Indoor location services promise to become a major growth opportunity, with retail sector revenues alone forecast to exceed $5 Billion by 2018. Many more applications offer increased employee or service productivity and other tangible benefits. A plethora of different technologies have been developed to achieve higher accuracy and reliability required. This is particularly relevant indoors, where Enterprise small cells could capitalise on this need, serving both requirements for location service and connectivity simultaneously.
A recent focus on indoor mobile location comes from US regulators and industry bodies who are converging on new requirements for in-building location for emergency calls. This could be another significant driver for in-building small cell adoption. Rich Lee of iPosi updated us on the latest industry developments.
New FCC rules for accurate in-building E911 mobile location
The US leads the way with strong mandates for location reporting for E911 calls from cellphones. When you make a call today, the emergency call centre agent should be notified of your position to within 150 metres for 90% of cases (where handset based techniques are used). A variety of techniques are used to achieve that, such as triangulating from measuring signal strengths and/or transmission delay from nearby cell towers or using a smartphone's on-board GPS receiver. This is even more important where emergency calls are made but the victim is unable to communicate or can't describe their exact position.
In the past, most emergency calls from indoors were made from fixed landline devices where a simple database lookup quickly mapped that to a street address and apartment number. Today over 2/3rds of emergency calls originate from smartphones – especially indoors. Earlier generation reverse geo-locating (looking up the Lat/Long co-ordinates and mapping these to street addresses) in the past has not been sufficiently reliable. Master address records (example) can easily become outdated. The error level varies widely between rural and metropolitan areas, but can exceed 10% and too often results in sending first responders to the wrong building.
This has led the FCC, wireless operators and public safety groups toward drafting new rules that reflect these realities. To this end, a joint roadmap has been published by the four national US Cellcos, CTIA (representing Cellular Equipment Vendors), APCO (Association of Public Safety Communication Officials) and NENA (National Emergency Number Association).
The Golden Standard of Dispatchable Location
However, it seems that everyone does agree on a definition and imperative of the "Dispatchable Location" as the prime indoor location objective – it might be a hotel room, a business suite, residential apartment or parking garage level – and this sets what Public Safety organisations would consider the Gold Standard.
A moving target: Voice calls migrate to VoLTE
The FCC originally proposed rules with separate timescales for today's predominantly 2G/3G voice calls as well as the new 4G VoLTE voice service. Their goal is to locate 50% of all wireless calls within 3 years and 80% of VoLTE calls within 6 years. This will be during a period where a large proportion of voice call traffic migrates to VoLTE and most users go through at least one smartphone upgrade cycle.
Existing technologies will be used where possible. A testbed will be established this year to evaluate these including (but not limited) to Wi-Fi, Bluetooth beacons, barometric pressure sensors (compensated and uncompensated), Small Cells, Assisted GNSS and OTDOA (Observed Time Difference of Arrival, part of LTE Release 9). Future smartphones may be mandated to incorporate some of these sensors or capabilities.
Databases and GIS enter the picture
One of the first issues is whether first responder computer systems can document in-building locations to this level of resolution. It's also fraught with error because of frequent moves and changes inside buildings and devices with dynamic MAC addresses (more below) and beacon "leakage" where beacon signals don't necessarily stop at the perimeter of in-building units, indeed can readily cross floors or units which introduces certain undesired location error possibilities. Experts are already looking into those matters. A National Emergency Address Database (NEAD) is to be established which maps Wi-Fi hotspot MAC addresses and Bluetooth beacons to Dispatchable Locations. This is similar to the MSAG (Master Street Address Guide) used for landline calls but more complex to manage given voluntary 3rd party record registration for instance.
Lee of iPosi takes the view that licensed and unlicensed Small Cells provide a vital gateway to automatically upgrade vast areas to usable indoor location including commercial security and safety in addition to seamless inclusion of E911 with ample accuracy, Dispatchable Location, at incremental low cost. Customers, small cell network equipment, operators and public safety all win with this approach.
During commissioning, small cell GNSS positioning provides a way to validate any voluntary, manually entered Dispatchable Addresses. iPosi is teaming with world class GIS system providers to present accurate pre-verified dispatchable location references based on accurate, continuously updated GNSS positions which match addresses and their precise (within lot or building) mapping coordinates. Pre-verification also has another critical value: the caller location is certain when E911 dispatch occurs based on a deterministic matched address.
Residential Small Cells already know their location
There is no mention of the NEAD holding addresses for small cells. This is probably because network operators already capture and hold this data – all residential small cells in the US have onboard GPS and report their location during initialisation. This is used partly to ensure they transmit on the correct spectrum (which varies by operator), to network sync for 3G CDMA or LTE, but also to determine emergency call routing and caller location.
Of course, Wi-Fi hotspots could also be equipped with GNSS receivers and report their location as small cells do but this reduces their remaining pricing advantage over small cells. It also risks moving the inherently ad-hoc, customer owned nature of Wi-Fi into more of a network operator owned/deployed/managed business model.
So at the moment, making an E911 emergency call from a residential small cell carries greater chance of accurate location fixing.
Verifying the "Z dimension" and Floor Level
Equally critical to E911 after obtaining the correct Dispatchable Address is determining the floor of the caller when there are more than two floors. This can be for residential apartments, office buildings and many other non-residential use cases.
Assisted GNSS is a key solution for solving indoor 3D and can calibrate or validate other techniques. Some might propose Wi-Fi but there are concerns that:
- not everybody has Wi-Fi switched on
- Wi-Fi hotspots can easily be moved around
- a significant percentage may not be accurately documented
- crowdsourced Wi-Fi location is not reliable since it tends to incorporate observations from distant points.
Barometric sensors are for high accuracy indoors in their early days though promising. Lee of iPosi believes that Small Cells could even include sensors to improve height accuracy since outdoor air pressure references are inaccurate when calibrating indoor mobiles – interior air column differences are well documented.
Prompted by the rising value of mobile indoor location, with further recent sharpening by operators and public safety in response to the FCC, the US cellular industry has set itself some tough milestones for in-building emergency location finding. Some of these may have benefits for commercial use (e.g. Apps), but I can't see the NEAD database being shared or used for non-emergency purposes. The US looks like it could well maintain its lead on emergency location reporting for cellular calls, especially as we move into the era of VoLTE voice calls. This could define the approach used elsewhere, especially if it affects the specifications for smartphone.
Having a denser network with more low powered radios, both licenced or unlicenced, should surely help improve the resolution of the mobile position found indoors. This bodes well for increasing enterprise, public venue and consumer in-building small cell deployments. It remains to be seen which of the other technologies proves successful on the testbed and is adopted by the industry for long term use.