You are here

Mobile, portable nodes to cover a large area event

16 posts / 0 new
Last post
kn5g
Mobile, portable nodes to cover a large area event
Our club has been asked to provide communications for a local "10K Run". We do not have the benefit of hills or high buildings to deploy nodes. We would like to provide live video feeds from some areas of the route back to the command post/finish line. We have been kicking around the possibility of using several nodes mounted on vehicles along the route. We will be dealing with some two and three story buildings, some hilly terrain and trees.
I'm wondering if anyone has tried using multiple, moving vehicles with attached nodes with any success? 
We have a selection of low, medium, and high powered nodes that could be mounted to various vehicles that could be staged along the route or moved along the route as needed to fill in the gaps in coverage.
 I suspect our mobile node's ranges will be limited but I do not have a feel for how limited or how well a moving node will work within the local mesh.
Any experience in trying a similar deployment would be helpful and appreciated!

Keith
KN5G
Henderson, TX.
nc8q
nc8q's picture
Mobile, portable nodes to cover a large area event

Suggestion:
Wait/look for a reply from groups that have experience in setting up portable stations for a public service event.
NWO: Chrissy?

For video, I suspect you will want 'high speed' data.
AREDN devices take 'minutes' to discover a valid 'maximum throughput' MCS.
I suspect that 'moving' nodes will not work well enough.
I suggest that you will need to do a 'site-survey'/'war drive' before the event to determine locations of where to
park and wait for the AREDN device to find valid MCS choices.

I hope this helps,
Chuck

 

kn5g
Mobile, portable nodes to cover a large area event
Thanks for the comments Chuck!
Good point I believe on the moving video source taking a time period to settle on an optimal data route. We may indeed need to choose the sites for video sources and stage those nodes before hand. Then we could perhaps use other mobile nodes on vehicles to move to more or less line of sight positions between the video sources and the command center. Have the vehicles stop or park while watching the signal strength between the nodes and quality of the video feeds. We will also have voice coms via 2 meters to facilitate the movement and positions of the mobile nodes. The RUN is not a fast paced event, in that we could deploy a camera to cover a length of street where the runners will be for several minutes while a second or third camera node is being positioned to cover the next leg of the route. In other words we would position the camera nodes in one spot for a period of time, then move our focus to the next camera feed along the route while the first camera is moved to the third position. Kind of a leap-frog approach, assuming we can deploy enough mobile nodes to create a path from the cameras back to the command post where the video will be viewed
I agree, a trial run is in order. I believe we will pick a weekend day and do a trial run to see if this idea is viable.

Keith
KN5G 
KK6BML
KK6BML's picture
Mobile, Portable nodes to cover large event area

Keith,
Our ACS (Laguna Niguel ACS) in Laguna Niguel, CA has been working for several years to develop protocols and procedures for this, both for disaster response and for our annual Holiday parade. We have done extensive development on mobile and portable mesh deployment. We have challenging line of site terrain and in a couple deployments did not have line of site between the field node and the command center and had to relay through a backbone node in our area.

First we learned a lot from the unit in a nearby city that sets on fixed location cameras along an event route (in a fairly open flat location) - this was primarily in the area of configuration of the camera nodes.

A tested relay network - even if temporary - of fixed nodes is the key to this. In our parade scenario we have fixed nodes at tested locations along the route that give a local backbone. We have other units that move along the route (although not in vehicles), one that fits in a backpack, including the node head that the operator just needs to stop walking, align the node and deploy the camera. we use a mixture of ip-addressable fixed and ptz cameras attached to the fixed stations and the ipcam app from cell phones for mobile units. all field units have both poe powered mesh nodes and a poe powered wifi access-point (either a Hap-lite or Ubiquiti AirGateway) powered by a 12 or 24v 15-30amphr battery-box (built in plastic ammo can enclosure). These can be used with a 12-2v vdc transformer or 24-12vdc inverter to have appropriate poe power for the different devices.

Over time we are building out a city wide local mesh network of fixed nodes plus mapping key high points with good line of site for "fixed" portable node locations for deployment relay sites. We have used iSpy (a windows based standalone app) and are now experimenting with zoneminder - a browser based app - for video monitoring from command center. Video capture is to server systems currently running on rPi's at our fixed bases but which can also be field deployed (still experimental and possibly not needed).

We would be happy to share further with you directly 

Bob Evans
KK6BML
Training Director Laguna Niguel ACS
kk6bml@outlook.com

kn5g
Mobile, Portable nodes to cover large event area
Thank you Bob for the comments! This gives me some encouragement that what we are purposing to try will indeed work to some extent. I do believe we may need to round up some more Nodes to deploy along the route since the roads will be closed to traffic during the race so we will not be able to slip in a vehicle with a node during the race itself, but we could have vehicles pre-deployed with attached nodes. In our case the use of vehicles helps with powering the nodes so that we do not have to move around a lot of 12 volt batteries or pairs of batteries but can relay on 12Volt to 120VAC inverters in the vehicles to power the nodes that like 24VDC power. I believe all the IP-Cams we have require 12V DC so that's an easy fix with some power splitters and Anderson Power Poles.
I am working towards getting a few of the local club members together this coming Sunday with our various Mesh nodes and doing some experimenting to see what we can cover with what we have on hand. The little "Travel Routers" that run off of 5VDC are quite interesting as they would be inexpensive and easy to power from a portable platform like a back pack, baby buggy, bicycle, ATV, etc. 
As we progress I hope that we will be able to permanently deploy a few nodes around the town that would help cover some of the more popular areas and be the beginning of a backbone for a local Mesh Network.

Thanks again!
Keith
KN5G
AJ6GZ
12v/24v
Most, if not all, of the MikroTik and Ubiquiti nodes we commonly use will run on 12V for up to about 25ft of Cat5/6 on a stiff power supply. Perhaps a little shorter on battery. If you have something that really needs 24V, a 12>24V DC-DC boost converter would be more effficent and less bulky than 12VDC > 120VAC > 24VDC.
k1ky
k1ky's picture
12=>24V
Mikrotik has a product for just that: https://mikrotik.com/product/mups  Just add 12v battery and you are good to go.
nc8q
nc8q's picture
Discontinued
Tom:
That web page shows 'Discontinued'.
:-|
Chuck
 
k1ky
k1ky's picture
MUPS Discontinued
Thanks Chuck - I didn't notice that.  Grab 'em up fast!
NM7B
12VDC in, 24V passive PoE injector

We use one of these for one of our full-time rooftop nodes that only has 13.8VDC available to it via a spare pair on our rotator cable:
https://www.amazon.com/dp/B07P5PGTY7/?coliid=I2E080LEPUO38&colid=3R6G5VT1EX2Z2&psc=1&ref_=lv_ov_lig_dp_it



 

k1ky
k1ky's picture
12VDC in, 24V passive PoE injector
That's pretty pricey!
AJ6GZ
Tycon
I've used several of Tycon's 12V > 48V 802.3af PoE. They make good stuff. You get what you pay for...
kf6iiu
AREDN Mobile Is Fun
Mobile mesh is a terrific use case for AREDN. I use the setup pictured below to "wardrive" various locations to scope out connectivity. The 1" PVC Sch 40 pipe, just by coincidence, slips over the 1" OD upright tube end of my Thule bike rack. There are a million ways to mount a mast like this on a vehicle. Inside the vehicle, the POE adapter and laptop are powered by an inverter. 

A moving station is going to be subject to dropping offline for a few seconds at a time as the signal and route configuration changes. I always operate while stopped. How your network might behave from a moving vehicle depends on a million things, you'll need to experiment. You should do a test deployment to see what work best.

Remember, when AREDN says "line of sight" they mean it. Any obstructions in the way will block your signal. Bearing that in mind, you should be able to connect over a half mile or so with 2 omni antennas, 2 or 3 miles between an omni and a sector antenna, and 5 or more miles with 2 sector antennas.

 Let us know how it works out. 
 
Wiley KF6IIU

w6bi
w6bi's picture
802.11n Mobile

A while back while digging into some technical details about the 802.11n protocol, I found the comment that due to the protocol's sensitivity to signal phase, it would be very sensitive to doppler shift.  No other details were given, so it would be interesting to see how SNR to a remote site (and maybe throughput) behave at different vehicle speeds.

Orv W6BI
 

kn5g
Mobile, Portable nodes to cover large event area, (Practice Run)
I do appreciate all the comments and suggestions!
This past Sunday afternoon I was able to get two other local Hams to meet me at a local public park and we tinkered with our collection of MESH Nodes. We had a couple of Ubiquity directional devices, a couple of TP-Link directional devices, and a handful of the GL-INet AR750 devices. One Ubiquity and both TP-Link devices had IP-Cams attached. The directional nodes were all mounted on 4-6 foot PVC mast with tripod stands.

This was a useful exercise from the beginning since we found out that part of our devices were configured for mesh channel -2 and others were configured for some other channel and were not seeing each other. We had a couple of Windows Laptops, an iPad, and Android and IPhones to monitor the Mesh Nodes with. Once we sorted out the different channel configuration we spent some time with the Ham who was not familiar at all with MESH Nodes and giving him a hands-on crash course. He was impressed and later texted me to let me know he had ordered a couple of the GL-iNet AR750s and an IP-Cam to play with.

We were able to stream video from the various cameras from around the park back to the picnic table we were using as a command post. We were also able to stream video from across the lake from a node and camera mounted in my truck as I drove along the road beside the lake. This was not a long distance, maybe a 1/4 of a mile or so. The camera in the truck was feeding into a TP-Link CPE210 which was pointed back to the GL-iNet AR750 that we were using as a wireless access point for the laptops and iPad to access the MESH Network. The CPE210 and the IP-Camera were powered by a 40Volt weed eater battery with a little 400 watt power inverter inside the truck. They could have been powered from the built in inverter in the truck but they were already wired up to the battery so I left them that way for our test.

We were able to maintain a connection between the Ubiquity unit and the TP-Link unit when I drove away from the park and we had a pretty dense stand of trees between us. The Link Quality did drop down to 50% and even less but we were still able to stream video. We had one of the GL-iNet AR750s set up so that we thought it would be a bridge between the Ubiquity and the TP-Link but after looking at some MESH Status screen shots later on we realized that the path between the Ubiquity and the TP-Link nodes was better than the link between them with the iNet between them. I believe the video was streaming directly from the TP-Link to the Ubiquity node even though we had the trees in the path. Of course the trees did not have any foliage on them this time of year so they were not attenuating the signal as much as they would have had they been covered in leafs.

While we did discover and correct some bugs in our setup we didn't really have enough participants to really check out a string of nodes along the roadway as I had hoped. We need to have a few more "Test Runs" with more participants so we can get a feel for how far apart our nodes can be and still maintain a network.
We were pleased with the ranges we were seeing between the omni-directional GL-iNet nodes since they are inexpensive, and can be powered with 12/5 volts easily. It would be a much easier sell to get other club members to purchase one or two of these than the more expensive devices.

We are looking and hopefully will find a few locations where we can mount some permanent nodes up high, but in the meantime we need to experiment some more with the ground level, and mobile nodes and learn how best to deploy them.

More to come, I'm sure!

Keith
KN5G
K6AH
K6AH's picture
Check out used gear...
For the price of a GL-INet AR750 (~$50) you could buy 1-2 used NanoStation M2 or M5s on eBay and extend your link distances to ~15 miles.

Andre, K6AH

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer