Wireless or radio technologies are making waves (sorry for the awful pun) at the moment. The telco world is abuzz with the changes being brought about by 5G, IoT, LEO satellite and other related tech. They all rely on radio frequency (RF) to carry their all-important communications information.
This article provides an example of preparing RF coverage maps by using the inventory module of our Personal OSS Sandpit Project as well as a freely available trial of Twinkler (www.twinkler.app). We’ve already shown some RF propagation in previous articles in this series, including:
- Fixed Wireless (including line of sight and viewsheds),
- IoT (including smart-city design)
- 5G prototyping (including physical, logical and virtual mappings) and
But in today’s article, we’ll show radio coverage planning across the three sectors of a cellular network site. We’ll also throw in a quick augmented reality (AR/VR) demo.
This RF planning capability is becoming more widely useful with the advent of these technologies and the new business models / use-cases they support. Whereas RF planning used to be the domain of mobile network operators, we’re now seeing more widespread use including:
- Neutral-host infrastructure owners
- Wireless ISPs
- TowerCo owners
- IoT networks
- Private mobile networks, especially for in-building or in-campus in-fill coverage (BTW, the Twinkler app allows both indoor and outdoor RF planning use-cases)
- Government radio networks (eg emergency services)
- Enterprise (eg mining, transport / logistics, agriculture, broadcast radio / TV, etc)
- Consulting services
- Managed service offerings
In this sand-pit demo, we’ll reverse-engineer an operator’s existing tower / assets, but the same approach can be used to design and configure new assets (eg adding antenna for millimeter wave 5G). Here in Australia, ACMA provides a public record of all licenced comms transmitter / receiver devices (as does RFNSA – the Radio Frequency National Site Archive). This link is an example of one site recorded in the ACMA database that we’ll use for this demo – https://web.acma.gov.au/rrl/site_search.site_lookup?pSITE_ID=204667 (Warrimoo Tower). We’ve developed queries to extract key ACMA data into our inventory database.
You may recall that we’d recently demonstrated building a 3D model of Warrimoo Tower. This was done by stitching together photos taken from a drone. Looking at the animated GIF below, you’ll notice that we’ve not just built a model of the tower, but also tagged the assets (antenna, combiners, radio units) attached to the tower in 3D space. If you look closely, you’ll notice the labels that appear at the end of the visual loop, which identify the names of each asset. We can prepare similar 3D models inside comms rooms, of devices and even of corridors (eg rail corridors) or areas (eg housing estates).
Whilst Warrimoo tower holds assets of many owners (different colours on the animation), we’ll focus specifically on one 3-sector cluster. We’ll specifically focus on 3 antennae owned by Optus transmitting in the 700MHz band (763 MHz centre frequency). These are highlighted in green in the 3D model above.
The steps we use for RF planning are as follows:
- Extract ACMA data into Kuwaiba, our inventory database
- Push data from Kuwaiba to Twinkler, our RF modelling tool.
- Visualise the radio coverage map
- Visualise the tower in 3D, including assets (and associated height, azimuth and tilt calculated from the 3D model for cross-referencing with ACMA data) and RF plumes
Step 1 – Tower / antenna data in inventory:
The following diagram shows the inventory of the tower in topological format. The 3-sector cluster we’re modelling has been circled in yellow. [Click on the image for a closer look]
You’ll also notice that we’ve specifically highlighted one antenna (in blue, which has the name “204667-ANT-81193-MG1-01 (Optus)” according to our naming convention). There’s a corresponding list of attributes in the right-hand pane relating to this antenna. Most of these attributes have been taken from the ACMA database, but could equally come from your OSS, NMS, asset management system or other network data sets.
Some of the most important attributes (for RF calculation purposes anyway) are:
- Device make/model (as this defines the radiation profile)
- Height (either above the base of the tower or sea-level, but above the base in our case)
- Azimuth (direction the antenna is pointing)
- Emission centre frequency (ie the frequency of transmission)
- Transmission power
Step 2 – Twinkler takes these attributes and builds an RF model/s
You can either use the Twinkler application (sign up to a free account here – https://twinkler.io). The application can visualise coverage maps, but these can also be gathered via the Twinkler API if you wish to add them as an overlay in your inventory, GIS or similar tools (API details can be found here: https://twinkler.io/TwinklerAPITestClient.html).
Step 3 – Visualise radio coverage diagrams
As you can see from the attributes in the inventory diagram above, we have an azimuth of 230 degrees for the first antenna in the 3-sector group. The azimuths of the other two antennae are 15 and 140 degrees respectively.
These give us the following radiation patterns (each is a separate map, but I’ve combined to make an animated GIF for easier viewing):
You’ll notice that the combined spread in the diagram is slightly larger because the combined coverage is set to -130dBm signal level whereas the per-sector coverages are to -120dBm.
Note: You may have spotted that the mapping algorithm considers the terrain. From this zoomed-in view you’ll see the coverage black-spot in the gully in the top-left corner more easily.
Step 4 – Visualise the tower in 3D, including assets and RF / EME Plume
The assets from Kuwaiba are cross-linked with the 3D model (as shown with the asset outlines below). Since antennae are identified in 3D space, it allows the height, azimuth and tilt to be identified. Furthermore, this allows the RF / EME plume (inner plume marked in red, outer plume marked in blue) emanating from the antenna. These plumes are used to mark the non-occupational exclusion zones.
If you look closely, you’ll also notice the yellow and blue lines emanating from the site at the bottom of the tower below. These lines represent the fibre backhaul and power feed links coming to site respectively (if you look really closely you’ll also notice the service pits that support them). If you refer back to step 1, you’ll also notice more details relating to the fibre backhaul including nearest ODF (Optical Distribution Frame), Rack, etc. It can also show the active equipment that feed the remote units (though not shown on that particular screenshot).
Augmented Reality (AR/VR) View of the Tower
Using the 3D model shown above, we can also generate an augmented reality view of the tower and even coverage maps (not shown in animated GIF below). The video (shaky – sorry) shows a tower projected onto our boardroom table as recorded through my phone. The placeholder and orange annotations are static in this example, but can easily show real-time data streams.
BTW. If you’re interested in having a closer look at this tower being projected into your space through your phone, tablet or headset, please leave us a note and we’ll send you details. No software installs are required on your side.
In Building Coverage (IBC)
We can also combine the radio coverage of Twinkler and AR of Appearition to perform radio coverage planning inside buildings, to identify blackspots in your building or on your campus before installing your access points. It’s great for IBC design.
I wish we had’ve had that capability on a project we did back in circa 2008. We had to provide coverage across an oil / gas processing plant where we fitted out 120+ buildings with comms and 30+ different systems (eg PAGA, BMS, security, etc). Our WiFi and radio coverage planning techniques were a lot more rudimentary back then!
Drive Testing, SON and other Capabilities
In addition to the examples above, we’re also in discussions with leading vendors to incorporate drive testing, SON / call-trace data visualisations, AI-driven network energy reduction (for sustainability) and other related capabilities to add to the list above.
I hope you enjoyed this brief introduction into how we’ve created a radio coverage map of an existing cellular network using the Inventory module of our Personal OSS Sandpit Project. Click on the link to step back to the parent page and see what other modules and/or use-cases are available for review.