kurt

RFID vs. GPS

Building a location-based mobile media content delivery system around GPS presents some huge problems.

It occurs to me that mobile phone manufacturers may not add GPS functionality to their handhelds for quite a while, or if they do the implementation may not be sufficient for our needs. I feel that we need to have a dialogue with the manufacturers so that we can understand their position on GPS technology and so that they can understand the service that we can potentially provide their customers.

It is primarily out of this frustration (as well as the problem of GPS reception in buildings) that I have tried to imagine an architecture design for our system that uses RFID tags instead of GPS.

An RFID tag is small and cheap. These are great features for the system that we are designing. They are mobile, they function perfectly well indoors and mobile phone manufacturers are already implementing RFID functionality in their phones: both tags in phones (here) and sensors in phones (here). This overcomes my three biggest concerns about GPS.

Our system should accomodate mobile objects. If not immediately, then as soon as technologically possible. Do not resist this.

Our system should work indoors. If not immediately, then as soon as technologically possible. Do not resist this.

Our system should use technology that mobile phones will have soon. If not immediately, then as soon as technologically possible. Do not resist this.

I'm not elaborating on these point as I think that they are obvious. I will discuss them point by point with anyone that feel they are debatable.

RFID tags can substitute for GPS. Just give the tag its coordinates in the database. We could even use the GPS coordinate system to keep consistency and to allow for GPS device compatibility. Best of all, mobile objects tracked with RFID could be automatically pinpointed by their proximity to other stationary RFID tags.

But there are some major problems with using RFID tags over GPS:

First, it is obvious that we would give up the platonic ideal of a network that is accesible worldwide from the day it is launched. Second (and a corollary of the first), we are limiting expansion of the network to physical "tagging" of spaces and objects. Thirdly, we are allowing for a large amount of instability in the network as physical RFID tags could be removed or vandalized. Fourthly, I am just not sure that the mobile user’s phone could be triangulated accurately from RFID tags in the environment. The user will not usually be standing on the tag itself so her position must be calculated from at least two stationary tags.

So what do we do? Perhaps we won’t have to make a choice at all and both technologies will become available simultaneously in mobile consumer devices. Then the advantages of each technology could be fused. But until that paradise emerges we should be prepared to compensate for the lack of one with an innovative use of the other.