Delivering data driven facilities management solutions using LoRaWan technology - Deep Dive
Hello everyone and welcome to another edition of our partner technical video series. Today we have the pleasure to have Richard Clarke from DXC Luxoft and Jonas Stokking from the Things Industries to talk about how Morgan Sindall is partnering with the things industries DXC looks soft and Microsoft to deliver. Data driven. LoraWan based solutions to their facilities management customers. My name is Oscar Naim your host today and in principle program manager in the Azure IoT engineering organization.
Richard, could you please introduce yourself? Sadly my name is Richard Clark. I'm a diss track attack at DXC Luxoft. Our company provides system integration solutions across multiple different verticals including automotive retail. I'm part of the global smart places and things. Team that provides IO T support to our clients and which Morganson was one of them. Excellent thank you, Richard. Now Johan, could you please introduce yourself and tell
us a bit about the things industries and what Laura one is all about? Particularly since Microsoft has just joined the reliance. Thank you Oscar. I'm really excited about that. So yeah. Hi everyone my name is Jan Stocking. I am CTO and Co.
Founder of the things industries. I'm also co-founder of The Things Network. More on that later and I co-founded the Things Network in 2015 when Laura, when was just released. And yeah, my co-founder and I were immediately immediately amazed by its potential and we decided to build a business around it, but also a Community initiative. So let me quickly explain what Laura and Laura went is for people who are new to this technology, so here's a great overview of where Laura is positioned. As compared to other wireless technologies that you might be familiar with and we really see it as complementary, complementary to what is already out there.
Lorawan and Laura in devices. They really focused on long range, low power and low bandwidth communications, so it's definitely not an IO T technology that's going to be suitable for all IoT use cases, but certainly it opens up a whole breed of new IoT use cases that were not possible before. So Lora, Lora Wan and the Lora alliance. That's an industry alliance that governs the Lorawan standard and certification program. So that combined this technology and the alliance and all the companies in the alliance.
And it really drives this unique combination of features that you see here listed and you won't see this combination in any other low power wide area networking technology currently on the market. So that's really that's really exciting. Now a typical Lorawan network deployments looks like this, so this is a typical topology, so you see the end nodes on the left and these can be anything, so these can be fridge sensors, parking sensors, desk presence sensors, cattle trackers, anything, because Lorawan is a generic messaging protocol. Then you have the gateways and gateways to forward all traffic to a network server and a network server manages the entire device fleet, including the end nodes and gateways. Most developers building solutions with Laura when they interact with the Lorawan devices on the application layer through integration to through integrations, and you see that on the right where you see the application server.
So the things network is a developer community and it's really focused on Laura win. The community has more than 150,000 members all over the world, and our community sets of gateways that can be used by other community members and they use the network to Community network for all kinds of things. Can be research and development, but also many not for profit use cases, and that's that is really nice. So this is really a free. Open network for everyone who who's interested in building Lorawan solutions. Besides that, and that's really what the things industries is really, the commercial wing that makes the things network sustainable and also provides enterprise services and solutions around lower wind.
And here you see our core products at the thing stack and the things that implements all the Lorawan specifications. So it implements the network server but also other components that are part of the lore of an ecosystem. They think stack is open core and you can grab a free distribution and browse the source code on our GitHub repository and this is also where you see the Azure IT integration positions. So you see on the right we have multiple integrations and they interact natively with the thing stack and the Azure IoT integration is really advanced because it also features device provisioning, telemetry and all sorts of advanced monitoring features.
Excellent thank you, Joanne. Now let's talk about Goldeneye and how Morgan Sindall is leveraging this data platform to deliver value to its customers. Oh Richard, please tell us more about the problem we want to solve for Morgan signal and how Goldeneye is helping on this morganson themselves our company. They provide property service management for various housing associations in the UK. It's also to do with county councils properties as well.
They go in and install equipment. They sort out properties. But in damp proofness, putting insulation and all sorts of features along those lines, but they had no ability to actually predict where they would need to go. Also they could offering this sort of service to their clients where they can say as a client their clients their the landlords in which they are responsible for dampness and air quality and living standards for their tenants.
So this is what the Goldeneye platform. Is about it allows. Their clients and their tenants to monitor the environment they're living in to see if there is any dampness or mold, or the temperatures were correct. Heating's on an efficient or not overheating, heating available to them, but one of the issues with this or low cost housing. It's also they didn't want to put sensors in that would use the local energy or electricity from the tenant or even there.
Bandwidth, so we decided on the law Wan network so we can have little sensors. Will talk to gateways of mounted on an external buildings and it'll all battery operated so they don't use the energy from the client. Excellent. Yeah, excellent,
so question to Richard. So that's the solution integrator. Can you paint us a picture of the overall architecture and the technical components that comprise the solution? I have covered most of the architecture around it. This slide really just shows the integration point for Goldeneye platform. We have specific sensors were using text, Alex sensors for air quality. We using MCF 88th and sense from the MCF 88 people who are monitoring temperature.
We also use their external sensors. Strategists ingest data so. We have to ingest the data into Azure.
Azure was selected by the client. With our help and decide in it, we then route this data into logic apps for transformation and process. Because different sensors from different manufacturers, the payload always different. So we wanted a common set payloaders entered into the system stream that's then pushed in through stream analytics for analysis. Is our weather temperatures reaching a threshold over multiple days or we got missing heartbeat sensors? 'cause sometimes the sensors might be in a black? Whole area so which we need. Idea where to go and actually get our engineers out on site to actually say the sensors needs replacing or repositioning and also the telemetry is then analyzed and pushed out into the data warehouse which are then the clients and self can view poo.
I burn mobile app or through our portal. And I I remember Richard that you know your solution. We will be evolving to leverage IoT Central as well, right? Oh yeah we are investigating IoT central for ease of management, easier manipulation of the modeling within the IO 2 central.
Also pushing that into digital twin. Eventually perfect so so Joe and this is a great opportunity to talk about Dtic integration with Azure IoT and Azure IO T Central. Could you please tell us more? Yeah so. The thing stack really has two integrations with Azure ID 14 IoT hub and the other for IoT central and customers can choose based on their requirements or what. What is the best fit for their existing architecture? So here is the IT hub integration and the IoT Hub integration allows for a lot of flexibility for developers so they can. They really have low level access to the event stream.
Now we have a Azure resource manager template to deploy these resources and you'll see that in a minute in the demo. These resources and this whole design is serverless, so there's no compute deployed in your Azure accounts, and that makes it really scalable and at the same time very low cost, especially if you don't have a lot of traffic in the beginning. So devices are onboarded automatically as they transmit data to IoT hub and the device twin also gets updated automatically with all lower wind metadata that one could think of. We also have an integration with IO T central. And the IoT central integration is really the batteries included variant is really plug and play.
There's just one Azure resource deployed in the customer's account and it supports the Azure device provisioning service for natively on boarding Lorawan devices securely and that scale. And so yeah, we made these Azure resource manager templates and that makes it really easy for customers to deploy them in their Azure accounts. Excellent, so another question to Johan.
Can you talk about how do we ensure security for all these sensors? Yeah, that's a great question, and it's a really important topic as well. So Laura win comes with mandatory security features, so you can't do lower win without using security features. But yeah, just like you shouldn't write your passwords on a post it you you should still use the Lorawan security features in a secure way.
Now Laura win comes at the heart of the protocol with Authenticity, integrity and confidentiality features. And here is a great overview of what that looks like. So you see, in red the red line is the Lorawan end to end encryption, so that's for confidentiality. That goes all the way from the end device to the application.
In blue you see the authentication and integrity that is maintained between the end device and the network server. And I mean you see the standards security features that are provided by using IP connected. Components so gateways typically have an Ethernet or a 4G or 5G back hole and they are just connected securely to cloud service and also between the network server. So the thing stack and the application server, so that's typically the IoT hub or IT central integration.
This is very useful. Thank you, Joanne. Now let's see a demo of the current solution that PTI DXC looks soft. And I have put together so John, why don't you get started? So this is the thing stack console which is our web interface where you can manage gateways. So these are covering your end devices.
You can also go to your applications. You can have multiple and in the applications you can have your end devices. So your sensors that are transmitting data.
So for the for the demo it can go here to our live data stream and this is all the metadata and all the details of all the traffic that the end devices are sending. So this is a lot of data. It's in Jason format. That you find here also the decoded payloads, but also a lot of other metadata that helps you.
With analyzing the coverage of urine devices. You can go to the integrations and here you'll find the Azure IoT hub integration and this is really easy to get started with. So first you generate an API key. You can copy that to your clipboard and then you can go to our documentation which contains a deploy button that makes it really easy to deploy the template in your Azure account. So this brings you straight to your Azure portal and that shows you a form where you fill out some details about the deployment.
So select your resource group and enter the API key and besides the API key you just enter your application ID and a few other things and that's it. And then you can go ahead with the deployment. So this typically takes up to five minutes. But when that is done, you can go to your resource group.
In any research group you see the resources that have been deployed in your Azure account. So this is what I just showed in the. In the on the slide now in VS code, you can really easily connect your IoT hub, that is that is now also deployed. And here you see the data stream. So these are the exact same Jason messages that I just showed in the things that console.
But then. On IoT hub. And this allows you, as an Azure developer, to work directly with the very same payloads and metadata as you are maybe already used to, as it thinks that customer or that you can take from here and build and use as part of your application. So now back to you Oscar to see what you can actually do with this. Great, let's wait now to Richard to talk about actionable data now. OK, thank you. I can see you on.
We built a two application groups. We've got the one per manufacturer. We take data ingested in from the sensors, pass it on to IoT hub, which then sends the telemetry or the event information off to a surface prosecute.
The service bus cube and then is triggered by within Logic app. The logic app will then process is to see if we need any new tax added to our device, twin if there was any failures who are raised and alert if the fail. On next stage it will then analyze what type of message we might have a time sync message which we need to report back to the actual sense of the current time, in which case we've sent update the desired properties within the.
Device twin and that then gets synchronized back to device with the new things industries integration. Once the telemetry sent through will get through this system. I would now show you some of the IoT events or IoT device to it itself.
The as you can see here we got a set of tanks. The first set is supplied by the things industries. The rest. There are our own popularity tags.
You can also see the decoded payload that will actually receive through the event. This is the standard message we get from the things industries I included and device twin. And for this particular sense, so you can see there's actually two payloads per message.
They're taking our 15 minutes apart. This is the actual event message when you actually receive sent through to the event hub from the things industry. It's got contains a lot more information around the meta data around the signal strengths that you get from the gateway itself. Because I'm pools of it and we're now going to have to look at the actual how you would do that. Transform how we transform the streams. I'm going to the logic app designer.
We pass the message coming in. We update the tanks, we set some internal properties around initializing the battery and property. As you can see, we check for the time sync message. Get South. He would send off the command. We use the measure function to do this.
We should then update the desired properties within the device Twain itself. Then we got two types of sensors are coming into this string transform this CO2 sensor and an external web sensor air quality. We update the battery percentage on the O2 sensor from percentage to actual million volts and we split their payload into two individual messages. As you can see on the transform payload off the external sensors, only one message needed to be sent.
We then write them to a storage account. That storage account helps with downstream processing where stream analytics will process it. He isn't a an existing long successfully went through. As you can see, the payload has certainly changed. We also record the signal strength information so we can do analysis later on.
Like we're gonna have a look at more simpler one. Now where the transform payload to only just needs minor changes. I say we look at the actual. Results are there. You see, these structure of the payload.
We actually right down to the storage account. BLOB is virtually the same, it's just an odd. Changes maybe nude. Different parameters are in it.
Here's a few examples of the transforms that we provide. Now we actually part of the process. We actually monitoring any alerts from the system.
Within a Q and it's these are just examples of the types of alerts we can actually retrieve from the system. That tip from this demo. Excellent, thank you very much, Richard. Now why don't you talk about a bit about
some of the main reasons why you have chosen Azure and Azure IoT services for building this solution? Certainly marketing sounds already an existing dynamics 365 customer, but we still have to show long the benefits of going through IoT hub event hub. What potentials of low code? No code patterns? We could potentially use what useful, serverless infrastructure there was wind Azure itself, so I didn't have to worry about virtual machines and keeping them running 24/7 for 365 days and all that sort of things also happen to show them how to use different cost models to save on actual expenses within it. We also showed them and use these server list compute functions. Some power apps and logic apps.
Excellent now Joanne, could you please comment on this as well? We'd love to hear your opinion. Yeah, sure. So we see many Lorawan solution developers struggling with the device provisioning telemetry, integrations, routing, data storage and that's just the start. You know this just just getting the data in the right place and and using Azure IoT really allows developers to focus on spending their time on on on building added value to. You know, fix the problem for their customers, so the integrations that we build, they bring Lorawan into the comfort zone of the existing Azure customers, but as well as you know, customers that are using dynamics or Office 365 like Morgan Sindall.
And another reason to use Azure IoT is really to combine all sorts of IoT technology. Technology is because lore awareness oftentimes just one piece of the puzzle, and developers need a development platform. And yeah, Azure it is just that.
We just provide 11 puzzle piece. Basically yeah exactly. Now now Johan. What are some of your recommendations that you have for solution builders out there? Yeah, so getting things working end to end in a way that is scalable, secure and future proof. That is really one thing to focus on.
I would say and and you know a hello world application for Lorawan really takes quite some effort and it comes with many many friction points and many existing guides that help you to build this Hello world application is is something you can basically throw away after after you know there's. Hello world on your screen because it was never designed in a scalable and secure way. But the integration here and that is also what I would recommend is to really start using building blocks that you can actually use when your application.
Need scale and need security and when it needs to become future proof and you know what you see here with the demonstration for the monk permanent single use case is actually the very same technology that you can build for a simple hello world test and you can build that further and skillet out. Perfect now. Richard and Joanne to close our conversation today. Could you please tell us what calls to action do you have for our listeners? And let's start with region right certainly. If they're interested in the Goldeneye platform itself,
we got there is a major website out there already which would show you the benefits of actually using the Goldeneye platform, how to monitor housing and everything. But if you need help with integrations of various items into different platforms then you can certainly contact us at look soft.com when there's certainly in the smart places and things and we can help with IoT. Integrations from different platforms, law around being one of the primary ones. Excellent thank you, Richard. Let's hear from Joanne now yeah, so the azurite hub integration that you saw in the central integration also are already available for our customers and it's even free for non commercial use on a community network. So you can get started with that today.
So yeah you can go to our website thethingsindustries.com you can sign up for a free discovery there you need you know any Laura win compliant hardware. It doesn't really matter which Gateway. Which in device it will just work. And yeah, you can follow the documentation and use this link to deploy the integration in your Azure account.
Excellent, so that's all for today. Thanks for watching.