The Industrial Internet of Things(IIoT): A Strategic Approach

Focusing on WiFi and LoRaWAN Technologies

Who hasn't heard of WiFi technology? Who hasn't connected a device to a WiFi network? WiFi is versatile and affordable, but when it comes to IoT, there are other factors to consider. This is why I've become an advocate for LoRaWAN technology and how it complements BLE devices. In the following sections, I will introduce you to the system's platform and the general knowledge that developers refer to as the Back-end. Later on, I will graphically and through practical examples show you how these technologies can merge to achieve a deployment that is reliable, affordable, and efficient. Let's begin:

1. ESP32 Devices: Utilizing multi-sensor equipped devices for data collection tailored to specific needs.

Fig. 1. The boards on the right and center with built-in antennas are ESP32 microcontrollers, known for their WiFi and BLE capabilities. The one on the right is a new model that also includes the Zigbee protocol, commonly used in Smart Homes, in addition to having a faster processor and more memory. In the center, the ESP32 is a versatile microcontroller often used in IoT projects and can be used to process images from the camera or control other devices.

The PCB on the left is an Arduino Nano 33 Sense, installed on a breadboard and equipped with an OV7670 camera. This device can host TensorFlow, which are deep learning tools used to capture images and perform basic image processing tasks like object/face detection or recognition. Additionally, combining the camera with the MPU 6050 sensor board allows for capturing both visual and motion data. This setup could be used for projects involving robotics, drones, or interactive installations.

2. MQTT Broker: Centralizing data exchange for enhanced efficiency.

MQTT (Message Queuing Telemetry Transport) is a lightweight publish-subscribe messaging protocol ideal for IoT applications. It enables efficient communication between devices with minimal network bandwidth and resource requirements.

Fig. 2. This diagram illustrates the architecture of an EMQX MQTT broker cluster and its interactions with various devices and applications.

How it Works:

1. Publishers publish messages: Devices send data to the broker on specific topics, like "temperature/living_room" or "light/bedroom/status".
2. Subscribers express interest: Devices or applications subscribe to topics they want to receive messages from, like "temperature/#" (all temperature readings) or "light/bedroom/#" (all messages related to the bedroom light).
3. Broker filters and distributes: The broker receives messages from publishers and checks which subscribers are interested in the topic. It then delivers the message to all matching subscribers.

Benefits of MQTT Broker in IoT:

• Scalability: Handles large numbers of devices efficiently.
• Decoupling: Publishers and subscribers are independent, enabling flexible system design.
• Reliability: Ensures message delivery even with unreliable connections.Lightweight: Minimizes network bandwidth and device resource usage.
• Flexibility: Supports various Quality of Service (QoS) levels to ensure message delivery based on application needs.

Examples of MQTT in IoT:

• Smart Home: Connecting sensors, appliances, and control systems.
• Industrial Automation: Monitoring and controlling industrial equipment.
• Environmental Monitoring: Gathering data from sensors deployed in the field.
• Asset Tracking: Tracking the location and status of assets in real-time.

Overall, the MQTT broker plays a crucial role in facilitating reliable and efficient communication in IoT environments, enabling seamless data exchange between devices and applications.

3.? InfluxDB: Serving as the cornerstone for data management.

InfluxDB is a specialized type of database built from the ground up to handle time series data. This means data that is organized by timestamps, like sensor readings, application metrics, or stock prices. It's optimized for high-performance ingestion, storage, and retrieval of this time-based information.

Key Features:

• Purpose-built for Time Series: Unlike traditional databases, InfluxDB's structure and functionalities are tailored to the unique needs of time series data, making it much more efficient and scalable.
• High Ingestion Rates: It can handle massive volumes of data points per second, making it ideal for real-time monitoring and analysis of fast-changing data streams.
• Powerful Query Language: InfluxQL, its query language, is designed specifically for time series data, allowing you to easily perform complex analyses, aggregations, and visualizations.
• Downsampling and Retention Policies: Manage storage space and optimize performance by automatically downsampling or deleting older data based on your needs.
• Built-in Visualization: InfluxDB offers basic graphing capabilities, letting you visualize your data directly within the platform.
• Open Source and Enterprise Options: Both a free, open-source version and a commercially supported enterprise version are available, catering to different needs and scales.

Common Use Cases:

• IoT Monitoring: Storing and analyzing sensor data from connected devices like temperature, humidity, pressure, etc.
• DevOps Monitoring: Tracking application performance metrics, system resource utilization, and infrastructure health.
• Real-time Analytics: Performing on-the-fly analysis of streaming data to identify trends, anomalies, and patterns.
• Financial Analysis: Storing and analyzing stock prices, trading volumes, and other financial time series data.

Why I Choose InfluxDB?

Choosing InfluxDB for my IIoT project offers numerous advantages, particularly in efficiently handling time-series data. Its design is optimized for high performance and can manage large volumes of data, making it ideal for IIoT applications that generate continuous data streams, such as sensor readings. InfluxDB supports scalable solutions, provides built-in tools for time-series analysis, and facilitates easy integration through its HTTP API and client libraries. It features a powerful SQL-like query language (InfluxQL) for complex data analysis and supports automatic data retention policies, ensuring optimal data management and storage. Additionally, the strong community and widespread industry usage provide robust support and resources, enhancing the development and maintenance of my project.

Fig. 3. The database is structured around the concept of "measurements." Each row represents a single measurement, which are the readings sent by the device to the database, and I can configure the frequency of data transmission.

The screenshot provides raw data. Integrating data visualization tools could enhance understanding and analysis of trends. Additionally, implementing mechanisms to trigger alerts based on predefined thresholds (e.g., high humidity levels) could be beneficial. Combining humidity data with other environmental factors (temperature, air quality) could provide a more comprehensive picture of the office environment.

Overall, the "GeekHomeLab Database" screenshot shows a well-structured approach to monitoring humidity using ESP32 devices and BME280 sensors. Expanding its functionalities and integrating data visualization could further enhance its utility in maintaining a comfortable and healthy office environment.

4. Grafana: Offering advanced visualization tools for data analysis.

Grafana is an open-source platform designed for data visualization and monitoring. It allows you to create insightful and interactive dashboards from various data sources, making it easier to understand and analyze complex information.

Key Features:

• Diverse Data Source Integrations: Grafana connects with a wide range of databases, including InfluxDB, Prometheus, Graphite, and many more. This flexibility allows you to bring all your data together in one place for unified visualization.
• Beautiful and Customizable Dashboards: Create visually appealing and informative dashboards with a variety of graph types, including line graphs, bar charts, heatmaps, histograms, and more. Customize colors, layouts, and styles to fit your needs.
• Powerful Alerting System: Set up alerts based on specific thresholds and conditions within your data. Receive notifications through various channels (email, Slack, PagerDuty, etc.) when critical events occur.
• Collaboration and Sharing: Easily share dashboards with your team or publicly on the web. Collaborate and discuss insights with built-in annotation features.

Benefits of using Grafana:

• Improved Data Visibility: Transform raw data into clear and actionable insights through interactive visualizations.
• Enhanced Decision Making: Identify trends, patterns, and anomalies more easily, leading to better-informed decisions.
• Proactive Monitoring: Detects potential issues before they become critical through real-time alerts and dashboards.
• Increased Collaboration: Facilitate communication and knowledge sharing among teams by providing a centralized platform for data visualization.

Common Use Cases:

• IT infrastructure monitoring: Visualize server health, network performance, and application metrics.
• DevOps monitoring: Track deployments, resource utilization, and application performance in real-time.
• IoT data analysis: Monitor sensor readings, device status, and environmental conditions.
• Business intelligence: Create dashboards to track key performance indicators (KPIs) and business metrics.

InfluxDB and Grafana:

InfluxDB and Grafana form a powerful duo in the realm of data management and visualization. InfluxDB excels at storing and querying time-series data, while Grafana specializes in turning that data into insightful visualizations. This seamless integration allows InfluxDB to serve as the robust data source, with Grafana acting as the front-end, enabling the creation of dynamic dashboards and effective alert systems. Together, they empower my projects by transforming complex datasets into accessible and meaningful visual presentations, leading to improved understanding, better decision-making, and enhanced monitoring capabilities.

Fig. 4. The provided Grafana dashboard offers a visual representation of environmental data collected from the GeekHome(my HomeLab server) server environment. The dashboard displays three charts: Temperature, Humidity, and Barometric Pressure. The accompanying graph shows fluctuations over the past two days, revealing a consistent pattern with some well-defined peaks and dips, particularly around the last 24 hours. This is due to triggers I set up that activate and deactivate a fan based on set maximum and minimum temperatures. This system helps maintain the environment within a desired range; constant forced cooling would bring more humidity and contamination to the system, while excessively high temperatures could risk damaging the electronic components.

Pressure Trend: The gradual rise in barometric pressure could be related to changing weather patterns. Monitoring this trend alongside weather forecasts could help anticipate potential environmental impacts on the server.

Fig. 5. This Grafana dashboard offers a comprehensive overview of resource utilization and the status of virtual machines (VMs) within the "GeekHome,” which is my Proxmox server hosting all my projects. I've created this dashboard to monitor my server, and as you can see:

Server CPU, Logical Cores, Total Memory, Memory in Use, Swap Total, Storage Pools, Running VMs, along with other indicators that are not visible in the screenshot but would be displayed further down on the screen.

The dashboard enumerates several running VMs, including "node01" (the main host), "GeekHome", "TrueNAS" (a network-attached storage solution), and others such as "Cloudflare," "Adguard," "Docker," “Grafana," and "InfluxDB", encompassing a wide array of applications and services.

The uptime for each VM is also depicted, providing information on how long they have been operational without interruption.

Graphs showcase the memory usage for both the host and individual VMs over the past three hours, illustrating patterns and trends.

Overall, Grafana is a valuable asset for server and environment monitoring, laying a strong foundation for performance tracking, resource utilization, and VM management. With the inclusion of additional metrics, historical data analysis, and more in-depth evaluations, it has the potential to become an even more effective instrument for optimizing resource distribution, maintaining system stability, and enhancing the overall efficiency of the infrastructure.

5. Alerting System: Integrating diverse communication platforms for instant alert dissemination.

?Grafana's alerting system is a powerful tool that enables you to proactively monitor your data and receive notifications when specific conditions are met. This helps you identify and respond to potential issues before they escalate into major problems:

• You can set thresholds on various metrics, including CPU usage, memory consumption, network traffic, and custom data from your data sources.
• Multiple conditions can be combined to create more complex alert triggers using logical operators like AND/OR. This allows for fine-grained control over when alerts are activated.

When an alert is triggered, notifications can be sent through various channels, including:

• Receive alerts directly in your inbox.
• Get notified in your Slack channels for team collaboration and visibility.
• Integrate with incident response workflows for critical alerts.
• Trigger actions in other systems or custom scripts based on alerts.(Webhooks, Whatsapp, Telegram, Twilio, SMS)

Alerts have different states that indicate their current status:

• The condition is not met, and everything is normal.
• The condition has been met for a specified duration, but the alert is not yet triggered, allowing for a grace period.
• The alert is active, and notifications are being sent.
• There is insufficient data to evaluate the alert condition.

Benefits of Grafana Alerting:

• Proactive Issue Detection: Identify potential problems early on, preventing downtime and service disruptions.
• Reduced Mean Time to Resolution (MTTR): Respond to issues quickly and efficiently with timely alerts and clear notifications.
• Improved Operational Efficiency: Automate monitoring tasks and reduce the need for manual data inspection.
• Enhanced Collaboration: Keep teams informed and coordinated with shared alerts and notifications.

Examples of Alerting Use Cases:

• Server Monitoring: Alert when CPU usage exceeds a threshold or disk space runs low.
• Application Performance: Trigger alerts if response times are slow or error rates increase.
• Network Monitoring: Get notified of network outages or unusual traffic patterns.
• IoT Sensor Data: Alert when sensor readings go outside of expected ranges.

?Fig. 6. The screenshot displays an email notification generated by Grafana's alerting system, indicating a potential issue with server temperature. Clicking the "View alert" button would likely redirect the user to the Grafana interface for further details about the alert rule and current status.

The options to "View dashboard" and "View panel" would allow the user to investigate the relevant visualizations and data associated with the server temperature within Grafana.

If the alert is expected or under investigation, the user could choose to "Silence" it temporarily to prevent further notifications.

IIoT Network Architecture

WiFi Network Environment

Fig. 7

?The image depicts a diagrammatic representation of an Industrial Internet of Things (IIoT) architecture utilizing Wifi Technology, this architecture is the integration of ESP microcontrollers that connect to a variety of sensors capturing temperature, vibration, and pressure data. These microcontrollers communicate via WiFi to a network that employs an MQTT Mosquitto Broker to facilitate a publish/subscribe messaging system.

The data flow is indicated by lines connecting the components, showing the path from data acquisition to storage and visualization. Data collected from the sensors are sent to the MQTT broker, then onto a Node-Red UI interface, illustrating the system's interactivity. The data is stored in an InfluxDB database, and this stored data is finally displayed in a Grafana dashboard that provides real-time data visualization.

Additionally, the system has notification capabilities, as denoted by the icons for SMS, WhatsApp, Telegram, and email, indicating that the system can send alerts through these channels. Hosting servers for this setup include a Proxmox server and Docker containers potentially deployed on a Raspberry Pi, highlighting the system's scalability and flexibility for different hosting environments.

This visualization encapsulates a holistic approach to monitoring and controlling an IIoT network, where all components work cohesively to provide a reliable, efficient, and cost-effective monitoring solution.

Developed Practical Cases?

1. ?Flow controller for an oil injection system in a Vane compressor

Fig. 8. Prototyping a display and flow controller for an oil injection system in a Vane compressor. Microprocessor ESP-32, also a Raspberry Pi Zero W was tested.?You can observe that the prototype operates wirelessly, connecting via WiFi and powered by a battery, which demonstrates its low power consumption.

1. Home Lab "GeekHome" Environmental Monitoring System

Fig. 9. This project is showcased in the development of InfluxDB and Grafana, aiming to monitor the environment of my personal server or home lab, "GeekHome." For this purpose, I utilized an ESP32 microprocessor connected to the WiFi network and employing the Back-End technology described above. Figures 4 and 5 illustrate the dashboard in question, along with the readings and alerts that can be generated.

1. Cost-Effective and Portable IoT System Solution with Raspberry Pi 4

Fig. 10. The image showcases a Raspberry Pi 4 equipped with dual SSD drives, representing an economical yet powerful solution for hosting an entire IoT or Home Assistant system. This setup stands out due to its portability, allowing for quick and versatile deployments suitable for businesses, homes, or small offices. The dual SSD configuration promises enhanced data processing speeds and reliability, vital for managing real-time data flows inherent in IoT operations. Moreover, the compact form factor of the Raspberry Pi ensures that this system can be easily integrated into various environments without requiring significant space, a testament to the project's commitment to delivering efficient technology solutions that are both accessible and robust.

1. Advanced Delivery Route Optimization System

Fig. 11. Pictured is a state-of-the-art solution featuring a Raspberry Pi 4 integrated with 4G LTE, GPS, and LoRaWAN connectivity. Developed for a client engaged in delivery route optimization studies, this portable yet powerful device captures routes using GPS and LoRaWAN technology. The collected data is then analyzed using Machine Learning algorithms to enhance the efficiency of drivers' delivery routes. This system exemplifies the intersection of IoT with AI to drive logistical improvements, offering significant benefits such as reduced fuel consumption, decreased delivery times, and overall operational cost savings. It represents the innovative use of technology in solving complex real-world challenges, contributing to smarter and more sustainable business practices.

LoRaWan Network Environment

LoRa and LoRaWAN are pivotal technologies in IoT, enhancing long-range communication with low power usage. LoRa, based on Chirp Spread Spectrum (CSS), efficiently transmits data over great distances and through challenging environments, making it perfect for low-bit-rate applications. It operates on free-to-use sub-gigahertz bands, ideal for devices like sensors and actuators in various settings. LoRaWAN, a protocol built on LoRa, defines network structure and communication, ensuring secure and efficient message delivery between devices and networks. It's maintained by the LoRa Alliance, promoting wide adoption through standards and certifications. LoRaWAN's capabilities extend to extensive applications: from smart agriculture and animal tracking to urban planning and health monitoring, proving its versatility and impact in modern technology.

Fig. 12. As can be seen, it is the same WiFi structure, but what sets the LoRaWAN network apart from WiFi is its low power consumption and the ability to cover vast ranges. There are network alternatives like Helium and The Things Network where you can subscribe at a low cost and achieve much greater coverage.

I have been working on these LoRaWAN projects:

1. Cattle and Pet Tracking: Utilizing GPS and LoRaWAN technology to develop tracking solutions for livestock and pets, enhancing the ability to monitor and manage their location and well-being effectively.

1. Hiker and Biker App: Creating a dedicated application for hikers and bikers that leverages the extended range and low power consumption of LoRaWAN networks, ensuring connectivity even in remote areas.

1. LoRaWAN for Smart Cities: Implementing LoRaWAN infrastructure to facilitate Smart City initiatives, supporting a wide range of applications from urban environmental monitoring to public services optimization.

1. LoRaWAN for IIoT: Deploying LoRaWAN networks for Industrial Internet of Things (IIoT) environments to enable remote monitoring, predictive maintenance, and efficient resource management across various industries.

Comparison of Wireless Technologies

Below is a comparative table that outlines the advantages and disadvantages of implementing an Industrial Internet of Things (IIoT) network using WiFi, LoRaWAN, and Cellular technology:

This table offers a snapshot comparison between WiFi, LoRaWAN, and Cellular networks in the context of IIoT implementations. Each technology has its own set of strengths and weaknesses, and the choice among them should be guided by the specific requirements of the IIoT application, such as coverage needs, power constraints, data throughput, and cost considerations.

Conclusion:

In today’s fast-evolving technological landscape, the Industrial Internet of Things (IIoT) stands out as a transformative force, particularly in sectors requiring robust and real-time data monitoring and analysis. This article delves into the complementary capabilities of WiFi and LoRaWAN technologies in powering IIoT systems, highlighting their distinctive advantages and suitability for different operational contexts.

WiFi’s ubiquity and high data rate capabilities make it ideal for complex tasks requiring substantial data transmission, such as video processing or large-scale sensor networks within limited areas. On the other hand, LoRaWAN emerges as a highly efficient choice for applications demanding long-range connectivity and low power consumption, suitable for remote or geographically dispersed assets.

The integration of these technologies presents a scalable and cost-effective solution, enhancing reliability and operational efficiency through sophisticated data exchange frameworks like MQTT and robust data handling platforms such as InfluxDB and Grafana. Such setups not only streamline processes but also pave the way for advanced data analytics and real-time decision-making.

Adopting a dual-technology approach enables industries to leverage the strengths of both WiFi and LoRaWAN, ensuring comprehensive coverage and connectivity for various IIoT applications. This strategic fusion empowers businesses to achieve higher productivity, improved safety, and ultimately, a significant competitive edge in the digital era.