The 3 P's of Location - Presence, Proximity and Positioning

The 3 P's of Location

When using any sort of Radio frequency to locate assets there are in general 3 types of location

that one would consider. Presence, Proximity or accurate Positioning. In a nutshell the way to

solve each consequent P is to add more locators gateways closer to each other. 

The more locator gateways the more accurate the position can be determined. 

Presence

 If you have one BLE gateway for instance that is sniffing BLE beacons then you will know

within 10's of meterswhether a device is near the gateway or not. This very inaccurate device ID

capturing would be considered Presence as you would know whether a beacon is near or in the

presence (field of view) of the gateway. This is helpful for instance where you want to know

whether a person or asset is in a warehouse but you don't need to know in which bin it is for

instance. 

Proximity

 If you were to place 4 gateways, one at each corner of a warehouse, you would be able to
calculate the proximity of the beacon to the closest gateway. By using triangulation or trilateration
you would be able to calculate by either using the RSSI - Received Signal Strength Indicator or
one of the other more technical algorithms to determine more accurately the proximity to one
of the gateway for each beacon. 

Position

 However, for more accurate positioning of the beacon you would have to place many gateways

over the whole warehouse at close intervals in a grid array. This would use the algorithms

mentioned above to calculate how close the beacon is to a set of gateways and then where

the beacon is spatially in relation to the gateways. This accurate positioning requires many

gateways to be successful. 

RFID

RFID having a much closer range than BLE or WiFi RF signals needs to have many more
gateways but the beacons (stickers) are extremely cheap compared to BLE beacons so the
gateways and antennas become a larger capital outlay but the cost of a few hundred
thousand tags quickly makes sense because each tag is so cheap. So large capital
outlay for antennas and readers but cheap stickers make this a good scenario for
certain use cases. 

 If you would like to be able to see which rack an asset is then you would have an antenna and

reader per rack with RFID, whereas with BLE or WIFI the beacons can send their signals way

further, sometimes 80 meters or more. So then you would use fancy algorithms to figure

out where each beacon is situated. 

Written by John K Weber - IoTdc

Here is an excellent article explaining a bit more in detail the points mentioned above by

an amazing tracking company called Quuppa. Click here for article

NOA Data Service - Making IoT device integration easier

Setting up a POC - Proof of Concept or POV - Proof of Value, for IoT devices can be extremely time consuming. If your development team doesnt understand how the devices interact with each other then it is going to take some time to integrate your devices and set up your POC.

At IoTdc, based in Centurion, we have realised that this type of integration takes up more time than most teams realise and when just trying to prove a concept, or trying to prove the value that the ioT solution might bring to your company, time is a commodity that can add up to a considerable amount super fast and in many cases the POC actually doesnt prove value and in that case the project wont progress much further than the POC. So time is not something you want to waste.

The faster you can get your data from your devices to a dashboard or alert system the sooner the end customer can see whether the project will add value to their company or not. 

For this exact reason we have created an integration system that will take your POC development time from months, weeks or days to hours and minutes. At a recent instalation of a few milesight devices at a restaurant , used to measure temeperature and humidity in the walk in fridges the devices were stuck in place and the dashboards were created literally the same day. The LoRaWAN network was setup at the restaurant and within in minutes data was being sent to the NOA Data Service backend (device integration system created by IoTdc) and within the next hour or so the dashoards were created and the client could see within the same day that one of their new fridges was actually not performing as expected and was actually being packed incorrectly. 

This is the insight you can only gain from seeing real data on a dashboard. I have included a snapshot of the dashboard just for clarity.

This data is comprised of a few LoRaWAN based Milesight EM-300TH and EM-300MCS devices. There is a Kerlink iFemto Evolution LoRa basestation with GSM backhaul capability and the data is first sent from the LoRaWAN gateway to a LNS - LoRa Network Server from where it is is then forwarded via an HTTP integration to the NDS backend server. The data is stored in an influx timeseries database and then visualised using Grafana. This layer could have even been the clients own BI system or visualisation layer. 

In many cases an existing client might want to see how the data can be visualised on their own backend system and then NDS - NOA Data Service will send the data in the correct payload format to their server and the client will do their own visualisation.

In all cases though the integration of these devices was already done by our team and the POC took hours to set up and for the client to start seeing the value.

In the case of this restaurant the reasons why this data is valuable is twofold. the first being insurance premiums can be lowered due to the visibility of how the walkin coldroom are being managed, hence lowering the risk and secondly for the maintenance teams that can now do maintenance scheduling based on their insights gained by the data gathered from these devices.

The Milesight EM300 series has many different sensors that are already all pre-integrated into NOA Data Service, which means setting up an y POC using these devices takes hours instead of days or weeks.

Some of these sensors like the Milesight EM300-UDL use and ultrasonic distance level sensor to tell the height of waste in a glass recycling bin as an example. This will allow the recycling company to do waste removel scheduling istead of having to drive to the same sites every day to empty bins that arent even full yet. This is a real use case where value can be measured in rands and cents due to the increase in petrol price here in South Africa. This ability to schedule pickups really is a great benefit for recyclers.