Matching Items (168)
Description
A workload-aware low-power neuromorphic controller for dynamic power and thermal management in VLSI systems is presented. The neuromorphic controller predicts future workload and temperature values based on the past values and CPU performance counters and preemptively regulates supply voltage and frequency. System-level measurements from stateof-the-art commercial microprocessors are used to get workload, temperature and CPU performance counter values. The controller is designed and simulated using circuit-design and synthesis tools. At device-level, on-chip planar inductors suffer from low inductance occupying large chip area. On-chip inductors with integrated magnetic materials are designed, simulated and fabricated to explore performance-efficiency trade offs and explore potential applications such as resonant clocking and on-chip voltage regulation. A system level study is conducted to evaluate the effect of on-chip voltage regulator employing magnetic inductors as the output filter. It is concluded that neuromorphic power controller is beneficial for fine-grained per-core power management in conjunction with on-chip voltage regulators utilizing scaled magnetic inductors.
ContributorsSinha, Saurabh (Author) / Cao, Yu (Thesis advisor) / Bakkaloglu, Bertan (Committee member) / Yu, Hongbin (Committee member) / Christen, Jennifer B. (Committee member) / Arizona State University (Publisher)
Created2011
Description
Negative bias temperature instability (NBTI) and channel hot carrier (CHC) are important reliability issues impacting analog circuit performance and lifetime. Compact reliability models and efficient simulation methods are essential for circuit level reliability prediction. This work proposes a set of compact models of NBTI and CHC effects for analog and mixed-signal circuit, and a direct prediction method which is different from conventional simulation methods. This method is applied in circuit benchmarks and evaluated. This work helps with improving efficiency and accuracy of circuit aging prediction.
ContributorsZheng, Rui (Author) / Cao, Yu (Thesis advisor) / Yu, Hongyu (Committee member) / Bakkaloglu, Bertan (Committee member) / Arizona State University (Publisher)
Created2011
Description
Sensing and controlling current flow is a fundamental requirement for many electronic systems, including power management (DC-DC converters and LDOs), battery chargers, electric vehicles, solenoid positioning, motor control, and power monitoring. Current Shunt Monitor (CSM) systems have various applications for precise current monitoring of those aforementioned applications. CSMs enable current measurement across an external sense resistor (RS) in series to current flow. Two different types of CSMs designed and characterized in this paper. First design used direct current reading method and the other design used indirect current reading method. Proposed CSM systems can sense power supply current ranging from 1mA to 200mA for the direct current reading topology and from 1mA to 500mA for the indirect current reading topology across a typical board Cu-trace resistance of 1 ohm with less than 10 µV input-referred offset, 0.3 µV/°C offset drift and 0.1% accuracy for both topologies. Proposed systems avoid using a costly zero-temperature coefficient (TC) sense resistor that is normally used in typical CSM systems. Instead, both of the designs used existing Cu-trace on the printed circuit board (PCB) in place of the costly resistor. The systems use chopper stabilization at the front-end amplifier signal path to suppress input-referred offset down to less than 10 µV. Switching current-mode (SI) FIR filtering technique is used at the instrumentation amplifier output to filter out the chopping ripple caused by input offset and flicker noise by averaging half of the phase 1 signal and the other half of the phase 2 signal. In addition, residual offset mainly caused by clock feed-through and charge injection of the chopper switches at the chopping frequency and its multiple frequencies notched out by the since response of the SI-FIR filter. A frequency domain Sigma Delta ADC which is used for the indirect current reading type design enables a digital interface to processor applications with minimally added circuitries to build a simple 1st order Sigma Delta ADC. The CSMs are fabricated on a 0.7µm CMOS process with 3 levels of metal, with maximum Vds tolerance of 8V and operates across a common mode range of 0 to 26V for the direct current reading type and of 0 to 30V for the indirect current reading type achieving less than 10nV/sqrtHz of flicker noise at 100 Hz for both approaches. By using a semi-digital SI-FIR filter, residual chopper offset is suppressed down to 0.5mVpp from a baseline of 8mVpp, which is equivalent to 25dB suppression.
ContributorsYeom, Hyunsoo (Author) / Bakkaloglu, Bertan (Thesis advisor) / Kiaei, Sayfe (Committee member) / Ozev, Sule (Committee member) / Yu, Hongyu (Committee member) / Arizona State University (Publisher)
Created2011
Description
Optical receivers have many different uses covering simple infrared receivers, high speed fiber optic communication and light based instrumentation. All of them have an optical receiver that converts photons to current followed by a transimpedance amplifier to convert the current to a useful voltage. Different systems create different requirements for each receiver. High speed digital communication require high throughput with enough sensitivity to keep the bit error rate low. Instrumentation receivers have a lower bandwidth, but higher gain and sensitivity requirements. In this thesis an optical receiver for use in instrumentation in presented. It is an entirely monolithic design with the photodiodes on the same substrate as the CMOS circuitry. This allows for it to be built into a focal-plane array, but it places some restriction on the area. It is also designed for in-situ testing and must be able to cancel any low frequency noise caused by ambient light. The area restrictions prohibit the use of a DC blocking capacitor to reject the low frequency noise. In place a servo loop was wrapped around the system to reject any DC offset. A modified Cherry-Hooper architecture was used for the transimpedance amplifier. This provides the flexibility to create an amplifier with high gain and wide bandwidth that is independent of the input capacitance. The downside is the increased complexity of the design makes stability paramount to the design. Another drawback is the high noise associated with low input impedance that decouples the input capacitance from the bandwidth. This problem is compounded by the servo loop feed which leaves the output noise of some amplifiers directly referred to the input. An in depth analysis of each circuit block's noise contribution is presented.
ContributorsLaFevre, Kyle (Author) / Bakkaloglu, Bertan (Thesis advisor) / Barnaby, Hugh (Committee member) / Vermeire, Bert (Committee member) / Arizona State University (Publisher)
Created2011
Description
Process variations have become increasingly important for scaled technologies starting at 45nm. The increased variations are primarily due to random dopant fluctuations, line-edge roughness and oxide thickness fluctuation. These variations greatly impact all aspects of circuit performance and pose a grand challenge to future robust IC design. To improve robustness, efficient methodology is required that considers effect of variations in the design flow. Analyzing timing variability of complex circuits with HSPICE simulations is very time consuming. This thesis proposes an analytical model to predict variability in CMOS circuits that is quick and accurate. There are several analytical models to estimate nominal delay performance but very little work has been done to accurately model delay variability. The proposed model is comprehensive and estimates nominal delay and variability as a function of transistor width, load capacitance and transition time. First, models are developed for library gates and the accuracy of the models is verified with HSPICE simulations for 45nm and 32nm technology nodes. The difference between predicted and simulated σ/μ for the library gates is less than 1%. Next, the accuracy of the model for nominal delay is verified for larger circuits including ISCAS'85 benchmark circuits. The model predicted results are within 4% error of HSPICE simulated results and take a small fraction of the time, for 45nm technology. Delay variability is analyzed for various paths and it is observed that non-critical paths can become critical because of Vth variation. Variability on shortest paths show that rate of hold violations increase enormously with increasing Vth variation.
ContributorsGummalla, Samatha (Author) / Chakrabarti, Chaitali (Thesis advisor) / Cao, Yu (Thesis advisor) / Bakkaloglu, Bertan (Committee member) / Arizona State University (Publisher)
Created2011
Description
One of the challenges in future semiconductor device design is excessive rise of power dissipation and device temperatures. With the introduction of new geometrically confined device structures like SOI, FinFET, nanowires and continuous incorporation of new materials with poor thermal conductivities in the device active region, the device thermal problem is expected to become more challenging in coming years. This work examines the degradation in the ON-current due to self-heating effects in 10 nm channel length silicon nanowire transistors. As part of this dissertation, a 3D electrothermal device simulator is developed that self-consistently solves electron Boltzmann transport equation with 3D energy balance equations for both the acoustic and the optical phonons. This device simulator predicts temperature variations and other physical and electrical parameters across the device for different bias and boundary conditions. The simulation results show insignificant current degradation for nanowire self-heating because of pronounced velocity overshoot effect. In addition, this work explores the role of various placement of the source and drain contacts on the magnitude of self-heating effect in nanowire transistors. This work also investigates the simultaneous influence of self-heating and random charge effects on the magnitude of the ON current for both positively and negatively charged single charges. This research suggests that the self-heating effects affect the ON-current in two ways: (1) by lowering the barrier at the source end of the channel, thus allowing more carriers to go through, and (2) via the screening effect of the Coulomb potential. To examine the effect of temperature dependent thermal conductivity of thin silicon films in nanowire transistors, Selberherr's thermal conductivity model is used in the device simulator. The simulations results show larger current degradation because of self-heating due to decreased thermal conductivity . Crystallographic direction dependent thermal conductivity is also included in the device simulations. Larger degradation is observed in the current along the [100] direction when compared to the [110] direction which is in agreement with the values for the thermal conductivity tensor provided by Zlatan Aksamija.
ContributorsHossain, Arif (Author) / Vasileska, Dragica (Thesis advisor) / Ahmed, Shaikh (Committee member) / Bakkaloglu, Bertan (Committee member) / Goodnick, Stephen (Committee member) / Arizona State University (Publisher)
Created2011
Description
A dual-channel directional digital hearing aid (DHA) front-end using a fully differential difference amplifier (FDDA) based Microphone interface circuit (MIC) for a capacitive Micro Electro Mechanical Systems (MEMS) microphones and an adaptive-power analog font end (AFE) is presented. The Microphone interface circuit based on FDDA converts the capacitance variations into voltage signal, achieves a noise of 32 dB SPL (sound pressure level) and an SNR of 72 dB, additionally it also performs single to differential conversion allowing for fully differential analog signal chain. The analog front-end consists of 40dB VGA and a power scalable continuous time sigma delta ADC, with 68dB SNR dissipating 67u¬W from a 1.2V supply. The ADC implements a self calibrating feedback DAC, for calibrating the 2nd order non-linearity. The VGA and power scalable ADC is fabricated on 0.25 um CMOS TSMC process. The dual channels of the DHA are precisely matched and achieve about 0.5dB gain mismatch, resulting in greater than 5dB directivity index. This will enable a highly integrated and low power DHA
ContributorsNaqvi, Syed Roomi (Author) / Kiaei, Sayfe (Thesis advisor) / Bakkaloglu, Bertan (Committee member) / Chae, Junseok (Committee member) / Barnby, Hugh (Committee member) / Aberle, James T., 1961- (Committee member) / Arizona State University (Publisher)
Created2011
Description
Phase locked loops are an integral part of any electronic system that requires a clock signal and find use in a broad range of applications such as clock and data recovery circuits for high speed serial I/O and frequency synthesizers for RF transceivers and ADCs. Traditionally, PLLs have been primarily analog in nature and since the development of the charge pump PLL, they have almost exclusively been analog. Recently, however, much research has been focused on ADPLLs because of their scalability, flexibility and higher noise immunity. This research investigates some of the latest all-digital PLL architectures and discusses the qualities and tradeoffs of each. A highly flexible and scalable all-digital PLL based frequency synthesizer is implemented in 180 nm CMOS process. This implementation makes use of a binary phase detector, also commonly called a bang-bang phase detector, which has potential of use in high-speed, sub-micron processes due to the simplicity of the phase detector which can be implemented with a simple D flip flop. Due to the nonlinearity introduced by the phase detector, there are certain performance limitations. This architecture incorporates a separate frequency control loop which can alleviate some of these limitations, such as lock range and acquisition time.
ContributorsZazzera, Joshua (Author) / Bakkaloglu, Bertan (Thesis advisor) / Song, Hongjiang (Committee member) / Ozev, Sule (Committee member) / Arizona State University (Publisher)
Created2012
Description
The non-quasi-static (NQS) description of device behavior is useful in fast switching and high frequency circuit applications. Hence, it is necessary to develop a fast and accurate compact NQS model for both large-signal and small-signal simulations. A new relaxation-time-approximation based NQS MOSFET model, consistent between transient and small-signal simulations, has been developed for surface-potential-based MOSFET compact models. The new model is valid for all regions of operation and is compatible with, and at low frequencies recovers, the quasi-static (QS) description of the MOSFET. The model is implemented in two widely used circuit simulators and tested for speed and convergence. It is verified by comparison with technology computer aided design (TCAD) simulations and experimental data, and by application of a recently developed benchmark test for NQS MOSFET models. In addition, a new and simple technique to characterize NQS and gate resistance, Rgate, MOS model parameters from measured data has been presented. In the process of experimental model verification, the effects of bulk resistance on MOSFET characteristics is investigated both theoretically and experimentally to separate it from the NQS effects.
ContributorsZhu, Zeqin (Author) / Gildenblat, Gennady (Thesis advisor) / Bakkaloglu, Bertan (Committee member) / Barnaby, Hugh (Committee member) / Mcandrew, Colin C (Committee member) / Arizona State University (Publisher)
Created2012
Description
Three-dimensional (3D) inductors with square, hexagonal and octagonal geometries have been designed and simulated in ANSYS HFSS. The inductors have been designed on Silicon substrate with through-hole via with different width, spacing and thickness. Spice modeling has been done in Agilent ADS and comparison has been made with results of custom excel based calculator and HFSS simulation results. Single ended quality factor was measured as 12.97 and differential ended quality factor was measured as 15.96 at a maximum operational frequency of 3.65GHz. The single ended and differential inductance was measured as 2.98nH and 2.88nH respectively at this frequency. Based on results a symmetric octagonal inductor design has been recommended to be used for application in RF biosensing. A system design has been proposed based on use of this inductor and principle of inductive sensing using magnetic labeling.
ContributorsAbbey, Hemanshu (Author) / Bakkaloglu, Bertan (Thesis advisor) / Kiaei, Sayfe (Committee member) / Goryll, Michael (Committee member) / Arizona State University (Publisher)
Created2012