Applications of Binary Neural Networks (BNNs) are promising for embedded systems with hard constraints on energy and computing power. Contrary to conventional neural networks using floating-point datatypes, BNNs use binarized weights and activations to reduce memory and computation requirements. Memristors, emerging non-volatile memory devices, show great potential as a target implementation platform for BNNs by integrating storage and compute units. However, the efficiency of this hardware highly depends on how the network is mapped and executed on these devices. In this paper, we propose an efficient implementation of XNOR-based BNN to maximize parallelization. In this implementation, costly analog-to-digital converters are replaced with sense amplifiers with custom reference(s) to generate activation values. Besides, a novel mapping is introduced to minimize the overhead of data communication between convolution layers mapped to different memristor crossbars. This comes with extensive analytical and simulation-based analysis to evaluate the implication of different design choices considering the accuracy of the network. The results show that our approach achieves up to $5\times$ energy-saving and $100\times$ improvement in latency compared to baselines.@en

While Resistive RRAM (RRAM) offers attractive features for artificial neural networks (NN) such as low power operation and high-density, its conductance variation can pose significant challenges when the storage of synaptic weights is concerned. This paper reports an experimental evaluation of the conductance variations of manufactured RRAMs at the memory array level. Working at the memory array level allows to catch cycle-to-cycle (C2C) as well as device-to-device (D2D) variability and, hence, to propose a realistic evaluation of the conductance variation. Variability is evaluated with respect to the RRAM low resistance state (LRS) and high resistance state (HRS) conductance ratio. This ratio is selected as the parameter of interest as it guarantees the proper operation of the RRAM: the larger the ratio, the more reliable and robust the RRAM cell is in storing and retrieving data. The measurement results show that the conductance ratio is heavily affected by variability. Large spatial and temporal variations are reported, making challenging RRAM-based analog weight storage.

@en

Binary Neural Networks (BNNs) have demonstrated significant advantages in reducing computation and memory costs, all while maintaining acceptable accuracy on various image detection tasks. Thus, BNNs have the potential to support practical cognitive tasks on resource-constrained platforms, such as edge computing devices. To realize this, SRAM-based digital Computation-in-Memory (CIM) has gained growing attention as it overcomes the analog CIM architecture bottlenecks such as limited computing accuracy due to process variation, non-linearity, power and area-hungry Analog-to-Digital Converters (ADCs), etc. However, digital CIM architectures are highly dominated by power-hungry adder-trees, which can nullify the benefits of SRAM-based digital CIM. To address this issue, this paper proposes an adder free SRAM-based digital CIM, AFSRAM-CIM, for BNN acceleration. The proposed CIM architecture utilizes a multi-functional 10-T SRAM cell-based crossbar array and a new energy-efficient approach to perform the popcount operation. Simulation results using the MNIST dataset show that the proposed architecture maintains the state-of-the-art inference accuracy of 99.21% with only 11.86 fJ energy per operation. Moreover, AFSRAM-CIM achieves over 3× energy and ≈17× area savings when compared to the conventional digital CIM approaches.@en

Resistive Random Access Memories (RRAMs) are now undergoing commercialization, with substantial investment from many semiconductor companies. However, due to the immature manufacturing process, RRAMs are prone to exhibit new failure mechanisms and faults, which should be efficiently detected for high-volume production. Some of those faults are hard-to-detect, and require specific Design-for-Testability (DfT) circuit design. This paper proposes a DfT based on a parallel-reference write circuit that can detect all single-cell RRAM array faults: strong faults (directly causing logic errors) as well as weak faults (caused by parametric deviations). The scheme replaces the regular write driver, and enables the monitoring and comparison of the write current against multiple references during a single write operation. Hence, it serves as a DfT scheme and as a normal write circuit simultaneously. In addition, it enhances production testing speed and online fault detection, while keeping the area overhead low. Furthermore, the DfT is configurable for efficient diagnosis and yield learning. The results of the simulations performed do not only show that the DfT can detect single-cell conventional faults (due to interconnects and contacts) as well as unique RRAM faults (based on silicon data) that have been demonstrated to exist, but also that the DfT is robust to process variations.@en

Recent theoretical and experimental spintronics developments clearly indicate that Spin Waves (SW) interference based Majority gates (MAJ3) open an alternative road towards ultra low-power circuit implementations potentially capable to outperform CMOS counterparts. However, hurdles still exist, e.g., gate cascading, as due to the very nature of SW interference MAJ3 gates are not input output coherent, i.e., produce output waves with different amplitudes corresponding to ‘weak’ and ‘strong’ majority, which precludes their direct cascading within the SW domain. State of the art designs address this issue by means of hybrid SW-CMOS systems that make use on domain converters, which are energy expensive and diminish if not nullify the ultra-low power promise of the SW-based computing paradigm. In this paper we propose a gate cascading technique that rely on the natural SW amplitude decay while propagating through a magnetic conduit. We demonstrate the concept by constructing the building block MAJ3(MAJ3(In1,In2,In3),In4,In5) and verifying its correct behaviour by means of micromagnetic simulations. We evaluate the proposed design in terms of energy consumption, delay, and area, and our calculations indicate that, when compared with its domain conversion counterpart, our proposal consumes 49.5% less energy, at the expense of 48% area and 76.5% delay overhead, respectively.@en

Resistive Random-Access Memories (ReRAMs) represent a promising candidate to complement and/or replace CMOS-based memories used in several emerging applications. Despite all the advantages of using these novel memories, mainly due to the memristive device's CMOS manufacturing process compatibility, zero standby power consumption, as well as, high scalability and density, the use of them in real applications depends on being able to guarantee their quality after manufacturing. As observed in CMOS-based memories, ReRAMs are also susceptible to manufacturing deviations, defects, and process variations, that can cause faulty behaviors different from the ones observed in CMOS technology, increasing not only the manufacturing test complexity but also the time required to perform the test. In this context, this paper proposes to study the use of temperature to facilitate fault propagation in ReRAMs, reducing the required test time. A case study composed of a 3x3 word-based ReRAM with peripheral circuitry implemented based on a 130 nm Predictive Technology Model (PTM) library was adopted. During the proposed study, a total of 17 defects were injected in different positions of the ReRAM cell, and their respective faulty behavior was classified into traditional and unique faults, considering three temperatures (25, 100, and -40°C). The obtained results show that the temperature can, depending on the position of the defect, facilitate fault propagation, which reduces the time required for performing manufacturing testing.

@en

Spiking Neural Networks (SNNs) are Artificial Neural Networks which promise to mimic the biological brain processing with unsupervised online learning capability for various cognitive tasks. However, SNN hardware implementation with online learning support is not trivial and might prove highly inefficient. This paper proposes an energy-efficient hardware implementation for SNN synapses. The implementation is based on parallel-connected Magnetic Tunnel Junction (MTJ) devices and exploits their inherent stochasticity. In addition, it uses a dedicated unsupervised learning rule based on optimized Spike-Timing-Dependent Plasticity (STDP). To facilitate the design of the SNN, its training and evaluation, an open-source Python-based platform is developed; it takes as input the SNN parameters and discrete circuit components, and it automatically generates the associated full netlist in SPICE; moreover, it extracts the simulation results and makes them available in python for evaluation and manipulation. Unlike conventional neuromorphic hardware that relies on simple weight mapping post-off-line training, our approach emphasizes continuous, unsupervised learning, ensuring an energy efficiency of 11.2nW per synaptic update during training and as low as 109fJ/spike during inference.@en

Memristor technology has shown great promise for energy-efficient computing [1] , though it is still facing many challenges [1 , 2]. For instance, the required additional costly electroforming to establish conductive pathways is seen as a significant drawback as it contributes to power and area overheads, and limited device endurance. In this work, we propose a novel forming-free HfO₂ -based ReRAM device with low operating voltages , multi-level capability , and less sensitivity to device-to-device (D2D) and cycle-to-cycle (C2C) variations. The device is fabricated using CMOS-compatible processes, excluding the undesirable complex steps mandatory to manufacture the state-of-the-art forming-free devices [3, 4, 5]. This is accomplished by utilizing the desirable formation energy of Pd-O bonds [6, 7], which creates conducting paths at room temperature while maintaining the analog switching ability of the devices. The proposed ReRAM device holds a great value for dense memories and energy-efficient compute architectures.

@en

The Advanced Encryption Standard (AES) is widely recognized as a robust cryptographic algorithm utilized to protect data integrity and confidentiality. When it comes to lightweight implementations of the algorithm, the literature mainly emphasizes area and power optimization, often overlooking considerations related to performance and security. This paper evaluates two of our previously proposed lightweight AES implementations using both profiled and non-profiled attacks. One is an unprotected implementation, and the other one is a protected version using Domain-Oriented Masking (DOM). The findings of this study indicate that the inclusion of DOM in the design enhances its resistance to attacks at the cost of doubling the area.@en

Current Spin Wave (SW) state-of-the-art computing relies on wave interference for achieving low power circuits. Despite recent progress, many hurdles, e.g., gate cascading, fan-out achievement, still exist. In a previous work, we introduced a novel SW phase shift based computation paradigm and demonstrated that an n-input Threshold Logic Gate (TLG) can be implemented with n + 1 phase shifters operating on the same SW. In this paper we further develop this concept by introducing a phase shift amount reading method by means of parametric amplification. We make use of 3-input Majority Gate (MAJ3) as discussion vehicle and introduce a novel majority function evaluation approach which postpone the threshold related calculations to the gate output readout stage. Subsequently, we verify this principle by means of micromagnetic simulations and discus the results. Finally, we utilize the proposed MAJ3 gate to implement a collection of representative logic circuits from the EPFL Combinational Benchmarking Suite and evaluate and compare their area, energy consumption, and Energy Area Product (EAP) with the ones of 7 nm CMOS technology node based counterpart imple-mentations. Our estimations indicate that EAP_CMOS/EAP_SW average value is 5.25 and 2.2 for a SW transducer feature size of 20 nm and 30 nm, respectively.

@en

Neuromorphic processors promise low-latency and energy-efficient processing by adopting novel brain-inspired design methodologies. Yet, current neuromorphic solutions still struggle to rival conventional deep learning accelerators' performance and area efficiency in practical applications. Event-driven data-flow processing and near/in-memory computing are the two dominant design trends of neuromorphic processors. However, there remain challenges in reducing the overhead of event-driven processing and increasing the mapping efficiency of near/in-memory computing, which directly impacts the performance and area efficiency. In this work, we discuss these challenges and present our exploration of optimizing event-based neural network inference on SENECA, a scalable and flexible neuromorphic architecture. To address the overhead of event-driven processing, we perform comprehensive design space exploration and propose spike-grouping to reduce the total energy and latency. Furthermore, we introduce the event-driven depth-first convolution to increase area efficiency and latency in convolutional neural networks (CNNs) on the neuromorphic processor. We benchmarked our optimized solution on keyword spotting, sensor fusion, digit recognition and high resolution object detection tasks. Compared with other state-of-the-art large-scale neuromorphic processors, our proposed optimizations result in a 6× to 300× improvement in energy efficiency, a 3× to 15× improvement in latency, and a 3× to 100× improvement in area efficiency. Our optimizations for event-based neural networks can be potentially generalized to a wide range of event-based neuromorphic processors.

@en

State-of-the-Art (SotA) hardware implementations of Deep Neural Networks (DNNs) incur high latencies and costs. Binary Neural Networks (BNNs) are potential alternative solutions to realize faster implementations without losing accuracy. In this paper, we first present a new data mapping, called TacitMap, suited for BNNs implemented based on a Computation-In-Memory (CIM) architecture. TacitMap maximizes the use of available parallelism, while CIM architecture eliminates the data movement overhead. We then propose a hardware accelerator based on optical phase change memory (oPCM) called EinsteinBarrier. Ein-steinBarrier incorporates TacitMap and adds an extra dimension for parallelism through wavelength division multiplexing, leading to extra latency reduction. The simulation results show that, compared to the SotA CIM baseline, TacitMap and EinsteinBarrier significantly improve execution time by up to ∼ 154× and ∼ 3113×, respectively, while also maintaining the energy consumption within 60% of that in the CIM baseline.

@en

Memristive devices have become promising candidates to complement the CMOS technology, due to their CMOS manufacturing process compatibility, zero standby power consumption, high scalability, as well as their capability to implement high-density memories and new computing paradigms. Despite these advantages, memristive devices are susceptible to manufacturing defects that may cause faulty behaviors not observed in CMOS technology, significantly increasing the challenge of testing these novel devices after manufacturing. This work proposes an optimized Design-for-Testability (DfT) strategy based on the introduction of a DfT circuitry that measures the current consumption of Resistive Random Access Memory (ReRAM) cells to detect not only traditional but also unique faults. The new DfT circuitry was validated using a case study composed of a 3x3 word-based ReRAM with peripheral circuitry implemented based on a 130 nm Predictive Technology Model (PTM) library. The obtained results demonstrate the fault detection capability of the proposed strategy with respect to traditional and unique faults. In addition, this paper evaluates the impact related to the DfT circuitry’s introduced overheads as well as the impact of process variation on the resolution of the proposed DfT circuitry.

@en

Effectively feeding a burgeoning world population is one of the main goals of sustainable agricultural practices. Digital technology, such as edge artificial intelligence (AI), has the potential to introduce substantial benefits to agriculture by enhancing farming practices that can improve agricultural production efficiency, yield, quality and safety. However, the adoption of edge AI faces several challenges, including the need for innovative and efficient edge AI solutions and greater investment in infrastructure and training, all compounded by various environmental, social and economic constraints. Here we provide a roadmap for leveraging edge AI at the intersection of food production and sustainability.

@en

Guaranteeing high-quality test solutions for Spin-Transfer Torque Magnetic RAM (STT-MRAM) is a must to speed up its high-volume production. A high test quality requires maximizing the fault coverage. Detecting permanent faults is relatively simple compared to intermittent faults; the latter are faults (caused by non-environmental conditions) that appear and disappear as a function of time, and are therefore hard to detect. Testing for such faults in STT-MRAMs is even worse considering the Magnetic Tunneling Junction inherent property ‘intrinsic switching stochasticity’, which results in inevitable random write errors. This paper presents a novel Design-for-Testability (DFT) scheme for detecting intermittent faults in STT-MRAMs; it is based on monitoring the write current. The strength of the write current is inversely correlated to the write error rate; when the write current is smaller than the specification, the device is considered faulty. A reduction in the write current can be caused by any defect in the write path of the memory (e.g., interconnects and contacts). Simulation results based on industrial design show that applying DFT yields a superior coverage of intermittent faults compared to functional test methods, such as march tests.@en

While Spin Waves (SW) interaction provides natural support for low power Majority (MAJ) gate implementations many hurdles still exists on the road towards the realization of practically relevant SW circuits. In this paper we leave the SW interaction avenue and propose Threshold Logic (TL) inspired SW computing, which relies on successive phase rotations applied to one single SW instead of on the interference of an odd number of SWs. After providing a short TL inside we introduce the SW TL gate concept and discuss the way to mirror TL gate weight and threshold values into physical phase-shifter parameters. Subsequently, we design and demonstrate proper operation of a SW TL based Full Adder (FA) by means of micro-magnetic simulations. We conclude the paper by providing inside on the potential advantages of our proposal by means of a conceptual comparison of MAJ and TL based FA implementations.@en

Artificial Intelligence (AI) supported by Deep Artificial Neural Networks (ANNs) is booming and already used in many applications, with impressive results, and we are still its infancy. For many sensing applications it would be advantageous if we could move AI from cloud to Edge. However this requires huge improvements in energy-efficiency. The CONVOLVE project (convolve.eu) aims at enabling smart edge devices through a concerted effort at all layers of the design stack. This ranges from using much more efficient models and mappings, like exploiting Spiking Neural Networks (SNNs), to new processing architectures, like compute-in-memory (CIM), use of approximation, and using new device technology, like memristors. However these latter changes make HW more susceptible to noise and other disturbances. Online continuous learning (i.e. adapting weights) may alleviate these problems. This paper shows several CONVOLVE developments in the crucial areas of CIM architectures, SNN accelerators and online learning.

@en