The AGILE IoT Gateway Recommenders

Would you like to know more about the recommenders we have put in place in our various pilot projects? Discover the video made by our partner TU Graz!

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.


Photo by Javier Allegue Barros on Unsplash

AirEx – Smart ventilation control to balance energy efficiency with air-exchange

AirEx is a novel smart home technology that helps reducing the home’s heat demand, without compromising moisture build-up and subsequently damp and indoor air quality. AirEx uses smart sensors to monitor and analyse environmental conditions while its cloud-based algorithms determine the vent state to automatically regulate airflow. It uses AI to predict occupants’ behaviour and weather patterns for more efficient airflow optimisation. Via improved thermal efficiency of the building fabric, the system can reduce heating energy bills by 15%, which means 2-3 years payback. AirEx’s key customers are Social Housing Associations who have the mandate to improve the energy efficiency of their building stock.

Within the scope of Agile IoT project, we had the opportunity to develop standard APIs using Agile IoT to interact to our product, AirEx and make it therefore more agnostic with regards to integration with third party connected devices and infrastructure. During this project the key component we utilised is ‘Connectivity and Communication Protocols’, where we managed to extend this component via implementing agile-rfm69 protocol to communicate using HopeRF RFM69 module.

We found that the communications component has a reasonable API and there is existing support for a number of useful technologies. For example, it provides a useful outline for integrating BLE sensors within Linux. During this Agile IoT project we have successfully developed communication protocols with Bluetooth-enabled low energy temperature and humidity sensors, which is a significant step in our development considering the increased need for connected devices.

One of the main benefits of using Agile IoT was that it did not require much effort for actual deployment: once the system is working, deploying new devices via requires minimal effort, since new devices will pull down the appropriate images automatically.

Furthermore, the use of Agile IoT does not require much memory: containers use limited RAM, some as little as 20MB and hence it works very efficiently on a Raspberry Pi. Nor did we experience any processor issues with using Agile IoT on Raspberry Pi 3 or even on Pi Zero.

The critical requirements are stability of the platform, extensibility and the ability to update remotely – all of which have been successfully managed using Agile IoT.

In the future we are keen to continue to use Agile-BLE in order to integrate additional sensors. In addition, once the Agile hub becomes LoRaWAN-enabled we will be very keen to explore the possibility of using Agile hub.

Our Software Lead is presenting some of the key outcomes of the project via this IoT Webinar:

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.

This other video demos AirEx on AGILE:

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.


– AirEx

Photo by Dane Deaner on Unsplash

Integrate Agile IoT gateways to Watson IoT platform

Sensinov is a start-up specializing  in smart building and industrial IoT setups offering an interoperability as a service which aims at coping with huge device diversity, lifecycle management, and continuous operations of IoT solutions.

The diversity  of devices and connectivity networks in smart building can only grow with new and emerging devices that are entering the mainstream. To ensure future proof and sustainable deployment of smart buildings, there is a need to:

  • Transform the collected data into uniform data formats ensuring their proper handling by applications and IoT platforms. By doing so, the management of the life cycle (changing device vendor or changing IoT platform provider) becomes an easy task for IoT project owners.
  • Process the data locally and send only processed/filtered data to the cloud. Indeed, huge amounts of data sets could be generated. It is neither beneficial nor economically viable to send all the data to the cloud. Processing data locally is an emerging technical and business imperative. It is usually performed through running application logic closer to the IoT devices: i.e. where the data is generated.

Sensinov was selected for an open call of H2020 Agile IoT project. Our role in the project was to integrate Agile IoT gateways in the Watson IoT platform. This was an excellent opportunity for Sensinov to demonstrate its interoperability platform in action. Among our findings:

Agile IoT provides us a perfect development environment to cope with industrial IoT requirements from real life projects.

These requirements include:

  • The management of diversity in  IoT devices and protocols
  • The enhancement of efficiency through local processing

Indeed, the role of the Agile IoT gateway is to perform several functions needed for running cost optimized IoT projects. These functions include

  • Device connectivity
  • Protocol adaptation
  • Data processing close to the devices
  • Securing IoT communications
  • Ensuring system extensibility through the installation of new applications

As a smart building specialist without any plan to build hardware, Agile IoT provides  a perfect solution for deployment scenarios where the need to add a gateway is necessary with the absence of local and extensible local processing capabilities. Sensinov is currently considering the inclusion of an Agile IoT compatible gateway in its portfolio, partnering with a hardware player who adheres to the Agile IoT principles. In interest to our roadmap and strategy, Agile IoT offers  us several interesting value propositions including:

  • A rich set of connectors (Enocean, ZWave, ZigBee, BLE, etc)
  • An open source approach which avoids lock-in for Sensinov as a provider and for our customers
  • The development of an  ecosystem of providers
  • Modular hardware and software to adapt to different deployment needs

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.


Monitoring the Supply Chain of Bottled Wine with AGILE

Maintaining appropriate conditions in the storage and transport of sensitive products is an age-old problem for food supply chains. Many types of goods can be easily damaged by improper variations in temperature while others can be severely affected by shocks or excessive duration of transport. The impact from unsuitable transport and storage for wine is far greater than the mere cost of damaged goods, as it also includes brand damage for companies such as wine producers, wholesalers, retailers and logistics companies. The size of the problem is very large as nearly 5% of bottled wine worldwide is damaged, up to 25% is negatively affected, and the uprise in  quality problems due to improper storage settings , along with 20% of the quality wine consumed worldwide is counterfeit.

ISTMOS monitoring platform allows the supply chain stakeholders of bottled wine to monitor the most critical parameters affecting wine quality during its storage and transportation. Delivering traceability information down to the level of each individual bottle, the platform includes complex-event processing, rule and alert management and an analytics engine. We employ monitoring stations (fixed for storage and mobile for transportation) and collect and analyze the information on our own IoT platform which relies on FI-WARE enablers, particularly  PERSEO complex event processing and ORION context broker.

Wineries and retailers already have various types of measurement instruments in place to control temperature and humidity in wine storage spaces. Soon, an increasing number of stakeholders in the wine supply chain will be equipped with a plethora of IoT devices from different hardware vendors to measure wine storage environments. These sensors will operate across a wide range of wireless communication protocols, notably BLE and LoRaWAN, where low-power consumption is important, but also WiFi or NB-IoT.

In this market, ISTMOS needs to offer an open architecture solution which doesn’t lock customers into using a restricted set of devices or protocols. The modular, multi-protocol gateway architecture proposed by AGILE represents a promising approach to address the interoperability challenge faced by ISTMOS.

Therefore, we have evaluated AGILE as an interoperability solution for ISTMOS gateways. We ported ISTMOS gateway controller logic to the AGILE architecture by developing the necessary containerized applications that  interact with several components of the AGILE stack.

We have used AGILE Node-RED for reading and collecting information from BLE and LoraWAN channels (sensors) and sending data via WiFi or NB-IoT to the external cloud services used by ISTMOS (based on FIWARE enablers).

The results of the ISTMOS-AGILE integration have been evaluated through a pilot case study in collaboration with two large Greek wineries.

Overall, AGILE offered a good approach for prototyping of devices. The interfaces to Cloud platforms are supportive, while remote fleet management is offering additional possibilities for the offer of  custom services. AGILE is a big step in the right direction and has offered us valuable experiences and knowledge that will be further exploited.


Accelerating Internet of Things Big-Data Analytics end-to-end applications development with Agile

The main objective of the experiment considers  the integration of Senseioty, the FlairBit software platform to accelerate the Industrial Internet of Things (IoT) Big Data Analytics, with the Agile IoT gateway and services. This integration allows Senseioty to benefit in flexibility both in terms of data ingestion and edge computing capabilities.

From a technical standpoint, joining the Agile-IoT ecosystem allowed FlairBit to benefit from the many innovative services provided by the Agile-IoT platform that  are also available in Senseioty after the integration carried out in the proposed project. The Agile IoT gateway modularity and compliance with many architectures at the hardware level provide  support for various wireless and wired IoT networking technologies. Senseioty benefits from these additional protocols as a way to create new data sources and IoT connectors which in turn creates  data diversity and heterogeneity. These features allow to speed-up the development of the integrated solution and to reduce the time-to-market for the applications built on-top-of Senseioty. We have also benchmarked other gateways, but they usually lack the rapid and smooth integration of  software components with third party systems.

In fact, at the software level, the Agile IoT stack provides data collection and management capabilities on the gateway, an intuitive interface for device management, a visual workflow editor for creating integration patterns with less coding, and an IoT marketplace  for releasing the resulting end-to-end applications. Senseioty generates data analytics applications and workflows to be pushed as close as  possible to the data sources and the devices, hence Agile IoT gateway represents the perfect match to deploy secure, scalable and reliable Senseioty edge-computing applications.

The overall system resulting from the integration of the Agile IoT stack and Senseioty allows

application developers to rapidly create and deploy innovative IoT applications without the necessity to customize a given IoT platform which in turn,, saves development time and resources.

Moreover, FlairBit, as an IoT platform provider, can now benefit from the Agile IoT ecosystem as a way to boost its user and developer base and create new revenue streams, since Senseioty can now be exploited and used in innovative cross-domain applications.

Furthermore, , in accessing the Agile-IoT ecosystem, FlairBit benefits from the different competencies, technologies, and use cases of the partners in the consortium. From a business perspective, it allows the generation of  partnership to speed-up the go-to-market strategy in new markets for FlairBit. In fact, Agile-IoT provides an interoperability framework to enable IoT platforms to open up access to their devices/services in a controlled and secure manner. This pursues  the acquisition of new revenue streams for offering added value services along with receiving missing ingredients that enrich FlairBit business offerings.

To conclude,  the open source software Agile-IoT consortium is developing with a  business-friendly license that allows FlairBit’s Senseioty IoT platforms to be Agile-IoT compliant, and at the same time to remain proprietary which allows FlairBit to increase its own product value and unlocking potential businesses in new markets. Simultaneously,  FlairBit open sourced software components extend the Agile IoT stack and enables Big-Data Analytics on-top-of Agile IoT gateway, based on open source technologies (e.g., Apache Cassandra, Apache Zeppelin, Apache Beam).


Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.

Company: FlairBit
Contact name: Luca Bixio
Contact e-mail:

Heptasense: Navigating through the fog to build an understanding

You always hear: “Data here. Data there.”

It’s funny that almost  no one talks about Information, yet everyone is connecting more and more devices to the internet and storing their data.  And if they are not extracting information from that data, why bother  storing it in the first place?

Seconds can be the differentiator  between fixed and broken, faulty -working, accident-good fortune , life or death.

Fog computing becomes the standard due to firstly   having the need of information in real-time but simultaneously , the need of data  to be correlated, something that is impossible, if our processing is centered in either our devices or our datacenter.

The Agile Gateway was incredible  in assisting us to understand how we could bring this decentralized  value to our business and platform which enabled us to offer a better service with better performance and responsiveness.

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.


Photo by Mika Baumeister on Unsplash

VizLore: Federated IoT blockchain fabric

The overall objective of the extension project

VizLore Labs Foundation (VLF) participated in the AGILE IoT second open call extension with the project titled “Federated IoT blockchain fabric”. The main objective of this extension project was to work on integrating VizLore IoT platform with the AGILE IoT platform through joint blockchain fabric based on the open source Hyperledger Fabric framework. The result is the first federated IoT blockchain established between two IoT platforms covering various IoT verticals. This integration enhances the AGILE IoT platform with robust blockchain enabled security and privacy policy enforcement framework while simultaneously broadening and enhancing the potential of the VizLore federated IoT blockchain solution.

Motivation behind the extension project

What the research community and industry are agreeing on is that IoT systems currently lack blockchain building frameworks which consider specific requirements of a typical IoT system including: a large number of microtransactions, unfeasible proof of work validation, real-time  validation, avoiding large transaction fees, IoT specific smart contracts, high node churn in case of hosting ledger on end-user personal devices. To address these challenges, VizLore built the IoT blockchain framework (based on Hyperledger Fabric) which enables federation of blockchains across multiple IoT systems and platforms. In order to properly validate the federation potential, we required collaboration with an IoT platform provider who offers open software stack that is capable of hosting our software agents and focusing on the edge and fog components of IoT systems. The AGILE IoT fulfilled both requirements. IoT gateway software architecture is very similar to that of our IoT controllers and cloud platform. AGILE IoT focuses on building new IoT systems with specific configuration and deployment of IoT gateways. This proved to be an ideal blockchain federation enabler since each new gateway instance (on both AGILE and VizLore Labs end) introduces a new node into the federated blockchain architecture and enhances its distribution span and robustness. The AGILE IoT project pilots require innovative privacy management approaches which do not limit the business potential based on data sharing. The VizLore extension project also focused on producing a set of smart contract templates, specifically tailored for IoT business cases that  are automatically stored on the federated blockchain. Therefore, the integration potential was very high and this remained true during the execution of the extension project.

Experience during the implementation of the extension project

During execution of the extension project and porting of the VizLore federated blockchain enablers onto the AGILE SW stack, we have worked most of the AGILE SW components and services (GW SE stack, ramp-up configurator, Node-RED, recommender system). The documentation and examples are excellent and allowed fast validation and prototyping. Together with the AGILE team, we have identified the need for supporting 64-bit OS architectures on AGILE GWs in order to enable full integration of the Hyperledger Fabric components. Integration with the ramp-up configurator and Node-RED environment helped us to validate the role of our Hyperledger Fabric components in fast IoT system prototyping framework. We have identified and addressed all the requirements that  our solution needs to fulfill in order to be an integral part of the AGILE fast IoT prototyping framework, thus bringing the blockchain technology closer to the IoT system integrators.

Achieved results and impact

We promote our solution as blockchain as a service (BaaS). Consequently, , “collective IoT” security and privacy management as a service is offered to IoT platform operators and system integrators. The VizLore IoT blockchain fabric can be deployed on IoT platforms (as part of VizLore IoT edge devices or hosted on 3rd party edge devices and servers) regardless whether  or not their services actively use the blockchain applications. The VizLore federated Hyperledger fabric for IoT systems is actively being exploited as an integral part of the VizLore’s business strategy and IoT service offerings. Our Smart Access Control IoT service is enhanced with the Hyperledger Fabric and smart contracts so that each virtual key is a smart contract stored on the federated blockchain fabric. This significantly enhances security and robustness, while providing solid ground for establishing innovative business models for entrance vouchers and remote access control.

Our Hyperledger agents, blockchain network generator, Node-RED component, and smart contract templates are available as part of the AGILE IoT software stack.


HAWK project: monitoring elder people living alone

The HAWK project is about monitoring the elderly living alone through off-the-shelve movement sensors, through  the development of a model of the elder’s home activity through artificial intelligence technologies. The activity being continuously acquired by the movements of the  elder through their home (in his/her daily activities) is compared against the last trained model to provide insights about routines and deviations from “what has been learned”, in order to  alert caregivers or professional service providers on warnings or alarms derived from those deviations.

The architecture of the system is the combination of AGILE gateway and S&C’s enControl IoT platform and smart home solution. At the AGILE gateway, we have developed a protocol for a commercial movement detector device (PIR), that one may  capture movement events detected to fit the activity model. The protocol has been encapsulated within a dedicated docker container, allowing any other AGILE component to work with the PIR sensor. The protocol pushes PIR events to enControl IoT platform,where the activity model has been developed and deployed. Results from the model evaluation against PIR events are broadcasted using messages in a pub/sub architecture within enControl. This pub/sub connectivity is used to connect this output back to AGILE (GW) by a developed microservice, so it can be exploited by other AGILE components. The microservice is also a docker container, and it opens a persistent connection to enControl by using lightweight web sockets  technology allowing real-time server push updates.

enControl is constantly monitoring devices data received from the gateway and, for example, when an anomaly is detected, it’s pushed immediately to the HAWK notifier microservice who then proceeds to parse the data and store it in a local database.

The HAWK project allowed us to explore a new gateway technology and architecture that provides more freedom on development for our enControl solution. Before using AGILE, we used a licensed gateway that constrained us in enlarging its existing functionalities, and, for instance, did not authorize  us to develop and integrate new device protocols. AGILE technology has empowered us to do so by opening a door to explore open software architectures. Additionally, thanks to the help of AGILE, we were enabled to develop novel elder activity models through AI technologies and explore against real home deployment. The overall  effort ended in a prototype showing full integration of the PIR device, the AGILE GW, and the enControl IoT platform by demonstrating the fully integrated functionality developed within the HAWK project. Thanks to AGILE, we have quickly arrived at a first working prototype and deployed our system in an elder’s home. The results of HAWK are important first steps for our business and continually encourage us to keep working to reach a commercial product.

Please accept YouTube cookies to play this video. By accepting you will be accessing content from YouTube, a service provided by an external third party.

YouTube privacy policy

If you accept this notice, your choice will be saved and the page will refresh.

– Sensing & control

Ex Machina – Industrial IoT Solutions for the Energy Industry

Ex Machina is a weather analytics and industrial IoT company, that offers forecasting services to the energy industry. The company uses a unique combination of highly accurate weather and energy data, weather sensors technology, and AI to deliver its market-leading services.

Within the scope of Agile IoT project, Ex Machina effectively progressed its  smart gateway hardware which is a critical component in the company’s industrial  IoT energy forecasting solution.Continue reading

An Overview of Research on Computational Resource Management in AGILE

We would like to take this opportunity to illustrate the challenges related to computational resource consumption such as memory, CPU or networking, in our gateway setup. First and foremost, we must admit the utmost importance of ensuring that services provided by the gateway have reasonable use of the computational resources. This is how we can ensure that there is space available for applications using core services from AGILE. However, this is not the only reason to keep an eye on resource consumption. There are also security implications and economic reasons to monitor and control the use of resources, i.e. especially when computation should be offloaded to the cloud.

We will discuss the release of an open-source component to measure resource consumption in the gateway plus other scenarios. We will then explain how we have used this component in the scope of the project to perform several research tasks. Applications of the resource measurement component have been used to study resource abuse scenarios from the perspective of security, detection of crypto-miners, and evaluation of strategies to balance the computation load between the gateway and the cloud.

Continue reading