Sean, do you literally want to make one? As in, pringles can, etc.? Or are you looking for the highest gain COTS one?
Nelly's Security | 07/24/13 11:34pm
ha ha, no I dont want to manufacture one, maybe someday though. Although I am not opposed to buying a third-party antenna to boost my signal. I want to know what specs I need to be looking for to achieve what I am wanting to achieve. What makes a good wireless access point "good" Keeping in mind, I want omni-directional, that will give me the farthest, strongest signal, even through walls.
For example, a person wants a high definition security camera. a 1080p camera will give him better quality than a D1 camera. a 5MP will give him even better quality. 16 MP wiill be even more high def.
There are two issues with high-gain omni antennas:
- They're big. The smallest on Tessco that I could find is 28" long. And that's only 7db gain. Keep in mind you'll probably have ~2-3db loss from connectors and cables, so a net of 4ish.
- They're vertically narrow beamwidth. While they're 360 degrees horizontally, that circle may only be 7-10 degrees vertically.
Nelly's Security | 07/25/13 12:00am
Very interesting info about the vertical beamwidth. Would have never known
One issue you might run into is finding an access point in the price range mentioned that can support an external antenna. Most of these use N-Type Connections so that would be something to check for when shopping.
Several factors are considered when you are going to decide if a access point is "good" compared to others. Looking at the enterprise grade product on the market (Cisco, Ruckus, Aruba, Moto, etc...) all use either Atheros or Broadcom chipsets. What makes most of their products unique are the management and software features as well as the quality of the other componets used, of these companies Ruckus is really the only one focusing on the antenna. Depending on your application one of these solutions may be more than you are looking for but each has their own claim to fame.
The best way to picture how a omni antenna works is to think of it as a light bulb. Transmitting energy (or signal) 360 degrees. While you would think increasing the gain on the antenna would be the best way to solve your issue it sadly is not that easy. It will give you extended range but at the same time it will also dramatically increases interference in every direction which will result in a decrease in how well the AP can listen to the client which is half of the battle.
Can you share what type of clients you will be connecting to?
Nelly's Security | 07/25/13 12:44am
Thanks. Basically trying to come up with a way to make a "universally good" wireless IP camera system for residential and small businesses. When I say "universally good", I want a good access point that will cover most residential and small business scenarios. Yeah I know ever situation is different, but if you had to speck out a good access point that would generally be good for most residential or small business applications, with say a 4-8 IP camera system with a total throughput of about, lets say no more than 50 Mbps, what would it be or what qualities would it have. Reason why I am looking for an access point is because I do not want to rely on the wifi connection of my customers many different routers. I would like to rely only on the access point that I am providing. It would be a point to multi-point setup. The clients are IP cameras with a cheap built in wireless B/G/N access point. I would be using the main access point as the "server"
So generally speaking, what would be some good strong qualities to look for in an access point with the scenario I stated above.
These answers are unacceptable :) :
- Every situation is different, it would depend on the situation. Yeah I know, but generally speaking.
- A wired installation is better, you should'nt do wireless unless you really have to. Yeah I know, thanks.
Sean, I'll add another answer you will find unacceptable:
- You are looking for the Holy Grail of wireless home surveillance, and unless you are Indiana Jones, I doubt you will find it :)
One alternative offered by Bohan is to use Sonos bridge / repeater units.
Finally, there appears to be 15db antennas that are 40 inches tall (e.g., this one on Amazon).
Nelly's Security | 07/25/13 02:20am
Nope, not looking for a miracle tool. Just trying to find out what specs to look for. I have a netgear wifi router that I have a few cameras wirelessly and an NVR connected to with access points, and I have pretty decent performance out of it actually. So I am guessing there has to be even better equipment out there then whats on my little consumer grade netgear router.
Is the netgear the holy grail of wireless routers? No but I bet it would work great for the majority of homes and small businesses when it comes to wireless networking computers, printers, etc. I am looking for an access point that would kind of do the same thing that would maybe get me even better performance because even on my netgear I get pretty decent performance.
In other words, I want to keep the wireless bandwidth off of peoples routers, and I dont want to rely on other people routers wifi connection, just in case they have some kind of piece of crap wifi router. Just trying to gain some knowledge is all, because I'll be honest, I dont know alot about the guts of wireless technology.
Nelly's Security | 07/25/13 02:47am
BTW, I am not looking for the stongest antenna as the title implies. I am looking for the best Access Point/Antenna to get me the best results of my intended usage.
First you have to understand something about FCC rules and regulations. The maximum gain for a point to multipoint is +36dbm or 4watts. This is calculated by taking you AP output and adding your antenna gain. You CAN NOT just add a high dbm antenna to an access point without doing some basic calculations of you might be breaking FCC rules and regulations and opening your self up to being fined.
If you are looking for some good inexpensive wireless you should look a ubiquiti.
dont forget most of the time its not the access point that is the signal problem but the client. Your only as strong as the client can tranmit
Nelly's Security | 07/25/13 06:04am
Thanks John, but if talking about a point to multi-point system, wouldnt the "server" access point be more important than the "client" access points? I figured you only as good as your server access point.
What do you guys think about this one for the application I am talking about: Engenius access points
and other than being outdoors, what makes it better than this one: Wireless Ethernet universal repeater
as far as wireless performance, just by a quick look at the specs, they look about the same.
Sean, if your goal is really this: "Basically trying to come up with a way to make a "universally good" wireless IP camera system for residential and small businesses."
You should be careful about the complexity of the components you use, because the potential troubleshooting mayhem and space of throwing in Ubiquiti or Engenius (or whoever) units around a house is an invitation to trouble.
I have never used the Sonos Bridge but it has strong reviews on Amazon and they claim it's a zero config, peer to peer mesh network. I suspect throughput is low but I don't know. However, the main benefit I see (theoretically) is that it is a lot cheaper to throw out another $45 deck of cards box than to configure and troubleshoot access points.
Nelly's Security | 07/25/13 06:36am
Thanks John. I'm not concerned about the complexity, we usually throw away the manufacturers instructions and write our own easy to understand instructions anyways. Our motto is "If a 90 year old granny cant understand our instructions, then they arent good enough." Also, most of our customers are installers who have a decent amount of tech knowledge.
I did find this article which I thought was very informative: Good Wifi Article
There's no amount nor clarity of instructions that will safely and consistently allow home users or small business users to set up wireless access points and repeaters. I think you are severely underestimating the complexity of such deployments.
I've seen legions of qualified security tech installers have huge problems deploying even modest wireless surveillance systems.
Nelly's Security | 07/25/13 06:53am
Trust me I know, but we have explained how to setup IP camera systems to people who dont hardly know what a left click is. Once they read the instructions, they usually can figure it out. The hard part is, getting them to actually read the instructions!
But like I said, this will be an extremely simple setup. One wireless access point that is wired into a router and then a few wireless cameras that have built in access point clients. For small areas. Not warehouses, hotels, etc where repeaters would be needed.
You are not really looking for the 'strongest omnidirectional signal' since your clients are IP cameras. So you are streaming video FROM the cameras TO the access point, meaning that the access point power output will have little impact on overall performance. What is actually important is the return path, i.e. you should be looking for an AP with a high performance low noise receiver to achieve teh highest possible data rate. This is generally one of the benefits from moving to an 'enterprise' from a 'consumer' access point.
Having a high antenna gain is not going to be a great help in an indoor non-line of sight environment. As mentioned above the only way to get horizontal gain in an omnidirectional antenna is to reduce the vertical beamwidth. Think of an antenna like a balloon full of a fixed amount of water. If you need it to have a gain in one particular direction the only way you can do it is to squash the balloon, reducing the gain in some other direction. If you need a generic design, how is that going to help if for example some of the cameras are installed upstairs? Also it's quite possible that the best signal return route may be via a ceiling void rather than through a solid wall.
3. Wi-Fi is a shared medium. Your signals will inevitably need to coexist with others. So you should be looking to mimimise external interference by selecting the best operating frequency and not increasing the level of antenna gain towards unwanted transmitters.
4. You are loooking to stream multiple sources. So you need to choose an access point that can effecitively handle and schedule multiple video streams.
5. You are seeking to stream up to 50Mbit/s. More knowledge of your client (camera) devices is required to know whether this is even feasible. If they are single stream, single antenna 2.4GHz only devices, their maximum raw data rate is only 72Mbit/s in the presence of other Wi-Fi signals (preventing 40MHz operation). So this data rate could only be got close to with zero external interference and all clients operating at full rate (most improbable unless they are all in the same room as the AP). If the cameras are dual band dual stream 2.4GHz/5GHz then we've got 450Mbit/s of raw data rate available - much easier to work with.
Anyway, my recommendation for this scenario would be to use Ruckus since they incorporate technology to overcome many of these issues without excessive complexity for the user. Specifically:
* Multielement adaptive antennas that continually adjust their pattern to improve wanted signal gain and reduce unwanted signal interference
* Polarisation diversity (on 7352, 7372, 7982 APs) to account for unknown alignment of client device antennas .
* Packet scheduling algorithms to optimsie streaming video.
Of course they are considerably more expensive than products such as Ubiquiti or Engenius
IPVMU Certified | 07/29/13 12:31am
I have really good luck with Ubiquiti products. I have installed the small airMAX point to point and have used the Ubiquiti Unifi series for multi-mesh network. The mesh and other features found in the 3.0 are still in beta but I haven't had any issues thus far. I am hosting the cloud controller on a VM from Digital Ocean and controlling 4 locations on the one VM. This way I can manage multiple locations from one controller which is the major benfit to the 3.0 version. It helps in deployment as well. I haven't used the Unifi for cameras like I have with the airMax but the outdoor AP does have an extra NIC with 24V. It could be used with an injector to connect 48V to a camera. I haven't tested this with the Unifi series, but I might give it a try. I understand there are many more expensive systems, but I have had good luck with Ubiquiti thus far. (TIP: On their toughswitch PRO don't turn on 48V and plug an intenna in by accident. It will fry the protection and you either take off the chip or find a replacement). I like the Toughswitch Pro 8 port switch as well. On my home test network, VLANS were very easy to setup compared to the Netgear where I had to go on multiple pages to set the PVID and tag settings.
I have also used Ubiquiti Airmax extensively for CCTV, but it is a totally different product line from their Unifi indoor WiFi range.
Andrew makes some very good points - I would add:
802.11 is not licensed - so your signals share space with others - if you come in and blow out existing installations - you will have unhappy neighbors (or your own customer if they have a system installed)
Megapixel streaming video - forget 2.4 Ghz completely. You need to be in the 5.8 Ghz "n" 40Mhz channel scenario.
In fact, forget 2.4 Ghz completely anyway - its a garbage band with massive interferenece and limited throughput.
There is no "universal" solution. Building construction and existing RF environment (neighbors and other sources of interferenece) requires system engineering - a site survey to determine the best design.
802.11 design theory is 180 degrees from your design intent. Its not about "how far can I reach" - its about small cells with better throughput due to fewer associated clients.
Nelly's Security | 07/29/13 04:49pm
Very good info. Thanks for all the input
IPVMU Certified | 07/29/13 05:18pm
Always remember the "What you do not measure, you cannot manage". With that said, I use a couple of tools to discovery, measure and manage my WiFi environments (see metageek).
I have and use Wi-Spy DBx and Wi-Spy 900x hardware along with the Chanalyzer Pro software
They have a free product called inSSIDER
There are other software tools out there on the internet that utilize your built in wireless interfaces on your laptops but for the best and more useful functions generally a dedicated USB/wireless interface module (such as Wi-Spy) is coupled with software for the best and more complete results.
IPVMU Certified | 10/22/13 05:51pm
Sean what did you decide on? I fight this wireless fight all the time with little expertise with AP's and signal manipulation.
Pro Focus LLC | 10/22/13 07:27pm
I don't mind using wireless as a backhaul, but if the goal is to not have to wire a small office, then wireless seems less valuable to me. If you already have to run power to the camera, why not just run low voltage?
I know that most areas require a licensed electrician to do commercial high voltage (120VAC) work, however in our area, Toledo, OH, we do not need permits, licenses, etc to run low voltage.
Also, the cost of CAT5e is far cheaper than MC or primary wire and conduit. There really isn't a good reason not to run wire inside of a single structure, unless it has a long distance between points that would require a midspan.
We have generally only resorted to wireless backhauls when we were faced with a direct burial / trenching option. The cost of having a crew trench and bury conduit far exceeds the cost of some Ubiquiti radios and some PoE switches. But, that is assuming that you don't have radio interference with other systems in the area.
So unless you already have power close to where the camera is going to be mounted, I don't think wireless will ever be needed at this aspect.
Nelly's Security | 10/22/13 07:36pm
I never really pursued it much more. I know dropcam is wireless and they are successful with it, but that is just used in small indoor spaces. I think the big issue would be having the cameras on the exteriors of the structures and the wireless dropping in and out there. I just dont want to open up a can of worms as far as a tech support nightmare so that was the main reason I dropped the idea.