Enabling the GeoWeb by Mass Market Geotagging

Location: Salon A-F Level: Novice

Geotagging is the labelling of things with their location. It is a powerful and increasingly popular means of looking at, sorting, finding, and sharing all sorts of things, from photographs and videos to points and events of interest. In the Web 2.0 age of user generated content it is a powerful means of linking, sharing and accessing GeoIndexed information on the Web.

“Capture and Process” is a new disruptive Geotagging GPS technology that can instantly capture locations, and is ideally suited to geotagging because of its instant response, low cost, and ultra low power consumption. A short <200ms sample of the GPS satellite signal is simply captured and stored in Flash memory. Later the user uploads the signal samples to their PC, and then accesses the Capture and Process Server using the Internet. The system then processes the signal sample to calculate the location where the samples were captured. The geotag can now be uploaded to the Web and used to plot the event on a map, for sorting or sharing the content or for other applications.

The new technology can be used to make small ultra low cost logging devices with long battery life (can run for years), or to conveniently include an easy-to-use Geotagging function in any high volume portable CE device like MP3 players, digital cameras, camcorders, and watches. The technology can equally be easily integrated in very small and light form factors, so that geotag modules can be embedded in everyday items such as clothes, running shoes, golf clubs, car license plates, credit cards, keychains, badges, and animal tags.


In 2007, an immeasurable amount of user-generated information is being collected and uploaded to the Web every day. Almost all this data, however, lacks a fundamental and key piece of information: location.

Location is an extremely powerful form of metadata, that enables powerful organization, presentation, and searching of data. Location is also key in bringing the Internet closer to the real world as seen in the incredible variety of applications seen at Where 2.0.

So why, despite its importance, isn’t everything around us geotagged? Our emails, documents, pictures, videos, and activities would all benefit if we knew the location in which they were created or edited. The reason is that capturing information with conventional location technology is complex and slow.

The dominant location technology is GPS, but this has been built around the real-time navigation use case, and is not designed for quickly acquiring locations on demand at all times. Consequently, the opportunities that geotagging creates are restricted to the technically able, and the committed who are willing to tag by hand.

NXP Software Location Technologies has developed a new technology and related service that will radically change this: Capture and Process. Locations can be captured in less than 200ms by simply storing the GPS signals raw and leaving the processing to any standard Internet-enabled PC at a later point in time.

With this new approach users can record locations at the press of a button. There is no need to wait for a fix, look at a display, or check the battery life.
Alternatively, a small coin cell operated device can be used in an automatic logger mode, taking captures at regular intervals for up to a year without needing a battery recharge.

At Where 2.0 we will demonstrate the potential of these universal loggers. The small low cost device will enable new opportunities for the geoweb, as the problem of gathering location information is solved. This technology will enable mass market generation of geotagged content, being it geotagged pictures, videos, wikis, blogs, emails, or any other user generated piece of content or information.

Ubiquity of location information, however, will not be achieved until location capture is embedded in all the devices we use. In Q4 2008 the first cameras will be released in the market with integrated Capture and Process—and for the first time geotagging photographs will require no changes to user behavior (just point and shoot). We will demonstrate the world’s first point and shoot, capture and process-enabled cameras at Where 2.0.

Where 2.0 is brimming with innovation—by providing the locations that underpin it, we will enable the geoweb revolution to reach the mass market.


The GPS system is well known 1, with 24 GPS satellites orbiting the earth at an altitude of 20,000km. By listening to their signals a GPS receiver can measure the time taken for the satellite signals to reach the receiver, can calculate the distance of the user from the satellite, and therefore find out where he is.

The tasks of a conventional GPS receiver are as follows:

Firstly the GPS receiver searches for the weak satellite signals, then it measures their relative timing accurately to give the distance to the satellites.
In order to calculate where the receiver is, the receiver must first find out precisely where the satellites are in space. This “ephemeris” information is transmitted slowly by the satellites, repeated every 30s. The need to demodulate this signal is what primarily determines the response time of a GPS receiver. Then, once it has decoded the information about the orbit for (at least) four satellites, the GPS receiver calculates a position fix. Subsequent position fixes when tracking can then be carried out quickly, first as the satellite orbit information is now known and can be reused, and second because since everything is now known it is easy to keep continuous track of the satellite signals.

This system design is fine for navigation on a journey, where the user is prepared to wait a minute or so to start with, and then wants to track position continuously, using energy while doing so. However, this is not suitable for the ad hoc use typical of geotagging applications. For example, camera application users expect instant “point and shoot” operation, want it to be small and light to carry around everywhere, and demand a long battery life.


Capture and Process is a new GPS technology for geotagging, by which the user can instantly capture location with a new approach to GPS.

In the device a small sample of the GPS satellite signal is captured and stored in memory. Then later, the signal samples are uploaded to a PC, and processed to calculate the location where the signal was captured. This means that the processing power of the PC is used to quickly and accurately process the signal in software and calculate the users’ location. This can be done when convenient, using the resources of the PC and assistance information provided by a server connected over the Internet. As a result the device itself acts instantly, can be small and lightweight, and can have a battery life of many months or even years.

The two steps, Capture and Process, are now described in more detail, using the case of a digital camera as an example.


When a photograph is taken, the following simple sequence of events takes place:

  • the camera requests the capture of GPS data.
  • The GPS radio in the camera is switched on, and RF signal is streamed into the GPS front-end radio, which down-converts the signal to an Intermediate Frequency and outputs the raw sampled signal
  • This raw signal is passed to the camera processor, which stores it in flash memory. Typically only a small sample of <200ms of signal is acquired each time a capture is taken.
  • The GPS radio is switched off again
  • A timestamp is written to the capture record
  • The GPS signal sample is put in a file, and embedded in the photograph or cross-referenced with the photograph image.

Note that the camera does not carry out any processing of the signal sample which is just raw GPS signal. The camera does not measure or know the latitude and longitude of the event, and the signal sample must be processed using additional information, before the location can be discovered. This approach minimizes the time, energy consumption, and complexity of the camera, by keeping everything in the portable device nice and simple.


The processing stage is where the algorithmic work is done, and the location calculated. It is initiated later when the camera is connected to a PC.

In order to avoid the difficult and time-consuming step in the device of receiving and demodulating the satellite “ephemeris” data about where it is in space, this is provided separately, from a server that has collected this information already.

The processing proceeds as follows:

  • The camera is connected to a PC, and the pictures are downloaded
  • As each photo is retrieved, the capture data is extracted from the image, and loaded into the Capture and Process Engine
  • For each capture the timestamp is extracted
  • The Server is contacted, and the positions of the GPS satellites at that time are downloaded over the Internet
  • The system then analyses the data to find the satellite signals, measures the “pseudo-range” distance to each satellite it can find, and calculates a geotag position fix of location and time.
  • The location is then provided to the photo mapping application.

Three innovations have been necessary to make this processing system practical:

  • A full historical archive of GPS satellite information, allowing fixes on samples collected anywhere in the world, at any time
  • Advanced algorithms to extract and measure the GPS satellite signals in only a small sample of captured signal
  • Tolerance of the signal processing to large errors in the camera clock, which only needs to be accurate to within 10 minutes.


As has been indicated, the service is necessary in order to provide the accurate information about the orbits of the GPS satellites in space, and to assist in the processing of the signal data in order to determine the location of the signal capture.
The server has the following tasks:

  • It collects detailed information about the GPS satellites and their orbit around the earth, the so-called “ephemeris” information
  • It archives this information, to build a database of the orbit and status of all GPS worldwide
  • It provides a timing service, to calibrate to UTC
  • It authenticates enquiries and matches them with the type of device being used.
    Then, when properly authorised to do so…
  • It provides assistance including ephemeris information about where the satellites at the time of the time stamps of the signal samples.
  • It then provides the location fix to the application


Geotagging with Capture and Process is illustrated with two applications.
The first, digital cameras, illustrates the value of performing an instant capture of the GPS signal. The second, a small lightweight logging accessory, illustrates the low power consumption, enabling weeks, months or years of operation.



For adding the Geotagging capability to a camera only a GPS antenna and radio receiver are needed—there is no baseband processing IC.

For each capture, every time a photograph is taken, the GPS radio receiver simply passes the GPS signal into the host processor, typically over an SPI bus, and the processor stores the signal in flash memory. (A control interface is often required to configure and activate the GPS receiver.) The sample is given a time stamp by the real time clock. When the images are downloaded from the Camera to the PC, typically over USB, then the GPS signal samples are also downloaded to the PC.


For camera applications minimum size is important. The small size that can be achieved in the Capture unit module is illustrated by a GPS front-end receiver module, the GRM6510 from Rakon 2. This module provides all the required GPS front-end functionality in a small package, including GPS radio receiver, filtering, and oscillator TCXO. The footprint of this module is just 6mm × 5mm, 30mm2. A GPS antenna is all that needs to be added.


Capture and Process for digital cameras has been tested by collecting samples from around the world and processing them.



A small low cost ($7 BOM) battery operated USB GPS logger capture device has been developed. Capture and Process for logging has been tested by taking these capture devices out and about, recording sets of samples, and processing these samples at a later time when the device is connected to a PC over USB. The resulting geotags can then readily be used and mapped.


Since no signal analysis is performed on the device, and the signal capture only takes <200ms, very little energy is used by the Capture device. In one experiment, a test devices was run recording a signal sample every 10 minutes 24 hours a day for 10 days. This was capturing signal samples for a total of 1000 position fixes, running for 10 days without recharging the coin cell battery.

The low energy consumption (10mJ/capture) to receive and store the signal sample means that a battery life of many months or even years can be achieved by a lightweight, convenient device. This would be undreamt of with conventional GPS technology.


Capture and Process is a new GPS technology for conveniently capturing GPS signals for geotagging GeoWeb applications. Location can be captured instantly—in <0.2s using a simple, small and convenient device to store the GPS signal, with negligable energy consumption and very long battery life.
Using the stored fragment of GPS signal the location is quickly and easily calculated later with the help of a server connected to the users PC. Trials worldwide have shown good accuracy (better than 10m), with a fix location produced in 96% of cases.

This is ideal for “geo-tagging” user-generated distributed content like e.g., photographs (Flickr, Panoramio), videos (YouTube), and also opens up many other Web 2.0 Geo-enabled applications. Capture and process technology will enable the GeoWeb by providing mass market geotagging.

1. Understanding GPS – Principles and Applications – E.D. Kaplan – Artech House – ISBN:0-89006-793-7
2. Rakon GRM6510 Data Sheet, Rakon Ltd

Johan Peeters


Johan joined Geotate after four years of incubating the Geotate business development at NXP Software from his base in Silicon Valley. Prior to that, Johan held the position of Director of Sales, the Americas for the Business Unit Digital Transmissions Systems of the Philips division Digital Networks, which followed his earlier position as General Manager Marketing and Sales of that business in Eindhoven, the Netherlands. Before 1997 Johan held various General Management positions with Philips Kommunikations Industries and Philips Business Communications in Belgium where his claim to fame is that he made the very first GSM phone call in Belgium after project managing the first installations of the GSM network infrastructure in Belgium.

  • Autodesk
  • Google
  • ESRI
  • Nokia
  • DigitalGlobe
  • Earthscape
  • EveryScape.com
  • LightPole
  • MapJack
  • MapQuest
  • MetaCarta
  • Microsoft
  • Poly9
  • Skyhook Wireless
  • TeleAtlas
  • Yahoo! Inc.
  • Zvents
  • BNet

Press and Media

For media-related inquiries, contact Maureen Jennings at maureen@oreilly.com.

Contact Us

View a complete list of Where 2.0 contacts.