Edge Machine Research Articles

Towards generating a reliable device-specific identifier for IoT devices

A significant number of IoT devices are being deployed in the wild, mostly in remote locations and in untrusted conditions. This could include monitoring an electronic perimeter fence or a critical infrastructure such as telecom and power grids. Such applications rely on the fidelity of data reported from the IoT devices, and hence it is imperative to identify the trustworthiness of the remote device before taking decisions. Existing approaches use a secret key usually stored in volatile or non-volatile memory for creating an encrypted digital signature. However, these techniques are vulnerable to malicious attacks and have significant computation and energy overhead. This paper presents a novel device-specific identifier, IoT-ID that captures the device characteristics and can be used towards device identification. IoT-ID is based on physically unclonable functions (PUFs), that exploit variations in the manufacturing process to derive a unique fingerprint for integrated circuits. In this work, we design novel PUFs for Commercially Off the Shelf (COTS) components such as clock oscillators and ADC, to derive IoT-ID for a device. Hitherto, system component PUFs are invasive and rely on additional dedicated hardware circuitry to create a unique fingerprint. A highlight of our PUFs is doing away with special hardware. IoT-ID is non-invasive and can be invoked using simple software APIs running on COTS components. IoT-ID has the following key properties viz., constructability, real-time, uniqueness, and reproducibility, making them robust device-specific identifiers.We present detailed experimental results from our live deployment of 50 IoT devices running over a month. Our edge machine learning algorithm has 100% accuracy in uniquely identifying the 50 devices in our deployment and can run locally on the resource-constrained IoT device. We show the scalability of IoT-ID with the help of numerical analysis on 1000s of IoT devices. Further, we discuss approaches to evaluate and improve the reliability of the IoT-ID.11This manuscript is an extension of the paper ‘IoT-ID: A Novel Device-Specific Identifier Based on Unique Hardware Fingerprints’ Vaidya et al. (2020) published in 2020 IEEE/ACM Fifth International Conference on Internet-of-Things Design and Implementation (IoTDI).

Технологии обработки материалов на основе стеклопластиков

The review of basic methods for glass-fiber material machining is shown, characteristic features are presented also advantages and drawbacks of different processing methods are mentioned. There is shown a description of glass-fiber material edge machining. The alternative methods such as hydro-abrasive cutting and laser working are considered.

AHMoSe: A knowledge-based visual support system for selecting regression machine learning models

Decision support systems have become increasingly popular in the domain of agriculture. With the development of automated machine learning, agricultural experts are now able to train, evaluate and make predictions using cutting edge machine learning (ML) models without the need for much ML knowledge. Although this automated approach has led to successful results in many scenarios, in certain cases (e.g., when few labeled datasets are available) choosing among different models with similar performance metrics is a difficult task. Furthermore, these systems do not commonly allow users to incorporate their domain knowledge that could facilitate the task of model selection, and to gain insight into the prediction system for eventual decision making. To address these issues, in this paper we present AHMoSe, a visual support system that allows domain experts to better understand, diagnose and compare different regression models, primarily by enriching model-agnostic explanations with domain knowledge. To validate AHMoSe, we describe a use case scenario in the viticulture domain, grape quality prediction, where the system enables users to diagnose and select prediction models that perform better. We also discuss feedback concerning the design of the tool from both ML and viticulture experts.

A Novel ECG Automatic Detection Using LongShort-Term Memory Network and Internet of Things Technology

The early detection of cardiovascular diseases based on electrocardiogram (ECG) is very important for the timely treatment of cardiovascular patients, which increases the survival rate of patients. ECG is a visual representation that describes changes in cardiac bioelectricity and is the basis for detecting heart health. With the rise of edge machine learning and Internet of Things (IoT) technologies, small machine learning models have received attention. This study proposes an ECG automatic classification method based on Internet of Things technology and LSTM network to achieve early monitoring and early prevention of cardiovascular diseases. Specifically, this paper first proposes a single-layer bidirectional LSTM network structure. Make full use of the timing-dependent features of the sampling points before and after to automatically extract features. The network structure is more lightweight and the calculation complexity is lower. In order to verify the effectiveness of the proposed classification model, the relevant comparison algorithm is used to verify on the MIT-BIH public data set. Secondly, the model is embedded in a wearable device to automatically classify the collected ECG. Finally, when an abnormality is detected, the user is alerted by an alarm. The experimental results show that the proposed model has a simple structure and a high classification and recognition rate, which can meet the needs of wearable devices for monitoring ECG of patients.

Explore-before-talk: Multichannel selection diversity for uplink transmissions in machine-type communication

Improving the data rate of machine-type communication (MTC) is essential in supporting emerging Internet of things (IoT) applications ranging from real-time surveillance to edge machine learning. To this end, in this paper we propose a resource allocation approach for uplink transmissions within a random access procedure in MTC by exploiting multichannel selection diversity, coined explore-before-talk (EBT). Each user in EBT first sends pilot signals through multiple channels that are initially allocated by a base station (BS) for exploration, and then the BS informs a subset of initially allocated channels that are associated with high signal-to-noise ratios (SNRs) for data packet transmission by the user while releasing the rest of the channels for other users. Consequently, EBT exploits a multichannel selection diversity gain during data packet transmission, at the cost of exploration during pilot transmission. We optimize this exploration-exploitation trade-off, by deriving closed-form mean data rate and resource outage probability expressions. Numerical results corroborate that EBT achieves a higher mean data rate while satisfying the same outage constraint, compared to a conventional MTC protocol without exploration.

ОСОБЕННОСТИ ОБРАБОТКИ ДЕТАЛЕЙ НЕФТЕГАЗОВОЙ ОТРАСЛИ, ВЫПОЛНЕННЫХ ИЗ ВОЛЬФРАМОВЫХ СПЛАВОВ

The purpose of the work carried out is to increase the effectiveness of hard alloy machining. Within the limits of the paper there is solved a matter of electro-erosion treatment and turning TC 3 alloy parts. Basic investigation methods are: - experiment and comparison, - computer simulation The work resulted in recommendations for the choice of cutters and cutting modes for hard alloy edge machining and also electrode material at electro-erosion treatment. One of the ways to increase productivity and operation quality equipment consists in the application of new materials and alloys including tungsten alloys. Within the limits of this work there are considered the peculiarities in machining TC3 alloy parts for oil-gas industry by the example of the “sleeve” part. For machining inner and outer surfaces there was used a turning operation, and for complex grooves – electro-erosion treatment as the only possible choice for processing because of a small size of the part. The peculiarities hard alloy machining through turning and basic reasons of hard alloy hard-to-workability are described. There are shown approaches to the choice of cutters, and the dependence of surface roughness and tool durability upon cutting mode choice is shown. During turning plate material must have higher strength than alloy under machining. One of the best choices for hard alloy machining is cubic boron nitride. For processing it is necessary to use tool holders with the increased rigidity. There are presented principles of electrode designing for hard alloy electro-erosion treatment and ways for process productivity increase. The application of copper-tungsten pseudo-alloys and m1 grade copper is compared from the point of view of economic purposefulness in their use. In spite of copper-tungsten alloy high durability which gives advantages in processing rate and economy in manufacturing electrodes its use is economically inexpedient and it is more rational to use copper which is cheaper.

Investigation of Optimum Grinding Condition Using cBN Electroplated End-Mill for CFRP Machining

Recently, carbon fiber reinforced plastics (CFRP) have been used in various applications such as airplanes and automobiles. In CFRP molding, there are unnecessary portions on the outer area. Therefore, a machining process is required to remove them. Cutting and grinding are conventionally used in the finish machining of CFRPs. End-milling allows the removal of most of these portions. However, uncut fibers easily occur during end-milling. In contrast, a precise machined surface and edge are easily obtained using a grinding tool. Therefore, this research has developed a novel cubic boron nitride (cBN) electroplated end-mill that combines an end-mill and a grinding tool. This is a versatile tool that can cut and grind CFRPs by changing the direction of rotation of the tool. In this study, the effectiveness of the developed tool is investigated. First, the developed tool machined the CFRP by side milling. Consequently, cBN abrasives that were fixed on the outer surface of the developed tool did not detach in certain cutting conditions. Next, in order to generate a sharp edge on the CFRP and restrict the increase in the CFRP temperature with the cBN electroplated end-mill, the optimum abrasive size and grinding condition were investigated through the design of experiments. Moreover, the effectiveness of the developed tool was verified by comparing it with a conventional tool. As a result, smaller burrs and uncut fibers were observed after final machining with the developed tool under the derived optimum condition than those with conventional tools. However, the desired surface roughness could not be achieved as required by the airline industry. Therefore, oscillating grinding was applied. In addition, the formula of the theoretical surface roughness while using the developed tool was derived using the theory of slant grinding. As a result, the oscillating condition that led to the required surface roughness was obtained by theoretical analysis. In addition, the required value for the airline industry was achieved by oscillating grinding.

Process design and optimization Of Two-Stage Axial Micro Turbine Air motor

Based on the general machining center, a high-efficiency, mass-produced micro-turbine processing technology and a process scheme for improving metal removal rate are proposed to improve and optimize the metal removal rate of micro-machining. Taking the two-stage axial micro turbine air motor as the research object, the machining route of clamping the nozzle rotor is developed, and the metal removal rates of the turbine with the expandable surface when using the rounded end mill and the ball end mill are compared by the single factor analysis method. In the three processes of rough milling blade passage, finish milling blade and finish milling wheel hub, the metal removal rate of the rounded end mill is 2.96 times, 10 times and 2 times higher than the ball end mill, respectively. The Swarf finishing strategy of Powermill software is used to processing the side and bottom of the blade both roughly and accurately, and the suitable cutting parameters and side edge machining parameters are obtained. The surface roughness Ra of the side and bottom surfaces of the blade is tested to be 0.4-0.8μm using a surface roughness measuring instrument SJ-410. With the trigger type probe of the model Omp60, the dimensional tolerance of the blade is within 0.015 mm, and the coaxiality is from 0.002 to 0.003mm. The two-stage axial-type micro turbine air motor is assembled by the hot-packing method, and the torque and rotor speed are tested by a special platform. The results show that when the compressed air inlet pressure is 300KPa, the designed maximum speed of 100000rpm is satisfied, which meets the design requirement of the turbine motor and verifies that the numerical control machining process plan is feasible. Besides, the side edge machining of the blade using the cylindrical milling cutter is efficient to improve the metal removal rate. It also lays a foundation for the mass production of the two-stage axial-type micro turbine air motor.

Wireless Edge Machine Learning: Resource Allocation and Trade-Offs

The aim of this paper is to propose a resource allocation strategy for dynamic training and inference of machine learning tasks at the edge of the wireless network, with the goal of exploring the trade-off between energy, delay and learning accuracy. The scenario of interest is composed of a set of devices sending a continuous flow of data to an edge server that extracts relevant information running online learning algorithms, within the emerging framework known as Edge Machine Learning (EML). Taking into account the limitations of the edge servers, with respect to a cloud, and the scarcity of resources of mobile devices, we focus on the efficient allocation of radio (e.g., data rate, quantization) and computation (e.g., CPU scheduling) resources, to strike the best trade-off between energy consumption and quality of the EML service, including service end-to-end (E2E) delay and accuracy of the learning task. To this aim, we propose two different dynamic strategies: (i) The first method aims to minimize the system energy consumption, under constraints on E2E service delay and accuracy; (ii) the second method aims to optimize the learning accuracy, while guaranteeing an E2E delay and a bounded average energy consumption. Then, we present a dynamic resource allocation framework for EML based on stochastic Lyapunov optimization. Our low-complexity algorithms do not require any prior knowledge on the statistics of wireless channels, data arrivals, and data probability distributions. Furthermore, our strategies can incorporate prior knowledge regarding the model underlying the observed data, or can work in a totally data-driven fashion. Several numerical results on synthetic and real data assess the performance of the proposed approach.

Recently evaluated Electrical Discharge Machining (EDM) process performances: A research perspective

Abstract Recently, cutting edge machining approaches are used extensively for answering multifarious issues in manufacturing sectors that consist of machining larger strength materials, manufacture of intricate shaped profiles, improve surface characteristics, and lower production time. In this scenario, the development of Electro Discharge Machining (EDM) is rapidly taking place. EDM is a non-conventional thermal-based machining procedure that is used for the machining of harder materials. At large, spark erosion is the main working principle performed in EDM. This meticulous review study provides based on the recently published EDM process articles where the objective is to attempt newer work material and advanced electrode that could significantly meet industrial manufacturing domain around the globe. In addition, the concluding observations turn out with defined underlines and technological research gaps making this review article useful to related academic and scientific research community with proper identification of EDM process factors, advance work materials, material electrode tools and sustainable dielectric for achieving favorable outcomes as a frontier machining technology.

Impact of machining-induced surface defects on the edge formability of commercially pure titanium sheet at room temperature

Despite the good properties of titanium, which have drawn the interest of various industries over the years, one of the major drawbacks of this material is its poor machinability. This has largely been attributed to its low thermal conductivity and elastic modulus. The ability to attain the optimum sheet edge performance during forming is dependent on the quality of the edges produced. Also, the demanding nature of aerospace part design has provoked the interest of both industry and academia to continually explore avenues tailored at enhancing part performance. The sort of edge surface integrity produced for aerospace part fabrication thus becomes a vital consideration in the quest to ensuring prime performance of components. This work seeks to study the influence of different machining-induced surface defects on the sheet edge performance of CP-Ti (grade 2) at room temperature. Hole expansion test was used to assess the edge surface formability of CP-Ti with different machining-induced edge defects. The research found that machining-induced surface defects act as stress concentration sites during the hole expansion test and have a major impact on the material flow. Electro-discharge machined edges were observed to exhibit high edge formability compared to laser and abrasive water jet cut edges due to the impact of machining-induced microstructural changes.

Energy-Efficient Processing and Robust Wireless Cooperative Transmission for Edge Inference

Edge machine learning can deliver low-latency and private artificial intelligent (AI) services for mobile devices by leveraging computation and storage resources at the network edge. This article presents an energy-efficient edge processing framework to execute deep learning inference tasks at the edge computing nodes whose wireless connections to mobile devices are prone to channel uncertainties. Aimed at minimizing the sum of computation and transmission power consumption with probabilistic Quality-of-Service (QoS) constraints, we formulate the joint inference tasking and the downlink beamforming problem that is characterized by a group sparse objective function. We provide a statistical learning-based robust optimization approach to approximate the highly intractable probabilistic-QoS constraints by nonconvex quadratic constraints, which are further reformulated as matrix inequalities with a rank-one constraint via matrix lifting. We design a reweighted power minimization approach by iteratively reweighted ℓ <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sub> minimization with difference-of-convex-functions (DC) regularization and updating weights, where the reweighted approach is adopted for enhancing group sparsity whereas the DC regularization is designed for inducing rank-one solutions. The numerical results demonstrate that the proposed approach outperforms other state-of-the-art approaches.

Thermal Effect on Poly(methyl methacrylate) (PMMA) Material Removal in the Micromilling Process.

Poly(methyl methacrylate) (PMMA) is of growing interest in the application of microfluidic devices and high precision optical elements due to its excellent moldability and formability. Micromilling is one of the micromachining methods which has been extensively used to manufacture polymer components. In this study, a high-speed micromilling method was used to manufacture polymer with high form accuracy and surface quality. The processing temperature effects on the surface quality were investigated in detail. The dynamic mechanical analysis (DMA) experiment was used to study the material mechanical property under different temperatures. According to the DMA results, the PMMA sample is in the glass and viscoelastic state during the milling process. The cutting chips under various processing temperatures are classified into three kinds according to their shapes: roll, sheet, and sinter. The surface roughness of samples with sheet and roll cutting chips is smaller than that of sinter cutting chips. To obtain a better machining bottom surface and edge shape, the processing temperature below 70 °C is recommended according to the results. This work is of great value for the study of polymer removal mechanism and optimization of processing parameters for the industry.

Surface generation mechanism of ceramic matrix composite in ultrasonic assisted wire sawing

Ceramic matrix composites (CMC) are highly required in many fields of science and engineering. However, the CMC parts always have poor surface finish. This study attempts to improve cutting performance of CMC material by combing the advantages of ultrasonic assisted cutting and diamond wire sawing. Cutting force, surface roughness, machined edge and tool wear are analyzed based on experimental results. It shows that the oscillatory movement of tool edges provides positive effect on particle ejection and residual material reduction. Ductile chip formation can be achieved due to the small tip radius of grits. Obvious decrease in cutting force, surface roughness and tool wear are obtained. Moreover, burrs, fuzzing and fracture are reduced. Meanwhile, both the surface characteristics and shape accuracy are significantly improved. These results provide a valuable basis for application of ultrasonic assisted wire sawing and understanding of CMC cutting mechanisms.

Data-Importance Aware Radio Resource Allocation: Wireless Communication Helps Machine Learning

The rich mobile data and edge computing enabled wireless networks motivate to deploy artificial intelligence (AI) at network edge, known as \emph{edge AI}, which integrates wireless communication and machine learning. In communication, data bits are equally important, while in machine learning some data bits are more important. Therefore we can allocate more radio resources to the more important data and allocate less radio resources to the less important data, so as to efficiently utilize the limited radio resources. To this end, how to define "more or less important" of data is the key problem. In this article, we propose two importance criteria to differentiate data's importance based on their effects on machine learning, one for centralized edge machine learning and the other for distributed edge machine learning. Then, the corresponding radio resource allocation schemes are proposed to improve performance of machine learning. Extensive experiments are conducted for verifying the effectiveness of the proposed data-importance aware radio resource allocation schemes.

Manufacturing of a ceramic groove part based on additive and subtractive technologies

Ceramic stereolithography Three-Dimensional (SL-3D) printing has a unique advantage in forming ceramic parts with complex shapes. However, because of the effects of the slicing layer thickness, laser spot diameter, and curing inhomogeneity, its manufacturing dimensional accuracy still lags far behind traditional processing methods. In this study, a manufacturing process based on additive and subtractive technologies was used to prepare ceramic groove parts. The manufacturing process of ceramic groove parts is composed of SL-3D printing, debinding, micro-milling, and sintering processes. In order to improve the dimensional accuracy of ceramic groove parts, a model based on shrinkage compensation and micro-milling was proposed. The dimensional change pattern in the curing phase was modelled and a shrinkage compensation method was established to control the dimensional accuracy of the finished parts. A method to characterize the micro-chipping on machined edges of the workpiece was investigated by applying a micro-milling cutter to process the 3D-printed part pre-sintered at 1150 °C. The effectiveness of the proposed dimensional accuracy control method was verified by the case study of a ZrO2 ceramic part.

Automatic CNN-Based Arabic Numeral Spotting and Handwritten Digit Recognition by Using Deep Transfer Learning in Ottoman Population Registers

Historical manuscripts and archival documentation are handwritten texts which are the backbone sources for historical inquiry. Recent developments in the digital humanities field and the need for extracting information from the historical documents have fastened the digitization processes. Cutting edge machine learning methods are applied to extract meaning from these documents. Page segmentation (layout analysis), keyword, number and symbol spotting, handwritten text recognition algorithms are tested on historical documents. For most of the languages, these techniques are widely studied and high performance techniques are developed. However, the properties of Arabic scripts (i.e., diacritics, varying script styles, diacritics, and ligatures) create additional problems for these algorithms and, therefore, the number of research is limited. In this research, we first automatically spotted the Arabic numerals from the very first series of population registers of the Ottoman Empire conducted in the mid-nineteenth century and recognized these numbers. They are important because they held information about the number of households, registered individuals and ages of individuals. We applied a red color filter to separate numerals from the document by taking advantage of the structure of the studied registers (numerals are written in red). We first used a CNN-based segmentation method for spotting these numerals. In the second part, we annotated a local Arabic handwritten digit dataset from the spotted numerals by selecting uni-digit ones and tested the Deep Transfer Learning method from large open Arabic handwritten digit datasets for digit recognition. We achieved promising results for recognizing digits in these historical documents.

An experimental study on fabricating microstructured surface by using LS-WEDM-prepared ball micro end mills

In recent years, there is an increasing demand for fabricating microstructures such as micro grooves, micro holes, micro dimples, and micro channels on various materials. Micro machining is an alternative method to manufacture microstructures due to its high efficiency and flexibility, and the micro end mill preparation is an essential step. In this study, the low-speed wire electrical discharge machining (LS-WEDM) is used to fabricate ball micro end mills with four cutting edges and radius of less than 200 μm, and the optimized parameters and multiple cutting strategies are adopted to improve machining surface quality and edge sharpness of ball micro end mills. Then, the micro dimples on soft and hard materials are formed by the designed and prepared ball micro end mills made of cemented carbide, respectively. The micro machining performance of ball micro end mills was evaluated by investigating the surface quality and microstructure of micro dimples, cutting force, and tool wear, and the influence of micro machining parameters on micro dimples is discussed in detail. Experimental results indicated that tool wear patterns mainly are adhesion friction and cutting edge damage, and the LS-WEDM-fabricated ball micro end mills with four cutting edges are potential to manufacture microstructures on hard and soft materials with acceptable surface quality and tool wear.

Protein Structure Classification Based on Distance Feature

In Bioinformatics field Protein Structure Classification is the hugest undertaking. The realized proteins have been requested subject to their level, feature, work, amino destructive and family and superfamily. Protein structure segregated into four sorts: all ? protein structure, all ? protein structure, ?+? protein structure, and ?/? protein structure. The use of a standard way to deal with perform plan is a very inconvenient and dreary task. The quantity of cutting edge Machine Intelligence enrolling strategies such Support Vector Machine, Random Forest, Artificial Neural Network, Decision Tree and Naïve Bayes Classifier had been proposed in the composition. Our objective right currently is to develop a system that performs better than anything past markers for protein structure gathering by thinking about the separation among the distinctive Amino Acid buildups. To take a gander at the display of proposed work particular datasets are used.

Application of an industrial robot for processing the edges of parts

The possibility of using a robotic complex when machining the edges of parts with solid polymer-abrasive radial brushes is studied. The dependences of the parameters of the machined edge and the roughness of the machined surface on the cutting speed, brush deformation, feed and the angle of inclination of the tool relative to the machined surface are determined. The possibility of processing the edges along the outer radius, in the oval hole, straight edges in the groove and on inclined sections is shown. Keywords edge, processing, robotic complex, polymer-abrasive brush, mode parameters. Dimov-Ura@yandex.ru