Referred to as the “missing component” in mobile location technology, Z-axis is a hot topic.
It claims to resolve the two-plus-decade search for 3D geolocation information. The final piece in the puzzle sending Responders to the right location. It suggests “floor level accuracy” for multi-storey buildings by providing additional altitude information.
For years the FCC has spoken of providing “dispatchable location”, the right door to knock on for Responders to find a mobile caller in distress.
So, the question is, does latitude (Y) + longitude (X) + altitude (Z) = dispatchable location?
There is no dispute, location inaccuracy of mobile calls still plagues the Emergency Number System around the globe. GPS, cell towers, triangulation and other location efforts using crowdsourced, noisy data estimate mobile location on a “best effort” basis as a “search area”. The problem has always been, how do we accurately locate a mobile caller in urban, and indoor environments? Because once a caller walks inside a multi-level building they become almost invisible.
And why does this matter?
Because more people live in apartment buildings, urban areas than ever before. More 9-1-1 calls are made through mobile devices than fixed ones. People need a 9-1-1 service. And thousands of lives globally are lost because of the problem.
Why Z-axis?
Z-axis refers to the “vertical” coordinate. GPS provides latitude and longitude, but for emergency calling there was nothing to identity a caller from anywhere other than ground floor. Hence the search for the Z component…to better assist Responders in narrowing down their searches.
But before moving on, here are some “vertical” definitions:
Altitude is classed as height above SEA LEVEL
Elevation is classed as height of the terrain above SEA LEVEL
HAE or Height Above Ellipsoid is synonymous with GPS. Ellipsoid refers to a mathematical shape of the earth's surface.
HAT or Height Above Terrain is your height above the ground.
How Z-axis works
Specific software is uploaded into smart devices already with barometric reading capability.
Mobile barometric readings are delivered to a cloud-based network.
Simultaneously, using preestablished physical network of “altitude base stations” reading are also taken to build a “picture of local environmental conditions”.
·This “localized base station data” is also sent to the cloud-based network.
The cloud-based network compares both readings, calculating a differential altitude reading.
·This final reading is relayed back to the mobile device typically as a “altitude” or Height Above Terrain.
Considerations
1. Location data is represented as altitude, HAE or HAT.
Typically location information is more useful in an address format. Reverse geocoding turns coordinates into a “physical address” (albeit with its own concerns). But this demonstrates that useable information is warranted and preferred. That’s what Responders are used to. Having anything other than an address raises the question “how useable or actionable is it?”. Plus is time and/or effort wasted trying to translate it?
If I told you I was at 180 meters altitude, or 202 ft HAT would you know where I am?
2. Accuracy of the latitude and longitude coordinates.
GPS has accuracy issues, especially in downtown and urban areas as we’ve discussed in previous articles. So the GPS “ground location” is still a search area, and GPS doesn’t work well inside buildings. Now add into this data “Height Above Terrain”, or HAE, or an altitude reading, and what you have is still a 3D search area.
If you look at the picture further into this article you’ll see a map. The left picture shows what is typically provided in a “best case” situation of around 50 metres or approx. 164 ft. The middle picture adds in z-axis, so we have an additional vertical search area to consider.
Depending upon the city, many streets or rights of way are around 66 ft (20.1m) wide. It is extremely feasible for an emergency caller to be located in one building, but with error rates on lat/long/alt they could show up as being across the street in a different building all together.
source: https://www.chicago.gov/dam/city/depts/cdot/StreetandSitePlanDesignStandards407.pdf
3. Accuracy of the vertical information
If Responders were provided with the correct building, they still have to search several floors to find the caller. The FCC state Z-axis benchmarks need to be +/- 3m, which is approximately 10ft “up or down”, meaning spanning a 3-floor search area. Better than before, but not a door to knock on. Also basements, parkades, lobbies are typically different ceiling heights, and there is usually no 13th floor in hotels or apartment buildings.
4. Barometric (air) pressure readings are effected by:
Air temperature
Altitude or Elevation
Moisture
Weather system changes
Also, sprinkler systems, fires, and HVAC inside buildings alter indoor barometric readings, irrespective of “base-station” data. This could lead to greater inaccuracies.
5. We need an altitude database
To determine the altitude of a mobile wireless device, and “equate it” to a vertical location, it’s necessary to create an “altitude database” for ALL x-y coordinate locations which have more than one possible altitude. Why?
Let’s duplicate an emergency scenario. You have 2 identical buildings, the emergency mobile caller is on the same floor, same time, same day, same network, same phone.
• Building 1 is at sea level, the barometric reading is 10219 Pa.
• Building 2 is 15 kms inland and has a barometric pressure of 10150 Pa.
Barometric pressure determines the altitude, even small increase/decreases creates a different result for caller vertical location: https://www.mide.com/air-pressure-at-altitude-calculator
Using the calculator in the link, the altitude for both barometric pressures yields results separated by approx. 43 metres. This doesn’t seem like much, but an apartment floor level is roughly 2.5- 3m.
Having a single barometric measurement for a single city isn’t enough. Close geographical locations require accurate representation of daily barometric pressure readings, to ensure accuracy. This would involve measuring and collecting large amounts of data (requiring infrastructure adaptations) plus storage and maintenance.
Simplified representations of current location, location + altitude and location as dispatchable address
6. Additional infrastructure of beacons, antennas, and stations
Following on from point 5, acquiring comparative barometric data involves using a large network of long-range, low-cost broadcast beacons established throughout the urban environment placed in locations such as cell towers and rooftops. This additional hardware would require expensive build out, installation and maintenance. You can read more about it here: https://www.nist.gov/system/files/documents/2020/07/13/pnt-0026.pdf
Summary
According to the FCC, “Dispatchable location means a location delivered to the public safety answering point (PSAP) with a 9-1-1 call that consists of the validated street address of the calling party, plus additional information such as suite, apartment number.”
GPS coordinates, represented as x -axis (latitude) and y-axis (longitude)
Z-axis (altitude)
County/State/Province, City/Town, and civic street address
Indoor location with useable information such as floor, room, apartment number
There continues to be speculation and lingering concerns around how actionable the Z-axis metrics are. Yes, it provides more information which may narrow down the search area, but;
Does it overload Responders with additional useless data?
Last year APCO International replied to the FCC with a “Petition for Clarification”. They stated:
“[The recently adopted z-axis rules could result in no meaningful improvements to 9-1-1 location accuracy.]” Read more information here: https://ecfsapi.fcc.gov/file/1025866713987/APCO%20ex%20parte%20911%20location%20Oct2019%20final.pdf
Furthermore, APCO said:
“Translating the Metric to Real-World Performance Operational challenges with the Commission’s proposed Z-axis metric are not the only significant problems. Adopting the proposed z-axis metric of +/- 3 meters for 80% of calls would not mean that the height of 9-1-1 callers will be known within 3 meters 80% of the time.
If field responders are attempting to match the 9-1-1 caller’s elevation using their own devices, several factors could inject additional, unquantified error beyond what is required by the proposed Z-axis metric or what would be reported by confidence and uncertainty information. Field responders who are attempting to match the elevation of a 9-1-1 caller will encounter variations in their own devices and z-axis solutions that will likely inject additional error.”
Additionally, FCC Commissioner (at the time) Jessica Rosenworcel commented that “[the height-above-ellipsoid (HAE) format that the carriers must use to deliver vertical-location information is not useful to Public Safety”].
She also said:
“There is not one 9-1-1 call center today that can take the raw numbers in height above ellipsoid and translate them into actionable dispatchable-location information. If we acted in this room today like the job is done, then we lied to you. And I’m afraid our decision was dishonest about whether the information is actionable. The fact so many 9-1-1 operators wrote into the agency and wrote to my office, telling me they were distressed that we would organize around information they could not use, I think is a testament to the fact that … we have a lot more work to do.”]
Read the full article here: https://urgentcomm.com/2019/11/23/fcc-adopts-new-rules-for-z-axis-vertical-location-of-wireless-911-callers-but-critics-question-their-practical-use/
So what’s the solution?
Emergency localization requires accuracy and reliability of data, useable information coming from existing infrastructures, providing dispatchable location at the time of the call. It should be interoperable with what we have and have the ability to evolve as NG 9-1-1 rolls out.
Let’s look at a Wi-Fi tag-based location system, where every Access Point within the existing Wi-Fi ecosystem becomes a verified location datapoint, or “tag”. It associates an authoritative address and placement information with the omnipresent Wi-Fi equipment, exceeding the limitations of estimated GPS coordinates and providing complete dispatchable address.
EML’s tag-based system was successfully demonstrated live at EENA and NENA 2021. The screenshot below demonstrates how dispatchable address is provided immediately at the time of the test 9-1-1 call.
EML provides outdoor and rural location, but also the indoor component with the exact floor and room number of the caller. It’s the “door to knock on” the industry has been looking for, and it’s available today.
