therefore, it is necessary to avoid magnetic navigation system use and surgical instrument disturbances during surgery or select surgical instruments that do not interfere with the system. Surgical instruments must be evaluated for electromagnetic interference before they can be used in surgery with a magnetic navigation system. [ABSTRACT FROM AUTHOR]

In this paper, we propose an innovative channel-sensitive autoencoder (CSAE)-aided end-to-end deep learning (E2EDL) technique for joint geometric probabilistic shaping. The pretrained conditional generative adversarial network (CGAN) is introduced in the CSAE which performs differentiable substitution of the optical fiber channel model under variable input optical power (IOP) levels. This enables the CSAE-aided E2EDL to design optimal joint geometric probabilistic shaping schemes for optical fiber communication systems at varying IOPs. The results of the proposed CSAE-aided E2EDL technique show that for a dual-polarization 64-Gbaud signal with a transmission distance of 5 × 80 km, when the modulation format is a 64-quadrature amplitude modulation (QAM) or a 128-QAM, the maximum generalized mutual information (GMI) level learned via CSAE-aided E2EDL is 5.9826 or 6.8384 bits/symbol under varying IOPs, respectively. In addition, the pretrained CGAN, as a substitution for optical fiber transmission model, accurately characterizes the distortion of signals with different IOPs, with an average bit error ratio (BER) difference of only 1.83%, an average mean square error (MSE) of 0.0041 and an average K-L divergence of 0.0046. In summary, this paper delivers new insights into the application of E2EDL and demonstrates the feasibility of joint geometric probabilistic shaping-based E2EDL for fiber optic communication systems with varying IOPs. [ABSTRACT FROM AUTHOR]