Research

12 pages, 1763 KiB

Open AccessArticle

Intellino: Processor for Embedded Artificial Intelligence

by Young Hyun Yoon, Dong Hyun Hwang, Jun Hyeok Yang and Seung Eun Lee

Electronics 2020, 9(7), 1169; https://doi.org/10.3390/electronics9071169 - 18 Jul 2020

Cited by 23 | Viewed by 5084

The development of computation technology and artificial intelligence (AI) field brings about AI to be applied to various system. In addition, the research on hardware-based AI processors leads to the minimization of AI devices. By adapting the AI device to the edge of [...] Read more.

The development of computation technology and artificial intelligence (AI) field brings about AI to be applied to various system. In addition, the research on hardware-based AI processors leads to the minimization of AI devices. By adapting the AI device to the edge of internet of things (IoT), the system can perform AI operation promptly on the edge and reduce the workload of the system core. As the edge is influenced by the characteristics of the embedded system, implementing hardware which operates with low power in restricted resources on a processor is necessary. In this paper, we propose the intellino, a processor for embedded artificial intelligence. Intellino ensures low power operation based on optimized AI algorithms and reduces the workload of the system core through the hardware implementation of a neural network. In addition, intellino’s dedicated protocol helps the embedded system to enhance the performance. We measure intellino performance, achieving over 95% accuracy, and verify our proposal with an field programmable gate array (FPGA) prototyping. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

19 pages, 7328 KiB

Open AccessArticle

Exploiting Recurring Patterns to Improve Scalability of Parking Availability Prediction Systems

by Sergio Di Martino and Antonio Origlia

Electronics 2020, 9(5), 838; https://doi.org/10.3390/electronics9050838 - 19 May 2020

Cited by 9 | Viewed by 2012

Abstract

Parking Guidance and Information (PGI) systems aim at supporting drivers in finding suitable parking spaces, also by predicting the availability at driver’s Estimated Time of Arrival (ETA), leveraging information about the general parking availability situation. To do these predictions, most of the proposals [...] Read more.

Parking Guidance and Information (PGI) systems aim at supporting drivers in finding suitable parking spaces, also by predicting the availability at driver’s Estimated Time of Arrival (ETA), leveraging information about the general parking availability situation. To do these predictions, most of the proposals in the literature dealing with on-street parking need to train a model for each road segment, with significant scalability issues when deploying a city-wide PGI. By investigating a real dataset we found that on-street parking dynamics show a high temporal auto-correlation. In this paper we present a new processing pipeline that exploits these recurring trends to improve the scalability. The proposal includes two steps to reduce both the number of required models and training examples. The effectiveness of the proposed pipeline has been empirically assessed on a real dataset of on-street parking availability from San Francisco (USA). Results show that the proposal is able to provide parking predictions whose accuracy is comparable to state-of-the-art solutions based on one model per road segment, while requiring only a fraction of training costs, thus being more likely scalable to city-wide scenarios. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

13 pages, 671 KiB

Open AccessArticle

Towards Near-Real-Time Intrusion Detection for IoT Devices using Supervised Learning and Apache Spark

by Valerio Morfino and Salvatore Rampone

Electronics 2020, 9(3), 444; https://doi.org/10.3390/electronics9030444 - 06 Mar 2020

Cited by 44 | Viewed by 4902

Abstract

In the fields of Internet of Things (IoT) infrastructures, attack and anomaly detection are rising concerns. With the increased use of IoT infrastructure in every domain, threats and attacks in these infrastructures are also growing proportionally. In this paper the performances of several [...] Read more.

In the fields of Internet of Things (IoT) infrastructures, attack and anomaly detection are rising concerns. With the increased use of IoT infrastructure in every domain, threats and attacks in these infrastructures are also growing proportionally. In this paper the performances of several machine learning algorithms in identifying cyber-attacks (namely SYN-DOS attacks) to IoT systems are compared both in terms of application performances, and in training/application times. We use supervised machine learning algorithms included in the MLlib library of Apache Spark, a fast and general engine for big data processing. We show the implementation details and the performance of those algorithms on public datasets using a training set of up to 2 million instances. We adopt a Cloud environment, emphasizing the importance of the scalability and of the elasticity of use. Results show that all the Spark algorithms used result in a very good identification accuracy (>99%). Overall, one of them, Random Forest, achieves an accuracy of 1. We also report a very short training time (23.22 sec for Decision Tree with 2 million rows). The experiments also show a very low application time (0.13 sec for over than 600,000 instances for Random Forest) using Apache Spark in the Cloud. Furthermore, the explicit model generated by Random Forest is very easy-to-implement using high- or low-level programming languages. In light of the results obtained, both in terms of computation times and identification performance, a hybrid approach for the detection of SYN-DOS cyber-attacks on IoT devices is proposed: the application of an explicit Random Forest model, implemented directly on the IoT device, along with a second level analysis (training) performed in the Cloud. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

17 pages, 605 KiB

Open AccessArticle

Classification of Transition Human Activities in IoT Environments via Memory-Based Neural Networks

by Giovanni Acampora, Gianluca Minopoli, Francesco Musella and Mariacarla Staffa

Electronics 2020, 9(3), 409; https://doi.org/10.3390/electronics9030409 - 28 Feb 2020

Cited by 3 | Viewed by 2871

Abstract

Human activity recognition is a crucial task in several modern applications based on the Internet of Things (IoT) paradigm, from the design of intelligent video surveillance systems to the development of elderly robot assistants. Recently, machine learning algorithms have been strongly investigated to [...] Read more.

Human activity recognition is a crucial task in several modern applications based on the Internet of Things (IoT) paradigm, from the design of intelligent video surveillance systems to the development of elderly robot assistants. Recently, machine learning algorithms have been strongly investigated to improve the recognition task of human activities. Though, in spite of these research activities, there are not so many studies focusing on the efficient recognition of complex human activities, namely transitional activities, and there is no research aimed at evaluating the effects of noise in data used to train algorithms. In this paper, we bridge this gap by introducing an innovative activity recognition system based on a neural classifier endowed with memory, able to optimize the performance of the classification of both transitional and non-transitional human activities. The system recognizes human activities from unobtrusive IoT devices (such as the accelerometer and gyroscope) integrated in commonly used smartphones. The main peculiarity provided by the proposed system is related to the exploitation of a neural network extended with short-term memory information about the previous activities’ features. The experimental study proves the reliability of the proposed system in terms of accuracy with respect to state-of-the-art classifiers and the robustness of the proposed framework with respect to noise in data. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

21 pages, 1479 KiB

Open AccessFeature PaperArticle

First Order and Second Order Learning Algorithms on the Special Orthogonal Group to Compute the SVD of Data Matrices

by Simone Fiori, Lorenzo Del Rossi, Michele Gigli and Alessio Saccuti

Electronics 2020, 9(2), 334; https://doi.org/10.3390/electronics9020334 - 15 Feb 2020

Cited by 2 | Viewed by 2526

Abstract

The present paper deals with neural algorithms to learn the singular value decomposition (SVD) of data matrices. The neural algorithms utilized in the present research endeavor were developed by Helmke and Moore (HM) and appear under the form of two continuous-time differential equations [...] Read more.

The present paper deals with neural algorithms to learn the singular value decomposition (SVD) of data matrices. The neural algorithms utilized in the present research endeavor were developed by Helmke and Moore (HM) and appear under the form of two continuous-time differential equations over the special orthogonal group of matrices. The purpose of the present paper is to develop and compare different numerical schemes, under the form of two alternating learning rules, to learn the singular value decomposition of large matrices on the basis of the HM learning paradigm. The numerical schemes developed here are both first-order (Euler-like) and second-order (Runge-like). Moreover, a reduced Euler scheme is presented that consists of a single learning rule for one of the factors involved in the SVD. Numerical experiments performed to estimate the optical-flow (which is a component of modern IoT technologies) in real-world video sequences illustrate the features of the novel learning schemes. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

12 pages, 1038 KiB

Open AccessArticle

System Log Detection Model Based on Conformal Prediction

by Yitong Ren, Zhaojun Gu, Zhi Wang, Zhihong Tian, Chunbo Liu, Hui Lu, Xiaojiang Du and Mohsen Guizani

Electronics 2020, 9(2), 232; https://doi.org/10.3390/electronics9020232 - 31 Jan 2020

Cited by 8 | Viewed by 2622

Abstract

With the rapid development of the Internet of Things, the combination of the Internet of Things with machine learning, Hadoop and other fields are current development trends. Hadoop Distributed File System (HDFS) is one of the core components of Hadoop, which is used [...] Read more.

With the rapid development of the Internet of Things, the combination of the Internet of Things with machine learning, Hadoop and other fields are current development trends. Hadoop Distributed File System (HDFS) is one of the core components of Hadoop, which is used to process files that are divided into data blocks distributed in the cluster. Once the distributed log data are abnormal, it will cause serious losses. When using machine learning algorithms for system log anomaly detection, the output of threshold-based classification models are only normal or abnormal simple predictions. This paper used the statistical learning method of conformity measure to calculate the similarity between test data and past experience. Compared with detection methods based on static threshold, the statistical learning method of the conformity measure can dynamically adapt to the changing log data. By adjusting the maximum fault tolerance, a system administrator can better manage and monitor the system logs. In addition, the computational efficiency of the statistical learning method for conformity measurement was improved. This paper implemented an intranet anomaly detection model based on log analysis, and conducted trial detection on HDFS data sets quickly and efficiently. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

17 pages, 11244 KiB

Open AccessArticle

Individual Behavior Modeling with Sensors Using Process Mining

by Onur Dogan, Antonio Martinez-Millana, Eric Rojas, Marcos Sepúlveda, Jorge Munoz-Gama, Vicente Traver and Carlos Fernandez-Llatas

Electronics 2019, 8(7), 766; https://doi.org/10.3390/electronics8070766 - 09 Jul 2019

Cited by 16 | Viewed by 4733

Abstract

Understanding human behavior can assist in the adoption of satisfactory health interventions and improved care. One of the main problems relies on the definition of human behaviors, as human activities depend on multiple variables and are of dynamic nature. Although smart homes have [...] Read more.

Understanding human behavior can assist in the adoption of satisfactory health interventions and improved care. One of the main problems relies on the definition of human behaviors, as human activities depend on multiple variables and are of dynamic nature. Although smart homes have advanced in the latest years and contributed to unobtrusive human behavior tracking, artificial intelligence has not coped yet with the problem of variability and dynamism of these behaviors. Process mining is an emerging discipline capable of adapting to the nature of high-variate data and extract knowledge to define behavior patterns. In this study, we analyze data from 25 in-house residents acquired with indoor location sensors by means of process mining clustering techniques, which allows obtaining workflows of the human behavior inside the house. Data are clustered by adjusting two variables: the similarity index and the Euclidean distance between workflows. Thereafter, two main models are created: (1) a workflow view to analyze the characteristics of the discovered clusters and the information they reveal about human behavior and (2) a calendar view, in which common behaviors are rendered in the way of a calendar allowing to detect relevant patterns depending on the day of the week and the season of the year. Three representative patients who performed three different behaviors: stable, unstable, and complex behaviors according to the proposed approach are investigated. This approach provides human behavior details in the manner of a workflow model, discovering user paths, frequent transitions between rooms, and the time the user was in each room, in addition to showing the results into the calendar view increases readability and visual attraction of human behaviors, allowing to us detect patterns happening on special days. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

23 pages, 3234 KiB

Open AccessArticle

Machine Learning Prediction Approach to Enhance Congestion Control in 5G IoT Environment

by Ihab Ahmed Najm, Alaa Khalaf Hamoud, Jaime Lloret and Ignacio Bosch

Electronics 2019, 8(6), 607; https://doi.org/10.3390/electronics8060607 - 30 May 2019

Cited by 45 | Viewed by 6426

Abstract

The 5G network is a next-generation wireless form of communication and the latest mobile technology. In practice, 5G utilizes the Internet of Things (IoT) to work in high-traffic networks with multiple nodes/sensors in an attempt to transmit their packets to a destination simultaneously, [...] Read more.

The 5G network is a next-generation wireless form of communication and the latest mobile technology. In practice, 5G utilizes the Internet of Things (IoT) to work in high-traffic networks with multiple nodes/sensors in an attempt to transmit their packets to a destination simultaneously, which is a characteristic of IoT applications. Due to this, 5G offers vast bandwidth, low delay, and extremely high data transfer speed. Thus, 5G presents opportunities and motivations for utilizing next-generation protocols, especially the stream control transmission protocol (SCTP). However, the congestion control mechanisms of the conventional SCTP negatively influence overall performance. Moreover, existing mechanisms contribute to reduce 5G and IoT performance. Thus, a new machine learning model based on a decision tree (DT) algorithm is proposed in this study to predict optimal enhancement of congestion control in the wireless sensors of 5G IoT networks. The model was implemented on a training dataset to determine the optimal parametric setting in a 5G environment. The dataset was used to train the machine learning model and enable the prediction of optimal alternatives that can enhance the performance of the congestion control approach. The DT approach can be used for other functions, especially prediction and classification. DT algorithms provide graphs that can be used by any user to understand the prediction approach. The DT C4.5 provided promising results, with more than 92% precision and recall. Full article

(This article belongs to the Special Issue Recent Machine Learning Applications to Internet of Things (IoT))

► Show Figures

Figure 1

Journal Menu

Journal Browser

Recent Machine Learning Applications to Internet of Things (IoT)

Share This Special Issue

Special Issue Editors

Special Issue Information

Published Papers (8 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI