Technology
Unique approaches and solutions powering our products
BlueZoo technology is protected by five United States patents and dozens of trade secrets. Our solutions integrate passive Wi-Fi sensors, cloud-based applications, and mobile apps for iPhone and Android. Nonetheless, the bulk of our engineering effort is invested in our scalable cloud systems to collect, analyze, and report real-time and historical consumer behavior.
BlueZoo goes to extreme lengths to protect consumer privacy. We do not resell data. We collect MAC addresses, but encrypt them on our sensors at the moment of collection and delete their encrypted form after transmission to our servers. We do not correlate MAC addresses with any personal information.
BlueZoo takes advantage of the ubiquity of mobile phones in the developed and developing worlds. We passively listen to spontaneous Wi-Fi emissions (“probes”) from mobile phones to measure the occupancy of a space and the aggregate flow of mobile phones through a space. The accuracy of our results depends on statistics and analytics and is optimal at many dozens or hundreds of mobiles, and less precise at very small numbers of phones.
In this section you can review the evolution of Wi-Fi in mobile devices and how mobile phone manufacturers have progressively moved from sharing static phone identifiers (i.e. “Global” MAC addresses) to ephemeral phone identifiers (i.e. “randomized” or “Local” MAC addresses).
Technology landscape
The technology landscape for measuring occupancy of a space is broad. First, technologies are divided between those that measure entrances and exits across a perimeter to infer occupancy, those that measure occupancy directly, and those that track consumers everywhere using mobile applications and GPS.
| Technique | Perimeter detection with Entrances & Exits | Direct Measurement | Inference by GPS Location Services |
|---|---|---|---|
| Applied technologies | Cameras over doorways, light beams, infrared sensors detect visitors crossing a threshold | Wi-Fi, Bluetooth, cameras (e.g. security cameras) detect visitors | “Spyware” payload carried by third-party mobile apps reports consumer location |
| Features | Detects consumers only when in range of sensors, beacons, or cameras | Detects consumers only when in range of sensors, beacons, or cameras | Tracks consumers everywhere |
| Requirements | Requires well-defined entrances / exits | Varies with technology | Requires installation of mobile apps |
Perimeter detection’s greatest limitation is that it requires a well-defined perimeter, ideally with a small number of entries/exits. The technology is not well-suited to open spaces like malls, lobbies, or locations with either very wide entrances or large numbers of entrances. Camera technologies also carry concerns over facial recognition, require suitable lighting that may change over the course of the day, and are very sensitive to field of view.
GPS location services are the most likely to seriously compromise consumer privacy. While a consumer may have clicked “allow” when asked if a mobile app could use location services, most consumers don’t understand the implications of allowing their flashlight application to track their location at every moment of the day. Either government regulation or action by Google and Apple are likely to constrain the behavior of geolocation tracking apps.
Among the technologies for direct measurement, advantages and disadvantages vary widely.
| Dimension | Wi-Fi | Bluetooth Beacons | Cameras |
|---|---|---|---|
| Range | Large to small: radius of 300 to 15 feet | Small, requiring many beacons in large spaces | Medium to small (depending on camera location and resolution) |
| Privacy | Excellent, because mobile phones use identifiers do not correlate to personal information | Poor, because installation of a mobile app on each consumer’s mobile phone is required | Poor, because cameras make possible facial recognition. |
| Accuracy | Good, because a direct line of sight is not required. | Varies: Bluetooth is good because the “cells” are small, but poor because few visitors install the mobile apps. | Poor, because visitors that leave from and return to the field of view are “new visitors” |
| Cost | Low, because a single sensor can cover up to 300,000 sqft | High, because a single sensor typically covers 1,000 sqft. | High, because many cameras are required to prevent unseen corners. |
Privacy protection
BlueZoo products respect the privacy of all consumers and have been certified compliant with General Data Protection Regulation (GDPR) by ePrivacy. Unlike mobile apps, BlueZoo products collect no personal information, and depend on the MAC (media access control) address broadcast by mobile phones to distinguish one mobile phone from another on a Wi-Fi network.
All mobile phones broadcast MAC addresses in order to share data on a Wi-Fi network. These addresses are identified by 12 hexa-decimal characters. To allow MAC addresses to be unique across all devices worldwide, allocation of MAC addresses is regulated by the Institute of Electrical and Electronics Engineers (IEEE). MAC addresses can be either “Local” or “Global”. Some modern mobile phones broadcast “Local” MAC addresses that are selected at random from a set of trillions of possible numbers.
Because two different mobile phones can broadcast the same MAC address, it is impossible to unequivocally identify a phone knowing only its broadcast Local MAC address. Further, because some mobile phones regularly change over time the MAC address that they broadcast, identifying an individual mobile phone over time is very difficult.
BlueZoo products do not seek to track individual mobile phones but only to statistically estimate the number of phones present in a space and how they relocate to other spaces during limited time intervals.
Almost all older mobile phones, and many new Android phones, broadcast Global MAC addresses, sometimes called “OUI (Organizational Unique Identifier)” MAC addresses. These addresses do not change over time and can uniquely identify a mobile phone… though they share no information about the owner/user of the mobile phone. Individuals whose phones broadcast Global MAC addresses can register their addresses with BlueZoo at no charge to ensure that the presence of their phone is not accounted for in our statistical estimates. Use this form to register the 12-character “Global” MAC address of your mobile phone.
All phones, whether broadcasting Local or Global MAC addresses can contribute to our real-time count of the number of visitors, visits, and dwell time. The 50% of mobile phones that broadcast Global MAC addresses contribute to our measurement of unique visitors and unique visitor recurrence. This 50% of phones provide a very accurate picture of the behavior of phone users. This share of mobile phones that broadcast Local MAC addresses is climbing slowly, led by the adoption of Local MAC addresses by Apple. Still, most new Android phones broadcast Global MAC addresses and we expect this trend to continue.
Wi-Fi
Mobile devices use primarily two technologies to communicate with other phone users: cellular networks and Wi-Fi networks. Wi-Fi networks are often preferred because Wi-Fi communications are widely available at no cost to the consumer and rarely impose connectivity limits. Over ninety-five percent (95%) of all mobile phones actively broadcast Wi-Fi probes, whether or not they are connected to a Wi-Fi access point. Mobile phones identify available networks by broadcasting “probes” that say “I am here; is anybody out there?”. The phone identifies itself by broadcasting its MAC address.
These probes prompt access points (and routers) to respond with their SSID (Service Set Identifier) and whether or not that access point requires a password. A mobile phone might broadcast a 12-hexadecimal-character MAC address (e.g. 4A:78:1E:95:FA:23) and in response learn that three access points are within range:
- Mary’s WiFi, password required
- Comcast Xfinity, password required
- Free Public Wifi, no password required
Mobile phones probe incessantly. Even when connections to access points (i.e. hotspots) are “disabled” by a phone’s user, virtually all mobile phones continue probing. Phones perform these probes to improve location services and improve data collection for Google and Apple. While engaging “airplane mode” should eliminate a phone’s Wi-Fi radio, the Wi-Fi radio can be re-enabled while airplane mode is still engaged.
BlueZoo Wi-Fi sensors passively listen for probes. Essentially, the more probes our sensors hear, the more phones are in the neighborhood. The higher the power received from the probes, the closer the probes are. Based on trillions of probes heard by our sensors worldwide, and the hundreds of thousands of probes heard by any individual sensor, our cloud systems are able to infer a real time count of mobile phones and the length of time a phone remains in a space, among other details. We don’t know who owns the phone or why they have visited, but we often know if they’ve been here before and sometimes where they have come from.
Wi-Fi has evolved enormously since it was first introduced in the 1990s. Originally, all MAC addresses were “Global” addresses consisting of a 6-character Organizationally Unique Identifier (OUI) that identified the company making the equipment, and a 6-character serial number. Starting in 2014, after revelations by Edward Snowden about the tracking of mobile phones using Wi-Fi, Apple iPhones began probing with “Local” MAC addresses. Google’s Android phones adopted “Local” MAC addresses in 2017. Phones from both companies are able to probe and connect to access points using Local MAC addresses–though many new phones do not provide these capabilities and others don’t use them by default.
Role of calibration
One challenge of counting the number of phones in a space is that spaces and phones vary. Some spaces are constructed of materials that block Wi-Fi signals, while others allow Wi-Fi signals to pass freely. Some spaces have internal obstacles (e.g. metal filing cabinets), while others do not. And few spaces are circular, a case that would allow signal strength to be a good proxy for the perimeter of the space. Some phones probe more aggressively and with higher power than other phones…though the variation among phones is surprisingly small.
As a result, BlueZoo sensors operate best if they are calibrated, a process that compares “ground truth”, the true number of persons in a space, with the Wi-Fi signals collected by BlueZoo sensors. Generally, we like to collect ground truth on several different occasions, both when a space is crowded, and when it is less crowded. We offer a purpose-built application (available in June 2021) that permits a group of people to simultaneously collect ground truth, for example at different entrances to a space. Our systems collate this ground truth and compare it to the values detected by our sensors to calibrate sensors for optimal sensitivity.
A single BlueZoo sensor can accurately count many hundreds of mobile phones in a space, such as a retail store, cafeteria, or hotel ballroom. Indeed, the hardest part of calibration is collecting the ground truth in these deployments. It’s not easy to count precisely the number of people in a hotel ballroom, with multiple entrances/exits and continuous flow of people in and out. Asking people to stand in place or taking dozens of (intrusive) photographs to be later examined are not practical options.
One technique that BlueZoo employs is to collect a series of entrance/exit data. We don’t require an exact occupancy count at the beginning of the entrance/exit series. Staff with our mobile app (or other highly-accurate counters) are placed at each entrance and count the number of persons entering and leaving the space. Our mobile app transmits this information to the BlueZoo cloud where we extract in/out traffic data for the time period when all entrances are covered.
First, we select one of our half-dozen computational methods that we have honed since 2016 and which are protected by trade secrets. We apply this computational method to the ground truth data to minimize the mean square error across all sample points.
Second, in the statistical counterpart of curve-fitting, we match the exact entrance/exit data with sensor data, scaling the data to achieve a good fit, and deducing the ground truth at the outset of the measured entrance/exit sequence.
Patents
The BlueZoo products are protected by six patents issued by the United States Patent and Trademark Office. Other patents are in process. The granted patents are listed here:
| Title | Application Number | Filing Date | Patent Number | Issue Date |
|---|---|---|---|---|
| MOBILE DEVICE DETECTION AND TRACKING | 15/006,057 | 2016-01-25 | 9,769,625 | 2017-09-19 |
| MOBILE DEVICE DETECTION AND TRACKING | 15/426,945 | 2017-02-07 | 9,936,357 | 2018-04-03 |
| MOBILE DEVICE DETECTION AND TRACKING | 15/426,953 | 2017-02-07 | 10,395,278 | 2019-08-27 |
| MOBILE DEVICE DETECTION AND TRACKING | 15/681,076 | 2017-08-18 | 10,156,167 | 2018-12-18 |
| PASSIVE AND ACTIVE TECHNIQUES FOR PEOPLE-COUNTING | 15/823,478 | 2017-11-27 | 10,440,505 | 2019-10-08 |
| MOBILE DEVICE DETECTION AND TRACKING | 16/249,760 | 2019-01-16 | 11,151,611 | 2021-10-19 |
One capability protected by our patents that is clearly differentiated is measurement of unique visitors in a manner consistent with GDPR certification. Measurement of unique visitors is especially valuable in many use cases because it permits individual behavior to be characterized in aggregate.
For example, in out-of-home advertising the number of consumer “impressions” is a useful measurement. These impressions can be measured using Wi-Fi or using video cameras. However, unlike video cameras, BlueZoo can also measure unique visitors and unique visitor recurrence which is a separate and useful dimension for advertisers.
