Sunday, July 19, 2015

Overview of indoor positioning

Author: Zhenqiang Su

 If you are new to this phrase -- indoor positioning, you will argue why we need this. Yes, the GPS works really well around the word. When you drive you car anywhere, GPS direct the way for you. It's true, but it does not work indoors! It even can not work well in down town when lots of skyscrapers stand there. So when you are in a big shopping mall or a big airport, you can't count on GPS anymore to get the directions. And that's why the indoor positioning springs up!



Nowadays, it seems that we will lost our way in the word without the GPS, just like we can't live without phone and computer! Similarly, we will lost the way in the building without indoor positioning.  Undoubtedly, indoor positioning will, actually it has been, be the next dominant network service to greatly facilitate our lives, also it's fashionable and cool!. 

But why, now in 2015, we still can not find out an real useful service in indoor positioning or its related area? Because currently, no company or an institution or an individual can provide a applicable solution to this service! I mean 'applicable', means that a solution which is accurate, economic and easy to install and use! Each term with an underscore is very important! Otherwise, it's not applicable (or no attractions) to consumers and therefore no benefits for a company to  build them. Or in other words, currently, in terms of technology, we are able to build a indoor positioning system with high accuracy but it's not economic. Or it's economic but not accurate. Or .... ..  

You may want to know why it's so difficult to build a system with all above characters. Anyway, it's indeed not any easy job. Give you some facts first:  
  • This area has been studied for at least a decade!! 
  • Many companies including many huge companies like Goolge, Micorsoft and many small and new start-ups have been engaged in this war for many years. 
  • Many institutions ...
After knowing these facts, you may have a impression how hard it's. Right now, we can still not be able to find a satisfying solution. So if you are able to find out a solution, you will be the next millionaire, undoubtedly. 

Again, why it's so hard?? why???

In terms of techs, it's a simple issue as we have several different available techs to achieve the accuracy (we say the accuracy should be at least around 1~1.5 meters to distinguish where you are in a big room), like the system using ultra-wide-band (UWB) signal. The accuracy with UWB can be up to centimeter level! BUT, BUT how can we use it?! Each of us, say, in a big shopping mall should carry a specific machine/device in order to locate where we are? We need to carry an additional device with us which is only used to do locating? It looks ridiculous and funny. And also, another important thing, who should pay the bill for this extra device? No one want. So this situation blocks the way to apply them for these available techs.  Maybe some of you are smart and think this way: why not integrate the available techs into the smartphone/mobile-device so that we don't need to carry such additional device. In theory, it's indeed a good solution! BUT,  AGAIN, who is the people who can persuade all or at least most of the phone/mobile-device companies to integrate them into their device?  At first, the additional function will definitely increase the cost of the mobile-device. Secondly, most of the mobile-device producers should agree to integrate it. If say, only Apple agrees to do it, that way only people with iphone/ipad can use the service while the others can't. For a shopping mall, they hope all customers with any mobile device can use the service. Thirdly, even all mobile device producer agree to integrate a tech into the phone, since there are several available techs, which one they should agree to use? A lot of these things make the way difficult to go.  

Therefore, you may have already known, currently the way or direction most of the companies/institutions followed to try to work out an solution to the indoor positioning is to try to make full use of the available resources to achieve this goal.  So what's the 'available resources'? The available resources here refer to the infrastructures and devices that have already been widely employed in the commercial market and/or in our daily lives OR some kinds of resources that although they are not widely employed, they can be easily and economically put into widely use.  The widely employed resources which may be used to achieve the indoor positioning are listed below (not necessarily restricted by this list):  
  • The telephony communication system include the networks and the cell towers (or base stations);
  • The internet networks and the access points (APs);
  • The mobile device like phone, pad, etc.;
  • The sensors employed in the smart phone, like inertial sensor system, cameras, etc. ;
  • Widely used Bluetooth techs
  • ... 
Now let's see how can we use above resources to achieve our goal -- indoor positioning? In fact, it's the same problem of 'how the  positioning is achieved?', as the basic methods/algorithms used in general positioning problems are the same as that in the indoor positioning. So we need to dive into a general problem: how positioning is achieved. There are many different kinds of methods/algorithms which have been applied to this problem. The following table shows a classification of major potential methods/algorithms/techs:
                           TABLE 1-1: classification of positioning

This table just gives you a general idea about how positioning (including outdoor and indoor positioning) is achieved. We will introduce the details later and now just keep going in top level. Currently, it's still difficult to say which method is better for indoor positioning, as almost all listed methods have been tried to do the indoor positioning. Each of them has its own merits and drawbacks. And some of them are still in investigation.  Usually, one institution or company only focuses on one method, or at most a hybrid of them due to the due to the limitation of funds and resources. But the ultimate goal is the same, the accuracy! 
Now it's time to give you a general impression about how many companies are working on this end. Following it's a list about companies VS. the methods they used to achieve the indoor positioning : 
WiFi Triangualtion 
WiFi Fingerprinting 
Beacons
BlueTooth 
Sensors
(quoted from: Indoor Location startups innovating Indoor Positioning. Only part of list is shown, please refer to the original link for the whole list.)

All of above companies in the list, an impressing company is the WifiSlam, who uses the WiFi combined with many different kinds of sensors integrated in the phone to do the positioning and achieve a reasonable accuracy. What you need to do it's just to down load its app, install and use it.  Another impressing company is Google, who is trying to use camera to extract the location information in indoor environment (combined with the indoor view -- an indoor version of street view )
Since the techs used in both indoor and outdoor positioning are the same, we will introduce the different techs listed in TABLE 1-1 one by one in the following sections.



Posted by at No comments:
Subscribe to: Comments (Atom)