Long-Distance WiFi Device

[Beausoleil]

Originally Posted by markpj23

Buy the short pigtail from Data-Alliance. Make sure you get the correct connectors for the antenna below if you go that route...

I searched their site (DA) for antennas and did not see an outdoor model I liked. I got the engenius 8db outdoor antenna that HUD mentioned at the beginning of this thread: EAG-2408 - EnGenius Datacom Products

I found it at Prometric.com. Not the absolute cheapest supplier but they had it in stock and were within $4 of the lowest cost dealer I found.

Mark,

Does that antenna have an N connector, or SMA?

[thatboatguy]

"GSKY High Power 500mW 27dBm 802.11b/g USB Wireless WiFi LAN Adapter With Bonus Screw-On Swivel 9dBi Rubber Antenna"

There is a seller at Amazon who is selling them for $31 (about $41 with shipping) with an extra 9dbi antenna. I just ordered one.

George

[markpj23]

Originally Posted by Beausoleil

Mark,

Does that antenna have an N connector, or SMA?

I got a pigtail RP-SMA male (Alpha end) to N-male (antenna end).

RP-SMA male to N male. 10-inch "pigtail" for wireless / WiFi network

[bbhflts]

Originally Posted by RBEmerson

Um, if you're under way, I suppose it would matter, but on the hook or tied up, if heeling's an issue, it's time to ask what the weather's doing or why the boat is that badly out of trim.

But why stop at 12 db when the Hawking antenna does even better (and, yes, 3 db does make a difference)?

On the issue of directionality and gain, TINSTAAFL applies to antennas, too. Gain comes from reducing the volume of coverage, no way around it. The trick is to find the compromise that fits your needs.

DISCLAIMER: I have no connection with Hawking Technology, save as a customer.

Guys. Watch out for dBi vs dBm. dBi is relative to an imaginary isotrophic antennae that only exists in theory. Therefore when you consider efficiency and real gain, the dBm will tell you what you need to know in terms of real power output. Even then, check the max output dBm, which will be determined by the max input minus the inefficiency loss.

Cheers.

Ben.

[thatboatguy]

That's a very good point Ben,

Can you back up a bit and put that in the form of a primer? A lot of folks are not going to understand the brief explanation. As mentioned earlier there are marketing types who take advantage of the less than straight forward and various measures of wifi power.

George

[Gashmore]

Not to confuse things but dBi and dBm are two differnt things but you need to know both. There are a lot of different dB's that measure different things but each increase of 3dB of any type doubles what ever it is measuring. dBm is a measure of relative power with 0dbm = 1mw. A 27dbm transmitter will deliver about 500mw at the output connector.

dBi is a comparison of "gain" by an antenna relative to an ideal theoretical antenna that radiates in an even sphere. By squeezing the sphere into a doughnut you reduce the power radiated along the antenna's axis and increase it perpendicular to the antenna. The flatter the doughnut the higher the dBi. Slicing the doughnut into a wedge as in a directional antenna increase it even more.

You can add dBm and dBi to estimate the "effective radiated power" (ERP) of a transmitter/antenna combination. For example, that 27dBm transmitter connected to a 3db rubber duck that came with it has an effective radiated power of a 30 dBm or 1 watt transmitter connected to that ideal theoretical antenna. Upgrading to a 9dbi antenna will increase the ERP by 4 times over the 3dBi antenna. Remember however that this is true only when the receiving antenna is in the plane of maximum radiation and there will be some loss in the coax cable that the stock antenna connected directly to the transmitter does not encounter.

Looking at that doughnut from the side you see that the edges are rounded. Similarly the amount of signal seen along the curve of the antenna's pattern drops as you move away from that plane of maximum radiation. The point where the power drops by half (3db) is called the beam width and is measured as the angle between the points at the top and bottom of the doughnut. Similarly, the pattern of a directional antenna is shaped like a cheap party balloon. Rounded at the end and tapering back to the antenna. The angle between the points where the signal drops by half both horizontally and vertically is the beam width.

You can squeeze the doughnut a lot and get a lot of range or slice it up and get even more but the more you concentrate the signal the more effort it takes to keep the receiving antenna inside the beam width. When one of the antennas is mobile that can be rather frustrating.

[Beausoleil]

Ben,

Gashmore's got it right. What you were thinking of was dBd (decibels referenced to a 1/2 wave dipole).

Marketers will always use dBi, since it makes their antennas seem to have higher gain, since a 1/2 wave dipole has a gain of 2.15 dB over a reference isotropic radiator. To the unwashed masses, a 10dBi antenna sounds better than a 7.85 dBd antenna, even though they have the same gain.

Even worse is the ham radio multiband HF antenna market - rarely will you see any gain advertized, since in the attempt to make them resonant a multiple frequencies, their gain is most often less than a resonant dipole.

[bbhflts]

I’ll try to summarise my thoughts here from the entire discussion, but also that of my IP networking and RF experience. I’ll try to keep it simple and digestible for the majority as many of the guys have already done a deep dive.

Key considerations for max performance, idiot proofing, ease of use, user friendlyness and lowest power consumption:

• KISS Principle. Try to limit it all to one box.
  • Caveat against the KISS above. Putting it all in one box could be a single point of failure, but does limit power consumption.
  • One box solution will give you firewall protection, encryption protection, repeater/bridge functionality (depends on your box) and lower overall power consumption.
• Security is critical. A single firewall wireless box with encryption to your laptops is a lot simpler than running a firewall application on each pc and much more secure. Also, running multiple personal firewalls increases each laptops power consumption. Lets be realistic. Most people other than us IT guys (of course) have clicked “ok” to a firewall app, virus warning or spyware app warning, only to be infected. Try saving your data, email, media and reinstalling windows + all apps when at sea. Not fun.
• Local WiFi rules, USB sucks except for simplicity (sort of).
  • You have to have the drivers set up for each laptop and with each different OS, you have more hassle and complexity.
  • USB up close to the mast, does the job, but you don’t have a marine rated solution with the dongle up the pole.
  • USB local to the laptop, with RF to the pole, you have to deal with RF loss on the coax and few USB dongles really output the optimal power to cope with the coax and antenna loss.
  • Most of us will have multiple laptops and multiple users. Logic says that your nav laptop should only be that and NOT multi functional. (KISS and avoid crashes etc., no silly apps other than required for running the boat.)
  • Multi-user: When you connect to WiFi, you are going to be anchored/berthed. When in WiFi range, there will be a mad rush to connect anything IP. Hence the skipper will need to download the latest nav stuff like charts, weather, software updates etc for the nav laptop. Then there is the missus (and yourself and any crew, especially the little ones) connecting on PC’s and downloading bucket loads of everything. If you have USB there will be a fight for the connection. You need WiFi locally on the boat for seamless multi user.
  • Other than multiple PC’s simultaneously, WiFi locally means that one doesn’t have to worry about cables or position on the boat (bed, saloon, cockpit, beach etc..).
  • WiFi locally on the boat means you can set up a PC as a media server, a Network Attached Storage (NAS) or third party media streaming solution and deliver it to anywhere on the boat even when NOT connected to the internet via the shore. Lets face it, we all have terabytes of downloaded movies, music etc.. and most smart phones, iphones, etc can use WiFi to stream content.
• Omni vs Directional Antenna. Unless you have one of those anti sway gadgets that hang of each side of the boat, I tend to agree with “Gashmore”, ease of use is probably more important than getting an extra mile or so. It’s nice just to find a signal, throw out the anchor and not have to do anything more except surf the net, check emails and fish etc..
• Minimising the coax run. This is a no brainer. You want to limit the loss of the RF on the way to the antenna as much as possible.
• Choice of Antenna. “There are three kinds of lies: lies, damned lies, and statistics” (Benjamin Disraeli and popularised by Mark Twain). Let me give you an example. You just bought a hugely powerful 100W stereo. The marketing people don’t tell you this is actually Peak Momentary Power Output (PMPO). The 100W PMPO is in fact over a sine wave, so the real Peak Power is only 50% of that. Then you need to find out the real or average power output other than the peaks. This is Peak Power (our example) 50W x .707 = 35.35W otherwise known as RMS or Root Mean Squared (lets just say average power). Now we realise it’s a stereo, so we have to divide 35.35W by two to get the real power per channel. Ok, you just bought a 17.675W per channel sound system. That’s a bit of a cry away from the 100W advertised.
• dB is a bit like the above. Confusing to most and easily used by marketing and vendors to mislead. First point. dB is not a measure of power, it is a unitless measure of gain. If you don't know the reference level, the dB power has absolutely no meaning. It’s like saying A x B where both A and B are unknown. For dB to have any meaning it must have a reference such as either A or B being known. Hence in the RF industry, you have dBm with the m being a reference to milliwatts. dBi being a reference to an imaginary perfect antenna and so the list goes on. Forget BS figures of dB, dBi etc unless it specifically states dBm output. This will tell you the maximum real output, but remember dB is a factor of efficiency. Therefore, whilst an antenna may have a very attractive dBm max output, if you don’t pump in the right amount or it’s not balanced, tuned or optimised correctly, you will never get the stated output. If the vendor is ethical, they will be clear about the conditions of input to produce such real dBm output. If you hear “dB” ALWAYS ask dB what? It must be relative to a baseline otherwise they are not measuring anything other than their imagination. Then check the input.
  • The Hawking antenna is like this. Actually it’s dBi stated, so you should subtract at least 3dB from this to get close to reality if you trust their figure in the first place based on how they presented themselves.
• Facts:
  • An antenna is a lossy device, it’s not 100% efficient and doesn’t magically produce “gain”. What you need to know is how efficient it is, what it’s tuned to for optimal efficiency.
  • Output power is only half the story. Sensitivity or the ability to receive signals and discern the noise from the good stuff in an intelligible fashion is just as important. For example, what is the point of yelling louder when you are deaf to make better conversation.
  • To work out your real output, start with the RF generating device. Then deduct the loss of transmission to the antennae, then deduct the loss of the antenna due to it’s inefficiency. Even then, you will probably be lost as to how good your system is as it depends on the foot print of your antenna such as omni or directional.
• Solution:
  • If you’re not an IP or RF engineer, don’t try to engineer it yourself unless you have a penchant for masochistic activities or an overly fat wallet. Follow the advice of one of the boffins in this forum that says something like: Connect device A to device B with connection X and connectors Y……and so on where they specify every bit of the path including hardware and software……and all the other boffins have agreed.
  • There are three solutions:
    • The Masochistic and possibly financially wasteful.
    • The cheap, but well proven…. but not that flexible etc., USB solution.
    • The optimally engineered cost effective high performance solution that you wait for the boffins to prove and describe in perfect detail.
• Considerations:
  • If undecided on the RF device, whilst power isn’t everything, you may want to consider that the FCC is a tad more lenient on output power than the Europeans. I can’t remember the exact figures off hand. The point is if you are going to rely on the box’s max configurable output, if it’s an EU box it will be slightly limited, even if it’s via the software.
  • To open up a can of worms. No one has mentioned 802.11n yet. A utopian solution that will be future proofed for at least 5 years will be 802.11a/b/g/n. This raises a few more issues such as cost of additional antenna’s, place to put them, product availability, the fact that to really utilise “n” you need two radios in the one box with one for 2.4Ghz and one for 5GHz plus three antenna’s each and true 40MHz channels, just to name a few,….. etc.. What it gives you is MIMO (nice for added performance and building penetration when you’re trying to connect to an indoor WiFi Access Point, better Non Line of Sight and buckets more speed up to 600Mb. Reality is that you probably don’t need the speed as few AP’s will ever give you more than a few Mb speed wise. Still food for thought re future proofing.
  • The idea of putting the box up the mast is an interesting one. I particularly like the comment on the on the counter lever effect. Either way, it’s nasty, hoisting such electronics, having it banged around and then worrying about the connections and managing the cable. I’m thinking I’ll be putting mine on the davit, regardless of height being king, unless the team concludes that the bullet product is the icing on the cake.
  • Antenna, I’m waiting for the verdict from the team as it’s a sea of misinformation out there on the best marine grade product.
  • The magical box that does everything. Again, I’m out on the verdict for now, but will feed back my thoughts once realised.

I say this after 20 years as an RF and IP engineer.

Hope this helps those confused out there.


Cheers.


Ben.

[motion30]

Originally Posted by Gashmore

Not to confuse things but dBi and dBm are two differnt things but you need to know both. There are a lot of different dB's that measure different things but each increase of 3dB of any type doubles what ever it is measuring. dBm is a measure of relative power with 0dbm = 1mw. A 27dbm transmitter will deliver about 500mw at the output connector.

dBi is a comparison of "gain" by an antenna relative to an ideal theoretical antenna that radiates in an even sphere. By squeezing the sphere into a doughnut you reduce the power radiated along the antenna's axis and increase it perpendicular to the antenna. The flatter the doughnut the higher the dBi. Slicing the doughnut into a wedge as in a directional antenna increase it even more.

You can add dBm and dBi to estimate the "effective radiated power" (ERP) of a transmitter/antenna combination. For example, that 27dBm transmitter connected to a 3db rubber duck that came with it has an effective radiated power of a 30 dBm or 1 watt transmitter connected to that ideal theoretical antenna. Upgrading to a 9dbi antenna will increase the ERP by 4 times over the 3dBi antenna. Remember however that this is true only when the receiving antenna is in the plane of maximum radiation and there will be some loss in the coax cable that the stock antenna connected directly to the transmitter does not encounter.

Looking at that doughnut from the side you see that the edges are rounded. Similarly the amount of signal seen along the curve of the antenna's pattern drops as you move away from that plane of maximum radiation. The point where the power drops by half (3db) is called the beam width and is measured as the angle between the points at the top and bottom of the doughnut. Similarly, the pattern of a directional antenna is shaped like a cheap party balloon. Rounded at the end and tapering back to the antenna. The angle between the points where the signal drops by half both horizontally and vertically is the beam width.

You can squeeze the doughnut a lot and get a lot of range or slice it up and get even more but the more you concentrate the signal the more effort it takes to keep the receiving antenna inside the beam width. When one of the antennas is mobile that can be rather frustrating.

i bought a hawkings omni15dbi and it seems to work no better then the rubber ducky ant the cane with the alpha unit, providing they are at the same elavation Why is that ?

[Gashmore]

I was with you up until "Forget BS figures of dB, dBi etc unless it specifically states dBm output. " I have never seen an antenna rated in dBm and if I had I would immediately discount it. Antennas are rated in dBi (reference to a theoretical isotropic antenna) or dBd (reference to an ideal dipole which is 2.15 db lower). It is a very usefull figure for comparing antenna performance if you know how to use it. You just have to be careful not to mix dBi and dBd. Virtually all wifi antennas are rated in dBi. While the specifications may state a maximum power an antenna can handle the should not state a dBm.

What you are after is how much power is radiated in the direction you want it to go. You get an approximation of that by adding dBm and dBi to get "Effective Isotropic Radiated Power" (EIRP). A 9dBi antenna will still have an EIRP 4 times that of a 3dBi antenna regardless of whether it is fed with a 200mw (23dBm) transmitter or a 500mw (27dBm) transmitter.

We are not talking laboratory accuracy here. Just good enough figured to make an intelligent choice. It is not rocket science. With a basic understanding of the principles, good equipment, strong mechanical mounting and connections and a little patience a relatively handy sailor can install a very good wifi system. Better in fact than some professional installations I have seen.

[bbhflts]

Hi Gashmore.

I won't agree or disagree on your comment "I have never seen an antenna rated in dBm and if I had I would immediately discount it. Antennas are rated in dBi (reference to a theoretical isotropic antenna) or dBd (reference to an ideal dipole which is 2.15 db lower)." I'm not a WiFi guru. I started in HF, VHF and UHF, then satellite prior to moving to IP. Currently I sell WiMAX solutions. Now considering that they are not going to be full mass market with roaming capabilities in the next 12 months, I won't drill into WiMAX.

My guess is that you are correct that most antenna's for WiFi that are sold individually are rated dBi or dBd, but I wouldn't know for sure as I'm not a seller of such antennas or a serious consumer. When selling a WiMAX solution though, especially in the Service Provider market, it's all about link budgets, both on the uplink and downlink. This is the real meat and proof point of any meaningful discussion on a system performance. It's what you get in real life, not just theoretical calculations of how much one thinks they can achieve. I'm not against the theory or marketing, but have seen many vendors BS'ing their way into deals by fudging the numbers, when at the end of the day it comes down to the real life link budget.... and that typically is a dBm figure correlated against the modulation scheme.

BTW, what are your thoughts on the "n" options?

Cheers.

Ben.

[bbhflts]

Addendum:

I've just reviewed the Ubiquiti Networks, Inc. site and noticed that they are a partner partner or my company. I'll have a chat with our engineering team internally and see what the view of the product is.

In the meantime, for the utopian solution assuming that the bullet gets around the feed issues to the antenna, there are still two unanswered questions.

1. What's the optimal box to act as a local WiFi router feeding ethernet to the bullet.
2. What's the optimal antenna.

All assuming of course that 802.11a/b/g is the way to go and not throw "n" into the mix.

I've looked at my companies product line and think that the enterprise solution is perhaps not self install consumer friendly enough. Then our consumer product line is in my view not robust enough or with enough tech support.

Assuming that we are now looking to the bullet, we have an indoor classified device to be a router, this opens up the options. Thoughts re ease of use, robustness, support and quality etc., much appreciated......as for the antenna.....

Cheers.

Ben.

[Therapy]

Thanks to all for the information so far.

I am so totally confused on what to do about wifi connectivity on the (maybe, soon to be, possibly) boat, I told the wife, sorry, try the laptop card and if you get a signal then great, otherwise I can't help you.

PS: I am the IT administrator (ever seen that prompt on Windows?) in my house. What a joke that is!

[Gashmore]

You are dealing with a complete system. In that situation effective dBm out of the antenna (actually EIRP) can be used as one valid measure but not for the antenna alone.

I am primarily a Caribbean sailor so to me "N' is sort of "soon come on island time". :-) By the time it becomes widely enough available to make the additional cost effective there will be some other latest and greatest spec. For now b/g is about the best you can expect on most islands. For cruising the US coast N will probably become more available much faster.

After 3 years of installing and repairing 3220s and figuring out their advantages and short comings I have developed several requirements for the ideal marine bridge. I haven't gotten my hands on one yet but the Bullet 2HP looks like it might meet most of them.

#1 Water and weather proof in any position with corrosion resistant fittings
#2 Minimum of 27dBm (500mw) output power @12Mps
#3 Minimum -90 db receive sensitivity @ 12Mps
#4 Easy to configure
#5 an easy to use a polling mode to select the AP you want to grab
#6 12VDC POE powered
#7 small and light enough to be mounted at the mast head
#8 strong enough to survive anything up to and maybe including a roll over. (But maybe not an encounter with the other kind of bridge.)

Antenna:
#1 Water and Weather Proof and corrosion resistant
#2 Omnidirectional 8 to 9 dBi for monos. Maybe 12dBi for cats.
#3 N connector for strength. No RP-SMA or pigtail. (A few inches of LMR400 would make a flexible relief point for that encounter with the other kind of bridge.)
#4 same as #8 above

The boat router is still an open question. I have been using Linksys routers because I an familiar with them but surely there is a better solution.

[bbhflts]

Interesting thread on the Bullet and water proofing here:
Ubiquiti Networks Forum :: View topic - Bullet2 Waterproof question...

I'm still searching their forum to see if anyone has found a router with POE on the outside port so you don't have to use the POE-15.

Cheers.

Ben.