Tuesday, April 28, 2009

Mobile internet visualized

It's a commercial but nicely puts numbers and uses of mobile internet into perspective:

and here's another one:

Monday, April 27, 2009

Impressions from IEEE RFID in Disney World/Orlando

I just got the chance to visit the IEEE RFID conference since together with Mikko, a Ph.D. student of ours, we were presenting two papers at this event in the area of RFID security [1][2].
It was quite striking to see how this conference grew. IEEE RFID has received 175 submissions whereof 47 had been accepted and presented in three tracks.

The conference was opened by a keynote from Chris Diorio, co-founder of Impinj and assistant professor at UW. After pointing out the history of the developments concerning UHF RFID he drew some interesting conclusions with regards to future challenges and research. Firstly, as development of RFID initially mainly aimed at replacing barcodes, security and privacy aspects had been neglected at first and only came in later. However, in his perspective especially privacy is not as bad as suggested, it is not primarily research, it rather has a lot do with media and education. Thus, researchers should rather focus on security technology, but the problem of privacy as such is not really research. It is about perception and should be solved by education.
Secondly, he mentioned reader zone confiment as a major research challenges. Today, it still is the number one reason for end-users users to give up with RFID when they do not manage to only read tags in the intended zone instead of random others due to artifacts stemming from the environment, e.g. reflection.

There was an interesting presentation about the development of a 2.4Ghz tag readable by a mobile phone's antenna [3]. This work really broadened the perspective of reading UHF and HF tags as we have discussed earlier [4] and even allows to remotely power sensors sitting on the passive tag by the emitting mobile phone. This work is part of the EU project Mimosa.
Another remarkable paper was presented by Rahul Bahattacharyya from the Auto-ID Lab at MIT and was about how to use RFID tags to sense displacement [5]: the basic idea is to place a metal plate in front of an RFID tag attenuating its power backscatter. The paper proves the idea in an experimental set-up by showing the monotonical relationship between power backscatter and distance from tag to the metal plate.
Another paper, awarded with the best-paper awared and presented by Alansson Sample [6], took a similar direction of turning the properties of bad tag performance under certain conditions into a benefit for sensing a specific phenomenon. He made use of the fact that users touching a dipole antenna of an RFID tag add capacity which slows down the tag's discharge behaviour. Accordingly, this effect can implement a power-free touch-based interface. Suggested applications could be powerless switches, remote controls and alike. The very same effect of changed capacity could be also applied to sense the presence of liquids with RFID tags.

Finally, the conference organizers also managed to really apply RFID for some practical purposes. First of all each conference badge contains a tag which can be read by exhibitors to collect your adress. It's convenient for them but from an interaction perspective it feels kind of odd when you look into an RFID 'gun' and get 'shot'. Second, if visiting a course of three vendors' booths one could win a video camera. You collected one RFID tag at each vendor and only if you had all three tags together the reader at the last station let you enter into the lottery. Nothing ground breaking but deployed...

Ariving in Orlando I was surprised that public transport actually was not that bad. They have buses every half hour into different directions, even an express one (golynx.com no. 111) directly to Disney. 40km for 2$ - not a bad deal. The little handsight of course, it's a little hard to find out where the buses depart and they end at the 'central ticket center' where private Disney buses take you to the hotel. When asking at the hotel where I could catch the public bus back to the airport they were really puzzled: "Public bus!? - never heared of that, you' re the first person asking for, unfortunately we don't have an affiliation with them!". This again shows that for successful logistics transportation and information have to go together...

[1] Lehtonen, M., Michahelles, F., Fleisch, E.: How to Detect Cloned Tags in a Reliable Way from Incomplete RFID Traces. In the IEEE RFID 2009 Conference, Orlando, Florida, April 2009.
[2] Lehtonen, M., Ruhanen, A., Michahelles, F., Fleisch, E.: Serialized TID Numbers – A Headache or a Blessing for RFID Crackers? In the IEEE RFID 2009 Conference, Orlando, Florida, April 2009.
[3] Yann T├ętu, Iiro Jantunen, Bertrand Gomez, Stephanie Robinet:Mobile-phone-readable 2.45 GHz Passive Digital Sensor Tag. In the IEEE RFID 2009 Conference, Orlando, Florida, April 2009.
[4] T. Wiechert, F. Thiesse, F. Michahelles, P. Schmitt, E. Fleisch: Connecting Mobile Phones to the Internet of Things: A Discussion of Compatibility Issues between EPC and NFC, Americas Conference on Information Systems (AMCIS), Keystone, Colorado, USA, 2007, [PDF].
[5] Rahul Bhattacharyya, Christian Floerkemeier, Sanjay Sarma: Towards Tag Antenna Based Sensing – An RFID Displacement Sensor. In the IEEE RFID 2009 Conference, Orlando, Florida, April 2009.
[6] Alanson Sample, Daniel Yeager, Joshua Smith: A Capacitive Touch Interface for Passive RFID Tags. In the IEEE RFID 2009 Conference, Orlando, Florida, April 2009.

Wednesday, April 8, 2009

Mobile Monday at ETH about camera phone apps

Mobile monday [1] has happened for the first time at ETH. The running topic was camera phones used for richer real world experience.

Juha Laurila from Nokia Research Lausanne started off with Nokia point and find [2] representing a compeling infrastructure for authoring travel guides and navigation services based on recognition of places by the user's camera-phone. Another, even more compelling project for me, was Nokia Image Space [3] which aims at sharing photos taken from Nokia phones together with their spatial relationship. That means besides the photo's location also the angle of the photograph is incorporated through the phones accelerometer and magnetometer. Accordingly, a 3D-representation can be calculated by overlaying various pictures from different angles - there is a nice video here. For me this is another nice example of user-generated content: instead of google-cars crawling around the world and taking pictures, Nokia users could do that for free...

Daniel Wagner was showing a couple of augmented reality "applications" - overlaying the real world seen from a phone's camera with items, animals, markers etc. While the shown examples did not really contain useful applications they gave great perspectives about what could be done in terms of embedding information on items and objects in the real world in a more compelling way than just text. They showed some demos were they really could overlay pictures with 3D images shown on the phone without using markers (e.g. see here).

While augmented reality appears to be a fascinating research topic, I'm still wondering what it will look like when in fufutre people will not only have blinking bluetooth ears but also tap through the world using their mobile phone as a white cane for "added-value information"...

Herbert Bay from Kooaba [4], an ETH-based spin-off, presented their approach of having visual search implemented on the iphone: you take a photo, send an MMS to their server, and out of 1 million objects so far you can receive further information.

Roger Fischer from Kaywa [5] presented 2D barcode applications. He explained the differences between Japan, where NTT Docomo has years ago moved forward and forced the other Japanese operators to follow spreading 2D barcode applications, whereas Europe still fights with a plethora of various barcode standards.

Mindaugas Stonys from Convision talked about smart camera applications based on the bee-tag 2d barcode standard. A nice feature of beetag is that you can place a logo inside the code. He proudly mentioned that their reader software runs on 11 platforms.

Finally, there was a 30s pitch about a start-up [5] empowering consumers to distinguish between orginal and faked products by using their mobile phones - an idea we've published about as well [6].

The mobile monday organizers had been so nice to put together a video playlist of compelling videos on youtube.

[1] http://www.mobilemonday.ch/
[2] http://pointandfind.nokia.com
[3] http://research.nokia.com/imagespace
[4] http://www.kooaba.com/
[5] http://www.originize.com/
[6] F. von Reischach, F. Michahelles, E. Fleisch: Anti-Counterfeiting 2.0 - A Consumer-Driven Approach towards Product Authentication , Late Breaking Results at the 9th International Conference on Ubiquitous Computing (UbiComp 2007), Austria, September 2007, [PDF].

Dissertation about NFC payment

Tom Wiechert finally had his disputation. He was analyzing the economics of contactless payments based on NFC. There, he compared the transaction costs and fees of various other payment systems (credit card, debit card, cash...) against NFC payment. Tom came up with a mathematical
model calculating different scenarios for retailers introducing NFC payment. His thesis [1] will be published soon on auto-id labs.org.
Throughout his thesis work he wrote an article [2] on the differences and communalities between EPC and NFC which got cited a quite bit, recently.

[1] (link to follow, search here)
[2] T. Wiechert, F. Thiesse, F. Michahelles, P. Schmitt, E. Fleisch: Connecting Mobile Phones to the Internet of Things: A Discussion of Compatibility Issues between EPC and NFC, Americas Conference on Information Systems (AMCIS), Keystone, Colorado, USA, 2007, [PDF].

Wednesday, April 1, 2009

Google Guntitude

When I recently was in the US I was really glad to have the new plugin Google Guntitude in my phone: I suddenly could see how many pistols, guns and other weapons had surrounded me. On the one hand that was frightning, on the other hand it made me feeling saver from unwanted surprise. How it works is that the mobile phone communicates with the (smart) guns [1] and displays the information on Google Maps.
I'm wondering about the next step: Since it's mostly deterrence anyway, it is enough to just buy a 'virtual mock-up' of a gun - it keeps criminals away from me since they already see my tremendous laser gun/tank/war plane equipment on their mobile phone. Accordingly, I myself neither have to carry this heavy devices with me nor do I have spend money on the real physical artifacts. This means higher deterrence with less guns, and as such a saver place...so Google Guntitude is the solution!

[1] Cliff Gromer: 'Smart' Guns: Dumb Idea, Popular Mechanics, July 2003 issue