J. Low Power Electron. Appl., Volume 12, Issue 4 (December 2022) – 17 articles

Wireless Sensor Networks (WSNs) are considered to be among the most important scientific domains. Yet, the exploitation of WSNs suffers from the severe energy restrictions of their electronic components. For this reason there are numerous scientific methods that have been proposed aiming to achieve the extension of the lifetime of WSNs, either by energy saving or energy harvesting or through energy transfer. This study aims to analytically examine all of the existing hardware-based and algorithm-based mechanisms of this kind. The operating principles of 48 approaches are studied, their relative advantages and weaknesses are highlighted, open research issues are discussed, and resultant concluding remarks are drawn. Full article

In this paper, the most suited analog-to-digital (A/D) converters (ADCs) for Internet of Things (IoT) applications are compared in terms of complexity, dynamic performance, and energy efficiency. Among them, an innovative hybrid topology, a digital–delta (Δ) modulator (ΔM) ADC employing noise shaping (NS), is proposed. To implement the active building blocks, several standard-cell-based synthesizable comparators and amplifiers are examined and compared in terms of their key performance parameters. The simulation results of a fully synthesizable Digital-ΔM with NS using passive and standard-cell-based circuitry show a peak of 72.5 dB in the signal-to-noise and distortion ratio (SNDR) for a 113 kHz input signal and 1 MHz bandwidth (BW). The estimated ${\mathrm{FoM}}_{\mathrm{Walden}}$ is close to 16.2 fJ/conv.-step. Full article

Computing data-intensive applications on the von Neumann architecture lead to significant performance and energy overheads. The concept of computation in memory (CiM) addresses the bottleneck of von Neumann machines by reducing the data movement in the computing system. Emerging resistive non-volatile memory technologies, as well as volatile memories (SRAM and DRAM), can be used to realize architectures based on the CiM paradigm. In this paper, we propose a hybrid cell design to provide the opportunity for CiM by combining the magnetic tunnel junction (MTJ) and the conventional 6T-SRAM cell. The cell performs CiM operations based on stateful in-array computation, which has better scalability for multiple operands compared with stateless computation in the periphery. Various logic operations such as XOR, OR, and IMP can be performed with the proposed design. In addition, the proposed cell can also operate as a conventional memory cell to read and write volatile as well as non-volatile data. The obtained simulation results show that the proposed CiM-A design can increase the performance of regular memory architectures by reducing the delay by 8 times and the energy by 13 times for database query applications consisting of consecutive bitwise operations with minimum overhead. Full article

A bootstrapping technique used to increase the intrinsic voltage gain of a bulk-driven MOS transistor is described in this paper. The proposed circuit incorporates a capacitor and a cutoff transistor to be connected to the gate terminal of a bulk-driven MOS device, thus achieving a quasi-floating-gate structure. As a result, the contribution of the gate transconductance is cancelled out and the voltage gain of the device is correspondingly increased. The technique allows for implementing a voltage follower with a voltage gain much closer to unity as compared to the conventional bulk-driven case. This voltage buffer, along with a pseudo-resistor, is used to design a linearized transconductor. The proposed transconductance cell includes an economic continuous tuning mechanism that permits programming the effective transconductance in a range sufficiently wide to counteract the typical variations that process parameters suffer during fabrication. The transconductor has been used to implement a second-order $G_m$-C bandpass filter with a relatively high selectivity factor, suited for multi-frequency bioimpedance analysis in a very low-voltage environment. All the circuits have been designed in 180 nm CMOS technology to operate with a 0.6-V single-supply voltage. Simulated results show that the proposed technique allows for increasing the linearity and reducing the input-referred noise of the bootstrapped bulk-driven MOS transistor, which results in an improvement of the overall performance of the transconductor. The center frequency of the bandpass filter designed can be programmed in the frequency range from 6.5 kHz to 37.5 kHz with a power consumption ranging between 1.34 $\mathsf{\mu }$W and 2.19 $\mathsf{\mu }$W. The circuit presents an in-band integrated noise of 190.5 $\mathsf{\mu }$V${}_{rms}$ and is able to process signals of 110 mV${}_{pp}$ with a THD below −40 dB, thus leading to a dynamic range of 47.4 dB. Full article

The edge computing paradigm for Internet-of-Things brings computing closer to data sources, such as environmental sensors and cameras, using connected smart devices. Over the last few years, research in this area has been both interesting and timely. Typical services like analysis, decision, and control, can be realized by edge computing nodes executing full-fledged algorithms. Traditionally, low-power smart edge devices have been realized using resource-constrained systems executing machine learning (ML) algorithms for identifying objects or features, making decisions, etc. Initially, this paper discusses recent advances in embedded systems that are devoted to energy-efficient ML algorithm execution. A survey of the mainstream embedded computing devices for low-power IoT and edge computing is then presented. Finally, CYSmart is introduced as an innovative smart edge computing system. Two operational use cases are presented to illustrate its power efficiency. Full article

Self-sustainable energy harvesting for Internet of Things devices is challenging since ambient energy may be sporadic and unpredictable. This situation leads to frequent power failures that lead to intermittent operations, which prevent the reliability of data communications. This article presents fundamental hardware circuitry that enables reliable intermittent communications over wireless batteryless node networks. We emphasize two main mechanisms that ensure energy awareness and reliability: energy status-sharing and synchronized operation. We introduce novel low-power and self-sustainable plug-and-play circuits to support these mechanisms. Full article

We present a transfer learning method inspired by modulatory neurotransmitter mechanisms in biological brains and explore applications for neuromorphic hardware. In this method, the pre-trained weights of an artificial neural network are held constant and a new, similar task is learned by manipulating the firing sensitivity of each neuron via a supplemental bias input. We refer to this as neuromodulatory tuning (NT). We demonstrate empirically that neuromodulatory tuning produces results comparable with traditional fine-tuning (TFT) methods in the domain of image recognition in both feed-forward deep learning and spiking neural network architectures. In our tests, NT reduced the number of parameters to be trained by four orders of magnitude as compared with traditional fine-tuning methods. We further demonstrate that neuromodulatory tuning can be implemented in analog hardware as a current source with a variable supply voltage. Our analog neuron design implements the leaky integrate-and-fire model with three bi-directional binary-scaled current sources comprising the synapse. Signals approximating modulatory neurotransmitter mechanisms are applied via adjustable power domains associated with each synapse. We validate the feasibility of the circuit design using high-fidelity simulation tools and propose an efficient implementation of neuromodulatory tuning using integrated analog circuits that consume significantly less power than digital hardware (GPU/CPU). Full article

Tunnel Field-Effect Transistors (TFETs) have been considered one of the most promising technologies to complement or replace CMOS for ultra-low-power applications, thanks to their subthreshold slope below the well-known limit of 60 mV/dec at room temperature holding for the MOSFET technologies. Nevertheless, TFET technology still suffers of ambipolar conduction, limiting its applicability in digital systems. In this work, we analyze through SPICE simulations, the impact of the symmetric and asymmetric ambipolarity in failure and power consumption for TFET-based complementary logic circuits. Our results clarify the circuit-level effects induced by the ambipolarity feature, demonstrating that it affects the correct functioning of logic gates and strongly impacts power consumption. We believe that our outcomes motivate further research towards technological solutions for ambipolarity suppression in TFET technology for near-future ultra-low-power applications. Full article

Convolutional Neural Networks (CNNs), due to their recent successes, have gained lots of attention in various vision-based applications. They have proven to produce incredible results, especially on big data, that require high processing demands. However, CNN processing demands have limited their usage in embedded edge devices with constrained energy budgets and hardware. This paper proposes an efficient new architecture, namely Ocelli includes a ternary compute pixel (TCP) consisting of a CMOS-based pixel and a compute add-on. The proposed Ocelli architecture offers several features; (I) Because of the compute add-on, TCPs can produce ternary values (i.e., −1, 0, +1) regarding the light intensity as pixels’ inputs; (II) Ocelli realizes analog convolutions enabling low-precision ternary weight neural networks. Since the first layer’s convolution operations are the performance bottleneck of accelerators, Ocelli mitigates the overhead of analog buffers and analog-to-digital converters. Moreover, our design supports a zero-skipping scheme to further power reduction; (III) Ocelli exploits non-volatile magnetic RAMs to store CNN’s weights, which remarkably reduces the static power consumption; and finally, (IV) Ocelli has two modes, including sensing and processing. Once the object is detected, the architecture switches to the typical sensing mode to capture the image. Compared to the conventional pixels, it achieves an average 10% efficiency on its lane detection power consumption compared with existing edge detection algorithms. Moreover, considering different CNN workloads, our design shows more than 23% power efficiency over conventional designs, while it can achieve better accuracy. Full article

Machine-learning accelerators rely on floating-point matrix and vector multiplication kernels. To reduce their cost, customized many-term fused architectures are preferred, which improve the latency, power, and area of the designs. In this work, we design a parameterized fused many-term floating-point dot product architecture that is ready for high-level synthesis. In this way, we can exploit the efficiency offered by a well-structured fused dot-product architecture and the freedom offered by high-level synthesis in tuning the design’s pipeline to the selected floating-point format and architectural constraints. When compared with optimized dot-product units implemented directly in RTL, the proposed design offers lower-latency implementations under the same clock frequency with marginal area savings. This result holds for a variety of floating-point formats, including standard and reduced-precision representations. Full article

In this study, a direct-grown helical-shaped tungsten-oxide-based (h-WO_x) selection device is presented for emerging memory applications. The selectivity in the selection devices is from 10 to 10³ with a low off-current of 0.1 to 0.01 nA. In addition, the selectivity of volatile switching in the h-WO_x selection devices is reconfigurable with a pseudo RESET process on the one-time negative voltage operations. The helical-shaped selection devices with the glancing angle deposition (GLAD) method show good compatibility, low power consumption, good selectivity, and good reconfigurability for next-generation memory applications. Full article

Memristive neuromorphic systems represent one of the most promising technologies to overcome the current challenges faced by conventional computer systems. They have recently been proposed for a wide variety of applications, such as nonvolatile computer memory, neuroprosthetics, and brain–machine interfaces. However, due to their intrinsically nonlinear characteristics, they present a very complex dynamic behavior, including self-sustained oscillations, seizure-like events, and chaos, which may compromise their use in closed-loop systems. In this work, a novel intelligent controller is proposed to suppress seizure-like events in a memristive circuit based on the Hodgkin–Huxley equations. For this purpose, an adaptive neural network is adopted within a Lyapunov-based nonlinear control scheme to attenuate bursting dynamics in the circuit, while compensating for modeling uncertainties and external disturbances. The boundedness and convergence properties of the proposed control scheme are rigorously proved by means of a Lyapunov-like stability analysis. The obtained results confirm the effectiveness of the proposed intelligent controller, presenting a much improved performance when compared with a conventional nonlinear control scheme. Full article

IoT embedded systems have multiple objectives that need to be maximized simultaneously. These objectives conflict with each other due to limited resources and tradeoffs that need to be made. This requires multi-objective optimization (MOO) and multiple Pareto-optimal solutions are possible. In such a case, tradeoffs are made w.r.t. a user-defined preference. This work presents a general Multi-objective Reinforcement Learning (MORL) framework for MOO of IoT embedded systems. This framework comprises a general Multi-objective Markov Decision Process (MOMDP) formulation and two novel low-compute MORL algorithms. The algorithms learn policies to tradeoff between multiple objectives using a single preference parameter. We take the energy scheduling problem in general Energy Harvesting Wireless Sensor Nodes (EHWSNs) as a case example in which a sensor node is required to maximize its sensing rate, and transmission performance as well as ensure long-term uninterrupted operation within a very tight energy budget. We simulate single-task and dual-task EHWSN systems to evaluate our framework. The results demonstrate that our MORL algorithms can learn better policies at lower learning costs and successfully tradeoff between multiple objectives at runtime. Full article

RISC-V is a modern Instruction Set Architecture (ISA) that, by its open nature in combination with a clean and modular design, has enormous potential to become a game changer in the Internet of Things (IoT) era. Recently, SystemC-based Virtual Prototypes (VPs) have been introduced into the RISC-V ecosystem to lay the foundation for advanced industry-proven system-level use-cases. However, VP-driven environment modeling and interaction have mostly been neglected in the RISC-V context. In this paper, we propose such an extension to broaden the application domain for virtual prototyping in the RISC-V context. As a foundation, we built upon the open source RISC-V VP available at GitHub. For a visualization of the environment purposes, we designed a Graphical User Interface (GUI) and designed appropriate libraries to offer hardware communication interfaces such as GPIO and SPI from the VP to an interactive environment model. Our approach is designed to be integrated with SystemC-based VPs that leverage a Transaction-Level Modeling (TLM) communication system to prefer a speed-optimized simulation. To show the practicability of an environment model, we provide a set of building blocks such as buttons, LEDs and an OLED display and configured them in two demonstration environments. Moreover, for rapid prototyping purposes, we provide a modeling layer that leverages the dynamic Lua scripting language to design components and integrate them with the VP-based simulation. Our evaluation with two different case-studies demonstrates the applicability of our approach in building virtual environments effectively and correctly when matching the real physical systems. To advance the RISC-V community and stimulate further research, we provide our extended VP platform with the environment configuration and visualization toolbox, as well as both case-studies as open source on GitHub. Full article

Object detection is a predominant challenge in computer vision and image processing to detect instances of objects of various classes within an image or video. Recently, a new domain of vehicular platforms, e-scooters, has been widely used across domestic and urban environments. The driving behavior of e-scooter users significantly differs from other vehicles on the road, and their interactions with pedestrians are also increasing. To ensure pedestrian safety and develop an efficient traffic monitoring system, a reliable object detection system for e-scooters is required. However, existing object detectors based on IoU loss functions suffer various drawbacks when dealing with densely packed objects or inaccurate predictions. To address this problem, a new loss function, balanced-IoU (BIoU), is proposed in this article. This loss function considers the parameterized distance between the centers and the minimum and maximum edges of the bounding boxes to address the localization problem. With the help of synthetic data, a simulation experiment was carried out to analyze the bounding box regression of various losses. Extensive experiments have been carried out on a two-stage object detector, MASK_RCNN, and single-stage object detectors such as YOLOv5n6, YOLOv5x on Microsoft Common Objects in Context, SKU110k, and our custom e-scooter dataset. The proposed loss function demonstrated an increment of 3.70% at $A{P}_S$ on the COCO dataset, 6.20% at AP55 on SKU110k, and 9.03% at AP80 of the custom e-scooter dataset. Full article

In the new Internet of Things (IoT) era, embedded Field-Programmable Gate Array (FPGA) technology is enabling the deployment of custom-tailored embedded IoT solutions for handling different application requirements and workloads. Combined with the open RISC-V Instruction Set Architecture (ISA), the FPGA technology provides endless opportunities to create reconfigurable IoT devices with different accelerators and coprocessors tightly and loosely coupled with the processor. When connecting IoT devices to the Internet, secure communications and data exchange are major concerns. However, adding security features requires extra capabilities from the already resource-constrained IoT devices. This article presents the FAC-V coprocessor, which is an FPGA-based solution for an RISC-V processor that can be deployed following two different coupling styles. FAC-V implements in hardware the Advanced Encryption Standard (AES), one of the most widely used cryptographic algorithms in IoT low-end devices, at the cost of few FPGA resources. The conducted experiments demonstrate that FAC-V can achieve performance improvements of several orders of magnitude when compared to the software-only AES implementation; e.g., encrypting a message of 16 bytes with AES-256 can reach a performance gain of around 8000× with an energy consumption of 0.1 $\mathsf{\mu }$J. Full article

This paper proposes a novel approach suitable to design energy-efficient approximate multipliers using both ASIC and FPGAs. The new strategy harnesses specific encoding logics based on bit significance and computes the approximate product performing accurate sub-multiplications by applying an unconventional approach instead of using approximate computational modules implementing traditional static or dynamic bit-truncation approaches. The proposed platform-independent architecture exhibits an energy saving of up to 80% over the accurate counterparts and significantly better behavior in terms of accuracy loss with respect to competitor approximate architectures. When employed in 2D digital filters and edge detectors, the novel approximate multipliers lead to an energy consumption up to ~82% lower than the accurate counterparts, which is up to ~2 times higher than that obtained by state-of-the-art competitors. Full article