نشریه صنایع الکترونیک
سال دوازدهم شماره 2 (پیاپی 46، تابستان 1400)

This paper presents a low-power and fast voltage level shifter that can convert sub-threshold logic levels to the conventional level of target technology. The proposed design performs fast and low-power operations due to the improved stacked-based cross-coupled structure. In order to reduce the power consumption caused by the short-circuit current at the output driver stage, the output inverter is modified to have phase-changed inputs on PMOS and NMOS transistor gates. Additional diode-connected transistors in the feedback loop are used to reduce the voltage swings in the feedback loop, and to make a difference between the on and off time of the output transistors. Post-layout simulation results on standard 180 nm CMOS technology show that the proposed level shifter can convert 0.4V to 1.8V at a frequency of 1MHz with a power consumption of 89.23nW and a propagation delay of 23.68ns correctly. The proposed design is suitable for Internet-of-Things (IoT) sensor tags due its low-power and converted voltage levels with emphasis on the boundary of low-voltage low-power and normal-voltage high-performance sub-units.

This paper presents a new estimator for the speech enhancement using codebook. Codebook-based speech enhancement method separates the noise and speech from each other and synthesizes the enhanced speech signal by optimally selecting the speech codebook indexes. This method can enhance the noisy speech with signal to noise ratio of less than zero decibel. In this method it is very important to select the correct codebook indexes. Therefore, in this paper, the maximum likelihood estimator is proposed for speech and noise by applying auditory quality-enhancing weights. The relation of this estimator is also used as a distance function in the design of codebooks. This method is simulated for different speakers and noises. The results show the proposed maximum likelihood estimator leads to better speech enhancement than the euclidean distance estimator. The proposed method is also more successful in dealing with non-stationary or stationary noises and negative or positive SNRs than other methods. The cost of the superior quality enhancement in this method is the requirement to a relatively time-consuming signal processing.

In this research, a fast responsive and highly sensitive hetero structure gas sensor is proposed. To fabricate Au/CNT/Ti heterojunction Schottky structure, vertically aligned carbon nanotubes (VACNTs) were deposited on silicon oxide substrate between two metal electrodes using plasma-enhanced chemical vapor deposition method. To study the surface characteristics of the structure, Scanning Electron Microscopy analysis was performed to show the density and well-aligned growth of CNTs. The Titanium and gold metals were used as the electrodes to form schottky and ohmic contacts, respectively. The fabricated sample was exposed to 500, 1000 and 1500 ppm concentrations of N2 and NO2 gases. The response and recovery times of 85 and 154 sec for NO2 and 97 and 190 sec for N2 were measured, respectively. On this basis compare to similar structures, the response and recovery times of the fabricated sensor are shorter. It is observed that the sensitivity of 13.5% and 33.5% occurs for 1500 ppm of NO2 and N2, respectively, which in comparison with the sensistivity to CO and H2S gases, proves that the fabricated structure is suitable for Nitrogen detection applications.

Forecasting the consumed electrical load is critical for reliable operation of power systems as well as for management of planning and load storage. In this paper, short-term load forecasting is performed by extracting the characteristics of the load history using deep neural networks. Recurrent neural networks, especially improved recurrent neural networks such as LSTM and GRU, are able to retain short-term and long-term memory to extract the relationships between load values from the time series. On the other hand, convolutional neural networks are able to automatically learn features and can directly generate a vector for prediction. The proposed method is to extract the load characteristics using convolutional neural networks and then extract the load sequence information using the GRU network. The results of experiments on three data sets, Toronto, ISO-NE, and North American Utility, show a decrease in the forecasting error of the proposed method compared to other competitor methods.

The DC microgrids in comparison with AC microgrids offer high efficiency, high reliability and also easy connection to most of energy resources. In DC microgrids, equal load sharing between resources and also regulation of DC bus voltage are important issues. Two control methods including droop control and centralized control have been introduced to handle these issues. Although the centralized controllers have high accuracy for microgrid control, this control method needs a high bandwidth communication link between the converters, hence the reliability of this method is low. For increasing performance of the droop control method, this paper introduces a new approach based on fuzzy logic control. In the proposed method, parameters of voltage restoration and current average controllers are determined by a fuzzy controller which causes the microgrid be properly controlled in different load conditions and line impedances. The proposed method has be simulated in MATLAB/Simulink environment. The simulation results show that the proposed controller causes enhanced load sharing accuracy and voltage restoration

This paper presents an 8-bit analog-to-digital converter (ADC) designed using an improved successive approximation Register(SAR). The proposed structure of the SAR converter uses an integrator embedded with a digital-to-analog converter (DAC). The presence of an integrator at the input of the ADC eliminates the need for a sample and hold circuit and is used to generate different voltage levels in the DAC. Thus, the successive approximation algorithm in the proposed structure has been modified to provide the required voltage levels with a few number of capacitors in the DAC. Therefore, the complexity of the circuit is reduced and less silicon is occupied due to a reduced capacitive array in DAC. The proposed ADC completes the analog input to digital code conversion in 10 clock pulses. In order to study the proposed method, the SAR ADC is designed and simulated at the transistor level in 0.18 μm CMOS technology at 1.8 V supply voltage. The simulation results show that the ratio of signal to noise and distortion (SNDR) for input bandwidth of 640 kHz is 48.3 dB, effective number of bit (ENOB) and power consumption are 48.3 dB, 7.74 bits and 0.85 mW, respectively.

In the structure of 3D ICs, several layers of components are placed on top of each other. The connections between these layers is through vertical layers of interconnections. One of the major challenges that can threaten the reliability of a grid-based system on a three-dimensional chip is crosstalk Fault. Crosstalk fault in 3D ICs, is depending on the ranking of their transition patterns that greatly challenges the reliability, efficiency and Energy of the 3D ICs. The purpose of this paper is to present and investigate coding, called disturbance coding. By reducing the most delayed pattern, this coding mechanism reduces crosstalk in 3D ICs with the least possible overhead and power consumption. The basic idea is that first the data to be transferred from layer x to layer y is placed in the intermediates in the layer x solutions, then the counter unit measures the amount and counts of 4C class transitions and gives a signal to the unit. The crossbar will specify the final arrangement of the data to be sent to the y layer. Simultaneously with sending this control signal, another signal will be sent through this unit to the decoder in the receiver to retrieve the original data by this section. The simulation results show the amount of overhead compared to other methods.

Making distance between the antenna and the transmitter-receiver, without disrupting the performance of wave propagation systems, is a key requirement for many exploiters, including military, medical, and information and communication organizations.In addition to all the studies done to achieve this important goal, the use of electro-optomechanical systems with special design and measures can be considered as a basic strategy in this field, but their use requires that the efficiency of the system is not compromised in different environmental conditions. Two environmental factors affecting efficiency are the effect of changes in air pressure and temperature. In this paper, the effect of these two key factors, as a case study on the efficiency of an electro-optomechanical system, has been studied. The results show that the effect of changes in air pressure on the efficiency of the system under study is very significant and the effect of temperature changes depends on the dielectric type of the system capacitor, which must be considered in their design and construction.