Unfairness of Random Access with Collision Avoidance in Industrial Internet of Things Networks

This paper is focused on the analysis of unfairness of random media access in Local Operating Networks (LON), which is one of the commercial platforms of the Industrial Internet of Things (IIoT). The unfairness in accessing the LON channel is introduced by a collision avoidance mechanism in the predictive p-persistent CSMA protocol adopted at the media access control layer. The study on the bandwidth share in predictive p-persistent CSMA calls for the analysis of multiple memoryless backoff. In this paper, it is shown that the channel access in LON systems is unfair in the short term for medium traffic load conditions, and in the long term for heavy loaded networks. Furthermore, it is explained that the average bandwidth allocated to a particular node is determined implicitly by the load scenario, while an actual node bandwidth fluctuates in time according to stochastic dynamics of the predictive p-persistent CSMA. Next, it is formally proven that the average bandwidth available to a node is a linear function of its backoff state and does not depend on backoff states of the other stations. Finally, it is demonstrated that possibly unfair bandwidth share in LON networks determined implicitly by load scenario is stable because, with lowering a fraction of actual network bandwidth accessible by a given station, the probability to decrease it in the future also drops.

Download Full-text

Deep Reinforcement Learning for Scheduling in an Edge Computing-Based Industrial Internet of Things

Wireless Communications and Mobile Computing ◽

10.1155/2021/8017334 ◽

2021 ◽

Vol 2021 ◽

pp. 1-12

Author(s):

Jingjing Wu ◽

Guoliang Zhang ◽

Jiaqi Nie ◽

Yuhuai Peng ◽

Yunhou Zhang

Keyword(s):

Reinforcement Learning ◽

Internet Of Things ◽

Manufacturing Systems ◽

Performance Metrics ◽

Scheduling Algorithm ◽

Edge Computing ◽

Stream Data ◽

Industrial Internet Of Things ◽

Network Bandwidth ◽

Industrial Internet

The demand for improving productivity in manufacturing systems makes the industrial Internet of things (IIoT) an important research area spawned by the Internet of things (IoT). In IIoT systems, there is an increasing demand for different types of industrial equipment to exchange stream data with different delays. Communications between massive heterogeneous industrial devices and clouds will cause high latency and require high network bandwidth. The introduction of edge computing in the IIoT can address unacceptable processing latency and reduce the heavy link burden. However, the limited resources in edge computing servers are one of the difficulties in formulating communication scheduling and resource allocation strategies. In this article, we use deep reinforcement learning (DRL) to solve the scheduling problem in edge computing to improve the quality of services provided to users in IIoT applications. First, we propose a hierarchical scheduling model considering the central-edge computing heterogeneous architecture. Then, according to the model characteristics, a deep intelligent scheduling algorithm (DISA) based on a double deep Q network (DDQN) framework is proposed to make scheduling decisions for communication. We compare DISA with other baseline solutions using various performance metrics. Simulation results show that the proposed algorithm is more effective than other baseline algorithms.

Download Full-text

An Integrated Legacy of Modbus Devices to Industrial Internet of Things: Approach for Smart Industries

10.31219/osf.io/54tje ◽

2020 ◽

Author(s):

Karthik Muthineni

Keyword(s):

Internet Of Things ◽

Industrial Revolution ◽

Digital Technologies ◽

Industrial Internet Of Things ◽

Intelligent Decision Making ◽

Internet Of Everything ◽

Industrial Internet ◽

Intelligent Decision ◽

Network Platform ◽

Smart Factories

The new industrial revolution Industry 4.0, connecting manufacturing process with digital technologies that can communicate, analyze, and use information for intelligent decision making includes Industrial Internet of Things (IIoT) to help manufactures and consumers for efficient controlling and monitoring. This work presents the design and implementation of an IIoT ecosystem for smart factories. The design is based on Siemens Simatic IoT2040, an intelligent industrial gateway that is connected to modbus sensors publishing data onto Network Platform for Internet of Everything (NETPIE). The design demonstrates the capabilities of Simatic IoT2040 by taking Python, Node-Red, and Mosca into account that works simultaneously on the device.

Download Full-text

MONITORING AND PREDICTIVE ANALYTICS OF TECHNOLOGICAL EQUIPMENT ON THE BASED OF INDUSTRIAL INTERNET OF THINGS

ВЕСТНИК ВОРОНЕЖСКОГО ГОСУДАРСТВЕННОГО ТЕХНИЧЕСКОГО УНИВЕРСИТЕТА ◽

10.36622/vstu.2020.16.5.001 ◽

2020 ◽

pp. 7-12

Author(s):

С.Л. Добрынин ◽

В.Л. Бурковский

Keyword(s):

Internet Of Things ◽

Data Acquisition ◽

Predictive Analytics ◽

Actual State ◽

Monitoring And Control ◽

Industrial Internet Of Things ◽

Machining Time ◽

Industrial Internet ◽

Acquisition Device ◽

Data Acquisition Device

Произведен обзор технологий в рамках концепции четвертой промышленной революции, рассмотрены примеры реализации новых моделей управления технологическими процессами на базе промышленного интернета вещей. Описано техническое устройство основных подсистем системы мониторинга и контроля, служащей для повышения осведомленности о фактическом состоянии производственных ресурсов в особенности станков и аддитивного оборудования в режиме реального времени. Архитектура предлагаемой системы состоит из устройства сбора данных (УСД), реализующего быстрый и эффективный сбор данных от станков и шлюза, передающего ликвидную часть информации в облачное хранилище для дальнейшей обработки и анализа. Передача данных выполняется на двух уровнях: локально в цехе, с использованием беспроводной сенсорной сети (WSN) на базе стека протоколов ZigBee от устройства сбора данных к шлюзам и от шлюзов в облако с использованием интернет-протоколов. Разработан алгоритм инициализации протоколов связи между устройством сбора данных и шлюзом, а также алгоритм выявления неисправностей в сети. Расчет фактического времени обработки станочных подсистем позволяет более эффективно планировать профилактическое обслуживание вместо того, чтобы выполнять задачи обслуживания в фиксированные интервалы без учета времени использования оборудования We carried out a review of technologies within the framework of the concept of the fourth industrial revolution; we considered examples of the implementation of new models of process control based on the industrial Internet of things. We described the technical structure of the main subsystems of the monitoring and control system to increase awareness of the actual state of production resources in particular machine tools and additive equipment in real time. The architecture of the proposed system consists of a data acquisition device (DAD) that implements fast and efficient data collection from machines and a gateway that transfers the liquid part of information to the cloud storage for further processing and analysis. We carried out the data transmission at two levels, locally in the workshop, using a wireless sensor network (WSN) based on ZigBee protocol stack from the data acquisition device to the gateways and from the gateways to the cloud using Internet protocols. An algorithm was developed for initializing communication protocols between a data acquisition device and a gateway, as well as an algorithm for detecting network malfunctions. Calculating the actual machining time of machine subsystems allows us to more efficiently scheduling preventive maintenance rather than performing maintenance tasks at fixed intervals without considering equipment usage

Download Full-text