Analog-digital Systems Research Articles

Modeling aperiodic sampling to increase the signal processing bandwidth of the sampling analog optical path

The paper discusses methods of aperiodic optical pulsed signal sampling in photonic analog-digital systems. We describe numerical modeling of aperiodic sampling and a method for reconstructing a spectrally sparse signal from aperiodicly selected samples. The described method is based on minimizing the l1 norm of the spectral representation of the reconstructed signal. Numerical simulations show that using aperiodic sampling, it is possible to reconstruct a spectrally sparse signal in the 5 GHz band using a sampling rate of 500 MS/s. The work also shows that using window functions it is possible to improve the quality of signal reconstruction from pseudo-random samples.

Echo state graph neural networks with analogue random resistive memory arrays

Recent years have witnessed a surge of interest in learning representations of graph-structured data, with applications from social networks to drug discovery. However, graph neural networks, the machine learning models for handling graph-structured data, face significant challenges when running on conventional digital hardware, including the slowdown of Moore’s law due to transistor scaling limits and the von Neumann bottleneck incurred by physically separated memory and processing units, as well as a high training cost. Here we present a hardware–software co-design to address these challenges, by designing an echo state graph neural network based on random resistive memory arrays, which are built from low-cost, nanoscale and stackable resistors for efficient in-memory computing. This approach leverages the intrinsic stochasticity of dielectric breakdown in resistive switching to implement random projections in hardware for an echo state network that effectively minimizes the training complexity thanks to its fixed and random weights. The system demonstrates state-of-the-art performance on both graph classification using the MUTAG and COLLAB datasets and node classification using the CORA dataset, achieving 2.16×, 35.42× and 40.37× improvements in energy efficiency for a projected random resistive memory-based hybrid analogue–digital system over a state-of-the-art graphics processing unit and 99.35%, 99.99% and 91.40% reductions of backward pass complexity compared with conventional graph learning. The results point to a promising direction for next-generation artificial intelligence systems for graph learning.

The guts of assessment: a digital architecture for machine learning and analogue judgement

ABSTRACT Educational assessment is inherently uncertain, where physiological, psychological and social factors play an important role in establishing judgements which are assumed to be “absolute”. AI and other algorithmic approaches to grading of student work strip-out uncertainty, leading to a lack of inspectability in machine judgement and consequent problems of trust and reliability. The technique of Adaptive Comparative Judgement (ACJ), in focusing on small-stakes binary comparisons provides an alternative approach to dealing with uncertainty in grading. Rankings can be produced which codify uncertainty, rendering machine judgements inspectable. However, ACJ demands resources in terms of time and the number of people making comparisons. Machine learning trained to make binary comparisons can help to make the process of human comparison more efficient. In combining ACJ and AI, we argue that the result is an analogue-digital system where the physiological/analogue processes of assessment can be coordinated with digital services which steer human assessment to those judgements which are most uncertain requiring most human deliberation. Drawing on our design of such a system developed for medical judgements, we describe a general architecture for human–machine assessment in education and discuss its potential in bridging the gap between analogue human cognition and digital machine learning.

Mixed‐Precision Continual Learning Based on Computational Resistance Random Access Memory

Artificial neural networks have acquired remarkable achievements in the field of artificial intelligence. However, it suffers from catastrophic forgetting when dealing with continual learning problems, i.e., the loss of previously learned knowledge upon learning new information. Although several continual learning algorithms have been proposed, it remains a challenge to implement these algorithms efficiently on conventional digital systems due to the physical separation between memory and processing units. Herein, a software–hardware codesigned in‐memory computing paradigm is proposed, where a mixed‐precision continual learning (MPCL) model is deployed on a hybrid analogue–digital hardware system equipped with resistance random access memory chip. Software‐wise, the MPCL effectively alleviates catastrophic forgetting and circumvents the requirement for high‐precision weights. Hardware‐wise, the hybrid analogue–digital system takes advantage of the colocation of memory and processing units, greatly improving energy efficiency. By combining the MPCL with an in situ fine‐tuning method, high classification accuracies of 94.9% and 95.3% (software baseline 97.0% and 97.7%) on the 5‐split‐MNIST and 5‐split‐FashionMNIST are achieved, respectively. The proposed system reduces ≈200 times energy consumption of the multiply‐and‐accumulation operations during the inference phase compared to the conventional digital systems. This work paves the way for future autonomous systems at the edge.

Machines that perform measurements

We review ten years of work in modeling the computation of analog-digital systems that express the physical reality in the limit, an outcome of our involvement in the conferences UC (Unconventional Computation) and UCNC (Unconventional Computation and Natural Computation).Computers are connected with sensors that measure physical quantities such as distance, electric current, mass, pressure, temperature, etc, interfering in the environment as experimenters. In this paper, we consider the experimenter (e.g. the experimental physicist) as a Turing machine – the digital component – and the experiment of measurement – the analog component – as an oracle to the Turing machine. The algorithm running in the machine abstracts the method of measurement (encoding the recursive structure of experimental actions) chosen by the experimenter. The interaction between the machine and the sensors is mediated by a protocol, that specifies the experimental error, and a schedule, that determines its running time. We then overview the computational complexity classes related to diverse protocols between the digital and the analogue parts of the digital-analogue machine.

ВИМІРЮВАННЯ ОПОРНОЇ ЧАСТОТИ ВУЗЬКОСМУГОВОГО РАДІОСИГНАЛУ ОБМЕЖЕНОЇ ТРИВАЛОСТІ

The work concerns analog-digital systems that work with radio signals emitted and received by the antenna, and these signals have a limited duration, ie are pulsed. The propagation conditions of such signals and the processes of formation of echo signals affect their amplitude, frequency and phase characteristics in such a way that they form classical narrowband signals. The operation of the system involves the determination of certain parameters of echo signals, and taking into account their pulse nature for such a definition is given a limited time interval. This means that the procedure for determining the parameters must meet the criterion of high speed, and therefore differ from the traditional, built on the use of phase-locked loop. The article is devoted to solving the problem of measuring the reference frequency of a pulsed narrowband radio signal. By analyzing the results of experimental studies, two types of errors in measuring the reference frequency of a narrowband radio signal were identified and the causes of their occurrence were established. These errors are provoked by the peculiarities of the structure of the narrowband signal. The reason for the error of the 1st type, the absolute value of which correlates with the duration of the half-cycle of the reference frequency, is the so-called phase jumps at the point of change of the sign of the bypass. The frequency of such errors within the duration of the echo signal is generally low, although it increases with increasing signal spectrum width. Type 2 errors occur due to the appearance of zones with almost completely suppressed signal amplitude, which may be the result of intrapulse interference and / or signal attenuation. Such errors also occur more often if the spectrum width is larger. A method for measuring the reference frequency of a narrowband pulsed radio signal is proposed, which is based on counting the number of half-cycles of the reference frequency on a time-limited measurement interval and removing from this procedure areas with completely suppressed signal amplitude. An auxiliary highly stable reference frequency is used to establish the numerical value of the reference frequency. The block diagram of the frequency meter and the algorithm of its operation are given, the implementation of which avoids these errors.

BiLiMO: Bit-Limited MIMO Radar via Task-Based Quantization

Recent years have witnessed growing interest in reduced cost radar systems operating with low power. Multiple-input multiple-output (MIMO) radar technology is known to achieve high performance sensing by probing with multiple orthogonal waveforms. However, implementing a low cost low power MIMO radar is challenging. One of the reasons for this difficulty stems from the increased cost and power consumption required by analog-to-digital convertors (ADCs) in acquiring the multiple waveforms at the radar receiver. In this work we study reduced cost MIMO radar receivers restricted to operate with low resolution ADCs. We design bit-limited MIMO radar (BiLiMO) receivers which are capable of accurately recovering their targets while operating under strict resolution constraints. This is achieved by applying an additional analog filter to the acquired waveforms, and designing the overall hybrid analog-digital system to facilitate target identification using task-based quantization methods. In particular, we exploit the fact that the target parameters can be recovered from a compressed representation of the received waveforms. We thus tune the acquisition system to recover this representation with minimal distortion, from which the targets can be extracted in digital, and characterize the achievable error in identifying the targets. Our numerical results demonstrate that the proposed BiLiMO receiver operating with a bit budget of one bit per sample achieves target recovery performance which approaches that of costly MIMO radars operating with unlimited resolution ADCs, while substantially outperforming MIMO receivers operating only in the digital domain under the same bit limitations.

Electronic neural interfaces

Devices such as keyboards and touchscreens allow humans to communicate with machines. Neural interfaces, which can provide a direct, electrical bridge between analogue nervous systems and digital man-made systems, could provide a more efficient route to future information exchange. Here we review the development of electronic neural interfaces. The interfaces typically consist of three modules — a tissue interface, a sensing interface, and a neural signal processing unit — and based on technical milestones in the development of the electronic sensing interface, we group and analyse the interfaces in four generations: the patch clamp technique, multi-channel neural interfaces, implantable/wearable neural interfaces and integrated neural interfaces. We also consider key circuit and system challenges in the design of neural interfaces and explore the opportunities that arise with the latest technology This Review Article examines the development of neural interfaces, which can provide a direct, electrical bridge between analogue human nervous systems and digital man-made devices, considering challenges and opportunities created with such technology.

Ultrastable Offset-Locking Continuous Wave Laser to a Frequency Comb with a Compound Control Method for Precision Interferometry.

A simple and robust analog feedforward and digital feedback compound control system is presented to lock the frequency of a slave continuous wave (CW) laser to an optical frequency comb. The beat frequency between CW laser and the adjacent comb mode was fed to an acousto-optical frequency shifter (AOFS) to compensate the frequency dithering of the CW laser. A digital feedback loop was achieved to expand the operation bandwidth limitation of the AOFS by over an order of magnitude. The signal-to-noise ratio of the interference signal was optimized using a grating-based spectral filtering detection unit. The complete system achieved an ultrastable offset-locking of the slave CW laser to the frequency comb with a relative stability of ±3.62 × 10−14. The Allan deviations of the beat frequency were 8.01 × 10−16 and 2.19 × 10−16 for a gate time of 10 s and 1000 s, respectively. The findings of this study may further improve laser interferometry by providing a simple and robust method for ultrastable frequency control.

Analogue part of multichannel highly productive analog-digital system on converters and switches of current

Analogue part of multichannel highly productive analog-digital system on converters and switches of current

This Was The End

In this article I describe the process of conceiving, creating and directing a performance called This Was The End, and the sound and video installation that archives it. This exploration of aging and memory has been a decade long project inspired by Uncle Vanya’s desperate thought ‘Suppose I live to be 60?’ This simple prompt lead to my collaboration with a quartet of older actors, whose careers in New York City’s downtown experimental theatre scene encompass Charles Ludlam’s Ridiculous Theatrical Company and Joe Chaiken’s Open Theater. I reflect on how their presence in the process opened up an exploration of memory on multiple fronts from the role memorization plays in conventional theater, to working with audio recordings of the actors’ memories of Chehkov’s play, to design considerations. I describe how the set, itself a recovered piece of architecture, is activated in performance with an integrated analog sound and digital video system inspired by the optical science behind Proust’s formulation of memory as the convergence of past and present. Through the scholarship of Roger Shattuck I explain how taking a Proustian approach to Chekhov’s dramatic trajectory lead to the deeper underlying question of the work - What do we do with our past? What can we make of it? Finally I consider how memorializing the fragility, resilience and persistence of these older performers as an interactive sculptural installation serves to deny erasure and remind us of a life before that continues.

Design and implementation of a jerk circuit using a hybrid analog–digital system

The third order systems are very attractive because they present a rich dynamics and can be applied in secure communications, cryptography and so on. A special class of this kind of systems is the so-called jerk systems. This type of systems is very interesting in the field of research, because of its simplicity and ease of construction. In this work, the design and implementation of a jerk circuit using a hybrid analog–digital system is presented. This is done by means of a combination of analog circuits, which include op-amps configured as integrators, and electronic passive components; and digital circuits, as microcontrollers and DAC/ADC boards. The combination of these components takes the best of each analog/digital devices allowing simplification of the traditional analog jerk circuit and also a fast reconfiguration provided by the digital part. The obtained experimental results show that the proposed hybrid circuit can reproduce the dynamics of the jerk system.

ВИСОКОЛІНІЙНІ БУФЕРИЙ МАСШТАБАТОРИ НАПРУГИ НА БІПОЛЯРНИХ ТРАНЗИСТОРАХ ІЗ НИЗЬКИМ ВХІДНИМ СТРУМОМ

Buffer devices and voltage scalers are widely used in various analog-digital systems, when it is necessary to match the signal in the form of voltage from a low-power sensor to a load, it consumes significantly more power. In this case, the voltage buffer is characterized by a voltage transfer coefficient close to unity, and must also have a high input resistance and sufficient load capacity. The voltage scaler, in contrast to voltage buffers, must additionally provide the necessary gain transfer ratio, which can be substantially more than one. The circuit design features of three variants of the construction of voltage buffer cores and voltage scaling are considered. It is proved that it is advisable to reduce the input zero bias current by using amplifying n-p-n and p-n-p transistors, as well as Shiclay transistors in the input stages. The static and dynamic characteristics of the voltage buffers and the voltage scaler must meet the system requirements of the device. The static characteristics should be attributed primarily to the error of the transfer characteristics of the scale, zero offset and linearity. The dynamics of these devices is determined by the frequency response and transient response. Static and dynamic characteristics are analyzed by computer simulation where it is shown that the scale errors of the voltage buffers and the voltage scaler do not exceed 10 µV in the range of the corresponding signal ± 5 V, and the linearity errors are 300 nV. A transient response was obtained which states that the slew rate of the output voltage will be no worse than 2000 V/µs. A comparison of the metrological characteristics of the voltage buffers and the voltage scaler in the form of a set of cores and output push-pull DC amplifiers. It is proved that the use of these amplifiers allows significantly (by 3-4 orders of magnitude) to improve the load capacity of the circuits while maintaining the level of the input zero bias current, as well as scale and linearity errors.

A unified direct approach for discretizing fractional-order differentiator in delta-domain

Fractional-order differentiator is a principal component of the fractional-order controller. Discretization of fractional-order differentiator is essential to implement the fractional-order controller digitally. Discretization methods generally include indirect approach and direct approach to find the discrete-time approximation of fractional-order differentiator in the [Formula: see text]-domain as evident from the existing literature. In this paper, a direct approach is proposed for discretization of fractional-order differentiator in delta-domain instead of the conventional [Formula: see text]-domain as the delta operator unifies both analog system and digital system together at a high sampling frequency. The discretization of fractional-order differentiator is accomplished in two stages. In the first stage, the generating function is framed by reformulating delta operator using trapezoidal rule or Tustin approximation and in the next stage, the fractional-order differentiator has been approximated by expanding the generating function using continued fraction expansion method. The proposed method has been compared with two well-known direct discretization methods taken from the existing literature. Two examples are presented in this context to show the efficacy of the proposed discretization method using simulation results obtained from MATLAB.

Hybrid Analog-Digital Processing System for Amplitude-Monopulse RSSI-Based MiMo WiFi Direction-of-Arrival Estimation

We present a cost-effective hybrid analog digital system to estimate the Direction of Arrival (DoA) of WiFi signals. The processing in the analog domain is based on simple well-known RADAR amplitude monopulse antenna techniques. Then, using the received signal strength indicator (RSSI) delivered by a commercial MiMo WiFi cards, the DoA is estimated using the so-called digital monopulse function. Due to the hybrid analog digital architecture, the digital processing is extremely simple, so that DoA estimation is performed without using IQ data from specific hardware. The simplicity and robustness of the proposed hybrid analog digital MiMo architecture is demonstrated for the ISM 2.45 GHz WiFi band. Also, the limitations with respect to multipath effects are studied in detail. As a proof of concept, an array of two MiMo WiFi DoA monopulse readers is distributed to localize the two-dimensional position of WiFi devices. This cost-effective hybrid solution can be applied to all WiFi standards and other Internet of Things narrowband radio protocols, such as Bluetooth Low Energy or Zigbee.

Development of a compact and cost effective multi-input digital signal processing system

A prototype digital signal processing system (DSP) was developed using a microcontroller interfaced with a 12-bit sampling ADC, which offers a considerably inexpensive solution for processing multiple detectors with high throughput. After digitization of the incoming pulses, in order to maximize the output counting rate, a simple algorithm was employed for pulse height analysis. Moreover, an algorithm aiming at the real-time pulse pile-up deconvolution was implemented. The system was tested using a NaI(Tl) detector in comparison with a traditional analogue and commercial digital systems for a variety of count rates. The performance of the prototype system was consistently superior to the analogue and the commercial digital systems up to the input count rate of 61 kcps while was slightly inferior to the commercial digital system but still superior to the analogue system in the higher input rates. Considering overall cost, size and flexibility, this custom made multi-input digital signal processing system (MMI-DSP) was the best reliable choice for the purpose of the 2D microdosimetric data collection, or for any measurement in which simultaneous multi-data collection is required.

Transverse beam feedback systems in the U-70 synchrotron

A method for analyzing the transverse beam feedback circuit in a ring accelerator and selecting its parameters using the concept of the active beam coupling impedance, which is externally imposed and controlled, is described. This approach is used for the physical substantiation of approaches to upgrading two operational transverse feedback systems that are available in the U-70 synchrotron of the Institute for High Energy Physics, National Research Centre Kurchatov Institute: narrow-band analog and broadband digital systems. Both systems were subjected to deep upgrading, which is based on this method, and were successfully commissioned in 2007 and 2011, respectively. A brief description of their updated engineering implementation is given. The presented experimental results confirm the adequacy of the impedance approach to designing and adjusting the transverse beam feedback circuits in the U-70 synchrotron.

CBM Experiment Local and Global Implications

Abstract The research area of the compressed baryonic matter - CBM experiment (FAIR/GSI in Darmstadt) is subnuclear physics, thus hadron-baryon and quark-gluon, and the essence of phase transitions in the area of hot nuclear matter, and dense strongly interacting matter. Our interest in this paper are mainly considerations on the impact of such large infrastructural experiments and possibilities they give to local, smaller but very active, university based research groups and communities. Research and technical input from such groups is depicted on the background of the CBM detector infrastructure and electronic instrumentation just under design and test fabrication for this experiment. An essential input to this research originates from Poland via the agreed in-kind contribution. The areas of expertise of these groups are: superconductivity, structural large scale cabling, precision machined parts, RF and microwave technology, analog and advanced digital electronics, distributed measurement and control systems, etc.

An Open-Source Tool Set Enabling Analog-Digital-Software Co-Design

This paper presents an analog-digital hardware-software co-design environment for simulating and programming reconfigurable systems. The tool simulates, designs, as well as enables experimental measurements after compiling to configurable systems in the same integrated design tool framework. High level software in Scilab/Xcos (open-source programs similar to MATLAB/Simulink) that converts the high-level block description by the user to blif format (sci2blif), which acts as an input to the modified VPR tool, including the code v p r 2 s w c s , encoding the specific platform through specific architecture files, resulting in a targetable switch list on the resulting configurable analog–digital system. The resulting tool uses an analog and mixed-signal library of components, enabling users and future researchers access to the basic analog operations/computations that are possible.

Towards a Neuromorphic Vestibular System

The vestibular system plays a crucial role in the sense of balance and spatial orientation in mammals. It is a sensory system that detects both rotational and translational motion of the head, via its semicircular canals and otoliths respectively. In this work, we propose a real-time hardware model of an artificial vestibular system, implemented using a custom neuromorphic Very Large Scale Integration (VLSI) multi-neuron chip interfaced to a commercial Inertial Measurement Unit (IMU). The artificial vestibular system is realized with spiking neurons that reproduce the responses of biological hair cells present in the real semicircular canals and otholitic organs. We demonstrate the real-time performance of the hybrid analog-digital system and characterize its response properties, presenting measurements of a successful encoding of angular velocities as well as linear accelerations. As an application, we realized a novel implementation of a recurrent integrator network capable of keeping track of the current angular position. The experimental results provided validate the hardware implementation via comparisons with a detailed computational neuroscience model. In addition to being an ideal tool for developing bio-inspired robotic technologies, this work provides a basis for developing a complete low-power neuromorphic vestibular system which integrates the hardware model of the neural signal processing pathway described with custom bio-mimetic gyroscopic sensors, exploiting neuromorphic principles in both mechanical and electronic aspects.