Abstract Background Epilepsy is a neurological disorder that usually detected by electroencephalogram (EEG) signals. Since manual examination of epilepsy seizures laborious and time-consuming process, lots automatic detection algorithms have been proposed. However, most the available classification for EEG signals adopted single feature extraction, in turn to result low accuracy. Although small account studies carried out fusion, computational efficiency reduced due too many features,...

The diagnosis and treatment of epilepsy continue to face numerous challenges, highlighting the urgent need for development rapid, accurate, non-invasive methods seizure detection. In recent years, advancements in analysis electroencephalogram (EEG) signals have garnered widespread attention, particularly area recognition. A novel hybrid deep learning approach that combines feature fusion efficient detection is proposed this study. First, Discrete Wavelet Transform (DWT) applied perform a...

Epilepsy is one of the most well-known neurological disorders globally, leading to individuals experiencing sudden seizures and significantly impacting their quality life. Hence, there an urgent necessity for efficient method detect predict in order mitigate risks faced by epilepsy patients. In this paper, a new seizure detection prediction proposed, which based on multi-class feature fusion convolutional neural network-gated recurrent unit-attention mechanism (CNN-GRU-AM) model. Initially,...

Brain tumors can be classified into many different types based on their shape, texture, and location. Accurate diagnosis of brain tumor help doctors to develop appropriate treatment plans save patients’ lives. Therefore, it is very crucial improve the accuracy this classification system for assist in treatment. We propose a deep feature fusion method convolutional neural networks enhance robustness while mitigating risk over-fitting. Firstly, extracted features three pre-trained models...

In this paper, a differential cross-coupled Colpitts voltage control oscillator (VCO) based on 1 µm GaAs HBT technology is presented, and response surface method (RSM) used to co-optimize the phase noise, center frequency, tuning range of VCO. The structure provide negative resistance applied for achieving low noise. Then, we choose five parameters VCO circuit through establishing RSM model, in order find optimal combination design obtain best performance. results show that frequency...

Brain tumors pose significant harm to the functionality of human nervous system. There are lots models which can classify brain tumor type. However, available methods did not pay special attention long-range information, limits model accuracy improvement. To solve this problem, in paper, an enhanced short-range and dependent system for classification, named as EnSLDe, is proposed. The EnSLDe consists three main modules: Feature Extraction Module (FExM), Enhancement (FEnM), Classification...

Based on the GaN HEMT 20-element small-signal model, an improved particle swarm algorithm is proposed in this paper for optimization of intrinsic parameters. The conventional Particle Swarm Optimization (PSO) prone to fall into local optimum solutions when parameter performed, turn affect accuracy experimental results. To solve above problem, dynamic evolutionary estimation, elite learning strategies and adaptive are added traditional PSO algorithm, yield Adaptive (APSO) paper. results show...

—In this paper, an improved genetic algorithm-based hybrid direct-optimal extraction method for small-signal model of GaN HEMT devices is proposed. Simulated annealing algorithm and fuzzy adaptive strategy are incorporated into parameter optimization, to overcome the shortcoming conventional that tend fall local optimum solutions. The validity proposed verified by using a 20-element equivalent circuit under different operating conditions. results show modeled S-parameters agree well with...

In order to effectively solve environmental problems induced by organic pollutants, the MoS2@ZnO heterostructure is successfully fabricated via a two-step hydrothermal method, in which MoS2 nanosheets grew on surface of ZnO nanorods. The synthetic demonstrated possess excellent photocatalytic activity degradation methylene blue (MB) under ultraviolet (UV) and visible light irradiation. corresponding photodegradation rate constant reaches up 0.02075 min−1 0.00916 min−1, are higher than that...

an improved small-signal model for the GaN pseudomorphic high electron mobility transistor (P-HEMT) process is proposed in this paper. Experimental studies have found that a positive bias voltage usually damages or destroys Schottky grid. In order to solve problem, method of extracting parasitic inductance and resistance using low gate technology proposed. research circuit parameters will change with voltage. New intrinsic (RL) (Lds) are introduced into equivalent suppress its dependence...

The Differential Evolution-Support Vector Regression (DE-SVR) algorithm is designed to model the small-signal intrinsic noise behavior of GaN HEMT. It not only overcomes local minimization shortcoming Back Propagation (BP) algorithm, but also uses DE (Differential Evolution) obtain best parameter c (punishment factor) and g (variance kernel function) Support (SVR) algorithm. In order validate superiority DE-SVR experiment compares modeling effects BP SVR in experimental results show that...

The WO3-Ag3PO4 heterojunction on carbon cloth (CC@WO3-Ag3PO4) was prepared through a facile two step approach. photocatalysts are evaluated in the degradation of rhodamine B (RhB) under visible light irradiation; CC@WO3-Ag3PO4 exhibited enhanced photocatalytic RhB efficiency. calculated k for over CC@WO3@Ag3PO4 is 0.03097 min−1, which higher than those reactions CC@Ag3PO4 (0.01254 min−1) and CC@WO3 (0.00935 min−1). efficiency significantly after formation heterojunction. This could be mainly...

The rational design of polysulfide electrocatalysts is vital importance to achieve longevous Li─S batteries. Notwithstanding fruitful advances made in elevating electrocatalytic activity, efforts regulate precatalyst phase evolution and protect active sites are still lacking. Herein, an situ graphene-encapsulated bimetallic model catalyst (CoNi@G) developed for striking a balance between activity stability sulfur electrochemistry. layer numbers directly grown graphene can be dictated by...

An improved GaN high electron mobility transistor (HEMT) small-signal equivalent model is proposed. To the effects of extrinsic parameters that vary with bias voltage, RL and Lds are added to model. Moreover, substrate-related (Rsub Csub) used substrate effect. In order avoid damaging gate under forward pressure conditions, an accurate parasitic parameter extraction method proposed, which improves reliability validity extraction. The S-parameters proposed in good agreement measured range...

Brain tumors are one of the most threatening diseases to human health. Accurate identification type brain tumor is essential for patients and doctors. An automated diagnosis system based on Magnetic Resonance Imaging (MRI) can help doctors identify reduce their workload, so it vital improve performance such systems. Due challenge collecting sufficient data tumors, utilizing pre-trained Convolutional Neural Network (CNN) models classification a feasible approach. The study proposes novel...

Neuromorphic computing that physically mimics human brain, is considered as one of the rebooting frontiers, promising in far-reaching applications like machine learning. Progress research on memristor-based synapses has far exceeded neurons and interconnects. Also essential parts physical implementation brain-inspired computing, neuron circuit cell interconnects may both greatly benefit from high flexibility parallelism 3D heterogeneous integration technologies based TSV (through...

For predicting the effects of gamma radiation on gallium-arsenide (GaAs) heterojunction bipolar transistors (HBTs), a novel model is presented in this paper, considering effects. Based analysis radiation-induced degradation forward base current and cutoff frequency, three semiempirical models to describe variation sensitive parameters are used for simulating within framework simplified vertical inter-company model. Its validity was demonstrated by experimental results GaAs HBTs before after...