Feedback Amplifier Research Articles

Time Complexity of In-Memory Solution of Linear Systems

In-memory computing with crosspoint resistive memory arrays has been shown to accelerate data-centric computations such as the training and inference of deep neural networks, thanks to the high parallelism endowed by physical rules in the electrical circuits. By connecting crosspoint arrays with negative feedback amplifiers, it is possible to solve linear algebraic problems such as linear systems and matrix eigenvectors in just one step. Based on the theory of feedback circuits, we study the dynamics of the solution of linear systems within a memory array, showing that the time complexity of the solution is free of any direct dependence on the problem size N, rather it is governed by the minimal eigenvalue of an associated matrix of the coefficient matrix. We show that, when the linear system is modeled by a covariance matrix, the time complexity is O(logN) or O(1). In the case of sparse positive-definite linear systems, the time complexity is solely determined by the minimal eigenvalue of the coefficient matrix. These results demonstrate the high speed of the circuit for solving linear systems in a wide range of applications, thus supporting in-memory computing as a strong candidate for future big data and machine learning accelerators.

During Aspergillus Infection, Monocyte-Derived DCs, Neutrophils, and Plasmacytoid DCs Enhance Innate Immune Defense through CXCR3-Dependent Crosstalk

SummaryAspergillus fumigatus, a ubiquitous mold, is a common cause of invasive aspergillosis (IA) in immunocompromised patients. Host defense against IA relies on lung-infiltrating neutrophils and monocyte-derived dendritic cells (Mo-DCs). Here, we demonstrate that plasmacytoid dendritic cells (pDCs), which are prototypically antiviral cells, participate in innate immune crosstalk underlying mucosal antifungal immunity. Aspergillus-infected murine Mo-DCs and neutrophils recruited pDCs to the lung by releasing the CXCR3 ligands, CXCL9 and CXCL10, in a Dectin-1 and Card9- and type I and III interferon signaling-dependent manner, respectively. During aspergillosis, circulating pDCs entered the lung in response to CXCR3-dependent signals. Via targeted pDC ablation, we found that pDCs were essential for host defense in the presence of normal neutrophil and Mo-DC numbers. Although interactions between pDC and fungal cells were not detected, pDCs regulated neutrophil NADPH oxidase activity and conidial killing. Thus, pDCs act as positive feedback amplifiers of neutrophil effector activity against inhaled mold conidia.

In‐Memory Eigenvector Computation in Time O(1)

In‐memory computing with cross‐point resistive memory arrays has gained enormous attention to accelerate the matrix‐vector multiplication in the computation of data‐centric applications. By combining a cross‐point array and feedback amplifiers, it is possible to compute matrix eigenvectors in one step without algorithmic iterations. Herein, time complexity of the eigenvector computation is investigated, based on the feedback analysis of the cross‐point circuit. The results show that the computing time of the circuit is determined by the mismatch degree of the eigenvalues implemented in the circuit, which controls the rising speed of output voltages. For a dataset of random matrices, the time for computing the dominant eigenvector in the circuit is constant for various matrix sizes; namely, the time complexity is O(1). The O(1) time complexity is also supported by simulations of PageRank of real‐world datasets. This work paves the way for fast, energy‐efficient accelerators for eigenvector computation in a wide range of practical applications.

A 1.7-dB Minimum NF, 22–32-GHz Low-Noise Feedback Amplifier With Multistage Noise Matching in 22-nm FD-SOI CMOS

A low-noise feedback amplifier (LNA) with interstage noise matching is implemented in 22-nm fully depleted silicon-on-insulator (SOI)-CMOS technology. Minimum noise figure (NF) is 1.7 dB centered at 28 GHz, and NF remains below 1.98±0.25 dB across a 10-GHz range. Peak gain of the two-stage LNA is 21.5 dB at 22 GHz, and the bandwidth (BW) for $|{S_{21}} |$ is 19–36 GHz. Input and output return losses are better than 10 dB across an effective LNA BW of 22–32 GHz. The third-order input intercept is −13.4 dBm at peak gain when dissipating 17.3 mW. Continuous dc power control on the fly is implemented using modulation of FET backgates. When dc power consumption is reduced 5.6 mW, NF increases by less than 0.5 dB, peak gain decreases by 3.6 dB, and input return loss remains better than 10 dB with no change in effective BW.

Model Following Control of a Higher Order Plant by a Lower Order Model

In this paper, based on Model Following Control (MFC) approach, a robust controller is used to control a flexible robot manipulator along a pre-defined trajectory. Here two degree of freedom plant is considered that has two different inertias. The plant is run by the single degree of freedom ideal model. Primarily, an ideal model is formulated from the mathematical expression and by selecting a suitable feedback amplifier gain a well-defined response is established. A reference input voltage is given to the model and the plant is driven by the errors, generated from the differences of the states between the plant and model. Here special attention is given to the fact that how precisely the states of the plant can follow the ideal states of the model. The proposed model following control (MFC) system may be used successfully in industrial robots.

Robust and efficient wireless power transfer using a switch-mode implementation of a nonlinear parity–time symmetric circuit

Stationary wireless power transfer has been deployed commercially and can be used to charge a variety of devices, including mobile phones and parked electric vehicles. However, wireless power transfer set-ups typically suffer from an inherent sensitivity to the relative movement of the device with respect to the power source. Nonlinear parity–time symmetric circuits could be used to deliver robust wireless power transfer even while a device is moving rapidly, but previous implementations have relied on an inefficient gain element based on an operation-amplifier circuit, which has inherent loss, and hence have exhibited poor total system efficiency. Here we show that robust and efficient wireless power transfer can be achieved by using a power-efficient switch-mode amplifier with current-sensing feedback in a parity–time symmetric circuit. In this circuit, the parity–time symmetry guarantees that the effective load impedance on the switch-mode amplifier remains constant, and hence the amplifier maintains high efficiency despite variation of the transfer distance. We experimentally demonstrate a nonlinear parity–time symmetric radiofrequency circuit that can wirelessly transfer around 10 W of power to a moving device with a nearly constant total efficiency of 92% and over a distance from 0 to 65 cm. A parity–time symmetric circuit that uses a switch-mode amplifier and current-sensing phase-delay feedback can wirelessly transfer around 10 W of power to a moving device with a nearly constant total efficiency of 92%.

Realization of current mode universal shadow filter

In this paper, a current-mode universal shadow filter is proposed using a new variant of current differencing transconductance amplifier (CDTA), namely current controlled current differencing cascaded transconductance amplifier (CC-CDCTA) and two capacitors. It realizes three shadow filter functioning; low pass (LPS), high pass (HPS) and band pass (BPS). Furthermore, it is extended to a universal filter by adding one second generation current conveyor (CCII) and CC-CDCTA building blocks. The proposed configuration realizes all the standard responses of a universal shadow filter such as LPS, BPS, HPS, band-reject (BRS), and all pass (APS) simultaneously. It does not use any resistor. The presented shadow filter utilizes the LP and BP outputs of the basic filter in the feedback amplifier to obtain the desired universal shadow filter functionality. The pole frequency (ωos) and quality factor (Qos) of the shadow filter is electronically tunable using the gain of the feedback amplifiers. The ωos and Qos can also be tuned without disturbing each other. It is also found to be suitable for full cascadability. The validation is done using 180 nm technology in Cadence. Experimental verification has also been done using IC AD844, and CA3080 and found satisfactory results.

Quality factor tuning of micromechanical resonators via electrical dissipation

Sensitive capacitive transduction of micromechanical resonators can contribute significant electrical dissipation, which degrades the quality factor of the eigenmodes. We theoretically and experimentally demonstrate a scheme for isolating the electrical damping of a mechanical resonator due to Ohmic dissipation in the readout amplifier. The quality factor suppression arising from the amplifier is strongly dependent on the amplifier feedback resistance and parasitic capacitance. By studying the thermomechanical displacement noise spectrum of a doubly clamped micromechanical beam, we confirm that electrical dissipation tunes the actual, not effective, quality factor. Electrical dissipation is an important consideration in the design of sensitive capacitive displacement transducers, which are a key component in resonant sensors and oscillators.

Two-Stage BJT Amplifier with Negative Voltage Series Feedback Analysis Program Proteus 8

In this paper, the common two-stage amplifier voltage series negative feedback amplifier circuit is analyzed theoretically. With negative feedback, the closed-loop voltage gain can be reduced and controlled so that the op-amp can function as a linear amplifier. Negative feedback is the basic method and essential means of improving amplifying circuit efficiency. Then using Proteus 8 professional as simulation software, the relationship between the gain obtained by open-loop and the simulation of the closed-loop and the pass-band is shown by comparing the real experimental data with the result of the circuit’s simulation and the waveform. Negative feedback opposes or subtracts from the input signals giving it many advantages in the design and stabilization of control systems. Basically the relationship consistent with the calculated results, that is in good agreement with the theoretical analysis.

Nanostrain Resolution Strain Sensing by Monocrystalline 3C-SiC on SOI Electrostatic MEMS Resonators

The paper reports on the fabrication and characterization of flexural resonators designed for resonant strain sensing and manufactured using 2 $\mu \text{m}$ thick monocrystalline 3C-SiC layers grown on Silicon-On-Insulator substrates. The resonators are designed with Double-Ended-Tuning Fork geometry and actuated electrostatically using lateral electrodes fixed on the sides of the resonator tines, through coupling gaps with average width of 1.03 $\mu \text{m}$ . They show excellent performance in vacuum environment with Q-factor around 30,000 and 18 dB high resonance peaks measured in open loop with DC bias voltage of 20 V. The closed-loop operation of the resonators with an external transimpedance amplifier feedback circuit is demonstrated and the stability of the resulting MEMS oscillator is analyzed using a microcontroller-based digital readout to measure its oscillation frequency. The results of the Allan deviation tests on the MEMS oscillator indicate a relative frequency stability better than 10−7 on a measurement time of 280 ms. The results of strain sensitivity experiments in vacuum performed after adhesive attachment of the resonator chip on Al are also reported, indicating a sensitivity of 41.7 Hz/ $\mu \varepsilon $ and a resolution of about 0.82 $\text{n}\varepsilon $ for the same measurement time. [2019-0183]

Simulations of beam loading compensation in a wideband accelerating cavity using a circuit simulator including a LLRF feedback control

Magnetic alloy cavities are employed in the J-PARC RCS to generate high accelerating voltages. The cavity, which is driven by a vacuum tube amplifier, has a wideband frequency response and the beam loading in the cavity is multiharmonic. Therefore, the tube must generate a multiharmonic output current. An LTspice circuit model is developed to analyze the vacuum tube operation and the compensation of the multiharmonic beam loading. The model includes the cavity, tube amplifier, beam current, and LLRF feedback control. The feedback control consists of the I/Q demodulator including low pass filters, PI control, and I/Q modulator. In this article, we present the implementation of the LLRF functions in the LTspice simulations. The preliminary simulation results are also presented. The simulations fairly agree with the beam test results.

Compact readout of large CLYC scintillators with silicon photomultipler arrays

Compact readout of a large (5×5×5 cm3) CLYC scintillator was demonstrated with a 8×8 array of 6mm silicon photomultipliers (SiPMs, specifically, a SensL ArrayJ-60035-64P) and an optimized amplification and summing circuit. Read out by the SiPM array, the crystal yielded an energy resolution of 5.5% at 662 keV and a figure of merit of 3.5, equivalent to the performance achieved with a 3-inch Hamamatsu R6233-100 PMT. To reduce channel count and avoid individual amplification of each SiPM in the array, the standard outputs of 16 SiPMs were passively summed then amplified. A current feedback amplifier was used for each channel to provide the required fast response time. The four amplified channels were then passively summed resulting in a single signal channel. A custom laboratory setup, based on the PSD8C ASIC, was used to capture the output and provide time-gated integration for PSD computation. The compact volume, low bias voltage, and reduced mass of the SiPM array readout are desirable for space and national security applications.

Techniques for vibration resistance increase of quartz pendulum accelerometer using digital feedback amplifier

Subject of Research. The paper considers approaches of vibration resistance increasing for quartz pendulum accelerometer with digital feedback amplifier and without further development of measuring element design or application of damping measures. Method. The proposed methods are implemented algorithmically by means of the control program of the digital amplifier controller and are based on the extension of the device bandwidth, the correction of the regulating coefficients during the accelerometer operation, and the introduction of an additional control channel to the feedback loop. Main Results. The proposed methods were studied experimentally on a prototype quartz pendulum accelerometer with a digital feedback amplifier and a measuring range of ±50 g. The studies were carried out under the effects of vibration actions: broadband random vibration in the frequency range from 20 to 2000 Hz, (RMS deviation of the vibration acceleration amplitude of 8 g) and sinusoidal vibration in the frequency range from 20 to 2000 Hz (vibration acceleration amplitude of 10 g) along the sensitivity axis. The efficiency of the methods was estimated by the value of device vibration error — the difference between the averaged value of the device readings before vibration and the averaged value of the readings during the period of vibration. It was experimentally established that the best solution is to combine a wide bandwidth of the device with an additional control channel in the feedback loop, that reduces the accelerometer vibration error by an order of magnitude. Practical Relevance. The proposed methods increase the vibration resistance of a quartz pendulum accelerometer by applying original algorithms in the digital amplifier controller program, that makes it possible to put to use such accelerometers in objects susceptible to vibration while eliminating the need to introduce physical design modifications or additional damping measures.

A 2–18 GHz Compressed Sensing Receiver With Broadband LO Chain in 0.13-$\mu$ m BiCMOS

We present an integrated 2–18 GHz four-channel compressed sensing receiver (CSRX), which performs a $128.5\times $ sub-Nyquist acquisition with 36-Gb/s pseudorandom bit sequence (PRBS) generators in 0.13- $\mu \text{m}$ BiCMOS. The integrated four-channel CSRX consists of broadband receiver chains, each of which includes a distributed RF buffer that operates as an active balun at 2–18 GHz, followed by a double-balanced passive mixer for high local-oscillator (LO)-to-intermediate frequency (IF) isolation at the IF band, and a 700-MHz variable gain amplifier (VGA). Four independent 36-Gb/s PRBS generators are integrated to perform the compressed sensing by mixing the PRBS signal as the random modulator. To deliver the generated wideband PRBS signal to the mixer, the LO driver conducts band-equalization by applying a shunt-shunt feedback amplifier to compensate for the high-frequency degradation of the PRBS LO tones through the LO generation stage. The chip consumes 195 mA per channel with a 2.5-V supply.

Flux Feedback Amplifier based on Superconducting Quantum Interference Device

We introduced the Flux Feedback Amplifier (FFA) to design simple and easy-to-use Flux-Locked Loop (FLL) used to operate Superconducting Quantum Interference Device (SQUID) for practical applications. The SQUID FFA is a hybrid operational amplifier (op amp) with SQUID as flux input end and feedback coil as flux output end. The room temperature circuit is simplified as a low noise differential amplifier with no more than two op amps, and operated with only one switch. Two SQUID FFA schemes were demonstrated and the dynamic performances were characterized and compared in the experimental results. The scheme with a monolithic op amp achieves one-pole open loop gain with slope of 20 dB/decade and is suited for the wideband FLL. The scheme with two op amps achieves two-pole open loop gain with slope of 40 dB/decade and is suited for the FLL with high slew rate (up to 1Φ0/μs).

Sensitive and Specific Whole-Cell Biosensor for Arsenic Detection.

Whole-cell biosensors (WCBs) have been designed to detect As(III), but most suffer from poor sensitivity and specificity. In this paper, we developed an arsenic WCB with a positive feedback amplifier in Escherichia coli DH5α. The output signal from the reporter mCherry was significantly enhanced by the positive feedback amplifier. The sensitivity of the WCB with positive feedback is about 1 order of magnitude higher than that without positive feedback when evaluated using a half-saturation As(III) concentration. The minimum detection limit for As(III) was reduced by 1 order of magnitude to 0.1 µM, lower than the World Health Organization standard for the arsenic level in drinking water, 0.01 mg/liter or 0.13 µM. Due to the amplification of the output signal, the WCB was able to give detectable signals within a shorter period, and a fast response is essential for in situ operations. Moreover, the WCB with the positive feedback amplifier showed exceptionally high specificity toward As(III) when compared with other metal ions. Collectively, the designed positive feedback amplifier WCB meets the requirements for As(III) detection with high sensitivity and specificity. This work also demonstrates the importance of genetic circuit engineering in designing WCBs, and the use of genetic positive feedback amplifiers is a good strategy to improve the performance of WCBs.IMPORTANCE Arsenic poisoning is a severe public health issue. Rapid and simple methods for the sensitive and specific monitoring of arsenic concentration in drinking water are needed. In this study, we designed an arsenic WCB with a positive feedback amplifier. It is highly sensitive and able to detect arsenic below the WHO limit level. In addition, it also significantly improves the specificity of the biosensor toward arsenic, giving a signal that is about 10 to 20 times stronger in response to As(III) than to other metals. This work not only provides simple but effective arsenic biosensors but also demonstrates the importance of genetic engineering, particularly the use of positive feedback amplifiers, in designing WCBs.

A Chopper-Stabilized, Current Feedback, Neural Recording Amplifier.

Advanced neural prosthetics requires high density neural recording and stimulation electrodes interfacing with the tissue. For an implantable device, area, power consumption, and noise performance are the key design metrics. Due to the low-frequency nature of the recorded signals, chopping technique is inevitable to satisfy the noise requirement while maintaining a small area and low power consumption. However, chopping leads to a significant drop in input impedance, which leads to potential attenuation of neural signals recorded from high impedance miniature electrodes, and an unacceptable large input current drawn from the tissue. This work presents a chopper stabilized, current feedback amplifier (CFA) with input impedance boosted to 3.0 GΩ. The amplifier has an adjustable voltage gain of 40-60 dB, and an adjustable high-pass cut-off frequency of 0.5 - 5 Hz, with a power consumption of 2.6 μW and noise efficiency factor (NEF) of 3.2.

All-Pass Negative Group Delay Function With Transmission Line Feedback Topology

A novel circuit theory of all-pass Negative group delay (NGD) function is investigated. The NGD function is implemented with unity direct chain feedback (UDCF) system. The active circuit is built with an operational amplifier in feedback with a lossy transmission line (TL). The UDCF topology S-parameter model is established versus TL parameters. The NGD analysis is elaborated from the frequency dependent transmission coefficient. The NGD behavior characterization is developed. The synthesis method allowing to determine the UDCF topology parameters in function of the targeted NGD values, gain and reflection coefficient is formulated. The all-pass NGD function is validated with a proof-of-concept (POC) design. Frequency and time domain simulations confirm the feasibility of the innovative all-pass NGD function. With S-parameter analysis, it was shown that the UDCF circuit exhibits NGD up to -7-ns with gain more than 0-dB and reflection coefficient -20-dB. More importantly, time-domain analysis illustrates how the transient tested voltage outputs can be in advance compared to the input.

Negative Group Delay Theory of a Four-Port RC-Network Feedback Operational Amplifier

An innovative negative group delay (NGD) theory based on a unity direct chain feedback (UDCF) circuit topology is developed in this paper. This NGD circuit is an active cell constituted by an operational amplifier in feedback with a four-port RC-network. This NGD circuit theory is developed based on the S-parameter model analytically established from the equivalent impedance matrix. The UDCF group delay frequency response is expressed as a function of the feedback RC-cell and the operational amplifier parameters. The NGD analysis of the developed UDCF cell is introduced. According to theoretical analysis, under a certain condition, the UDCF topology is susceptible to behave as a low-pass NGD function. The UDCF cell NGD characteristics are defined theoretically. The theoretical prediction is verified numerically and experimentally in both the frequency- and time-domain by designing and fabricating an active PCB prototype. The simulations and experimentations show that the UDCF circuit exhibits an NGD of approximately -38 ns with NGD cut-off frequency of about 5.5 MHz. More importantly, it is demonstrated in the time-domain that the low-pass NGD effect enables the UDCF cell to generate advanced output with sinc waveform input voltages.

The Transimpedance Amplifier [A Circuit for All Seasons

Many of today's communication systems incorporate a transimpedance amplifier (TIA). Although the TIA concept is as old as feedback amplifiers [1], it was in the late 1960s and early 1970s that TIAs found widespread usage in optical coupling and optical communication receivers. In a patent filed in 1967, Miller proposes the circuit shown in Figure 1 [2], which consists of two TIAs for converting a photodiode's current to a differential output voltage. Additionally, these amplifiers have become popular in radio-frequency (RF) receivers in the past 10 years [3]. This article deals with TIA design for optical and RF applications.