[Fourlanders] Re: Thoughts on wireless network ? - Long Response

[Brian McCarthy]

Bert,

Yes, I am late in my posting here. I just did not have a chance to pull 
all of my thoughts together on this immediately.

I have a a multi-layer network of PC's at home. Part of that network 
includes an all-in-one nearly indestructible PC for my almost 2-year old 
daughter with a PCI wireless ethernet card in it. I also ahve a PCMCIA 
wireless card that I slide into my work laptop to connect to the home 
network. It is convenient simple and very handy.

I have been running these for about 6 months now. I have had no 
significant problems, but I do not operate ANY high power ham gear at 
home. I also do not operate on 2304.... Most of the 802.11b hardware is 
in the 2.4GHz range. There are 11 channels and it is possible to select 
your channel to reduce the possibility of interference and/or 
detrimental mixing products. I just do not know what the interaction 
might be.

I almost brought my company laptop, its wireless card and my wireless 
access point to the mountain. I had thought that it would be nice to 
have a portable to use for real time score and log review as well as for 
going to a vehicle and logging the odd FM contact. It would have been 
very handy if we had found a lot of activity on 1296MHz. for logging at 
your truck.

Line of site connectivity is spendid with these cards. When you take 
them into a real house though, the distance of connectivity degrades 
quickly. Since the direction of the group is to have more of the 
stations setup inside cargo trailers (big shielded box), any wireless 
network would require external antennas for solid connectivity. I am not 
aware of any PCMCIA format wireless cards that have a detachable 
antenna. All that I have seen of the PCMCIA types are built with surface 
mount antenna arrays encasulated into the plastic packaging of the card. 
There are some USB wireless cards, but connecting one of these to the 
older (almost disposable) laptops we use on the mountain might be 
troublesome to impossible. For the purposes of EMI/RFI, we discourage 
the use of conventional cathode ray tube monitors on the mountain. They 
are huge RF noise sources. This pretty well rules out the use of 
 desktop or tower case PC's on the mountain, unless you really want to 
bring a $500+ flat screen monitor up there?!

In our configuration this past contest it would not have been a problem 
to slide a PCMCIA adapter into the 6M laptop. We could have arranged to 
put the access point at the 144/222/432 trailer, but it would have been 
outside. Sure, we could have put it under a plastic bucket on top of the 
trailer, but I would not have wanted to with all the constant rain. If 
6m had been in a metal trailer, I would not have expected it to work.

My thoughts are as follows:

1) The laptop and station at the 6M tent had some issues with stability 
and RF into the complete system. We did not find any correlation of 6M 
transmitting to ethernet dropping though. More investigation would be 
prudent.

2) Wireless networking of the stations is unproven on the moutain and 
should be tested on a limited basis (with conentional backups on hot 
standby) before really investing in the technology.

3) The Bit-Error-Rate (BER) of our long runs of ethernet cabling has 
never been tested. I do check the cables at home on the 100baseT side of 
my home network before each contest. (100BaseT requires more pins/wires 
and is more sensitive to cable faults.) I will look for some software to 
do BER testing on the cables.

4) I will be shopping for shielded twisted pair (STP) CAT-5 or better 
cable for use in the future. (...but where would we connect the drain 
wires?) Using STP should further reduce the possibility of RF and other 
noise slowing or halting our network.

5) Now that we are going to enclosed metal trailers, to fully utilize 
the benefits of the shielding we need to implement proper single point 
grounding of all the equipment to the trailer and the adjacent tower. 
This might actually be of more benfit than STP for the ethernet cable.

6) We should continue to cap the cabled network at 10Mbits. There is no 
reason to go to 100BaseT as we do not generate anywhere near enough 
traffic to slow down 10BaseT. The hub we have used and that I intend to 
continue using is an 8-port metal encased unit that will only do 10BaseT 
(self limiting.)

Sorry if this was too long. I guess I really must be the right person 
for overseeing the networking for me to have this much to say about it! ;-)

I will post the log results and summary this evening.

Later,
Brian
NX9O


Bert Rollen - K4AR wrote:

> Hi all,
>
> Considering the ups and downs of the network during the contest, is a 
> wireless LAN feasible ? Considering the level of RFI between the 
> stations, I doubt that we would even notice a contribution (if any) 
> from the LAN.
>
> Thoughts ?
>
> Bert
>
>
>