Getting beyond the WiMAX hype

Getting beyond the WiMAX hype

Summary: WiMAX (World wide Interoperability for Microwave Access) is a standardization process for the IEEE 802.16 Ethernet standards, much like Wi-Fi is the standardization process for the IEEE 802.

SHARE:
TOPICS: Networking
31

WiMAX (World wide Interoperability for Microwave Access) is a standardization process for the IEEE 802.16 Ethernet standards, much like Wi-Fi is the standardization process for the IEEE 802.11 standards.  WiMAX has been one of the most prolific buzz words of wireless networking for nearly two years and it is probably one of the more hyped and misunderstood technologies in recent memory.

Some of the misunderstandings of WiMAX are:

  • WiMAX is longer range than Wi-Fi
  • WiMAX is faster than Wi-Fi
  • WiMAX will compete with and replace Wi-Fi
  • WiMAX is more secure than Wi-Fi

 

WiMAX is longer range than Wi-Fi:
WiMAX is designed to operate in both the licensed radio band and unlicensed radio band.  When WiMAX operates in the licensed radio band, it is free to broadcast with many times more channels and power than any unlicensed radio solution so long as the broadcaster owns the license for that radio band.  This means that licensed WiMAX operation will typically be used by Wireless ISP operators.  Wi-Fi is designed to only work in the unlicensed radio frequencies in the limited 2.4 GHz and 5 GHz spectrum and is restricted by the FCC to very low broadcast power levels typically in the 1 mW to 100 mW range.  This is about one million times less power output than a typical licensed band WiMAX implementation putting out 100 kilowatts.  Is it any wonder that a WiMAX access point broadcasting at 1 million times the power levels of a Wi-Fi access point would have longer range?

While it's true that WiMAX can obviously have longer range in licensed operation, it doesn't mean that you get to put up your own kilowatt WiMAX access point willy-nilly even if you could afford the equipment.  You must own the licensed radio spectrum much like a music radio station must own their FCC license to broadcast their FM signals over the airwaves.  Put WiMAX in the unlicensed radio band like Wi-Fi and you're instantly restricted to the same power levels and scarce channels as Wi-Fi and the range advantage vanishes.  WiMAX like Wi-Fi is a wireless technology based on radio frequencies and must abide by the same physical engineering constraints.  Any technological advantage in WiMAX can easily be adopted in Wi-Fi and vice versa.  While WiMAX might for example take advantage of multi-path signals whereas original Wi-Fi implementations didn't, modern pre-N MIMO based Wi-Fi products make exquisite use of multi-path signaling.

 

WiMAX is faster than Wi-Fi:
WiMAX proponents often boast of its speed advantages.  While WiMAX claims to offer 70 mbps theoretical maximum bandwidth, modern Wi-Fi MIMO implementations can boast a theoretical maximum throughput of 108 mbps and sustain a real world throughput of 45 mbps.  While I can go to the local computer store and buy a production Wi-Fi MIMO access point and verify the 45 mbps sustained throughput myself, production WiMAX access points don't exist yet.  Again, the same engineering constraints and laws of physics apply and it's simply too early to talk about WiMAX throughput.

Wireless ISP implementations of WiMAX will have the same contention problems that any current wireless ISP has because the massive range of a licensed band WiMAX implementation measured in kilometers is a double-edged sword.  The fact that the coverage area is so large means that there are that many more users contending for the same spectrum and bandwidth.  Even though the wireless ISP will broadcast on multiple independent channels, there will still be many times more users on a single WiMAX channel than any Wi-Fi channel.  Every wireless ISP technology that I have ever dealt with from wireless microwave ISP access to CDPD to EDGE to Satellite all have the same over-contention and QoS (Quality of Service) issues.  Severe latency and round-trip ping times hovering erratically between 200 to 2000 milliseconds is completely unacceptable for VoIP, Video Conferencing, gaming, or any other real-time application.  While there is talk of WiMAX having QoS for VoIP, actual implementations and their effectiveness remains to be seen.  Also keep in mind that there are proven implementations of VoIP QoS for Wi-Fi in the form of Spectralink and the upcoming 802.11e wireless QoS standard.  Wireless ISP WiMAX implementations will always be slower than any unlicensed WiMAX or Wi-Fi access point because there is no way the ISP can afford to dedicate an entire channel to a few individual subscribers.

Whether or not private unlicensed WiMAX implementations will be faster than private Wi-Fi implementations will depend on specific product implementations.  While they would theoretically be the same in an apples to apples comparison with equal amounts of broadcast power and radio frequency, my money is on Wi-Fi because of the abundance and maturity of modern Wi-Fi products.  Wi-Fi has such a head start in the unlicensed market that it's difficult to imagine that WiMAX would be successful here.  This is why the vast majority of effort for WiMAX is headed for the wireless ISP market.

 

WiMAX will compete with and replace Wi-Fi:
The craziest of the WiMAX proponents even argue that WiMAX will entirely replace Wi-Fi and make it obsolete.  Such beliefs are over optimistic at best and irrational at worst.  WiMAX will have its place consolidating and standardizing the wireless microwave ISP market, but it will not directly compete against most Wi-Fi implementations.  WiMAX will primarily be focused on the licensed wireless ISP market while Wi-Fi will continue to dominate the private unlicensed wireless market such as the corporate or home wireless LAN.  The only place that WiMAX will overlap and compete with Wi-Fi is in the case of commercial Wi-Fi hotspots.  WiMAX clearly has the advantage of universal coverage because of its massive continuous coverage area measured in kilometers while Wi-Fi hotspots are made up of a patch work of fragmented coverage spots measured in meters.  But it is still possible that Wi-Fi hotspots will continue to thrive because of market share momentum and the fact that it can be faster with lower latency because it is short range with few subscribers to contend with.  How well WiMAX ISPs implement QoS and fight the urge to over subscribe will be the determining factor in the overtaking of Wi-Fi hotspots.  If it's anything like the over-subscribed and high-ping wireless ISP implementations of today, it will not entirely eliminate the need for Wi-Fi hotspots.

 

WiMAX is more secure than Wi-Fi:
One of the myths propagated about WiMAX is that it is more secure than Wi-Fi.  WiMAX will actually use a similar authentication and encryption scheme to the Wi-Fi WPA2 standard.  The minor difference is that WiMAX security will use either 3DES or AES encryption and a flavor of EAP called PKM-EAP.  Wi-Fi WPA2 on the other hand will typically use PEAP authentication along with AES encryption.  No one should make blanket statements that one is more secure than the other because good security usually depends on proper implementation of the technology if it's implemented at all.

 

It is important for the public to know what WiMAX is and isn't.  WiMAX and the 802.16 standards is important because it is the next logical evolutionary step for the wireless ISP industry, but it is not the be-all end-all for anything wireless.  WiMAX or any other wireless technology will always complement but never entirely replace the need for wired technology.  Wireless solutions will always be more mobile and convenient while wired solutions will always be faster and more reliable if it is possible to lay the cabling.  We need to move beyond the all the WiMAX hype and appreciate the technology for what it is.

Topic: Networking

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

31 comments
Log in or register to join the discussion
  • One sided commentary...

    George, if someone read just your commentary they would be left with the idea there is no reason to develop WiMAX. That just isn't true.
    No_Ax_to_Grind
    • Did you read it carefully?

      "WiMAX will have its place consolidating and standardizing the wireless microwave ISP market, but it will not directly compete against most Wi-Fi implementations."

      Is that reason enough?
      george_ou
      • I could see it competing with Wi-Fi

        Though it's not some great leap there are benifits to going with Wi-MAX on unlicensed wave lenghts. It is slightly better that current technology.
        voska
        • Any improvements in one will be adopted by the other

          I have a lot of startups working on smart antenna technology bombarding me weekly. All of the new technologies for wireless networking apply to WiMAX or Wi-Fi. Radio is radio, it's all in the implementation. The momentum is clearly on the side of Wi-Fi because of its head start and market size.

          Read my MIMO blogs (http://blogs.zdnet.com/Ou/?p=74) and see if you're so sure that unlicensed WiMAX solutions will be necessarily better than Wi-Fi.
          george_ou
        • WiMAX FAQ

          http://wimax.com/education/faq/faq06

          See what WiMAX.com says about WiMAX competing against Wi-Fi.
          george_ou
    • I disagree

      He puts WiMAX in a realistic light.

      Basically it opens the door for ISP using licensed bandwidth. It could replace Wi-Fi by being just a little bit better on the unlicensed bandwidths. I'd say that's a good reason to develope WiMax. Especially if you are an ISP using current wireless technologies.
      voska
  • George, are you dismissing theWiMAX grid?

    Where commuication hops from one WiMAX system to the next as needed?

    True it isn't here yet but isn't that the ultimate goal?
    No_Ax_to_Grind
    • I believe you're talking about WiMAX mesh/backhauls

      I believe you're talking about the concept of WiMAX mesh/backhaul networks. Like I've always said, wired backhauls are always the prefered medium because of raw throughput and reliability, but sometimes that is not possible or feasible.

      What will most likely happen is that WiMAX backhauls will be used where fiber backhauls are too expensive to run. At some point you have to connect to the highspeed wired world. My fundamental premise is that Wireless and Wired communications complement each other, and I stand by that. It's not that one will obsolete the other because both will thrive.
      george_ou
  • why wi-fi or wi-max

    the fcc is supposed to be giving bandwidth in the tv spectrum the opportunity to become wireless access. I believe that cisco has expressed interest in this frequency to create wireless broadband. These frequencies are particularly inviting due to the penetrating ability (it goes through buildings, think of a tv's reception) and the ability to be used as a carrier frequency for sideband broadcast, with the ability for 2 sidebands (one up and one down) at a much higher data rate than the carrier frequency would support, certainly in the megahertz range and perhaps higher allowing data transfer in the Mbit to Gbit range. Broadcast ranges are typically 50 miles or more. I have seen some things about narrow bandwidth implementations, which would allow for many more channels than present "tv" broadcasting permits.
    In any technology endeavor it is not whether or not your technology will be replaced, it will, it is a matter of how long it will last, and how much money you can make before the party's over.
    The implementation I described is about 10 years out, although I'm sure once the switch to digital tv occurs in 2008? it will gain a lot of steam.
    pesky_z
    • It will be auctioned off I think

      Again, even if this TV spectrum is used for broadband internet access, it will not compete or eliminate the need for unlicensed wireless LANs. It's sort of a knee jerk reaction to assume that one technology will automatically negate another, especially when they serve different needs.
      george_ou
      • I would agree with you on this.

        Different wireless spec's serve different needs (Bluetooth is PAN (Personal, 5 meters), Wi-Fi is LAN (Local, 100 to 500 feet) and WiMAX is MAN (Metropolitan, about 30 to 50 km)). As more goes digital and services converge (FTTD: Fiber to the door, WiMAX or WiMAN (802.20), WCDMA/UMTS, etc.), broadband Internet will assume more services and possibly freeing up spectrum later. There will always be some overlap, though.
        B.O.F.H.
  • I suppose

    you could put dark glasses on and stand in the middle of the road and say "Gee, where's this so-called steamroller?".

    Basically your whole article comes down to power. You say that typical WiFi access points put out 1-100mW of power, and then say the WiMAX access "towers" put out 100kW. That's quite a range! And yes, I agree that it all comes down to power, but there's a lot of "wiggle room" in the middle of this range.

    Citizen's band radios (legally) allow up to 4 or 5 watts of power - with up to 12 watts on SSB channels. If WiMAX routers could (legally) use those power levels, then the long range aspect of WiMAX would be realized for EVERYONE.

    By utilizing MIMO-type multiple channel broadcasting, reliability and speed would reach the stated WiMAX levels (70mbps).

    [Some of the misunderstandings of WiMAX are:]

    * WiMAX is longer range than Wi-Fi
    It is DESIGNED with longer range in mind. If WiFi and WiMAX have the same power levels and frequency - I think its impossible for one to have a longer range than another (that Physics thing . . .)
    * WiMAX is faster than Wi-Fi
    WiFi has had a few generations to get faster and faster. WiMAX hasn't even been released. What speed will WiMAX2 or WiMAX3 run at?
    * WiMAX will compete with and replace Wi-Fi
    Lets see, ONE WiMAX spec can run at whatever frequency you want. 802.11a runs at one freq, and 11b, and g run at another. 802.11n will run at 11a frequencies. You know what I see here? If you want to change frequencies with WiFi, you have to REPLACE all of your gear. WiMAX will be adjustable . . .
    * WiMAX is more secure than Wi-Fi
    I doubt that you could ever make wireless secure, but WiMAX will certainly be equipped with the latest and greatest (WPA2 and AES).

    BOTH WiFi and WiMAX are comparable in all ways, but what makes WiMAX so exciting is the steamroller! The WiMESH, self-configuring networking capabilities have limitless possibilities. Some day, your house will have one cable comming in - electricity. Your telephone, television, radio, etc. will all be delivered via wireless IP. You COULD have WiMAX ISP connectivity, WiFi "local" house networking, and Bluetooth device connectivity all in that house - but WHY? Each of these connections could be WiMAX - by varying the frequency and power of those connections. Not only that, these connections will all "self-configure" on the network - bringing me to the piece-de-resistance. Your house connects to my house which connects to No_Ax's house, connects to Berlind's house ad-infinitum. This is the WiMESH, and it will replace ISPs and the internet. THIS is the true POWER of the WiMAX steamroller. FREE INTERNET (mesh) FOR ALL!

    A pipe dream? Well maybe. The WiMAX steamroller might become the WiMEDIA steamroller - you never know who wins in the ultimate marketplace, but the WiMESH will most certainly come into existence - and people will flock to it FASTER than the VoIP steamroller going on now.
    Roger Ramjet
    • The answer to IT job security

      This sounds like just the thing to solve the issue of job security in the IT field.

      Every family will need their own Network Security Administrator to keep up with the security patches lest their "smart kitchen" get a virus, and the breadbox and the toaster conspire to order 500 loaves of bread from the e-grocery and force the hapless homeowner to wade through two feet of toast one morning.
      RichN_z
    • sharing

      I would hate to be the homeowner on the high spot in the neighborhood with a line of sight to the next node in the mesh. Everytime I blow a fuse I can see the mob of torch weilding citzenry come to "straighten things out."
      Sorry to be a bad neighbor, but no mesh for me thanks. It seems like a nice implementation in buildings where it is difficult/impossible to install the necessary wiring, although I have some reservations as to whether or not it would work well in high rise buildings with a lot of inherent background noise (static) and large masses of concrete and steel to deflect the signal. That is exactly the reason safety responders are first on the list for bandwidth in the tv bands. Microwaves are nice in the country, with wide open spaces, but not so good in city environments, where the electrical feed itself can contain enough noise to swamp your power supply. I have already experienced so much interferance that my wireless keyboard wouldn't work. (bad grounding and a cb transmitter with a linear across the street)
      pesky_z
  • Wimax : the user side

    Hi George,

    If you said that Wimax will be able to reach kilometres due to the power output that reach 100 kW, than I'm wondering how can the user send back their signal if their equipment is not provided with the same power output ?
    abimanyu
    • Good question

      I'm not a hardware Engineer and I could be wrong. Obviously, a PC or notebook could never come close to outputting 1 kW let alone 100 kW. From what I understand and have observed, the clients don't need to transmit that much power. A large smart antenna can pick up a faint signal from a long distance away (think Pringles can). But still, it's been difficult to package WiMAX gear in to a PCMCIA or miniPCI format.

      I hope an engineer could answer this question more precisely for you. But unlicensed WiMAX towers do operate at 1 kW to 100 kW.
      george_ou
      • Correction

        "But unlicensed WiMAX towers do operate at 1 kW to 100 kW."

        I'm sorry. I meant to say licensed WiMAX base stations are designed to operate at 1 kW to 100 kW.
        george_ou
    • It is a good question...

      This has a lot to do with antenna location. You'll notice that cellular towers are generally very high (hundreds of feet in the air) and positioned to allow maximum line of sight to the widest area possible. The fact is, with good line of sight, you can transmit a very long way with very low power. For instance, 1 W will reach the moon. A cellphone transmits at 300 mW (.3W) and as long as it has reasonable line of sight to the tower, a connection can be established. Recently we set up a Cisco wireless system which runs at WiFI speeds (100 mW) and yet with good line of sight can be setup to have a maximum range of 42 km. The difference with WiMax will be, better connection quality and speed capability, because with licensed frequencies, you can go above the 100mW ceiling on (unmodified, heh heh) WiFi gear, in both the case of the transmitter base station (ISP tower) and the client gear. 1W, 2W would do it - but also be aware of the health dangers involved in transmitting microwave frequencies at that power. You run the risk of turning your house into a microwave oven (Cancer anyone?) This is why the TV-band range frequencies, soon to be auctioned off, will be in such high demand - lower frequencies propagate better at lower power and do not have the health risks associated with microwave frequencies (which, as they pass through your body, can knock electrons off of associated molecular structures, causing those structures to change into other substances. This causes mutations in DNA, which are chains of amino acids, a molecular structure. Once the DNA has become corrupt, it can turn cancinogenerate, that is, cancer causing.)
      Kamikaze_Ohka
      • Be careful of practicing science without a degree

        There are no scientific studies that prove any cause-and-effect between electromagnetic radiation and cancer. The only thing studies have ever shown is that it will heat up tissue by just a little. Be careful about spreading the cancer FUD.
        george_ou
        • Tell that to the microwave techs...

          In Canada who worked on the country-spanning MW telecomm network built in the 60's. They all died of cancer before they were 50. Dude, be careful of not parroting the corporate "microwaves are SAFE, SAFE SAFE!" line put out there by execs who could care less. The people who designed the microwave oven discovered that mw's heat water at the right harmonic frequency (1520 MHz or so) because they were setting their coffees on the tops of the radios they were working on (totally unshielded) and they were getting hotter not cooling off. And don't tell me what degrees I do and do not have, George.
          Kamikaze_Ohka