Is VOIP Over WLAN DOA?

prostoalex writes "Voice-over-IP in Wireless LAN environment - a futurist's dream of always-on always-connected service. Guy Kewney from eWeek tests the technologies that try to satisfy this market today and finds nothing but disappointment. " The best result we got was that just once, I heard his voice with a delay of about 15 seconds, saying "You just have to speak up!"--which was part of a 20-second burst of speech from him. The rest was lost.""

Comments

sataphone

[Cyberglich]

Well i have used stantaphone over my home wifi worked ok (1-1.5 sec delay) normal for anything of that nature.

Re:sataphone

[Loualbano2]

1 or 1.5 seconds is not ok for any type of real telecom, maybe for walkee-talkees but not phones.

The telecom industry spent a lot of money to find out what people find is OK. The two main factors are delay and Mean Opinion Score (MOS).

For delay anything past 300ms people will notice, around 500ms you start to hear echo. Most phone service worth it's salt will keep it below 300ms.

MOS is a 1-5 score placed on the quality of the voice through a connection 1 being low, 5 being excellent. More info:

http://www.tech.plymouth.acuk/spmc/people/lfsun/ mo s/

So when the author states that it's not ready for prime time its because a 1 second delay is actually at least 3 times too long. If you can deal with it, more power to you, but the telecom industry would laugh at any company who would try to bring 1 second delays to market.

ft

Re:sataphone

[dasmegabyte]

I've used iChat over my wireless lan to my boys in Cinci who ALSO use wireless lan. Not only was there no lag at all, the clarity was better than any phone I've used.

No Mr. Enderle.

[Raven42rac]

VOIP over WLAN is not DOA. If you tweak the QOS etc settings and don't just throw things together haphazardly, then it works beautifully. Personally, I just wire VOIP to a cordless phone, then let the phone handle the wireless part. Enough of the ____ is dead articles.

Re:No Mr. Enderle.

[LostCluster]

If you tweak the QOS etc settings and don't just throw things together haphazardly, then it works beautifully. Personally, I just wire VOIP to a cordless phone, then let the phone handle the wireless part. Enough of the ____ is dead articles.

WiFi is always haphazard compared to wired. You never know if a packet's going to make it safely or crash into something else in 2.4gHz land.

Phone calls work better when circuit switched... we only do VoIP because packet-based is less wasteful.

Re:No Mr. Enderle.

[nacturation]

Enough of the ____ is dead articles.

Yeah, no kidding. This guy had a problem with getting his VOIP implementation working. Well, let me tell you about my vehicle experiences. I once tried to start up a Lada car and drive it down the street. Well, within a few seconds the vehicle crashed into the median and then careened over an embankment and the car was destroyed. Therefore, according to the logic employed by the "VOIP is dead" reporter, all vehicles will never work and we might as well just forget about this whole "driving thing".

Oh, and a non-sarcastic note to others considering writing such drivel: just because your *implementation* sucks, it doesn't imply that the technology itself won't work. Case in point: most of the large long distance providers are already moving to VOIP, such as AT&T. So now when you make a long distance phone call, it's being routed through VOIP. Make that same phone call on a cell phone and now you're doing VOIP on a Wireless Local Area Network.

Re:No Mr. Enderle.

[bconway]

You expect the average user to tweak QoS settings? The same people that don't change the default password on their Linksys wireless router?

too early

[Chanc_Gorkon]

First off, it's eWeek for crying out loud! Second, it's still too early to proclaim VOIP over anything (WiFi or Ethernet or whatever) dead. Second, I have seen it work and it worked wonderfully. Declaring VOIP over WiFi dead is like saying Apple's dead because it does nto have the market share that Dell does.

Re:too early

[BWJones]

Shoot, we used iChatAV rather successfully from Auckland, New Zealand to Salt Lake City for remote collaboration in a lab environment rather successfully with hardly any delay whatsoever. In fact, I routinely used (and still use) iChatAV with a wireless connection, so I do not understand what this is all about.

Re:too early

[Rimbo]

I agree it's too early to pronounce it DOA. The guy's obviously trolling for readership; things like this aren't going to work overnight.

As far as what you've seen work, he does make a good point: Once you get a bunch of people trying to use it at once, things are going to become an issue, especially with the limitations of Access Points.

The future is mesh networking, where rather than an access point, all of the 802.11 devices run in ad hoc mode. Then, routing software, such as AODV, will automatically generate routes with failover.

I work for a company called Kiyon that has been doing ad-hoc mesh networking for a couple of years. By dynamically generating routes between nodes, we can extend range through multiple hops and reliability by changing the route mid-stream. The lightweight nature of the protocol improves latency as well!

So we're pretty happy about what we've got. It'll solve a lot of the problems of VoIP out of the box.

Re:too early

[sg3000]

> shoot, we used iChatAV rather successfully from Auckland,
> New Zealand to Salt Lake City for remote collaboration in a lab
> environment rather successfully with hardly any delay
> whatsoever

Yeah. I do this all the time.

Ingredients:
1. PowerBook G4
2. Mac OS X 10.3
3. iChat AV
4. AirPort (802.11b version)
5. Comfy bed, little computer lap tray, Collie sitting on your feet (all optional)

Results: no problem at all. No delay noticeable. Voice quality was fine. Voice quality was so good, the whole thing was kind of anti-climatic.

Re:too early

[WasterDave]

Likewise, I gave an interview via iChat AV from Wellington NZ to someplace in California.

However, I would say that iChat and VoIP are actually very different things. The point is (IIRC) that VoIP has to use much shorter frames and more or less *has* to run over H323 to be compatible with other bits of VoIP gear.

To those of you who say "well, to hell with VoIP and POTS numbers and all that, we'll use our modified IM networks and be DAMNED!!" ... I can only suggest that, well, you're probably right.

Dave

Re:too early

[hashashin]

You may not see a point, but I do. Sure, I can use a standard consumer 900MHz phone at home, but where this technology really shines is in mobile phones.

The idea is: you have a mobile phone that you carry around with you. When you are not in range of WiFi, then you use the normal GSM/CDMA network and pay the normal price for it. But when you are in range of a WiFi network, like at the office, you can switch to VoIP and talk for free (more or less).

It may not be perfect today, but I think it's pretty close, and in a few years it will be obvious.

The 802.11 basic standard does not support voice!

[Ho Kooshy Fly]

802.11 standard was modeled around having a CSMA/A algorithm that tried to be as much like Ethernet as possible. There is no provision in the BASIC standard to provide for clients to shut up for higher priority voice clients at all! This means that a data client can blow the voice guy to kingdom come.

There are extensions to the 802.11 standard like 802.11e and WME that will allow priority queuing and some minimalistic scheduling to take place. Other companies play tricks with the protocol to allow for voice clients to perform better under the BASIC standard but there are drawbacks.

In the end, it is too early to judge VOIP over WLAN because clients and access points have yet to adopt extensions to the basic standard.

-Ho

VOIP over DSL isn't much better

[EmbeddedJanitor]

A collegue of mine has VPN over DSL to a corporate network. They do all their phones via VOIP. If you send him a ~1MB email while he's on the phone, the call goes down the toilet. Not exactly a "new millenium experience".

Re:VOIP over DSL isn't much better

[mcrbids]

Your post is an excellent example of what's wrong with VOIP at current.

To use a telephone, you go to the local thrift store, spend $5 on a phone, and plug it into the wall to get reliable telephone service.

To use VOIP, you have networks, hubs, routers, wireless cards, firewalls, switches, and enough power bricks to saturate two 6-plug power strips so that you end up with something that must be tweaked to operate smoothly at all, in order to get something with the range of a cordless telephone. (Wifi uses the exact same frequencies as a cordless telephone - it's essentially a fancy cordless telephone modem)

Never mind what happens when any of those various boxes have a security vulnerability found...

In order to be truly successful, the technology needs to GET OUT OF THE WAY for the common user. Bridges are technology that are so reliable we never think about using one. So are telephones. VOIP will have "arrived" when it's use is automatic and reliable.

Re:VOIP over DSL isn't much better

[Yaztromo]

To use VOIP, you have networks, hubs, routers, wireless cards, firewalls, switches, and enough power bricks to saturate two 6-plug power strips so that you end up with something that must be tweaked to operate smoothly at all, in order to get something with the range of a cordless telephone. (Wifi uses the exact same frequencies as a cordless telephone - it's essentially a fancy cordless telephone modem)

You'd have to install a whole pile of even more expensive equipment if your phone company wasn't providing all of the necessary switches and other hardware leading up to your home.

I just moved into a new home less than two weeks ago. The house didn't have any telephone service coming into it, so I had to order it from the local telephone company (Bell Canada). Unfortunately, for new homes they'll only setup your service from the curb to an outside wall of your home. You have to do the rest (or pay them quite a bit to have them do it).

Between the cabling, the phone line, the basic 1x9 phone distribution panel, the wall box to contain it, and various bits and pieces of necessary hardware, it's cost nearly $200 CDN in parts. And none of these components has any processing capability -- it's all simple electronics.

Sure it's easy to think of POTS as being easier than VOIP, but that's typically because the telephone company or someone else has done all the work for you, giving you a wall jack as your interface. If the phone company did nothing but give you a cable with access to their network, you'd have to invest a pile of money into the necesary equipment to make that useful.

VOIP may be in the same boat one day. One of my previous employers went all VOIP for a new office of about 2000 employees, and setup a seperate network just for handling the telephone traffic. For those of us using the system, it was as easy as plugging the phone into the jack marked "phone" -- no different than with a POTS phone in any home (but mine. As it happens, the builder appears to have screwed up much of the phone cabling theey roughed in internally, as thus far only 1 in 5 is actually working :P).

Brad BARCLAY

That does it

[worst_name_ever]

VOIP over WLAN is DOA? WTF? I'm going back to DTMF over POTS ASAP!

Cisco 7920

[Acidangl]

At work i use my Cisco 7920 wireless VoIP phone. We are in the middle of a remodle of the IT department and it works great. I've had very few issues.

The Mike

[dj245]

A lot of some peoples troubles has to do with the Microphone. Most people use the bargain basement $8 model with no standard over how far away it is from their mouth, how loud they speak, etc. A good microphone headset is worth its weight in gold. I remember just a few weeks ago I went to a baptism/christening, and at the mass the microphone was so good you could actually hear the water flowing from one cup to another at the mass, and the little "crunch" when the priest broke the bread. After hearing the little slurp when he took a drink, I fully realized the value of a good mike and taking the time to set up the correct voice activation thresholds.

Me? I use Zalman headphones with a logitec webcam microphone duct taped on the right side. But I just use it for gaming. If I had to communicate anything other than "Our base is 0wn3d!" I would probably get something better.

VOIP over WLAN?

[Cytlid]

Is it just me, or am I the only one who ever thought this was completely stupid? Some company came to where I work and had a big presentation on VOIP over wireless... I thought it was ridiculous.

Let's see... you take your voice, turn it into packets to be reassembled a short distance away (espcially indoors). Ummm this has been around for millions of years... voices going "wirelessly" over the air... it's called "yelling".

Works fine at my house

[Gaewyn L Knight]

My main phone line comes over a 6.1 mile 802.11b link. I use Asterisk PBX with the IAX protocol to bridge the calls.

And my Grandstream SIP phone works great attached to a Linksys WET-11 client bridge.

And my Ipaq runs IAXComm just fine over it's wireless card to use as a netphone.

Does the battery life suck... yes... does it work and show promise... YES!

Just because people have problems with these cheap (as in quality)(usually SIP or H.323 based) piece of crud phones doesn't mean the technology and possibilities are not still there. SIP is VERY prone to problems from NAT (which many wireless networks use of course).

Anyways... for my 2 cents though I say... just give it time.

VOIP WLAN DOA? (YAFIYGI)

[joelparker]

IMHO the VOIP RFC for WLAN was OK, FWIW.
But IIRC, its FUBAR WRT the FCC and maybe IRS.
A FOAF says the MSFT will BSOD it b/c its w/o DRM.
IAC, the US DOJ, FBI, CIA & IRS also dislike it.
So FTTB VOIP WLAN is SOL. HTH... HAND! -JPH

Complete bollocks

[GuyFawkes]

I've used Skype ( http://www.skype.com/ ) quite extensively (windows only at the moment but they have a linux version in the works) over my LAN and via my cable connection to people ranging from 160 miles away to people in other countries.

Sure, there is a slight "houston, this is tranquility base) type of delay, but within a couple of hours use this becomes second nature.

Many of the calls I made exceeded one hour in duration, god alone knows what they would have cost via telephone.

Every call was end to end encrypted, yes, even the voice signal.

To call what is effectively a brand new technology which is basically still in public beta DOA is nothing other than complete and utter bollocks and a sure sign that whoever is applying such a label to VOIP is either...

a/ terminally fucking clueless
b/ blunkett (UK) / cheney or rice (US) / a telco shitting themselves.

BT has just started rollout of 21CN which will involve the ENTIRE NETWORK moving over to IP based traffic routing, so some 30,000,000 telephones in the UK alone will be, guess what, VOIP within a few years... link here http://www.theregister.co.uk/2004/06/09/bt_ip_netw ork/

Slashdot is rapidly declining to the point where Pissy World (UK) / Fry's (US) sales staff will start calling what THEY percieve as stupid clueless customers as "slashdotters" as a term of generic abuse.

"News for Nerds" ??? Give me a fucking break, Twaddle for Teletubbies is more accurate a decription of the content lately.

Test your connection out

[fiji]

You can simulate a VoIP call and get the MOS voice quality score. So if you want to see how your Wireless setup fares, visit testyourvoip.com.

Even if you don't care about VoIP, it is a useful test of the latency and bandwidth of your connection. VoIP is pretty sensitive to late packets so this tool highlights connectivity problems.

-ben

I just put a Cisco VoIP system in...

[eric2hill]

...one of our plants in Ohio. The install was a little rocky, and many of the features you'd find in any circuit-based system were simply non-existant or poorly implemented.

Now, that said, I put the system on its own POE switches and isolated network. Nearly 100 phones and the voice quality is superb. As a matter of fact, I had to introduce some comfort-noise because if nobody was talking, you couldn't tell you were even connected to anyone. It was really that clear. The POTS connection was done with a single PRI span, so calls were digital end-to-end.

I had to place two of the ephones on a remote end of a 10MB fiber link. They worked flawlessly. I then tried a single phone on a WIFI bridge, and it worked flawlessly.

Back to the article... The protocol the phones talk to each other using is g729. It uses roughly 9.6K worth of bandwidth, and sends packets every 20ms or so. A quality secured WiFi connection without any interference can support at least 25 to 30 phones before you start having channel speed or bandwidth problems.

In summary, a properly architected system has NO problems, whereas a system implemented over old, crappy hardware will have problem after problem.

The equipment matters

[Effugas]

First of all, full disclosure: I work at Avaya, for their security practice. What I'm about to say may seem pretty self-serving for the company, but I can only hope my posting history establishes me as some sort of credible source.

Second warning -- I'm actually raving about my own company's gear here. I'm way more likely to get in trouble over this, but heh -- can't let an entire nascent industry get tarred over a temporary generation of *ahem* lesser performing equipment.

So, warnings aside -- I was at Hivercon last year. Hivercon's a fun show, set in the middle of Dublin, Ireland (which, btw, is a fantastic city.) I'm sitting there, on:

1) My laptop
2) Wireless
2.1)HOTEL wireless
3) VPN (IPSec w/ 50% packet overhead!)
4) An international link
5) VoIP into a conference call

By all rights, the quality should have been awful. I mean, it had every right to be...

Now, we have VoIP at home too -- Vonage, to be specific. Our Vonage link runs over an 1.5mbit SDSL line provided by Speakeasy/COVAD, is QoS'd at our firewall, and connects directly to our home telephone wiring.

The quality on the international, wireless, IPSec'd, laptop'd conference call through my Avaya softphone exceeded what I was used to from our home VoIP provider. It was basically landline equivalent -- yes, it was even better than my cell phone.

I was _shocked_. I remember PowWow, FreeSpeak, and all those other systems that ran VoIP over Modem lines. In what alternate universe did VoIP become a quality leader under difficult network conditions?

Turns out that implementation matters. I went and harassed some of the people who worked on the phone equipment (heh Brian) and asked how this system could possibly be working at all. Apparently Avaya got a bunch of the people from Bell Labs (it came from Lucent, which came from AT&T, which itself came from Ma Bell), so there was all this knowledge lying around already in how to manage reliable communications like lives depended on it. The big things being used were:

1) Error Concealment
2) Dynamic Jitter Buffers

Error concealment is simple -- there's necessarily 50 packets per second on a 20ms-latency link (1000ms / 50 packets per second = 20ms of audio per packet), and speech is massively redundant. So rather than simply dropping out when packets were missing, the voice client was "filling in the gaps" with neighboring content. Since the overall frequency profile was kept relatively consistent, short term drops were kept outside the range of human perception. Neat -- obvious, and not entirely unique to this particularly implementation (there's direct support for concealment in some of the G.72x codecs), but neat.

The dynamic jitter buffers are cooler. The basic idea here is that some links are high quality and others are less so, and sometimes the quality of the link will change in the middle of a call. As a response, the Avaya architecture will negotiate a longer buffer for packets to be stored before they're output to the listener to be heard. This buffer starts at ~10ms and can scale up to ~300ms -- distracting, but users have been accustomed to higher latency through their love of cell phones (sad, but true). The key is that the human auditory system can't easily detect speed changes at subsecond resolutions, so you simply execute a non-pitch-shifting slowdown of output speech over a second or two and now you've got a jitter buffer far more tolerant of inclement network conditions. Mind you, this is an absolute nightmare for automated testing equipment, which expects time to be constant, but it's great for everything else -- even TTY's! You'd think a 150bps modem could travel over any link, but apparently not...

Anyway, we keep hearing about how Motorola and Avaya are putting out some kind of VoIP phone, so I'm actually pretty hopeful that we'll see a GOOD VoIP/WiFi solution sometime in my lifetime. I can say this much, though -- simply spurting ulaw on the wire and calling it VoIP ain't my idea of a good time.

--Dan

Just not ready yet

[emorphien]

Maybe one day, but right now the technology isn't there, and the need/reason/means isn't that strong in a lot of ways. People have cell phones, and for most those accomplish all that is necessary. There are already devices connecting over cellular networks that can accomplish everything needed for many people.

Down the road I bet the networks will mesh together, and the wifi, cellular and others will start to be one big network operating in small clusters to keep things running smoothly. We can't handle that kind of bandwidth and that many users now, but who's to say we won't be able to in 20 years?

I just doubt that the separate wifi and cellular (and other) standards will persist side by side for all that long as convergence quickens.

It works well.

[tarak.org]

Its all in the connectivity and the speed of the device getting the call packetized and placed on the wire. I use my softphone on my laptop with a 802.11b PCMCIA card at home through a Juniper SSL VPN with consistant success and quality. I have installed VoIP phones that connected to a gateway for call setup and tear down over a wireless bridge with good success. I have also experienced the mediocre quality of the DialPad's/Net2Phones etc.. If the device doing the codec compression can get the packet on the wire fast enough and there is enough throughput for the call. The call will succeed and be quality, WLAN's included. If the network is already taxed and the device doing the codec compression/and or call setup and teardown is taxed (in the case of some of the VoIP providers) the call will suffer.

VoIP over WLAN has just started. If you work for an intrenched Wireless Phone provider or a Baby Bell you wish that VoIP over WLAN was dead and this is probably just the beginning of the FUD from these guys and their pawns in an effort to hold on to their customers. So my answer is no.. its not dead.

Acronyms

[gad_zuki!]

> VOIP over WLAN is not DOA. If you tweak the QOS etc

IANAE, but perhaps change the TCP/IP settings so you dont waste so many SYN/ACKs then upgrade the WIFI to 802.11g perhaps with a VPN/IPSEC or maybe SSH over RF ASAP. FWIW the WAP shoud be IEEE standards compliant, the company who sells it should be ISO9000 compliant too. HTH.

To be fair

[Effugas]

Yes, all the carriers use VoIP on their backbones, about as controlled an environment as you'll find. The real question is VoIP to the endpoints, particularly over wireless. The answer is -- it's doable, but you need a much more aggressively correcting implementation than what's commonly deployed.

Cell nets aren't LANs, btw -- they're either MANs or WANs. There are real differences -- in protocol, in problems, in nature.

--Dan