Hobbyist Open Source Satellite Receiver - Raspberry Pi 2???

[Titanium]

tcrn.ch/1EYSDw1

Anyone interested in discussing and working together on a hobbyist, collaborative designed, open source satellite receiver based on a Raspberry Pi 2? The new Pi 2 base cost is $35 and more powerful than most STBs on the market. Might be a great platform to develop a new breed of receiver based on the hobbyists likes and wants rather than a receiver adopted from an STB designed for service providers or single satellite viewing.

Imagine a platform that grows and components transform with technology changes rather than be a fixed format device that is discarded with each cycle.

 

[wallyhts]

I'm in. I was watching tv on the old Pi with a tevii usb when playing around with outernet.

Sold Out every where. Have one on back order at mcm.

So I would think finding a Good remote is a must?

 

[KE4EST]

Yeah they released the Pi 2 yesterday and sold out every where in a few minutes. I tried early but was not up early enough. I have one on back order too.
I'm all for designing a receiver around one. The specs are awesome for something like this. Faster than some of cheapo laptops you can get a Walmart.

 

[Babadem]

Great idea. Worth looking into.

 

[Titanium]

I will start the conversation with questions and ideas.

Should the unit be a STB, a server or incorporate both?

For those in the know... is there any way to use PCIe cards (tuners such as the TBS cards) with the Raspberry Pi 2? I know that folks are already using USB tuners with some success, but the idea of being able to use the latest tuners and chipsets would open many additional options. Yes, TBS will likely introduce USB versions eventually, but ....

Should we investigate building our own daughter boards with Tuner chipsets or keep it more open source by using off-the-shelf products?

If it is to be a STB, what would be the most flexible, yet powerful video output option?

If it is to be a STB, what would be the most flexible and comprehensive audio output option?

Maybe use a touchscreen or smartphone to interface instead of a dedicated hardware remote control unit?

Tuning options including calibrated spectrum display (like on SDR Touch) to see a satellite frequency range and choose carriers to tune and view?

Throw in your ideas and let's focus on what we would like to see and think outside the box....

 

[KE4EST]

From the get go let's start with a STB/Server solution. This way when later it is decided to try and add one or the other,it won't be hard trying to make code work with what is already there. From scratch think like this while writing the code and putting together the hardware.

Don't know about PCIe

I say keep the it open source for the Tuner, it is the nature of the Pi community.

Touch Screen and Smartphone both should be very easy to implement.

Good chance to build your dream FTA box!!!

 

[Santa Clause]

I like the STB idea and the spectrum display would be a great feature. This may be thinking way outside the box but my wish for the ultimate satellite receiver is to have it auto detect and scan all available satellites with just a simple request. Perhaps something like that would be too difficult to implement but I can dream.

 

[wallyhts]

Smartphone connects via webserver running on Pi to remove having to build apps and dealing with App Store process.

I think pciE is way to fast for GPIO. Perhaps a usb bridge may be a solution.

So if folks want a GUI perhaps just use the livetv in XMBC. I think it would be cool to be able to ssh into the Pi and change to different PIDs etc.

 

[KE4EST]

I think you are right on the USB bridge solution. I have been doing some research and PCIe is for sure way to fast and wide for GPIO. However, PCI may be possible will have to research further. Maybe someone with more knowledge of PCI buses can come in and comment.

 

[wallyhts]

I think you are right on the USB bridge solution. I have been doing some research and PCIe is for sure way to fast and wide for GPIO. However, PCI may be possible will have to research further. Maybe someone with more knowledge of PCI buses can come in and comment.

Does TBS make a model that Brian just bought in PCI?

 

[. Raine]

Sounds good, will think on this. I have two B+ Pi's and one B right now. I missed the boat on the Pi-2 also, when they become available I'm definitely getting one or two. On the remote, I'd definitely want a standard remote, maybe implement both smartphone and IR remote control.

Power might be a consideration in using a PCIe card, it would need a power source separate from the Pi, a 12 and 5VDC source ? Unless the new Pi-2 has better power and distribution than it's predecessors. Of course, I guess just build a supply with the Pi and card and all in a case, so that's not an issue, I guess.

 

[Martyn]

But will it do 4:2:2

Seriously though, this is a good idea. I've been looking at some of the IPTV forums too and thinking about getting into that, so any cross over would be good. In fact, one of the most attractive parts of a PC tuner is the added visibility into the technical parameters of the streams, which seems like something that would be fun to mess around with.

Anyway, I look forward to the discussion. I'm not much of a coder so probably can't help in a direct way, but I stand open to testing and throwing in helpful and not-so-helpful suggestions.

 

[KE4EST]

Does TBS make a model that Brian just bought in PCI?

No not that one.

 

[bpalone]

Don't know if my coding skills are up to the task. But, as time permits, I would be willing to help out provided my skills are up to snuff. Any thought of what language to use? I would vote for "C" and assembly, the latter used to handle the things that tend to bog down processors, as it doesn't get any faster than well written assembly. I have never been a fan of C++, as it tends to drag along to much baggage.

Ke4est, is probably correct in that both STB & Server from the get go. Much easier than trying to shoehorn in code later. I agree that a wish list is needed, but the first order of business is to get a bare bones system working, then start on the wish list.

This would be a great opportunity to create a LEAN AND MEAN receiver, with the ability to add bells and whistles as wanted. Maybe, design in such a manner, that it is modular. That way if one wants a feature they include it, if not, they don't include it.

 

[Cham]

I think it is a great idea... I actually don't have one of the rasberry pi units so far but it wouldn't take long to find (make up) a use for one.

I'm more inclined to go with the full motherboard/quad processor thing though mostly so I can use newer pcie cards etc.

Will be watching!

 

[Bongu]

For adding PCIe, This looks interesting. ...
http://www.xmos.com/startkit

 

[wallyhts]

Thought of something that would be cool there are a few motor control add-on boards for the PI could add one to control your Actuator and servo motor

 

[bpalone]

Imagine a two block, block diagram.

Block one (1) being the RF portion.

Block two (2) being the software portion.

Block one (1):

Just how geeky or nerdy do we wish to make the project as whole?

If, pretty geeky, then we need someone with the ability to design and layout PCBs, so we could have PCBs produced. One catch is that I would say that no surface mount technology be used on the PCB, as most of us are not able to handle that task. I would guess that most of us are not afraid of a soldering iron, but SMT is not for the faint of heart or as most of us are getting older, the eye sight isn't what it used to be either.

The plus side of going geeky, is having total control over what is done in the RF portion.

The negatives are several. Did we just exclude a bunch of people because they wouldn't think of soldering anything together? Would the group always have someone that could layout a new PCB when technology moves forward? Is it even possible to get the required chips in non SMT format? Then, we also open up a whole new can of worms as some people may have problems assembling such a board, because at the frequencies we are talking about, if you clip your leads to long you have just created an antenna and probably some stray RF issues.

Personally, I think it would be cool and fun to build the RF portion, but from a public project standpoint, I would have to say probably not.

So, that would leave us with the prospect of using off the shelf stuff for the RF portion. I would guess that USB would probably be the best and easiest way to communicate between the Pi and the RF portion. Is there a STANDARD for the USB tuners? I looked at it a bit here a while back and it seems that TBS did have an API, but not all units supported all of the functions. Now, could we come up with an USB to PCI(e) interface, that would allow us to use cards rather than USB receivers?

Block two (2):

I am assuming that we would use Rasbian (sp?) version of Linux for the OS. Or, maybe even the XBMC version, so as to not be totally reinventing the wheel.

The software portion is going to have to communicate with the RF portion in order to control it and take the output and process it as needed. Those needs are going to be defined by the wish list, but the primary is going to be to deliver the digital signal to the decoder (which could be XBMC (whatever its new name is)).

I think we should keep the Unix philosophy of do one thing and do it well, deeply in mind in the design phase.

UpDateLee has done a lot of work and we could probably either get him involved or utilize some of his code for a great starting point. His stuff now requires that one recompile the Kernel, which isn't that hard, but turns a lot of people off. I think we should do our best to live with a stock kernel, so as to keep more people interested in the project.


All the above is intended to help move the project forward, and are entirely my thoughts. We can use it as starting point to get things better defined, throw it out or whatever the group decides.

I think we should keep it simple as possible, so as to not exclude users that might not want to get their hands dirty.

 

[Titanium]

Here is an email that I received from a hobbyist who reads, but doesn't post on SatelliteGuys.

- - - - - -

Feel free to share the below with the FTA gang....

here's some of my initial thoughts about the RaspPi2 STB project. Please understand I'm at the novice level on FTA systems, however in last year have put together 3 or so different Ku band setups in different parts of the great white North (Maine) using the MicroHD, Amiko Nano and Amiko HD SE STBs with 3 different antenna setups. Also am amateur Exrtra (ham 50 years) and have just retired after 46 years in military electronics related work covering DC to light (lots of RF/antenna stuff) for background. So here's some preliminary system ideas (not in any order):

Basic functions assumed to be similar to Amiko or MicroHD for a starting baseline (keep ability to have old fashioned composite video (480) as that is type of monitor I take up the ladder or into the woods...potentially to use for simple RF UHF/VHF modulator for distro also..

a) Signal strength Meter function:
To facilitate system setup/dish alignment for the FTA hobbyist that doesn't need/want a $300 meter that won't be used very often. On screen display and meaningful audio feedback (varying tones of micro are ok, beeps of Amiko not so good) for signal quality tweaking while outside balanced from ladder with hands full of tools (and swatting black flies)..... The microHD comes a lot closer than Amiko's signal quality display for this. It would be great to also have a SNR /db level (CNR or bit error rate) type of readout also, to permit that ever on-going need to tweek for that ever so important last 0.1 db.... . Speed of alignment updates are too slow on both Micro & Amiko, but scrolling bar display of Micro HD is much better than the Amiko bar graph method.

b) Spectrum Analyzer function:
Some type of simplified IF type of swept display across the band (SDR like?) . Should update fast enough to provide possible use with the alignment function, might help out with adjacent satellite (or terrestrial) interference for the mini budders) Potentially could be combined with the signal meter function above...(but it would be a great toy by itself) to provide a sense of what's active on any satellite, across the band, LNB functioning etc....

c)LAN/Wifi Interfaces: Since I not a digital guy; simple way to get in/out of our LAN/Wifi that doesn't require a software geek to configure software loads and patches every other day (avoid the always challenging "is it a hardware or software problem".... failure mode?) . Would be great to have it provide simplified compatibility (read as plug-play it works!) with a ROKU-HDMI at each of possibly 3 or 4 remote TVs.

d)Remote control: Need to have 3 possible system interfaces to command all functions; 1) Conventional IR 2) RF (with decent range) 3) LAN/Wifi

e) 4:2:2 If feasible, nice to have but lower priority.

f) Assume that you(Ti) or a sub would offer a package or kit of built up modules that could have one overall power supply/enclosure with basic system components Pi 2 CPU (pre-configured with basic firmware functionality/Receiver/IR control. The enclosure would have provisions for adding different modules for the selection of optional functions desired, ie Wifi module with external ant, and potentially add I/O ports if needed. External 12VDC battery operation for use at the dish very important. Also great feature would be either a built in LCD (480 or more) display or small separate display with sufficient resolution and brightness to function outside. Ever try reading the smaller font Amikos on a 12V black and white composite monitor? (the 8's and zeros all blend together on the fuzzy 5" dim CRT, colors are useless).

Great Project!

 

[Scott Greczkowski]

I would assume that you would need to use a USB tuner, and if we wanted this project to succeed we would need to pick one that is the best so that everyone is on the same page hardware wise for a receiver.

Then next step is software, what can be done on the free that does not require licensing such as Dolby Digital audio and other common satellite receiver standards.