In recent years, oscillators are considered as inevitable blocks in many electronic systems. They are commonly used in digital circuits to provide clocking and in analog/RF circuits of communication transceivers to support frequency conversion. Nowadays, CMOS technology is the most applicable solution for VLSI and especially for modern integrated circuits used in wireless communications. The main purpose of this project is to design a high performance voltage-controlled oscillator (LC VCO) using 65nm CMOS technology. To meet the state-of-the-art requirements, several circuit solutions have been explored and the design work ended-up with a Quadrature VCO. The circuit operates at center frequency of 2.4 GHz. The phase noise of QVCO obtained by simulation is -140 dBc/Hz at 1MHz offset frequency which is 6 dB less compared to conventional LC VCOs. The power consumption is 3.6mW and the tuning voltage can be swept from 0.2 V to 1.2 V resulting in 2.25 GHz - 2.55 GHz frequency range.

With increased complexity of the contemporary very large integrated circuits the need for onchip test addressing not only the digital but also analog and mixed-signal RF blocks has emerged. The standard production test has become more costly and the instrumentation is pushed to its limits by the leading edge integrated circuit technologies. Also the chip performance for high frequency operation and the area overhead appear a hindrance in terms of the test access points needed for the instrumentation-based test. To overcome these problems, test implemented on a chip can be used by sharing the available resources such as digital signal processing (DSP) and A/D, D/A converters to constitute a built-in-self-test. In this case, the DSP can serve both as a stimuli generator and response analyzer.

Arbitrary test signals can be achieved using DSP. Specifically, the ΣΔ modulation technique implemented in software is useful to encode a single- or two-tone stimulus as a onebit sequence to generate a spectrally pure signal with a high dynamic range. The sequence can be stored in a cyclic memory on a chip and applied to the circuit under test using a buffer and a simple reconstruction filter. In this way ADC dynamic test for harmonic and intermodulation distortion is carried out in a simple setup. The FFT artifacts are avoided by careful frequency planning for low-pass and band-pass ΣΔ encoding technique. A noise shaping based on a combination of low- and band-pass ΣΔ modulation is also useful providing a high dynamic range for measurements at high frequencies that is a new approach. However, a possible asymmetry between rise and fall time due to CMOS process variations in the driving buffer results in nonlinear distortion and increased noise at low frequencies. A simple iterative predistortion technique is used to reduce the low frequency distortion components by making use of an on-chip DC calibrated ADC that is another contribution of the author.

Some tests, however, like the two-tone RF test that targets linearity performance of a radio receiver, require test stimuli based on a dedicated hardware. For the measurement of the thirdor second-intercept point (IP3/IP2) a spectrally clean stimulus is essential. Specifically, the second- or third-order harmonic or intermodulation products of the stimulus generator should be avoided as they can obscure the test measurement. A challenge in this design is the phase noise performance and spurious tones of the oscillators, and also the distortion-free addition of the two tones. The mutual pulling effect can be minimized by layout isolation techniques.

A new two-tone RF generator based on a specialized phase-locked loop (PLL) architecture is presented as a viable solution for IP3/IP2 on-chip test. The PLL provides control over the frequency spacing of two voltage controlled oscillators. For the two-tone stimulus a highly linear analog  adder is designed to limit distortion which could obscure the IP3 test. A specialized feedback circuit in the PLL is proposed to overcome interference by the reference spurs. The circuit is designed using 65 nm CMOS process. By using a fine spectral resolution the observed noise floor can be reduced to enable the measurement of second- or third-order intermodulation product tones. This also reflects a tradeoff between the test time and the test performance. While the test time to collect the required number of samples can be of milliseconds the number of samples need not be excessive, since the measurements are carried out at the receiver baseband, where the required sampling frequency is relatively low.

This paper addresses a built-in-self-test (BiST) to characterize IP3 linearity of a RF receiver front-end. A two-tone stimulus is generated by a phase-lock loop (PLL) in GHz frequency range. The PLL is designed to keep the frequency difference between the two tones under control and in this way to avoid a possible injection-locking. One of the oscillation frequencies and the difference (beat) frequency can be externally controlled. According to the test requirements the phase noise and nonlinear distortion of the two-tone generator are considered as a merit for the VCO and analog adder design. A highly linear analog adder with output referred IP3 of more than +15 dBm is used to generate the RF stimulus. The two-tone power across 50 Ω receiver input impedance can be more than -25 dBm with very low intermodulation distortion of P_IM3 = -75 dBc. The receiver performance is not affected significantly by the test set-up. Simulations for linearity and noise performance of the PLL designed in 65nm CMOS show sufficient potential for on-chip IP3 measurements in the GHz frequency range.

This paper investigates the feasibility of pulse width modulation technique (PWM) for dynamic test of ADCs used for high speed applications. The requirements and limitations of digital PWM signal to noise ratio (SNR) are discussed in terms of pulse-width resolution corresponding to the choice of the carrier- and clock frequency of a pulse-width generator. The PWM SNR response is measured by FFT using coherent sampling for different PWM resolution. Low-pas filtering removing high frequency PWM components is introduced as well to improve PWM SNR and prevent intermodulation effects, which tend to hamper the harmonic distortion test (HD). As an example a 4-bit first-order SigmaDelta ADC under dynamic test is simulated and the requirements for PWM resolution with respect to SNR and HD measurements are identified.

This work presents a cancellation technique of non-linear distortion components of one-bit digital stimulus sequence which is generated in software by a ΣΔ modulator. The stimulus is stored in a cyclic memory and applied to a circuit under test through a driving buffer and a simple lowpass reconstruction filter. The distortion components originate from buffer imperfections which result in a possible asymmetry between rising and falling edges of a NRTZ waveform representing the encoded stimulus. We show that the distortion components can be cancelled by using a simple predistortion technique. In addition an on-chip DC-calibrated ADC can be used to identify the second-order nonlinear products of the driving buffer. This procedure allows for cancellation of all the second-order distortions before the actual test and it can be extended to the third order terms as well.

Application of the ΣΔ modulation technique to the on-chip spectral test for high-speed A/D converters is presented. The harmonic HD2/HD3 and intermodulation IM2/IM3 test is obtained with one-bit ΣΔ sequence stored in a cyclic memory or generated on line, and applied to an ADC under test through a driving buffer and a simple reconstruction filter. To achieve a dynamic range (DR) suitable for high-performance spectral measurements a frequency plan is used taking into account the type of ΣΔ modulation (low-pass and band-pass) including the FFT processing gain. Higher order modulation schemes are avoided to manage the ΣΔ quantization noise without resorting to a more complicated filter. For spectral measurements up to the Nyquist frequency, we propose a dedicated low-pass/band-pass ΣΔ modulation scheme that limits spreading of the low-frequency quantization noise by ADC under test that tends to obstruct the test measurements at high frequencies. Correction technique for NRTZ encoding suitable for ADCs with very high clock frequencies is put in perspective. The presented technique is illustrated by simulation examples of a Nyquist-rate ADC under test.

This paper presents application of the ΣΔ modulation technique to the on-chip dynamic test for A/D converters. The wanted stimulus such as a single- or two-tone signal is encoded into one-bit ΣΔ sequence, which after simple low-pass filtering is applied to the circuit under test with low noise and without distortion. In this way a large dynamic range is achieved making the performance harmonic- and intermodulation dynamic test viable. By a systematic approach we select the order and type of a ΣΔ modulator, and develop the frequency plan suitable for spectral measurements on a chip. The technique is illustrated by simulation of a practical ADC under test.

Most of today’s microwave circuits are designed for specific function and special need. There is a growing trend to have flexible and reconfigurable circuits. Circuits that can be digitally programmed to achieve various functions based on specific needs. Realization of high frequency circuit blocks that can be dynamically reconfigured to achieve the desired performance seems to be challenging. However, with recent advances in many areas of technology these demands can now be met.

Two concepts have been investigated in this thesis. The initial part presents the feasibility of a flexible and programmable circuit (PROMFA) that can be utilized for multifunctional systems operating at microwave frequencies. Design details and PROMFA implementation is presented. This concept is based on an array of generic cells, which consists of a matrix of analog building blocks that can be dynamically reconfigured. Either each matrix element can be programmed independently or several elements can be programmed collectively to achieve a specific function. The PROMFA circuit can therefore realize more complex functions, such as filters or oscillators. Realization of a flexible RF circuit based on generic cells is a new concept. In order to validate the idea, two test chips have been fabricated. The first chip implementation was carried out in a 0.2μm GaAs process, ED02AH from OMMIC^TM. The second chip was implemented in a standard 90nm CMOS process. Simulated and measured results are presented along with some key applications such as low noise amplifier, tunable band pass filter and a tunable oscillator.

The later part of the thesis covers the design and implementation of broadband RF front-ends that can be utilized for multistandard terminals such as software defined radio (SDR). The concept of low gain, highly linear frontends has been presented. For proof of concept two test chips have been implemented in 90nm CMOS technology process. Simulated and measurement results are presented. These RF front-end implementations utilize wideband designs with active and passive mixer configurations.

We have also investigated narrowband tunable LNAs. A dual band tunable LNA MMIC has been fabricated in 0.2μm GaAs process. A self tuning technique has been proposed for the optimization of this LNA.

This paper presents a self-tuning technique for optimization of a dual band LNAthat can be used in a flexible RF front-end suitable for IEEE 802.11a/b/g WLANapplications. With this tuning technique the LNA can perform self-calibrationfor the optimal performance. A possible shift in resonance frequency due toprocess and temperature variations can be compensated by this method. Theproposed self-tuning technique is implemented by using a simple RF detector atthe LNA output. Based on the DC value provided by this detector the LNA istuned for a maximum gain through the tuning loop, which incorporates ADC,digital base-band and DAC. We show that the tuning error can be within halfLSB of ADC provided the DAC and ADC resolutions are constraint by aspecified condition. For 4-bit case this value corresponds to a gain error of0.4 dB. The LNA has been implemented in 0.2μm GaAs process offered byOMMIC^TM. In measurements the LNA achieves a gain of 15.1 dB and 21.6 dBin the upper and lower band, respectively, with corresponding NF of 3.8 dB and2.8 dB. In the lower band the measured IIP3 is -3 dBm and 1dB_CP is -8 dBm.

This paper presents a dual band LNA that can be switched between two bands (2.4 GHz & 5.2 GHz) for IEEE 802.1 la/b/g WLAN applications. The LNA is also tunable within each band and the tuning is incorporated by on-chip varactors. The test chip consists of two fully integrated narrow-band tunable LNAs along with SPDT switch. For power saving one LNA can be switched off. The technology process is 0.2 mum GaAs offered by OMMIC. The LNA can achieve a relatively good performance over the two bands as demonstrated by simulation. With a 3V supply, the LNA has a gain of 26.2 dB at 2.4 GHz and 21.8 dB at 5.2 GHz and the corresponding NF varies between 2.07 dB and 1.84 dB, respectively. The LNA has an IIP3 of -7 dBm at 2.4 GHz and -1.6 dBm at 5.2 GHz.

This paper describes the use of programmable microwave function array (PROMFA) for different microwave application. The PROMFA concept is based on an array of generic cells, in which a number of different functions can be realized. Each PROMFA cell is a four-port circuit, that can either be programmed independently or collectively according to a specific need. Specifically, the phase shift capability in a single PROMFA cell, useful for a new type of phase shifter design is discussed. The paper also presents the functionality of this new architecture as a beamforming network. As an example case an active corporate feed network and a tunable recursive filter is demonstrated. Simulated and measured results are presented.

This paper presents a design approach for flexible RF circuits using Programmable Microwave Function Array (PROMFA) cells. The concept is based on an array of generic cells that can be dynamically reconfigured. Therefore, the same circuit can be used for various functions e.g. amplifier, tunable filter and tunable oscillator. For proof of concept a test chip has been implemented in 90nm CMOS process. The chip measurement results indicate that a single unit cell amplifier has a typical gain of 4dB with noise figure of 2.65dB at 1.5GHz. The measured input referred 1dB compression point is -8dBm with an IIP3 of +1.1dBm at 1GHz. In a single unit cell oscillator configuration, the oscillator can achieve a wide tuning range of 600MHz to 1.8GHz. The measured phase noise is -94dBc/Hz at an offset frequency of 1MHz for the oscillation frequency of 1.2GHz. A single unit cell oscillator consumes 18mW at 1.2GHz while providing -8dBm power into 50Ω load. In a single unit cell filter configuration, the tunable band pass filter can achieve a reasonable tuning range of 600MHz to 1.2GHz with a typical power consumption of 13mW at 1GHz. A single unit cell has a total chip area of 0.091mm2 including the coupling capacitors.

This paper presents design considerations for low power, highly linear currentmode LNAs that can be used for wideband RF front-ends for multistandardapplications. The circuit level simulations of the proposed architecture indicatethat with optimal biasing a high value of IIP3 can be obtained. A comparison ofthree scenarios for optimal bias is presented. Simulation results indicate thatwith the proposed architecture, LNAs may achieve a maximum NF of 3.6 dBwith a 3 dB bandwidth larger than 10 GHz and a best case IIP3 of +17.6 dBmwith 6.3 mW power consumption. The LNAs have a broadband input match of 50Ω. The process is 90nm CMOS and with 1.1V supply the LNAs powerconsumption varies between 6.3 mW and 2.3 mW for the best and the worst caseIIP3, respectively.

This paper presents the design and implementation of a low power, highly linear, wideband RF front-end in 90nm CMOS. The architecture consists of an inverter-like common gate low noise amplifier followed by a passive ring mixer. The proposed architecture achieves high linearity in a wide band (0.5-6GHz) at very low power. Therefore, it is a suitable choice for software defined radio (SDR) receivers. The chip measurement results indicate that the inverter-like common gate input stage has a broadband input match achieving S11 below -8.8dB up to 6GHz. The measured single sideband noise figure at an LO frequency of 2GHz and an IF of 10MHz is 6.25dB. The front-end achieves a voltage conversion gain of 4.5dB at 1GHz with 3dB bandwidth of more than 6GHz. The measured input referred 1dB compression point is +1.5dBm while the IIP3 is +11.73dBm and the IIP2 is +26.23dBm respectively at an LO frequency of 2GHz. The RF front-end consumes 6.2mW from a 1.1V supply with an active chip area of 0.0856mm².

This thesis addresses the potential of implementing watt-level class-AB Power Amplifier (PA) for WLAN in 65 nm CMOS technology, at 2.4 GHz frequency. In total, five PAs have been compared, where the examined parameters were output power (P_out), linearity, power added efficiency (PAE), and area consumption. Four PAs were implemented using conventional cascode topology with different combination of transistors sizes in 65nm CMOS, and one PA using a high-voltage Extended Drain MOS (EDMOS) device, implemented in the same 65 nm CMOS with no process or mask changes. All schematics were created using Cadence Virtuoso CAD tools. The test benches were created using the Agilent's Advance Design System ( ADS) and simulated with the ADS-Cadence dynamic link.

The simulation results show that the EDMOS PA (L=350 nm) has the smallest area, but has harder to reach the required P_out. Cascode no. 3 (L= 500,260 nm) has the best P_out (29.1 dBm) and PAE (49.5 %). Cascode no. 2 (L= 500,350 nm) has the best linearity (low EVM). Cascode no. 1 (L=500,500 nm) has low P_out (27.7 dBm). Cascode no.4 (L=500,60 nm) has very bad linearity.

The thesis also gives an overview for CMOS technology, discusses the most important aspects in RF PAs design, such as P_out, PAE, gain, and matching networks. Different PA classes are also discussed in this thesis.

A domino logic circuit contained within an integrated circuit includes a dynamic logic circuit and an intermediate logic circuit. The intermediate logic circuit includes a pull-up transistor having a source terminal coupled to a source voltage line and an n-block transistor having a source terminal connected to a low ground voltage lin

A domino logic circuit contained within an integrated circuit includes a dynamic logic circuit and an intermediate logic circuit. The intermediate logic circuit includes a pull-up transistor having a source terminal coupled to a source voltage line and an n-block transistor having a source terminal connected to a low ground voltage line.

n/a

A novel circuit technique for reducing leakage currents through the read-path of large register files in which a negative gate-source voltage is forced on a critical pass transistor between a cell read transistor and a local bitline such that when the cell is in a first state, the leakage current from a dynamic node of the cell read transistor is reduced. The reduced leakage current increases the robustness and performance of the read operation.

A method for controlling a local clock includes receiving a reference clock and generating a phase-shifted version of the reference clock. The two clocks are synchronized using a closed-loop method that produces a control signal. The control signal is smoothed during the closed-loop method and the smoothed signal is then used, instead of the control signal, in generating the phase-shifted clock

Increasing leakage currents combined with reduced noise margins are seriously degrading the robustness of dynamic circuits. This paper describes a dynamic implementation of a 256X32b 4-read/write-port Register-File for ~6GHz operation at 1.2V in a 0.13 utilize an efficient conditional keeper-technique, where a large fraction of the keeper is turned remains are able to improve upon all-low-Vt performance by 4%, while maintaining Dual-Vt usage. Thus, the robustness is improved by 96% and the active leakage power is reduced by 5X.

A dynamic circuit with a conditional keeper for burn-in. In the described embodiments, a conditional keeper is provided which is active only during the burn-in test, where the conditional keeper is sized larger than the standard keepers so as to compensate for additional leakage currents in the dynamic circuit.

A static receiver having a first inversion threshold for received signals undergoing a HIGH-to-LOW transition, and a second inversion threshold for received signals undergoing a LOW-to-HIGH transition, where the first inversion threshold is greater than the second inversion threshold. One embodiment comprises a static receiver, a pFET, and a nFET, where when a HIGH-to-LOW transition is being received at the receiver's input port, the pFET is coupled to the input port so as to contribute to raising the inversion threshold, and when a LOW-to-HIGH transition is being received at the input port, the nFET is coupled to the input port so as to contribute to lowering the inversion threshold. Other embodiments are described and claimed.

A voltage-level converter and a method of converting a first logic voltage level to a second logic voltage level are described. In one embodiment, a voltage-level converter connects a first logic unit connected to a first supply voltage to a second logic unit connected to a second supply voltage. The voltage-level converter includes at least one transistor connected to the second supply voltage. The at least one transistor has a threshold voltage whose absolute value is greater-than-or-about-equal to the absolute value of the difference between the second supply voltage and the first supply voltage. In an alternative embodiment, a method for converting a first logic voltage level to a second logic voltage level includes transmitting a logic signal from a logic unit having an output voltage swing of between a first voltage level and a second voltage level, receiving the logic signal at a logic circuit having a pull-up transistor and an output voltage swing between a third voltage level and a fourth voltage level, and turning off the pull-up transistor when the logic signal has a value slightly greater than the difference between the third voltage level and the first voltage level.

Embodiments of the present invention relate to memory circuits with heavily loaded bit-lines, and where either the effect of leakage current in the read access or pass transistors is reduced, or leakage current is reduced.

Embodiments of the present invention relate to memory circuits with heavily loaded bit-lines, and where either the effect of leakage current in the read access or pass transistors is reduced, or leakage current is reduced.

Embodiments of the present invention relate to memory circuits with heavily loaded bit-lines, and where either the effect of leakage current in the read access or pass transistors is reduced, or leakage current is reduced.

Embodiments of the present invention relate to memory circuits with heavily loaded bit-lines, and where either the effect of leakage current in the read access or pass transistors is reduced, or leakage current is reduced.

Embodiments of the present invention relate to memory circuits with heavily loaded bit-lines, and where either the effect of leakage current in the read access or pass transistors is reduced, or leakage current is reduced.

Increasing leakage currents combined with reduced noise margins significantly degrade the robustness of wide dynamic circuits. In this paper, we describe two conditional keeper topologies for improving the robustness of sub-130-nm wide dynamic circuits. They are applicable in normal mode of operation as well as during burn-in test. A large fraction of the keepers is activated conditionally, allowing the use of strong keepers with leaky precharged circuits without significant impact on performance of the circuits. Compared to conventional techniques, up to 28% higher performance has been observed for wide dynamic gates in a 130-nm technology. In addition, the proposed burn-in keeper results in 64% active area reduction

A dual-rail static logic gate with a self cut-off mechanism is disclosed. In an embodiment, the output of the first rail is coupled to the input of the pull-up device of the second rail and vice versa. The cross-coupling allows the self cut-off mechanism of the static gate to function properly and provides for components which have lower capacitance than conventional static gates. The lower capacitance results in a faster static gate.

A dual-rail static logic gate with a self cut-off mechanism is disclosed. In an embodiment, the output of the first rail is coupled to the input of the pull-up device of the second rail and vice versa. The cross-coupling allows the self cut-off mechanism of the static gate to function properly and provides for components which have lower capacitance than conventional static gates. The lower capacitance results in a faster static gate.

In this paper, interconnection length estimation is discussed and a general, simple, fast and efficient estimation technique is proposed. In contrast to traditional average length estimation techniques, such as the one based on Rent's rule, the new technique utilizes the topological information of the actual netlist and estimates the length of each interconnection separately. The result of the estimation can be directly used to assign a reasonable R and C to each interconnect, including long and wide buses. Consequently, the new technique enhances the accuracy of power and delay estimations at higher design levels of abstraction

In this paper, we present a new technique which indirectly separates and extracts the total short-circuit power consumption of digital CMOS circuits. We avoid a direct encounter with the complex behavior of the short-circuit currents. Instead, we separate the dynamic power consumption from the total power and extract the total short-circuit power. The technique is based on two facts: first, the short-circuit power consumption disappears at a Vdd close to VT and, secondly, the total capacitance depends on supply voltage in a sufficiently weak way in standard CMOS circuits. Hence, the total effective capacitance can be estimated at a low Vdd. To avoid reducing Vdd below the specified forbidden level, a polynomial is used to estimate the power versus supply voltage down to VT based on a small voltage sweep over the allowed supply voltage levels. The result shows good accuracy for the short-circuit current ranges of interest.

A sense amplifier for reading memory cells in a SRAM, the sense amplifier comprising two gate-biased pMOSFETs, each corresponding to a selected bitline. The gates of the two gate-biased pMOSFETs have their gates biased to a bias voltage, their sources coupled to the selected bitlines via column-select transistors, and their drains coupled via pass transistors to the two ports of two cross-coupled inverters, the cross-coupled inverters forming a latch. After a selected bitline pair has been pre-charged and the pre-charge phase ends, one of the two gate-biased pMOSFETs quickly goes into its subthreshold region as one of the bitlines discharges through its corresponding memory cell, thereby cutting off the bitline's capacitance from the sense amplifier. When the pass transistors are enabled, the other of the two pMOSFETs allows a significant bitline charge to transfer via its corresponding pass transistor to its corresponding port, whereas a relatively much smaller charge is transferred to the other port. This charge transfer scheme allows a differential voltage to quickly develop at the ports, thereby providing a fast latch and read operation with reduced power consumption. Bitline voltage swing may also be reduced to reduce power consumption.

n/a