BerlinRoofNet: Difference between revisions

From
Jump to navigation Jump to search
No edit summary
 
mNo edit summary
Line 42: Line 42:
Technical Papers:
Technical Papers:


* [[Hacking the Linksys WRT54GS Wireless Router]] (Tutorial)
* [[Programming the Linksys WRT54GS Wireless Broadband Router]] (Tutorial)
* [[Image:Click_modular_router.pdf|Click Modular Router]] (pdf)
* [[Image:Click_modular_router.pdf|Click Modular Router]] (pdf)
* [[Image:WiMax-LOS.pdf|WiMax's technology for LOS and NLOS environments]] (pdf)
* [[Image:WiMax-LOS.pdf|WiMax's technology for LOS and NLOS environments]] (pdf)

Revision as of 12:20, 18 October 2004

Q: What is Berlin Roof Net?

A: 'Berlin Roof Net' is a project run by volunteer students of the Computer Science Department at Humboldt University Berlin. The goal is to construct a network of nodes (access points) which share Internet access over wireless radio connections. The access points are run independently by the students with their own equipment.

The project was inspired by the MIT RoofNet project in Boston, USA.


Q: Which wireless communication technologies are used?

A: One technology that is economically feasible for our purpose is IEEE 802.11b, also known as WiFi or Wireless LAN. However, the communication range of this technology is rather limited, especially in cities with no line of sight (NLOS) between the participating nodes. We would like to augment our system with 802.16a (WiMax) technology, which provides longer communication ranges and higher data rates, as soon as such equipment becomes available at a reasonable price.


Q: Is it safe for my health?

A: WaveLAN operates at 0.25 watts in a bursty mode (i.e. only when data is actually transmitted). Cellular phones go up to 3 watts and microwave ovens leak even more - and they are generally considered 'safe enough'.

All devices on the US market are approved by the FCC, which has a good public document on RF Safety.


Q: What is the vision?

A: This is a research project. Researchers often take a simple good idea and push it to its logical extreme. From this they learn a lot and finally understand what's possible and what is not. Then, taking one step back from the extreme approach, real good practical systems can be built. Here is our extreme story:

Traditional communication networks are owned by operators; i.e. companies who invest money into the deployment of cables and devices, and who operate and maintain those resources in order to offer services to users. To recover their investments (and to make some profit) - they charge their customers.

This may change: Falling hardware prices combined with ever increasing functionality may make it feasible to outsource the function of an operator (company) to the service users. The software that normally runs on operator-owned computers may be split into components that are executed at end-user owned equipment (such as computers, access points or mobile phones). Cables may be replaced by wireless links - which are even free if operated in an unlicensed frequency spectrum, such as 2.4GHz or 5GHz .

Easy? Not so fast! There is still a lot of plumbing and configuration necessary to build and run a reliable network - just that it has to happen automatically now. And, in the event that something goes wrong, the system has to be abele to automatically recover from ANY failure by itself - there is no operator to help. ... and this is exactly where our research comes in. We want to see how far the idea of 'outsourcing the network operator to the end-user' can go with today's technology. If we are successful, those community owned networks may become a vital component in making Internet communication accessible to more people, for little or no money.

Is this the death of telephone operator companies? By no way! Certain people and businesses will always require high quality communication services - and they are willing to pay for that extra quality. Those services will continue to be provided by commercial operators - at least for now :-)


Similar projects in other cities:


Technical Papers: