Finding the value in mobile location data

Wireless editor James Atkinson reports from the recent Cambridge Wireless Conference which explored how mobile location data can be harnessed to add value to public and private services

Finding the value in mobile location data

Location-based services are expanding rapidly in many different sectors where they are helping the emergency services to improve safety and enabling public sector organisations and enterprises to add value to their services.

Attendees at a recent Cambridge Wireless event in London (25 October 2013) also heard how mobile location data could be used to predict future locations and events. This could prove vital for anticipating problems in public safety, as well as offering revenue opportunities in the enterprise and consumer world.

Solving indoor location issues

Introducing the event, Hamid Ahmadi, VP and chief innovation officer at CSR, which sponsored the conference, noted that thanks to GPS technology outdoor location based services have expanded almost everywhere. However, he said the next big thing was to improve indoor location services.

‘There are two different approaches to indoor location. You either have to install new dedicated infrastructure, as GPS doesn’t work indoors, or you use what’s there in terms of the existing infrastructure, such as cellular and Wi-Fi networks,’ said Ahmadi.

Key sectors for indoor location services include: smart buildings for office automation (e.g. lower/raise the temperature in response to the number of people in a room); advertising and marketing; retail; smart IT BYOD; healthcare; inventory asset monitoring; and people tracking.

There are a number of challenges that have to be overcome, however. The cost and complexity of the infrastructure have to be balanced against the return on investment. Other questions include: how scalable does the system need to be; are you looking at tracking or positioning; what level of accuracy is required; and what are you positioning – people or things, mobile or specialised?

‘Maybe positioning is not the answer,’ said Ahmadi. ‘Proximity services appear to be the current trend.’

In terms of the technical challenges, Ahmadi said that multi-path is a big problem - trilateration, triangulation and so on, and that existing communication networks are not designed as positioning systems, which can hamper accuracy of readings. The personal privacy aspects are also a key concern.

Ahmadi said that indoor maps are a very attractive possibility, but they are difficult to construct. ‘Where do the maps come from; how do you keep them up to date; who pays for them; and is the presentation suitable for navigation purposes?’ he asked.

He pointed out that presentation is a personal point of view. Crowd sourcing data via mobile devices is one possible way of gathering mapping data, but again this calls for the aggregation of personal points of view. In addition, the information is unstructured and privacy issues come into play.

Turning to potential business models, Ahmadi argued that revenues are not based on locations, but on the services unlocked by these technologies. He speculated that location information might become a commodity and asked:

‘Who owns what – the analytics, data, transactions v transport (service providers) – is the value in the transport or in the transaction you are trying to sell?’

He added: ‘What incentive is there for companies to put infrastructure in place if they can’t derive any return on it? The revenue opportunities probably come from mash ups of information, combining Google maps with real estate prices, for example. There is a revenue opportunity there, but is it coming from the core technology or what you’ve done with it? Is it better to go after the data and sell it, or extract revenue opportunities from the data?’

As an example he cited Google maps, which is used as an enabler for advertising; Google is not selling the maps. Then again he noted that where the core technology becomes pervasive, such as Wi-Fi and Bluetooth, in those instances it is the technology that is being charged for.

He concluded. ‘The Indoor Location Alliance is trying to create a standard, but I think this only defines the interfaces and APIs and not the services themselves, which is where the value add is.’

Transport for London iBus location system

Simon Reed, Head of Technical Services Group at Transport for London (pictured below), explained how TfL is providing more detailed and accurate information to its customers by making use of location data from its iBus system (a Siemens system originally – now Trapeze), installed on over 8,500 London buses carrying more than 6.3m passengers every day.

 

‘We have 43 control centres controlling all the routes and it costs us £1.6bn to run the bus network, so we track buses to ensure we make good use of public money and for when there are emergencies,’ said Reed.

The TfL bus network also has access to an analogue two-way radio system built around 10 professional mobile radio base stations, mobile (in-vehicle) and hand portable radios supplied by Tait Communications.

The iBus system uses advanced algorithms based on GPS and GSM bus location data updated every 30 seconds, along with other relevant traffic information to provide accurate positioning and arrival times.

‘We have five inputs to get the location right as GPS is not enough, including nets, odometers, GPS, maps and gyros. It is hard to ‘find’ bus stops, but if we are measuring them in relation to where the bus is we have to know. We define a corridor around the route, so we can pick up if the bus goes off route, and use that information to try and recalculate timings. So, we use multiple inputs to deliver the service,’ explained Reed.

This information along with service updates is available via SMS and mobile and fixed web services as well as new Countdown screens being installed at some 2,500 stops. TfL has also launched an Application Programme Interface (API) that feeds the same data to third-parties that has already led to the development of over 190 smartphone ‘apps’. All buses are fitted with audio syncs and visual displays to tell passengers where they are and what the next stop is.

The iBus system also enables TfL to manage the performance of bus operators as well as communicate with traffic signals for bus priority, help to identify and monitor congestion and even deliver ‘low bridge’ driver alarms.

Reed said that there were several reasons for deploying the iBus system including providing confidence and security to passengers, especially lone ones late at night. London buses are franchised so operators are set service level requirements and performance payments for rewards and fines are calculated using the iBus system to check they are hitting the targeted schedule of services.

“London operates one of the largest bus networks in the world and capturing precise location data and delivering accurate passenger information is helping to improve performance and increase passenger confidence and numbers,” said Reed. “We are also looking at new technologies such as 4G/LTE and the ability to harness more vehicle and passenger data to deliver more accuracy and value,” said Reed.

Location technology for Fire & Rescue Services

Olaf Baars, acting chief fire officer at Royal Berkshire Fire and Rescue Service, stressed that in situations full of variables and uncertainty, the delivery of effective emergency services revolves around reliable and accurate location information.

Baars pointed to a range of technologies from satellite navigation, intelligent routing, Automatic Vehicle Location Systems (AVLS) and dynamic mobilising to personal radio and GPS based systems that are being used to improve caller and incident location.  

He said: ‘APLS (automatic person location system) using radio (TETRA) and GPS based systems is not as good as AVLS because of urban canyon effects and lack of in-building where GPS doesn’t work. The fire service doesn’t use GPS therefore. But once at the incident, how should location based systems aid the rescue effort?

Baars said that the Royal Berkshire Fire and Rescue Service is therefore interested in the use of new indoor positioning technology to provide accurate firefighter location and tracking, particularly in large complex buildings where GPS does not work.

Consultancy firm Inperium, set up by Olaf Baars, is assisting UK based Omnisense, which is currently using 2.4GHz wireless sensor network (WSN) technology and patented cluster positioning techniques, to provide accuracy to a metre or less without the  need for an expensive fixed infrastructure.

“Firefighters can find themselves in harsh, unfamiliar environments with extreme heat and humidity as well as zero visibility and limited air supply, so being able to know where they are and the available escape routes can be matter of life or death,” said Olaf Baars.

“What is equally, if not more important is the ability for commanders and safety officers to be able to track the progress of personnel within the building to inform tactical decision making and manage firefighter safety.”

Baars noted that tracking and navigation in a normal house is not too difficult, but it is much trickier in complex buildings such as airports, tower blocks, multiple basements and shopping malls.

He also said that any system has to take into account the fact that building materials affect a fire and how it spreads. They can also affect radio signals and radio systems in the building, which would impact on the rescuers’ communications.

Looking into the future, he added: ‘When a fireman is operating in an unfamiliar building in extreme conditions it would be useful to have a heads up display on your helmet with a map showing you where you and your colleagues are. This would be seen by commanders as well.’

In the meantime, he said that any new location tracking technology needs to be lightweight and preferably integrated into clothing, so as to not increase the physiological burden on the firefighter. It obviously needs to be reliable and enhance communications both inside and outside the building.

‘The data must be graphically represented and easily comprehensible to the firefighter and it must be able to withstand the environment,’ said Baars.

Location prediction

Other speakers at the event included Dr. Rob Harle, from the University of Cambridge Computer Lab, who talked about new infrastructure-free indoor location systems using inertial and opportunistic positioning; and Mirco Musolesi, from the School of Computer Science at the University of Birmingham, who presented research on the possibilities of location prediction.

“Mobile phones are increasingly equipped with sensors, such as accelerometers, GPS receivers, proximity sensors and cameras, which can be used to sense and interpret peoples’ behaviour in real-time,” said Musolesi.

“Information extracted from sensors can be used to model and predict movement patterns and provide a very rich set of multi-dimensional data, for applications such as personalised marketing, real-time support for policy-makers and health interventions.”

Written by Wireless magazine
Wireless magazine

Leave a Comment