I need some help. Essentially this is a WIFI question over 160km ranges (ham groups wont help)

Thread Starter

born2dive00

Joined Oct 24, 2016
285
In brief I am working on a non profit project to prevent poaching. I am an old time analog groupie so some of this new tech I don't understand. I have built an air frame to observe and prevent poaching, I have tested the system with analog systems on 433mhz for the controls of the air frame and 1.2ghz for video transmitter. Currently we have got out to 160km with RC control and also with standard def. video using basic equipment.

The governing body of this project, has delivered a unanticipated new requirements that I believe can only be done with a Digital LAN setup.

1st. They now want High Def video which based on my research can only be done on 2.4ghz and it must be available at a range of 160km.
2nd. They want bi-directional communications between the base station and the observation platform, so they can update or change waypoints on the fly.
3rd. They want ( and I don't know if this is even possible to do) (there IS NO CELLPHONE COVERAGE) to take a cell phone to make a call to the park rangers on the ground using the airframe as the relay. Better create a cell service at the base station, dial a phone number, the base station sends the signal to the airplane which relays the call to a cell phone on the ground in the hands of the park rangers while the airframe is circling over head.

4th the signals must be encrypted to prevent the poachers from tapping in and either seeing what the park rangers at the base station are seeing, or to prevent them from taking control of the aircraft.

5th because this is flying a airframe from ground the lowest latency possible must be kept

Now I have been looking at using this 30dbi Parabolic antenna http://www.l-com.com/wireless-antenna-24-ghz-30-dbi-steel-grid-antenna-n-female-connector combined with this 25W Amplifier, (this is where the Ham heads were having conniptions) for this project we have received government exemption to exceed 1w also we are using the 2.4ghz unlicensed. www.l-com.com/multimedia/datasheets/DS_HA241025GAI-NF.PDF

What I do not know (as I said I am an analogue old timer) is how to get the signals in and out using I guess a router to create a LAN Network.

Having done some digging in the commercial airframes they say the data transfer rate is 12mb/s @100km, it is encrypted at 256bit, and is 802. and uses Ethernet. This to me sounds like they are using a router tied into a parabolic antenna with an amplifier. On the air frame they are using whip omnidirectional antennas. To increase my range as I did with the analogue system, I was going to use directional antennas on the airframe. and for the cell phone transfer to use a Omni directional antenna

So I have 3 questions 1 how do I get the router on the air frame to go from digital to analogue to control the servos for the control surfaces, 2 How do I connect the router to send the data commands to the various systems, i.e. sending the signal to the autopilot to change way points, 3rd how do I set up a cell signal to use the cellphones to call the park rangers??

As I have said I am an old time analogue and I even with reading some things I still don't understand what I need and how to get it to work. Please let me know as the project's future is hanging in the balance.

Any information that you can share would be great, a list of parts and or a drawing of how things are connected would be awesome.

Thank you all.
 

olphart

Joined Sep 22, 2012
78
Seems like you've got the comm path part, but I don't see any processing power in the airframe.
You may want some for encrypting, but will need it for command decoding / interface to the various systems.
For ranger messaging, I'd look into audio to data packetizers and internet intercoms.
Tough project, Good Hunting <<<)))
 
Last edited:

Papabravo

Joined Feb 24, 2006
14,702
I'm guessing that there is no off the shelf solution to your problem. The design of such a device represents an enormous challenge in design, fabrication and testing. If I had a budget of 2-4 million USD and 18-36 months -- maybe. One of the problems is finding the spectrum to use at appropriate power levels. Where in the world this will operate is another challenge since many governments have pathological ideas about the use of radio for any purpose.

My first thought on the relevant spectrum would be the whitespace abandoned by terrestrial VHF broadcasting. A 50 kW transmitter for channel 2 or 4 could probably cover the area to the airframe at some reasonable altitude. Not sure you'd want the expense of operating such ancient equipment. Useful radius from transmitter to receiver on the ground was about 50 miles. Aircraft radios in a light aircraft can reach 75-80 miles on 7-12 watts. So it seems there is a long shot on meeting your requirements even there. The useful bandwidth of an AM signal is about 20 Khz so you won't be pumping tons of data through that pipe, but modern modulation techniques may get you where you need to be. Two 6 MHz. wide VHF channels will get you the 12 Mb/s data transmission bandwidth you need.

When you say (ham groups won't help) is that a capability problem or an attitude problem on their part?
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Faa exemptions for initial testing, + these will be used in the game reserves and national parks in several African countries having issues with poaching
 

BR-549

Joined Sep 22, 2013
4,938
I believe you are really mixed up. What is the "air frame"? A satellite? You are trying to relate....un-relatable situations. Point to point LOS is NOT an air frame. It took a 14,000 ft mountain. And a constant bearing.

Does your air frame has two steerable parabolics on it? What kind of power is available on air frame?

Forget the cell phones.

This whole project is a star trek dream.

You can only do what you want at short range.
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Well Considering that the UAV Factory routinely does 120km flight flying at 6000' with a 2w transmitter, a 28db antenna on the ground, with antenna tracking of the air frame, and the plane uses only a 5w Omni directional antenna to achieve 120km on 2.4ghz with a 12mbs bi directional transfer rate

I believe that this would be do able, as we already have the antenna tracker system, the dish and transmitters are not hard to come by. and antennas can be purchased anywhere.

When I am referring to when I say setting up a wifi system, is this, How do you connect the system to the CAN network? The airframe is using IP, on a Controller Area Network. getting the signal to and from the ground station to the airframe and back for me Is the easy part.

I do not understand the digital aspect of things, like how to get a controller imput on the ground station to convert this to a WIFI signal, and how to build a CAN system.

There is available power generator of 100w@12v on the airframe
Below is the antennas used by UAV factory
DDeoqRLWsAAjJKE.jpg AO4055-TNC-.jpg this combined with a 5w transmitter has achieved 120km at 12mbs.

I believe you are really mixed up. What is the "air frame"? A satellite? You are trying to relate....un-relatable situations. Point to point LOS is NOT an air frame. It took a 14,000 ft mountain. And a constant bearing.

Does your air frame has two steerable parabolics on it? What kind of power is available on air frame?

Forget the cell phones.

This whole project is a star trek dream.

You can only do what you want at short range.
 
Last edited:

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Hello Papa bravo
The ham groups have confirmed that this is definitely do able, as many other aerial technology companies have done and routinely continue to do this such as the UAV Factory and awhile back has done this with a 28dbi parabolic antenna with tracking aircraft tracking, on 2.4ghz with 12mbs at routine ranges of 120km or 70 miles with using only a 5w Ethernet transceiver.

Where I am getting the reluctance is when they hear that the system is portable namely when they hear that it is being put on an airframe they drop the subject like a hot potato. The ones who are "willing to help" tell me to first get my HAM license, as my work has taken me over seas I am not currently living in the US at the moment and can not do the ham license. and a ham license while helpful for analogue (which I already know), it would not help me set up the LAN/ CAN as this is computer based communications, for the network. this is where I need the help.

I am an old analogue guy, a transmitter at the ground station on 433 MHz for sending instructions to the pilot, and to control the camera gimbals (so the pilot can worry about flying) , and a transmitter on the airframe to send back the video to the receiver back the computer and monitors. I have done this in analogue out to 160km already without much issue.

BUT the parks and reserves are now wanting digital High definition to be able to identify identifiable characteristics and have more detail for the courts, as they want to be certain that the person in the video is the person facing charges.

So as I said before they now want to go to High Def Digital, now based on my research 2.4ghz would allow this. as far as the cameras are concerned, Nikon has donated 2 very powerful cameras and lenses for this project which can see capture a face of a person at 6000' with extreme clear detail so taking photos from a high elevation with the lenses is not an issue.

Any help you can offer PapaBravo please let me know,


I'm guessing that there is no off the shelf solution to your problem. The design of such a device represents an enormous challenge in design, fabrication and testing. If I had a budget of 2-4 million USD and 18-36 months -- maybe. One of the problems is finding the spectrum to use at appropriate power levels. Where in the world this will operate is another challenge since many governments have pathological ideas about the use of radio for any purpose.

My first thought on the relevant spectrum would be the whitespace abandoned by terrestrial VHF broadcasting. A 50 kW transmitter for channel 2 or 4 could probably cover the area to the airframe at some reasonable altitude. Not sure you'd want the expense of operating such ancient equipment. Useful radius from transmitter to receiver on the ground was about 50 miles. Aircraft radios in a light aircraft can reach 75-80 miles on 7-12 watts. So it seems there is a long shot on meeting your requirements even there. The useful bandwidth of an AM signal is about 20 Khz so you won't be pumping tons of data through that pipe, but modern modulation techniques may get you where you need to be. Two 6 MHz. wide VHF channels will get you the 12 Mb/s data transmission bandwidth you need.

When you say (ham groups won't help) is that a capability problem or an attitude problem on their part?
 

Papabravo

Joined Feb 24, 2006
14,702
There several ways to approach digital control depending on which parts of the implementation you have control over. It all comes down to encapsulation. If we have a payload of control information, imagine tacking on a couple of bytes of header information and a couple of bytes of trailer information and passing it up to the next higher (lower) layer (in the communications architecture). At the destination the process is reversed and that control payload is passed to the actuator. What the network does in between is generally of no particular concern as long as the throughput meets the reliability and accuracy constraints.
 

Papabravo

Joined Feb 24, 2006
14,702
CAN is a fairly straightforward protocol that works over a copper wire physical layer. It is a robust protocol in the sense that common errors are easily detected and retries normally get around the problem. The data payloads are 0-8 bytes long and you can implement a Master/Slave type system or a peer to peer system. Each message is encapsulated in a header of 11+ or 29+ bits that is used to do bus arbitration. This fixes one of the major problems with Ethernet systems of two transmitters talking at once. It is CS/MA-CA. That is Carrier Sense/Multiple Access with Collision Avoidance. This is actually a clever way to do things.
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Hello Papa bravo,
I understand the very fundamentals of digital (it is a signal, just like analogue is a signal) and I understand from working with my home laptop about mbs and latency, and some of the jargon. it is how to get the system to work together Is what I don't get. and to get the different systems to talk together. this is where I need help. On both the transmitting end and receiving end, I know how to set up a router for a computer, (connecting it) but how do I get the together parts of the network to talk to the router on CAN and how do I get the digital signal to move a mechanical item.


There several ways to approach digital control depending on which parts of the implementation you have control over. It all comes down to encapsulation. If we have a payload of control information, imagine tacking on a couple of bytes of header information and a couple of bytes of trailer information and passing it up to the next higher (lower) layer (in the communications architecture). At the destination the process is reversed and that control payload is passed to the actuator. What the network does in between is generally of no particular concern as long as the throughput meets the reliability and accuracy constraints.
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Ok I am starting to get it. please continue.

CAN is a fairly straightforward protocol that works over a copper wire physical layer. It is a robust protocol in the sense that common errors are easily detected and retries normally get around the problem. The data payloads are 0-8 bytes long and you can implement a Master/Slave type system or a peer to peer system. Each message is encapsulated in a header of 11+ or 29+ bits that is used to do bus arbitration. This fixes one of the major problems with Ethernet systems of two transmitters talking at once. It is CS/MA-CA. That is Carrier Sense/Multiple Access with Collision Avoidance. This is actually a clever way to do things.
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
The biggest challenge that I am facing is it needs to be really off the shelf items that are cheap and readily accessible so if there is a component failure in the field it can be fixed or swapped out in the field.

I have the base station setup with analogue now using rc joysticks to a brain which converts the analogue to a pwm signal, which is currently connected to a 433mhz transmitter, so how can I take the analogue output or the pwm output and connect this to a Ethernet 801... which sends an ip encrypted signal to a 2.4ghz transmitter??

On the airframe side what do I need to convert the ip 2.4ghz signal to a pwm signal that is usable by the Autopilot and servos?

CAN is a fairly straightforward protocol that works over a copper wire physical layer. It is a robust protocol in the sense that common errors are easily detected and retries normally get around the problem. The data payloads are 0-8 bytes long and you can implement a Master/Slave type system or a peer to peer system. Each message is encapsulated in a header of 11+ or 29+ bits that is used to do bus arbitration. This fixes one of the major problems with Ethernet systems of two transmitters talking at once. It is CS/MA-CA. That is Carrier Sense/Multiple Access with Collision Avoidance. This is actually a clever way to do things.
 

Papabravo

Joined Feb 24, 2006
14,702
So just spit-balling here; suppose we have a control payload of eight bytes which contain four numbers. The four numbers represent {Throttle, Elevator, Aileron, Rudder} settings. A value of 0 represents a neutral setting for the UAV. A positive number means RIGHT or UP. so if I send the following packet:

{+1, 0, +1, 0} it means increase the throttle position one tick above neutral, elevators neutral, ailerons starting a right turn, rudder neutral. You send a continuous stream of these packets to the controller and the controls move the way you want them to. No problem with sending redundant packets either. That's a starting point at least.
suppose the next packet looks like
{+3, +1, +3, +2} it means we are increasing the throttle position for more power to compensate for the increased drag on the high wing, pulling back on the stick for more lift, increasing the roll rate, applying right rudder to keep the turn coordinated. You get the idea.
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Hello Papabravo
I get this part, very well explained though, it would be the digital pack of a pwm code. what I am trying to do/envision is the following.
I have a joystick, connected to a computer, the joy stick sends a digital signal to the computer.
From the computer on the ground I am sending control signals, live autopilot commands and updates from the computer software (to what???) a wifi? router? cellular sim (a private cell network if you will) ... From XXXX the signal is sent on 2.4ghz to the XXXX which then is sent to a receiving computer then to the Autopilot and servos??? or can it the computer on board be skipped.

Like I said before the actual sending of the signal over the air is not the problem it is how it is connected to the antenna.

I was thinking of something like this. https://wiredcraft.com/blog/drone-copter-uav-4g-network but this guy is skipping the base station computer.
What I need it to do is the following.

independent remote control for take off and landings
Lap top to send out mission updates to the autopilot
ability to change from hand held controller to joystick controller to pilot the airframe like this
Now while all that is going on and being sent up to the UAV,
I need the autopilot to receive the updates and monitor the EFI system of the engine
Then send down stream from the UAV to the base station.
I can currently do this with simple analogue transmitters or PWM but without the Live AP updates, and the High def video.

I was thinking of simply taking a lap top gutted of all the bloat ware, connecting the joysticks then using a (router?? connected to a amplifier and the parabolic antenna) then another router connected to something like a single board computer like ODroid XU4 which can have it own IP address and would have more than sufficient power for high def video and monitor the EFI system. But I don't know 1 how to do this directly without a http page, nor how to connect a router on the software side to the high power antennas....




So just spit-balling here; suppose we have a control payload of eight bytes which contain four numbers. The four numbers represent {Throttle, Elevator, Aileron, Rudder} settings. A value of 0 represents a neutral setting for the UAV. A positive number means RIGHT or UP. so if I send the following packet:

{+1, 0, +1, 0} it means increase the throttle position one tick above neutral, elevators neutral, ailerons starting a right turn, rudder neutral. You send a continuous stream of these packets to the controller and the controls move the way you want them to. No problem with sending redundant packets either. That's a starting point at least.
suppose the next packet looks like
{+3, +1, +3, +2} it means we are increasing the throttle position for more power to compensate for the increased drag on the high wing, pulling back on the stick for more lift, increasing the roll rate, applying right rudder to keep the turn coordinated. You get the idea.
 

Thread Starter

born2dive00

Joined Oct 24, 2016
285
Papa Bravo would not +1, 0 +1, 0 mean to increase to x value of +1 to neutral 0 back to increased x value back to 0 neutral value?? Most of the setups need to have at least 1270 bit resolution for smooth operations. Let me know I might have missed something

So just spit-balling here; suppose we have a control payload of eight bytes which contain four numbers. The four numbers represent {Throttle, Elevator, Aileron, Rudder} settings. A value of 0 represents a neutral setting for the UAV. A positive number means RIGHT or UP. so if I send the following packet:

{+1, 0, +1, 0} it means increase the throttle position one tick above neutral, elevators neutral, ailerons starting a right turn, rudder neutral. You send a continuous stream of these packets to the controller and the controls move the way you want them to. No problem with sending redundant packets either. That's a starting point at least.
suppose the next packet looks like
{+3, +1, +3, +2} it means we are increasing the throttle position for more power to compensate for the increased drag on the high wing, pulling back on the stick for more lift, increasing the roll rate, applying right rudder to keep the turn coordinated. You get the idea.
 
Top