Wearable systems for the continuous and real-time monitoring of cardiovascular diseases are becoming widespread and valuable assets in diagnosis and therapy. A promising approach for real-time analysis of the electrocardiographic (ECG) signal and the detection of heart conditions, such as arrhythmia, is represented by the transformer machine learning model. Transformers are powerful models for the classification of time series, although efficient implementation in the wearable domain raises significant design challenges, to combine adequate accuracy and a suitable complexity. In this work, we present a tiny transformer model for the analysis of the ECG signal, requiring only 6k parameters and reaching 98.97% accuracy in the recognition of the 5 most common arrhythmia classes from the MIT-BIH Arrhythmia database, assessed considering 8-bit integer inference as required for efficient execution on low-power microcontroller-based devices. We explored an augmentation-based training approach for improving the robustness against electrode motion artifacts noise, resulting in a worst-case post-deployment performance assessment of 98.36% accuracy. Suitability for wearable monitoring solutions is finally demonstrated through efficient deployment on the parallel ultra-low-power GAP9 processor, where inference execution requires 4.28ms and 0.09mJ.

Deep convolutional neural networks (CNNs) obtain outstanding results in tasks that require human-level understanding of data, like image or speech recognition. However, their computational load is significant, motivating the development of CNN-specialized accelerators. This work presents NEURA ghe , a flexible and efficient hardware/software solution for the acceleration of CNNs on Zynq SoCs. NEURA ghe leverages the synergistic usage of Zynq ARM cores and of a powerful and flexible Convolution-Specific Processor deployed on the reconfigurable logic. The Convolution-Specific Processor embeds both a convolution engine and a programmable soft core, releasing the ARM processors from most of the supervision duties and allowing the accelerator to be controlled by software at an ultra-fine granularity. This methodology opens the way for cooperative heterogeneous computing: While the accelerator takes care of the bulk of the CNN workload, the ARM cores can seamlessly execute hard-to-accelerate parts of the computational graph, taking advantage of the NEON vector engines to further speed up computation. Through the companion NeuDNN SW stack, NEURA ghe supports end-to-end CNN-based classification with a peak performance of 169GOps/s, and an energy efficiency of 17GOps/W. Thanks to our heterogeneous computing model, our platform improves upon the state-of-the-art, achieving a frame rate of 5.5 frames per second (fps) on the end-to-end execution of VGG-16 and 6.6fps on ResNet-18.

Increasing miniaturization is posing multiple challenges to electronic designers. In the context of multi-processor system-on-chips (MPSoCs), we focus on the problem of implementing efficient interconnect systems for devices which are ever more densely packed with parallel computing cores. Easily seen that traditional buses can not provide enough bandwidth, a revolutionary path to scalability is provided by packet-switched network-on-chips (NoCs), while a more conservative approach dictates the addition of bandwidth-rich components (e.g. crossbars) within the preexisting fabrics. While both alternatives have already been explored, a thorough contrastive analysis is still missing. In this paper, we bring crossbar and NoC designs to the chip layout level in order to highlight the respective strengths and weaknesses in terms of performance, area and power, keeping an eye on future scalability

Convolutional Neural Networks (CNNs) are extensively used in a wide range of applications, commonly including computer vision tasks like image and video classification, recognition and segmentation. Recent research results demonstrate that multi-layer (deep) network involving mono-dimensional convolutions and dilation can be effectively used in time series and sequences classification and segmentation, as well as in tasks involving sequence modeling. These structures, commonly referred to as Temporal Convolutional Networks (TCNs), represent an extremely promising alternative to recurrent architectures, commonly used across a broad range of sequence modeling tasks [1]. While FPGA based inference accelerators for classic CNNs are widespread, literature is lacking in a quantitative evaluation of their usability on inference for TCN models. In this paper we present such an evaluation, considering a CNN accelerator with specific features supporting TCN kernels as a reference and a set of state-of-the-art TCNs as a benchmark. Experimental results show that, during TCN execution, operational intensity can be critical for the overall performance. We propose a convolution scheduling based on batch processing that can boost efficiency up to 96% of theoretical peak performance. Overall we can achieve up to 111,8 GOPS/s and a power efficiency of 33,8 GOPS/s/W on an Ultrascale+ ZU3EG (up to 10x speedup and 3x power efficiency improvement with respect to pure software implementation).

The Internet of Medical Things (IoMT) paradigm is becoming mainstream in multiple clinical trials and healthcare procedures. Cardiovascular diseases monitoring, usually involving electrocardiogram (ECG) traces analysis, is one of the most promising and high-impact applications. Nevertheless, to fully exploit the potential of IoMT in this domain, some steps forward are needed. First, the edge-computing paradigm must be added to the picture. A certain level of near-sensor processing has to be enabled, to improve the scalability, portability, reliability and responsiveness of the IoMT nodes. Second, novel, increasingly accurate data analysis algorithms, such as those based on artificial intelligence and Deep Learning, must be exploited. To reach these objectives, designers, and programmers of IoMT nodes, have to face challenging optimization tasks, in order to execute fairly complex computing tasks on low-power wearable and portable processing systems, with tight power and battery lifetime budgets. In this work, we explore the implementation of a cognitive data analysis algorithm, based on a convolutional neural network trained to classify ECG waveforms, on a resource-constrained microcontroller-based computing platform. To minimize power consumption, we add an adaptivity layer that dynamically manages the hardware and software configuration of the device to adapt it at runtime to the required operating mode. Our experimental results show that adapting the node setup to the workload at runtime can save up to 50% power consumption. Our optimized and quantized neural network reaches an accuracy value higher than 97% for arrhythmia disorders detection on MIT-BIH Arrhythmia dataset.

The ever-shrinking lithographic technologies available to chip designers enable performance and functionality breakthroughs; yet, they bring new hard problems. For example, multiprocessor systems-on-chip featuring several processing elements can be conceived, but efficiently interconnecting them while keeping the design complexity manageable is a challenge. Traditional buses are easy to deploy, but cannot provide enough bandwidth for such complex systems. A departure from legacy architectures is therefore called for. One radical path is represented by packet-switching networks-on-chip, whereas a more conservative approach interleaves bandwidth-rich components (e.g., crossbars) within the preexisting fabrics. This paper is aimed at analyzing the strengths and weaknesses of these alternative approaches by performing a thorough analysis based on actual chip floorplans after the interconnection place&route stages and after a clock tree has been distributed across the layout. Performance, area, and power results will be discussed while keeping an eye on the scalability prospects in future technology nodes