The nameedge intelligence, also known asEdge AI, is a recent term used in the past few years to refer to the confluence of machine learning, or broadly speaking artificial intelligence, with edge computing. In this article, we revise the concepts regarding edge intelligence, such as cloud, edge, and fog computing, the motivation to use edge intelligence, and compare current approaches and analyze application scenarios. To provide a complete review of this technology, previous frameworks and platforms for edge computing have been discussed in this work to provide the general view of the basis for Edge AI. Similarly, the emerging techniques to deploy deep learning models at the network edge, as well as specialized platforms and frameworks to do so, are review in this article. These devices, techniques, and frameworks are analyzed based on relevant criteria at the network edge, such as latency, energy consumption, and accuracy of the models, to determine the current state of the art as well as current limitations of the proposed technologies. Because of this, it is possible to understand the current possibilities to efficiently deploy state-of-the-art deep learning models at the network edge based on technologies such as artificial intelligence accelerators, tensor processing units, and techniques that include federated learning and gossip training. Finally, the challenges of Edge AI are discussed in the work, as well as the future directions that can be extracted from the evolution of the edge computing and Internet of Things approaches.

This work presents a detailed study of the photothermal ablation of Kapton^® polyimide by a laser diode targeting its electrical conductivity enhancement. Laser-treated samples were structurally characterized using Scanning Electron Microscopy (SEM), Raman spectroscopy, X-ray Photoelectron Spectroscopy (XPS), as well as Diffuse Reflectance Infrared Fourier Transform (DRIFT) spectroscopy. The results show that the laser-assisted ablation constitutes a simple one-step and environmental friendly method to induce graphene-derived structures on the surface of polyimide films. The laser-modified surface was also electrically characterized through the Transmission Line Method (TLM) aiming at the improvement of the conductivity of the samples by tuning the laser power and the extraction of the contact resistance of the electrodes. Once the laser-ablation process is optimized, the samples increase their conductivity up to six orders of magnitude, being comparable to that of graphene obtained by chemical vapor deposition or by the reduction of graphene-oxide. Additionally, we show that the contact resistance can be decreased down to promising values of ∼2 $\mathsf{\Omega }$ when using silver-based electrodes.

This paper illustrates the application of the discrete wavelet transform (DWT) for wandering and noise suppression in electrocardiographic (ECG) signals. A novel one-step implementation is presented, which allows improving the overall denoising process. In addition an exhaustive study is carried out, defining threshold limits and thresholding rules for optimal wavelet denoising using this presented technique. The system has been tested using synthetic ECG signals, which allow accurately measuring the effect of the proposed processing. Moreover, results from real abdominal ECG signals acquired from pregnant women are presented in order to validate the presented approach.

Finding an inexpensive and scalable method for the mass production of memristors will be one of the key aspects for their implementation in end-user computing applications. Herein, we report pioneering research on the fabrication of laser-lithographed graphene oxide memristors. The devices have been surface-fabricated through a graphene oxide coating on a polyethylene terephthalate substrate followed by a localized laser-assisted photo-thermal partial reduction. When the laser fluence is appropriately tuned during the fabrication process, the devices present a characteristic pinched closed-loop in the current-voltage relation revealing the unique fingerprint of the memristive hysteresis. Combined structural and electrical experiments have been conducted to characterize the raw material and the devices that aim to establish a path for optimization. Electrical measurements have demonstrated a clear distinction between the resistive states, as well as stable memory performance, indicating the potential of laser-fabricated graphene oxide memristors in resistive switching applications.