Hello Linuxers,
An internet connection for the common man, that provides a download speed of 1 GBytes per second is still a dream. Although satellite based net connections exist, they have limitted subscribers due to high costs and limitted bandwidth.
Nowadays the satellite broadcast frequencies are already in several Ghz, for example the Ku band that transmitts DTH signals. Suppose we have a satellite that transmitts at a frequency of say 10 Ghz. This translates to a digital bit rate of 10 Gbits per second. After a little frequency beating, we get an output signal of 8 Gbits per second or 1 GByte per second. This signal inputs into the computer using a gigabit LAN.
The problem with normal internet is that it is an instantaneous system where the data is downloaded as per the user requirement and different users have different data to be simultaneously downloaded, thereby reducing bandwidth per user. Suppose instead of this we have a system where the data is not downloaded on demand but is serially transmitted at a high speed like 1 Gbytes per second. This means that in 24 hours, a channel of 1 Gbytes/sec can transmitt 24 x 3600 Gbytes of data. This corresponds to 86400 Gbytes or 86.4 terabytes of data in 24 Hours. The user has no choice but to wait for 24 hours for the next broadcast in case he/she has just missed it. If the satellite has 24 transponders then this data is split into 24 slots of one hour each. So now the waiting peroid for the next download is only one hour. If the transponders are 48 then this further reduces repeat download time to half an hour only.
The final result is that the downloadable files are available to *everyone* within the satellite footprint at a speed of 1 GBytes/second subject to a maximum waiting period of 1 hour or 30 minutes. So everyone benefits equally and parallely. This will bnefit remote schools, villages and anyone in any remote part of the country.
All data will be linked to an index file that will be transmitted as a pilot file. This file will be a mini website cum search engine classified into topics and subtopics. These will link to the actual file and its corresponding transponder and its broadcast time. A timer in the receiver will mark this file and download it at its arrival time into a buffer harddisk or directly into the computer depending on the hardware design. This will be very usefull for downloading large files, cds, dvds and other bulky data. Entire newspaper websites can be downloaded for liesure reading. Imagine downloading 4 CDs of Linux within 3 seconds flat.
The cost for setting this system up will not be cheap but considering the India-China-Asia-Pacific region alone, the number of subscribers will be in millions or a billion. The receiver should not cost more than a DTH system that is now available for Rs. 3000/-. Lets put it at 5000/- rupees. This is just a one time investment. If every subscriber is charged max. 100 rupees or 2$ per month, then this translates to approx. a billion dollars every month from the entire region. This makes it the system for the common man. There will be additional income from the website owners just as they normally pay for web space and from advertisers. Ads will now be fullfledged video films not dancing gif images. This system will peacefully co-exist with the normal internet which will be still used for general purpose light surfing, email, uploads, chat, online payments and registration.
The entire software for this can be developed using Linux which is stable and open-source and the techies are available right here in India so no need to import them. Equipment can be procured from China or elsewhere.
Please send your suggestions and criticisms and if you like the idea, you can very well use it and help to provide a real high speed net for everyone. Many of you guys are in high positions in big organisations and you can set the ball rolling.
Thanks and Regards,
Rony.
On Wed, 8 Sep 2004 00:56:11 +0530, Rony Bill ronbilly@hotpop.com wrote:
Hello Linuxers,
The entire software for this can be developed using Linux which is stable and open-source and the techies are available right here in India so no need to import them. Equipment can be procured from China or elsewhere.
check out moviebeam. Its a service provided in the western us, where they provide a set-top box with a satellite reciever that is configured for serial continuous download of movies that the user pays for.
On Wed, 8 Sep 2004 00:56:11 +0530 "Rony Bill" ronbilly@hotpop.com wrote:
Hello Linuxers,
An internet connection for the common man, that provides a download speed of 1 GBytes per second is still a dream. Although satellite based net connections exist, they have limitted subscribers due to high costs and limitted bandwidth.
Nowadays the satellite broadcast frequencies are already in several Ghz, for example the Ku band that transmitts DTH signals. Suppose we have a satellite that transmitts at a frequency of say 10 Ghz. This translates to a digital bit rate of 10 Gbits per second. After a little frequency beating, we get an output signal of 8 Gbits per second or 1 GByte per second. This signal inputs into the computer using a gigabit LAN.
Hey Rony,
The transponder transmits at 10GHz, but does not have a 10GHz bandwidth. These are two different things - center frequency and frequency bandwidth. Wish we could have that kind of frequency allocation though!
Keep the ideas coming. You never know, we must just hit onto the next big thing !
Cheers,
Zainul.
Surprising how none of us noticed this oversight. :)
- Navneet
Hey Rony,
The transponder transmits at 10GHz, but does not have a 10GHz bandwidth. These are two different things - center frequency and frequency bandwidth. Wish we could have that kind of frequency allocation though!
Keep the ideas coming. You never know, we must just hit onto the next big thing !
Cheers,
Zainul.
Hi! Zainul,
I know. I was thinking in line with something like a carrier only, without a modulating signal riding on it, having data bits mounted on it in digital form in every oscillation. Would it still have bandwidth? Is it possible? Something that will create points in a single frequency sine wave to distinguish between low and high or 0 or 1.
Regards, Rony
----- Original Message ----- From: "Zainul M Charbiwala" zainul@ee.iitb.ac.in
Hey Rony, The transponder transmits at 10GHz, but does not have a 10GHz bandwidth. These are two different things - center frequency and frequency bandwidth. Wish we could have that kind of frequency allocation though! Zainul.
Hello All,
Modulation is used to transport a signal on a carrier. In the earlier days frequencies were low and for sat. transmission they had to be mounted on a carrier. But today the freq. that is used for communication is so high that it is a carrier in itself. Eg. a gigabit LAN passes data in gbytes per second. Someone told me, in China they use terabit LAN. Sat. freqs. are also in GHz. Nowadays digital formats in broadcasting have narrowed bandwidth utilisation yet provide high volume content. They are able to cut down on side bands.
Can we have a satt. carrier signal that is a fixed frequency but is made up of a constant data train? As of now, what is the highest download speed achieved by satt. internet? What is the highest speed currently available in the satt. 'pipes' that big corporates use for international inter-office links?
Anyway, this is the technical part which will be looked after by satt. designers and other techies.
What I want to know mainly is whether the concept of having such a ***high speed-high volume-repeatitive-serial file download facility*** is acceptable and worth implementing. Many have mistaken this for another television broadcast. It will be similar but not the same. TV is only live entertainment. This is a normal serious ultra high speed file download system for the common man at a very low monthly rental, for an *infinite number of users* within the footprint. That is the whole issue. How good and usefull is this idea?
If the Govt. opens up the airwaves to private net operators then the entire 58 (60 - existing 2) terrestrial TV channel spectrum could be utilised for digital serial file broadcasting in a similar way but at a lower download speed than a satt.. Major companies like Red Hat, Novel, Microsoft and many many others can provide their heavy software files on the airwaves to be downloaded in our homes and offices in a fast and economical way. No wiring, no cable.
Regards, Rony.
----- Original Message ----- From: "Srinivasan Krishnan" skrishnan@primus.ca
Wouldn't this be modulation?
On 08/09/04 00:56 +0530, Rony Bill wrote: <snip>
The problem with normal internet is that it is an instantaneous system where the data is downloaded as per the user requirement and different users have different data to be simultaneously downloaded, thereby reducing bandwidth per user. Suppose instead of this we have a system where the data is not downloaded on demand but is serially transmitted at a high speed like 1 Gbytes per second. This means that in 24 hours, a channel of 1 Gbytes/sec
So whose data takes priority? Remember that some of us really really don't care about downloading gigabytes of data, but we want low latency interactive sessions. Half an hours waiting? We start screaming with more than 1 sec of response time in the round trip.
can transmitt 24 x 3600 Gbytes of data. This corresponds to 86400 Gbytes or 86.4 terabytes of data in 24 Hours. The user has no choice but to wait for 24 hours for the next broadcast in case he/she has just missed it. If the satellite has 24 transponders then this data is split into 24 slots of one hour each. So now the waiting peroid for the next download is only one hour. If the transponders are 48 then this further reduces repeat download time to half an hour only.
The final result is that the downloadable files are available to *everyone* within the satellite footprint at a speed of 1 GBytes/second subject to a
This sounds so much like TV, rather than Internet. News flash: The Internet is not television.
maximum waiting period of 1 hour or 30 minutes. So everyone benefits equally and parallely. This will bnefit remote schools, villages and anyone in any remote part of the country.
What happens if none of the content being broadcast is what I want?
All data will be linked to an index file that will be transmitted as a pilot file. This file will be a mini website cum search engine classified into topics and subtopics. These will link to the actual file and its corresponding transponder and its broadcast time. A timer in the receiver will mark this file and download it at its arrival time into a buffer harddisk or directly into the computer depending on the hardware design. This will be very usefull for downloading large files, cds, dvds and other bulky data. Entire newspaper websites can be downloaded for liesure reading. Imagine downloading 4 CDs of Linux within 3 seconds flat.
I can. I still prefer fibre to the house.
<snip>
Please send your suggestions and criticisms and if you like the idea, you can very well use it and help to provide a real high speed net for everyone. Many of you guys are in high positions in big organisations and you can set the ball rolling.
Unhappily, its actually cheaper to roll out fibre and recover your costs over a slightly longer period of time. Plus, what happens to different protocols like the peer to peer applications?
Devdas Bhagat
Unhappily, its actually cheaper to roll out fibre and recover your costs over a slightly longer period of time. Plus, what happens to different protocols like the peer to peer applications?
Dyu have any statistics to backup your statement ? I would be really interested in knowing how fibre to the home is actually cheaper than copper in reasonable recovery times.
On 07/09/04 23:54 -0600, Adil Kodian wrote:
Unhappily, its actually cheaper to roll out fibre and recover your costs over a slightly longer period of time. Plus, what happens to different protocols like the peer to peer applications?
Dyu have any statistics to backup your statement ? I would be really interested in knowing how fibre to the home is actually cheaper than copper in reasonable recovery times.
Fibre scales a lot better than Copper. Media costs are not much different in fibre and copper, termination costs are. With different frequencies in the same media, fibre has a lot more carrying capacity.
With copper, your limitation is 100 m of FE. Fibre is a few km. The trick is to combine both media, with copper at the edge connecting to a fibre network.
User -- Copper --- Fibre===========
Of course, the initial costs of both fibre and copper laid out correctly are high, but once you have the network installed and your cost recovered, the rest of the time you are making pure profits (your running costs is extremely low, as compared to other media).
Devdas Bhagat
What applications are you targetting with this ?
Any application that needs data to be pushed can benefit from your system, but not an interactive browsing system like the internet.
So, if the idea is to push information like news, weather, defined programmes ( as on TV or radio ), this medium will benefit. For others I don't see how this benefits.
What do you think ?
- Navneet
Sometime Today, Rony Bill assembled some asciibets to say:
have a satellite that transmitts at a frequency of say 10 Ghz. This translates to a digital bit rate of 10 Gbits per second. After a little frequency beating, we get an output signal of 8 Gbits per
what about latency?
The problem with normal internet is that it is an instantaneous system where the data is downloaded as per the user requirement and different users have different data to be simultaneously downloaded, thereby
this is the primary use case.
where the data is not downloaded on demand but is serially transmitted at a high speed like 1 Gbytes per second. This means that in 24 hours, a channel of 1 Gbytes/sec can transmitt 24 x 3600 Gbytes of data. This corresponds to 86400 Gbytes or 86.4 terabytes of data in 24 Hours. The
OR, you could package a few terabytes of data onto magnetic tapes and fly them to your destination. You'd get a higher data throughput. Do not underestimate the bandwidth of a station wagon full of magnetic tapes speeding down the highway.
user has no choice but to wait for 24 hours for the next broadcast in case he/she has just missed it. If the satellite has 24 transponders
Sounds like TV no?
The cost for setting this system up will not be cheap but considering the
Doordarshan already has the necessary infrastructure set up. Talk to them to reuse. I know there were plans with the dept of agriculture, something along the lines of regional programming slots for local farmers (anyone remember Amchi Mati Amchi Mansa?)
--- Philip Tellis philip.tellis@gmx.net wrote:
Doordarshan already has the necessary infrastructure set up. Talk to them to reuse. I know there were plans with the dept of agriculture, something along the lines of regional programming slots for local farmers (anyone remember Amchi Mati Amchi Mansa?)
I DO!!! pretty neat and informative program - used to be primetime too at 4pm i think..
heh!
__________________________________ Do you Yahoo!? New and Improved Yahoo! Mail - 100MB free storage! http://promotions.yahoo.com/new_mail
kishor bhagwat wrote:
--- Philip Tellis philip.tellis@gmx.net wrote:
Doordarshan already has the necessary infrastructure set up. Talk to them to reuse. I know there were plans with the dept of agriculture, something along the lines of regional programming slots for local farmers (anyone remember Amchi Mati Amchi Mansa?)
I DO!!! pretty neat and informative program - used to be primetime too at 4pm i think..
It was actually prime time stuff. They used to show it at 7:30 or 8:00 pm in the evening. Since there was no other channel to watch........
Regards saswata
Its a beautiful idea but not a good replacement for current system.Is`nt it better if we start a nice "low cost downloading center" with this architecture?
That would be nice. I had written to Digit magazine to start a service to download software on their high speed network and pass it on into recordable CDs for a small fee. Along with that another suggestion was given, to make a facility for paying for paid software in INR to digit and on our behalf, it will pay in dollars to the foreign site as everyone does not keep a credit card. There was no response from them. Maybe they want us to buy mags. to get software.
If ILUG-BOM can start something like this it will benefit many people. A one stop centre for all transactions and download utilities and of course a meeting point too.
Regards, Rony.
----- Original Message ----- From: "niranjan sathe" satheniranjan@gmail.com
Its a beautiful idea but not a good replacement for current system.Is`nt it better if we start a nice "low cost downloading center" with this architecture?
Morning All,
Commercials are always a touchy subject for a NPO like ours. Yes i still see great logistical issues with respect to availability and affordabliity of Free Software in India. There just aren't many options available in our city where one can just walk in and buy a copy of Open Office or Fedora Core 2.
I would like to discuss these and other issues at the upcoming meet such that we may formalise some self sustaining structure towards providing media at various outlets across the city.
Investment to get this running isn't an issue cause i am sure there will be more than a handful ready to contribute for such a cause.
To request a commercial entity like Jasubhai to perform commercial activities that have margins thinner than paper is being too harsh on them.
Put forth your ideas/feasbility model if you would like to work towards getting an activity like this up and running rather than requesting someone else to take up the baton for our sake.
Have a great weekend all.
Trevor
===== ( >- -< ) /~\ ______________________________________ /~\ | ) / Scaling FLOSS in the Enterprise \ (/ | |_|_ \ trevorwarren@yahoo.com / _|_| ____________________________________/
__________________________________ Do you Yahoo!? Yahoo! Mail - 50x more storage than other providers! http://promotions.yahoo.com/new_mail
-
Adil Kodian -
Devdas Bhagat -
kishor bhagwat -
Navneet Karnani -
niranjan sathe -
Philip Tellis -
Rony Bill -
Saswata Banerjee & Associates -
Srinivasan Krishnan -
Trevor Warren -
Zainul M Charbiwala