TuxPhone is a project to develop open source (hardware and software) GSM/GPRS cellphone. Our objective is to create an open (in every sense of the word) cellphone platform that is convenient for creating novel applications. For instance, someone could take this reference design and integrate it with a small RFID reader to create a RFID enabled cellphone. Or, for that matter someone come up with a software that finds the cheapest way to make a phone call based on the available connectivity - VOIP or GSM.
As one could see from the above picture, it is neither compact nor slick. But, it is open and free. You can make it uglier, smaller, nicer, smarter or whatever you like it to be. We hope we will be able to create a useful open cellphone platform.
During the early part of 2005, at TagSense Inc., we (Deva Seetharam and company) were trying to design RFID readers for various standard cellphones. TagSense is in the RFID business and had no problems building the reader. However, interfacing with existing mobile phones was a major issue. Not because the technology is difficult, but because handset manufacturers won't expose the interface and API details without exorbitant licensing fees."I felt oppressed," quotes Deva. "Since then I have been thinking about building my own cell phone so I can use its hardware and software in whatever way I/we want," he continues, "can someone say freedom?"While Deva was "talking the talk", Surj Patel, the idea man, "walked the walk" and conceived the do-it-yourself Linux cell phone project. Surj and Deva started working on the project together in November 2005. Deva built the first prototype of Tuxphone in January 2006 with multiple boards and presented the design at the O'Reilly Emerging Telephony Conference. While Deva and Surj were putting TuxPhone together, Matt Hamrick started the Homebrew Mobile Phone Club to provide a forum for hackers interested in building cell phones. Matt liked the TuxPhone idea and started working with Surj and Deva.Colin Cross,a great hacker, who had worked with Deva at TagSense, joined the project and created a breakout board for the TuxPhone and presented the new design at Oreilly Foo Camp in August 2006.Leo Bonanni, an excellent product designer, who had worked with Deva at France Telecom, has recently joined the project to create a nice case for the TuxPhone.
The design overview is available in Etel Presentation. The schematics, bom and layout files are available on SourceForge. Consider schematics to be right, if there is a discrepancy between the presentation and schematics. For instance, in the presentation we show a key encoder, but, in the schematic you will see a PIC microcontroller for encoding the matrix keypad into a serial output.
You can download the files from the SourceForge TuxPhone project page.
- 1 Where do I find information about the TuxPhone project?
- The latest information about the project can be found at the OpenCellPhone site. The latest hardware and software files can be downloaded from the SourceForge TuxPhone project page.
- 2 What is the current status?
we have built a two-board (breakout and gumstix) version of the phone. we are able to make and receive calls on this. Colin cross demonstrated this phone at O'Reilly Foocamp 2006.
- 3 Why would I want to build my own mobile phone when I can buy a subsidized phone from my network operator?
- Good question. Some people think we're building our own phones because we think existing SmartPhones are too expensive. That's not really the main reason for wanting to build one's own phone. SmartPhones are a little pricey, but you're likely to spend at least as much on a DIY mobile as you would on a new mid-range SmartPhone. Building your own phone is about control of the platform, not cost. Think about the "homebrew beer" community. Some people like mass-produced lagers. But other people don't. For the longest time in the United States, if you didn't like commercially brewed Pilsners, your only choice was to "brew your own." Over time a vibrant "homebrew" community emerged that supported people's desire to have a nice Ale once in a while. Eventually the beer distributors and major breweries began to realize there was a market for things other than Pilz. In the same way, we are building our own phones because we want to support user experiences different than those offered by the traditional carriers. Having an open platform allows us to control what features are included in the device. We hope that over time our efforts will "inform" product decisions at commercial carriers.
- 4 Why would I want to build my own mobile phone when it's clear that VoIP over WiFi (or WiMax) is the future of personal telecommunications?
- Another good one. VoIP, WiFi and WiMax are all very cool technologies that will have a disruptive effect on the telecom market. However... depending on network conditions, VoIP can have some problems with latency and dropped packets. It's a testament to the skill of the guys running the core 'net systems that we don't have more problems. WiMax and other long range wireless protocols have similar latency problems. Modern mobile phone network technologies like CDMA and GSM do not have such severe problems with latency (though I'm sure we've all experienced bad signal and dropped calls on mobile phones.) Not all applications care about latency, but if you've ever had the "joy" of trying to carry on a regular conversation over a high latency voice line, then you know that it's not an especially happy experience. Plus... there don't seem to be any widely deployed solutions for mobile IP or for "handing off" a VoIP call from one wireless access point to another. (Not that it's impossible, mind you, it's just that there don't seem to be any standards that all the AP vendors support.) Mobile phone networks on the other hand, routinely hand off calls from one cell to another when subscribers are on the move. There's nothing wrong with VoIP over WiFi (or WiMax.) But there's probably several years of experimentation before WiFi VoIP phones can handle seamless call hand-offs between access points. It's also not entirely clear that latency on WiMax networks will ever be low enough for all voice applications. The TuxPhone project is focused on the "here and now." We're focusing on the mobile phone as a personal communicator. When we get that functionality down, maybe we'll worry about VoIP over WiFi (or WiMax, or Bluetooth, or ZigBee, or ...)
- 5 Aren't you guys just trying to hack the carrier's networks?
- Nope. We're using pre-approved hardware modules that are designed for network interoperability. We're not writing software to attack commercial networks. That's bad, very bad. We're trying to build better software for mobile phones. Period.
- 6 Okay, how do I get started?
- Well... first you probably want to peruse the Open Cell Phone site. We have details of projects and people that are working on open mobile phones. Once you've done this, it's time to get out the soldering iron, order parts from suppliers like GumStix and SparkFun, and start putting them together. The TuxPhone project will have a software distribution you can load onto your device. See also the getting started page.
- 7 Are you creating a single board solution?
- Nope. we are just combining the GM-862 GPRS module and gumstix connex 400. See also our future projects.
- 8 Wouldn't that be bulky?
- yes, definitely. if someone funds the development, we can create a slick and small TuxPhone.
- 9 Wouldn't you still need the providers' networks?
- Yes, absolutely. perhaps, in the not so distant future, probably somone will come up with a pure p2p cellphone network that doesn't depend on any central authority. (i know several complex technical and policy problems must be solved before we can build decentralized cellular networks.) The good news is that there are people working on this.
- 10 How can you solve the original problem of creating rfid readers for standard cellphones using TuxPhone?
- We can't. however, we believe this phone is at least an alternative to closed phones.
- 11 What next?
- add the audio components such as biasing and amplification circuitry to the board. currently, we depend on the headset.
- make the phone more compact. this requires using smaller keypad,gsm module and antenna, and optimizing board layout.
- develop a well-defined application platform
- develop a few useful applications
- Imagineer - Surj Patel - surj AT media DOT mit DOT edu. he envisioned the opensource cellphone.
- System architect - Deva Seetharam - deva AT alum DOT mit DOT edu. he created the system design and created an immobile TuxPhone with multiple boards literally stitched together.
- Hardware engineer - Colin Cross - colin AT colincross DOT com. he created the first mobile cellphone with a 2-board design.
- Project manager - Matthew Hamrick - mhamrick AT cryptonomicon DOT net. he keeps us together and motivates us.
- Designer - Leonardo Bonanni - amerigo AT media DOT mit DOT edu. Leo recently joined the project. all the ugliness in the cellphone is our fault.
we got some press coverage:
- New Scientist. Republished by TMCNet as Welcome to the open source cellphone.
- CNET: Build it Yourself Cell Phones.
- Wired: DIY Cell Phone.
for the last several weeks, this wiki was pointing to porn sites as it was vandalized. our sincere apologies! (btw, we don't check the site regularly. if you see any issues, pls. feel free to email us.)