Addressing non-idealities in Resistive Random Access Memories (RRAMs) is crucial for their successful commercialization. For example, the inherent resistance drift that occurs during consecutive read operations can induce Read Disturb Faults (RDF), leading to functional errors. This paper analyzes and characterizes the resistance drift and the RDF based on data measurements and presents a physics-based RRAM compact model that incorporates these non-idealities. Additionally, an in-field mitigation scheme is proposed, leveraging bidirectional read operations to balance the resistance. The scheme is implemented and validated through circuit simulations, both for RRAM used as memory and for RRAM-based computation-in-memory microarchitectures for deep neural networks. The results demonstrate that RRAM without any mitigation scheme can start failing after 8,000 consecutive reads, while our mitigation scheme ensures that the memory remains functional even after 10⁶ consecutive reads. Furthermore, the results indicate that using the MNIST dataset as a case study, the accuracy can drop significantly from 86% to as low as 12.5% without any mitigation scheme. In contrast, the proposed mitigation scheme improves this accuracy up to 84.2%.

The goal of the NEUROKIT2E project is to create an open-source Deep Learning framework for edge and embedded AI built around an established European value chain. This framework, called AIDGE, supports a wide range of application areas that operate independently and serve a global user community. It provides easy and fast full-stack solutions from Neural Network design and optimization to AI application development all the way down to hardware implementations while enabling code generation for application-specific targets. This platform provides flexibility for academic users in the AI domain to explore and innovate while allowing them the possibility to prototype systems, ensuring their work aligns well with industrial needs. This paper presents the results and achievements of the first part of this three-year project, along with its roadmap and expected outcomes.

Advancements in Artificial Intelligence (AI) and Internet-of-Things (IoT) have increased demand for edge AI, but deployment on traditional AI accelerators, like GPUs and TPUs, using von Neumann architecture, suffer from inefficiencies due to separate memory and compute units. Computation-in-Memory (CIM), utilizing non-volatile memristor devices to leverage analog computing principles and perform in-place computations, holds great potential in improving computational efficiency by eliminating frequent data movement. However, standard implementation of CIM faces several challenges, primarily high power consumption and subsequently induced nonlinearity, debating its viability for edge devices. In this paper, we propose C3CIM, a novel memristor-based CIM micro-architecture, featuring a new bit-cell and array design, targeting efficient implementation of Neural Networks (NN). Our architecture uses a constant current source to perform Multiply-and-Accumulate (MAC) operations with a very low computation current (10 to 100 nA), thereby significantly enhancing power efficiency. We adapted C3CIM for Spiking Neural Networks (SNN) and developed a prototype using TSMC 40nm CMOS node for on-silicon validation. Furthermore, our micro-architecture was benchmarked using two SNN models based on N-MNIST and IBM-Gesture datasets, for comparison against current state-of-the-art (SOTA). Results show up to 35x reduction in power along with 6.7x saving in energy compared to SOTA, demonstrating promising potential of this work for edge AI applications.

Timely identification of cardiac arrhythmia (abnormal heartbeats) is vital for early diagnosis of cardiovascular diseases. Wearable healthcare devices facilitate this process by recording heartbeats through electrocardiogram (ECG) signals and using AI-driven hardware to classify them into arrhythmia classes. Spiking neural networks (SNNs) are well-suited for such hardware as they consume low energy due to event-driven operation. However, their energy-efficiency and accuracy are constrained by encoding methods that translate real-valued ECG data into spikes. In this paper, we present an SNN-based ECG classification architecture featuring a new adaptive multi-threshold spike encoding scheme. This scheme adjusts encoding window and granularity based on the importance of ECG data samples, to capture essential information with fewer spikes. We develop a high-accuracy SNN model for such spike representation, by proposing a technique specifically tailored to our encoding. We design a hardware architecture for this model, which incorporates optimized layer post-processing for energy-efficient data-flow and employs fixed-point quantization for computational efficiency. Moreover, we integrate this architecture with our encoding scheme into a system-on-chip implementation using TSMC 40 nm technology. Our approach provides up to 5.1x energy-efficiency compared to state-of-the-art SNN-based ECG classifiers, with high accuracy.

Computational-In-Memory (CIM) is an energy-efficient paradigm that integrates computation directly within memory arrays, reducing the bottleneck associated with data transfer. This approach is beneficial for Artificial Intelligence (AI) applications that require on-chip learning for real-time processing. However, implementing on-chip learning in CIM architectures remains challenging due to limited throughput and energy-efficiency during both online training and inference. In conventional architectures, weight updates necessitate the inference process to halt to avoid unintended computation outcomes. To overcome this limitation, this paper presents a novel Spin-Orbit Torque (SOT)-based CIM architecture tailored for continuous on-chip learning applications, which enable weight updates without interrupting the inference. The proposed SOT bit-cell utilizes two read ports and one write port (2R1W) configuration, where one read port (1R) is dedicated to inference and one read and one write (1R1W) for on-chip learning that enables concurrent read and write operations. Our proposed architecture is evaluated at the system-level using the Generic-PDK 45 nm technology node, demonstrating 2.4× improvement in energy-efficiency and 5.4× improvement in throughput compared to state-of-the-art solutions, with minimal overhead.

Approximately one-third of individuals with chronic epilepsy, a condition resulting from uncontrolled brain activity, do not respond to medication. Animal models are widely used to investigate the mechanism underlying epilepsy, so better drug treatments can be developed for this disease. In such studies, epileptiform activity, assessed by EEG recordings, can be used as a marker for the development of the disease. However, the analysis of EEG recordings is typically done manually, which is time-consuming, subject to observer bias, error-prone, and lacks consistency and efficiency. In this paper, we develop a novel automated methodology for detecting and classifying epileptiform activity, which is tested using the intrahippocampal kainic acid (IHKA) mouse model, a representation of human temporal lobe epilepsy. For that, EEG/LFP recordings are obtained from biological experiments using the IHKA mouse model for data acquisition. We use a spike detection method that combines an improved version of the nonlinear energy operator (NEO) with the automatic NEO thresholding (ANT) algorithm. The proposed method is implemented in Python as an automated and time-efficient algorithm, given its adaptability to different spike and epileptiform event criteria, making it suitable for use in preclinical and potentially future clinical studies. Using our proposed methodology, we achieve a 93.1% accuracy in detecting epileptiform events and a 95.8% accuracy in classification. Moreover, the time for analysis of EEG recordings was reduced by 98.8% compared to manual analysis. Additionally, to demonstrate the potential of the algorithm for brain–machine interfaces (BMI) applications, we develop a hardware architecture and implement it using both an application-specific integrated circuit (ASIC) and a field programmable gate array (FPGA). The FPGA shows the feasibility of near real-time implementation, and for our ASIC implementation, we achieve a post-layout area of 9114 µm² with a dynamic power consumption of 16.09 μW using TSMC 40 nm technology.

The IEEE European Test Symposium (ETS) has been facilitating progress in electronic systems testing since its launch in 1996. On the occasion of its 30th anniversary, this collaborative paper gathers sections by 21 ETS teams to outline their influential ideas and milestones. Each team's section highlights historical perspective, current research, frameworks and projects as well as forward-looking research agendas in the area of electronic-based circuits and systems testing, reliability, safety, security and validation. This anniversary summary documents how research of various ETS teams, exemplifying the test community, has been evolving and transitioning from concepts to practical standards and Electronic Design Automation (EDA) tools and flows. This legacy is a strong base to drive the next generation of advances in electronic systems testing.

Memristor-based Computation-In-Memory (CIM) has emerged as a compelling paradigm for designing energy-efficient neural network hardware. However, memristors suffer from conductance variation issue, which introduces computational errors in CIM hardware and leads to a degraded inference accuracy. In this paper, we present a hardware-aware quantization to mitigate the impact of conductance variation on CIM-based neural networks. We achieve this using the inherent characteristics of fixed-point arithmetic in CIM hardware. By tuning the bit-precision of weights, we align the conductance variation-induced errors with lower-order output bits. This reduces their numerical impact on the fixed-point output. We further decrease the residual errors by selectively discarding bits with low information and high error. This leads to error-free computations and a high inference accuracy. Our proposed methodology achieves 5.6× correct operations per unit energy compared to the conventional approach, while incurring very low hardware overheads.

The investigation of neural activity in the murine brain through electrophysiological recordings stands as a fun-damental pursuit within the domain of neuroscience. A specific area of keen interest within this field pertains to the scrutiny of Purkinje cells, nestled within the cerebellum, in order to gain insights into the mechanisms underlying brain injuries and the impairment of motor functions. Notably, Purkinje cells manifest two distinct types of spikes - complex and simple - a pivotal aspect for subsequent classification purposes. However, a critical challenge has persisted in the experimental paradigm: the prevailing setups necessitate the use of wired connections linking the mouse's head stage to data acquisition systems. This constraint substantially curtails the mouse's natural behavior during the course of experimentation, limiting the ability to study essential neural processes and motor function aspects over extended periods. In this paper, we propose a new architectural framework for the detection and classification of neuronal spikes originating from Purkinje cells. This system is engineered to exploit the distinct attributes of these neural entities, effectively winnowing out extraneous data while retaining the pertinent information. The resultant output is a refined dataset, amenable to convenient storage within the mouse's head stage, obviating the need for unwieldy wiring configurations. Our proposed implementation attains a classification accuracy of up to 98% on an in-vivo dataset. Furthermore, its compact form factor en-sures unhindered mobility for the experimental mouse, fostering naturalistic behaviors during the course of scientific inquiry.

Current Artificial Intelligence (AI) computation systems face challenges, primarily from the memory-wall issue, limiting overall system-level performance, especially for Edge devices with constrained battery budgets, such as smartphones, wearables, and Internet-of-Things sensor systems. In this paper, we propose a new SRAM-based Compute-In-Memory (CIM) accelerator optimized for Spiking Neural Networks (SNNs) Inference. Our proposed architecture employs a multiport SRAM design with multiple decoupled Read ports to enhance the throughput and Transposable Read-Write ports to facilitate online learning. Furthermore, we develop an Arbiter circuit for efficient data-processing and port allocations during the computation. Results for a 128×128 array in 3nm FinFET technology demonstrate a 3.1× improvement in speed and a 2.2× enhancement in energy efficiency with our proposed multiport SRAM design compared to the traditional single-port design. At system-level, a throughput of 44 MInf/s at 607 pJ/Inf and 29mW is achieved.

Computation-in-memory (CIM) using memristors can facilitate data processing within the memory itself, leading to superior energy efficiency than conventional von-Neumann architecture. This makes CIM well-suited for data-intensive applications like neural networks. However, a large number of read operations can induce an undesired resistance change in the memristor, known as read-disturb. As memristor resistances represent the neural network weights in CIM hardware, read-disturb causes an unintended change in the network’s weights that leads to poor accuracy. In this paper, we propose a methodology for read-disturb detection and mitigation in CIM-based neural networks. We first analyze the key insights regarding the read-disturb phenomenon. We then introduce a mechanism to dynamically detect the occurrence of read-disturb in CIM-based neural networks. In response to such detections, we develop a method that adapts the sensing conditions of CIM hardware to provide error-free operation even in the presence of read-disturb. Simulation results show that our proposed methodology achieves up to 2× accuracy and up to 2× correct operations per unit energy compared to conventional CIM architectures.

Diabetic retinopathy (DR) is a leading cause of permanent vision loss worldwide. It refers to irreversible retinal damage caused due to elevated glucose levels and blood pressure. Regular screening for DR can facilitate its early detection and timely treatment. Neural network-based DR classifiers can be leveraged to achieve such screening in a convenient and automated manner. However, these classifiers suffer from reliability issue where they exhibit strong performance during development but degraded performance after deployment. Moreover, they do not provide supplementary information about the prediction outcome, which severely limits their widespread adoption. Furthermore, energy-efficient deployment of these classifiers on edge devices remains unaddressed, which is crucial to enhance their global accessibility. In this paper, we present a reliable and energy-efficient hardware for DR detection, suitable for deployment on edge devices. We first develop a DR classification model using custom training data that incorporates diverse image quality and image sources along with improved class balance. This enables our model to effectively handle both on-field variations in retinal images and minority DR classes, enhancing its post-deployment reliability. We then propose a pseudo-binary classification scheme to further improve the model performance and provide supplementary information about the model prediction. Additionally, we present an energy-efficient hardware design for our model using memristor-based computation-in-memory, to facilitate its deployment on edge devices. Our proposed approach achieves reliable DR classification with three orders of magnitude reduction in energy consumption over state-of-the-art hardware platforms.

Electrophysiological recordings of neural activity in a mouse's brain are very popular among neuroscientists for understanding brain function. One particular area of interest is acquiring recordings from the Purkinje cells in the cerebellum in order to understand brain injuries and the loss of motor functions. However, current setups for such experiments do not allow the mouse to move freely and, thus, do not capture its natural behaviour since they have a wired connection between the animal's head stage and an acquisition device. In this work, we propose a lightweight neuronalspike detection and classification architecture that leverages on the unique characteristics of the Purkinje cells to discard unneeded information from the sparse neural data in real time. This allows the (condensed) data to be easily stored on a removable storage device on the head stage, alleviating the need for wires. Synthesis results reveal a >95% overall classification accuracy while still resulting in a small-form-factor design, which allows for the free movement of mice during experiments. Moreover, the power-efficient nature of the design and the usage of STT-RAM (Spin Transfer Torque Magnetic Random Access Memory) as the removable storage allows the head stage to easily operate on a tiny battery for up to approximately 4 days.

Deep Learning (DL) has recently led to remark-able advancements, however, it faces severe computation related challenges. Existing Von-Neumann-based solutions are dealing with issues such as memory bandwidth limitations and energy inefficiency. Computation-In-Memory (CIM) has the potential to address this problem by integrating processing elements directly into the memory architecture, reducing data movement and enhancing the overall efficiency of the system. In this work, we propose CIM architecture using three distinct emerging technologies. Firstly, a CIM architecture utilizing Ferroelectric Field-Effect Transistors (FeFET) is shown and the resulting errors from the analog compute scheme are injected into the emerging algorithm of Hyperdimensional Computing. Subsequently, we explore Vertical Nanowire Field-Effect Transistors (VNWFETs) based CIM within a 3D computing architecture, demonstrating improved energy efficiency and reconfigurability for CIM. Additionally, we improve the accuracy of the Resistive Random Access Memories (RRAM) based CIM architecture using two mapping-based solutions. These three technologies exhibit non-volatile characteristics, and when integrated into the CIM architecture, they yield significant advantages, including enhanced energy efficiency, reliability, and accuracy in computing processes.

Timely detection of cardiac arrhythmia characterized by abnormal heartbeats can help in the early diagnosis and treatment of cardiovascular diseases. Wearable healthcare devices typically use neural networks to provide the most convenient way of continuously monitoring heart activity for arrhythmia detection. However, it is challenging to achieve high accuracy and energy efficiency in these smart wearable healthcare devices. In this work, we provide architecture-level solutions to deploy neural networks for cardiac arrhythmia classification. We have created a hierarchical structure after analyzing various neural network topologies where only required network components are activated to improve energy efficiency while maintaining high accuracy. In our proposed architecture, we introduce a severity-based classification approach to directly help the users of the wearable healthcare device as well as the medical professionals. Additionally, we have employed computation-in-memory based hardware to improve energy efficiency and area consumption by leveraging in-situ data processing and scalability of emerging memory technologies such as resistive random access memory (RRAM). Simulation experiments conducted using the MIT-BIH arrhythmia dataset show that the proposed architecture provides high accuracy while consuming average energy of 0.11 $\mu$J per heartbeat classification and 0.11 mm2 area, thereby achieving 25× improvement in average energy consumption and 12× improvement in area compared to the state-of-the-art.

This paper addresses one of the directions of the HORIZON EU CONVOLVE project being dependability of smart edge processors based on computation-in-memory and emerging memristor devices such as RRAM. It discusses how how this alternative computing paradigm will change the way we used to do manufacturing test. In addition, it describes how these emerging devices inherently suffering from many non-idealities are calling for new solutions in order to ensure accurate and reliable edge computing. Moreover, the paper also covers the security aspects for future edge processors and shows the challenges and the future directions.

Analog computation-in-memory (CIM) architecture alleviates massive data movement between the memory and the processor, thus promising great prospects to accelerate certain computational tasks in an energy-efficient manner. However, data converters involved in these architectures typically achieve the required computing accuracy at the expense of high area and energy footprint which can potentially determine CIM candidacy for low-power and compact edge-AI devices. In this work, we present a memory-periphery co-design to perform accurate A/D conversions of analog matrix-vector-multiplication (MVM) outputs. Here, we introduce a scheme where select-lines and bit-lines in the memory are virtually fixed to improve conversion accuracy and aid a ring-oscillator-based A/D conversion, equipped with component sharing and inter-matching of the reference blocks. In addition, we deploy a self-timed technique to further ensure high robustness addressing global design and cycle-to-cycle variations. Based on measurement results of a 4Kb CIM chip prototype equipped with TSMC 40nm, a relative accuracy of up to 99.71% is achieved with an energy efficiency of 115.1 TOPS/W and computational density of 12.1 TOPS/mm2 for the MNIST dataset. Thus, an improvement of up to 11.3X and 7.5X compared to the state-of-the-art, respectively.

With the rise of deep learning (DL), our world braces for artificial intelligence (AI) in every edge device, creating an urgent need for edge-AI SoCs. This SoC hardware needs to support high throughput, reliable and secure AI processing at ultra-low power (ULP), with a very short time to market. With its strong legacy in edge solutions and open processing platforms, the EU is well-positioned to become a leader in this SoC market. However, this requires AI edge processing to become at least 100 times more energy-efficient, while offering sufficient flexibility and scalability to deal with AI as a fast-moving target. Since the design space of these complex SoCs is huge, advanced tooling is needed to make their design tractable. The CONVOLVE project (currently in Inital stage) addresses these roadblocks. It takes a holistic approach with innovations at all levels of the design hierarchy. Starting with an overview of SOTA DL processing support and our project methodology, this paper presents 8 important design choices largely impacting the energy efficiency and flexibility of DL hardware. Finding good solutions is key to making smart-edge computing a reality.

Emerging device technologies such as Resistive RAMs (RRAMs) are under investigation by many researchers and semiconductor companies; not only to realize e.g., embedded non-volatile memories, but also to enable energy-efficient computing making use of new data processing paradigms such as computation-in-memory. However, such devices suffer from various non-idealities and reliability failure mechanisms (e.g., variability, endurance, and retention); these negatively impact the memory robustness and the computation accuracy. This paper discusses the non-idealities and reliability failure mechanisms for RRAM devices, provides an overview on the most popular ones. In addition, it reports detailed anlysis of some of these based on data measurements. Finally, it presents two different mitigation schemes for RRAM based accelerators; one is based on RRAM non-ideality aware quantization and conductance control for neural network accuracy enhancement while the second is based on reliability-aware biased training technique.