13,990 research outputs found
Compact modelling in RF CMOS technology
Full text linkWith the continuous downscaling of complementary metal-oxide-semiconductor (CMOS) technology, the RF performance of metal-oxide-semiconductor field transistors (MOSFETs) has considerably improved over the past years. Today, the standard CMOS technology has become a popular choice for realizing radio frequency (RF) applications. The focus of the thesis is on device compact modelling methodologies in RF CMOS. Compact models oriented to integrated circuit (ICs) computer automatic design (CAD) are the key component of a process design kit (PDK) and the bridge between design houses and foundries. In this work, a novel substrate model is proposed for accurately characterizing the behaviour of RF-MOSFETs with deep n-wells (DNW). A simple test structure is presented to directly access the substrate parasitics from two-port measurements in DNWs. The most important passive device in RFIC design in CMOS is the spiral inductor. A 1-pi model with a novel substrate network is proposed to characterize the broadband loss mechanisms of spiral inductors. Based on the proposed 1-pi model, a physics-originated fully-scalable 2-pi model and model parameter extraction methodology are also presented for spiral inductors in this work. To test and verify the developed active and passive device models and model parameter extraction methods, a series of RF-MOSFETs and planar on-chip spiral inductors with different geometries manufactured by employing standard RF CMOS processes were considered. Excellent agreement between the measured and the simulated results validate the compact models and modelling technologies developed in this work
Synthesis and characterization of spherical amorphous alumo-silicate nanoparticles using RF thermal plasma method
No full textThe RF thermal plasma synthesis route was used for the preparation of alumo-silicate spherical particles. Homogeneous solid mixtures of raw materials (Al2O3, SiO2 and KOH) were used as precursors. SiO2 powders with different particle sizes (6 μm and 40 μm) were taken for this synthesis. For the characterization of obtained materials, scanning electron microscopy (SEM), transmission electron microscopy (TEM), X-ray powder diffraction (XRD) analysis, infrared spectroscopy (FTIR) and thermogravimetric analysis (TG/DTA) were used. Nanosized amorphous ceramic particles were formed via two-step RF thermal plasma processing. It was demonstrated that the size of the SiO2 particles plays a significant role in the formation of the alumo-silicate nanoparticles
A novel design for an RF MEMS resistive switch on PCB substrate
No full textCopyright @ 2008 Stimulation Action on MEM
RF-MEMS switch actuation pulse optimization using Taguchi's method
No full textCopyright @ 2011 Springer-VerlagReliability and longevity comprise two of the most important concerns when designing micro-electro-mechanical-systems (MEMS) switches. Forcing the switch to perform close to its operating limits underlies a trade-off between response bandwidth and fatigue life due to the impact force of the cantilever touching its corresponding contact point. This paper presents for first time an actuation pulse optimization technique based on Taguchi’s optimization method to optimize the shape of the actuation pulse of an ohmic RF-MEMS switch in order to achieve better control and switching conditions. Simulation results show significant reduction in impact velocity (which results in less than 5 times impact force than nominal step pulse conditions) and settling time maintaining good switching speed for the pull down phase and almost elimination of the high bouncing phenomena during the release phase of the switch
RF-Adaboost model’s structure.
No full textAnalyzing customers’ characteristics and giving the early warning of customer churn based on machine learning algorithms, can help enterprises provide targeted marketing strategies and personalized services, and save a lot of operating costs. Data cleaning, oversampling, data standardization and other preprocessing operations are done on 900,000 telecom customer personal characteristics and historical behavior data set based on Python language. Appropriate model parameters were selected to build BPNN (Back Propagation Neural Network). Random Forest (RF) and Adaboost, the two classic ensemble learning models were introduced, and the Adaboost dual-ensemble learning model with RF as the base learner was put forward. The four models and the other four classical machine learning models-decision tree, naive Bayes, K-Nearest Neighbor (KNN), Support Vector Machine (SVM) were utilized respectively to analyze the customer churn data. The results show that the four models have better performance in terms of recall rate, precision rate, F1 score and other indicators, and the RF-Adaboost dual-ensemble model has the best performance. Among them, the recall rates of BPNN, RF, Adaboost and RF-Adaboost dual-ensemble model on positive samples are respectively 79%, 90%, 89%,93%, the precision rates are 97%, 99%, 98%, 99%, and the F1 scores are 87%, 95%, 94%, 96%. The RF-Adaboost dual-ensemble model has the best performance, and the three indicators are 10%, 1%, and 6% higher than the reference. The prediction results of customer churn provide strong data support for telecom companies to adopt appropriate retention strategies for pre-churn customers and reduce customer churn.</div
Wafer-scale 3D integration of silicon-on-insulator RF amplifiers
No full textRF amplifiers are demonstrated using a three- dimensional (3D) wafer-scale integration technology based on silicon-on-insulator (SOI) CMOS process. This new 3D implementation reduces the amplifier size and shortens interconnects for smaller loss and delay. In addition, 3D integration allows the stacking of wafers fabricated using different process technologies to optimize the overall circuit performance at the lowest cost. In RF amplifier examples, MOSFETs and passive components are placed on separate tiers to reduce the size. Measured amplifier performance agrees well with simulation and footprint reduction of approximately 40% comparing to conventional 2D layout can be achieved.United States. Defense Advanced Research Projects Agency (Air Force Contract FA8721-05-C-0002
Sensor de temperatura integrado alimentado por RF
Full text linkDissertação (mestrado) - Universidade Federal de Santa Catarina, Centro Tecnológico, Programa de Pós-Graduação em Engenharia Elétrica, Florianópolis, 2013.Com o aumento do interesse na pesquisa em dispositivos eletrônicosinstalados no corpo humano, que se beneficiam de métodos nãoconvencionaisde captação de energia, o estudo e aprimoramento de taismétodos se torna necessário. Um desses métodos é a transferência de energiapor sinais eletromagnéticos de radiofrequência (RF). Tendo isso em vista,este trabalho apresenta o desenvolvimento de um sensor de temperaturaCMOS alimentado por RF aplicado na medição de temperatura do corpohumano. O sensor recebe energia via um sinal RF emitido por um dispositivoleitor. Uma vez que o sensor armazenou energia suficiente, eleenvia informação sobre a temperatura medida para o leitor. Para executartal função, os seguintes circuitos foram desenvolvidos: retificador, limitadorde tensão, fonte de referência, seletor de modo de operação, regulador detensão, oscilador e dispositivo de modulação de carga. Foi desenvolvidoum sistema que opera com um sinal RF de entrada com potência maior que-10dBm e frequência 900MHz, utilizando a tecnologia de fabricação IBM130nm. O sistema possui consumo de corrente igual a 8,5µA no modo ativoe 4,9µA no modo standby. Além disso, foi implementado um método decalibração do sensor, projetado para obter erro de medição de temperaturamenor que 0,2oC. Nesta dissertação, o projeto e simulação desses blocos sãodetalhados, bem como o teste de alguns blocos que foram fabricados. Abstract : With the increasing interest in research in biomedical electronic devices,which benefits from non-conventional energy transfer and harvestingmethods, the study and development of such methods becomes necessary.One of those methods is the wireless energy transfer. This work presents thedevelopment of a wirelessly powered CMOS temperature sensor, designedto measure temperatures in the human body temperature range. The sensorreceives energy through an RF signal emmited by a reader device. Once thesensor has enough energy, it sends data about the measured temperature tothe reader. The system was designed to operate with signal levels as low as-10dBm centered at 900MHz. The sensor device is formed by the followingcircuits: rectifier, voltage limiter, reference source, operating mode selector,voltage regulator, oscillator and backscattering device. The system presented8.5µA current comsumption in active mode and 4.9µA in standby mode.The developed sensor contains a calibration method, which was designed toachievemaximumtemperaturemeasurement error of 0.2oC. In this work, thedesign and simulation of these circuits are detailed, as well as the test of someblocks that were fabricated
Dynamic changes of touch- and laser heat-evoked field potentials of primary somatosensory cortex in awake and pentobarbital-anesthetized rats.
No full text
A simple and effective process for noise reduction of multichannel cortical field potential recordings in freely moving rats.
No full text- …
