Real world measurements show muni Wi-Fi networks outperform WiMAX and cellular

Two important issues in large-scale municipal wireless have been: (1) can a given technology provide a usable data communications service and (2) how much does it cost to deploy such a service?

A useful network service provided at an affordable price is a necessary condition for a successful network offering. Many of the early muni Wi-Fi networks were hampered by poor service AND a higher cost to deploy than expected.

In seeking to answer these questions, Novarum structured its Wireless Broadband Review to provide  information on the performance of Wi-Fi, cellular and pre-WiMAX networks.

During 2007 and early 2008, we tested cellular, Wi-Fi and pre-WiMax networks in the following cities: Anaheim CA (2x), Brookline MA, Chico CA, Cupertino CA, Daytona FL, Eugene OR, Galt CA, Longmont CO, Madison WI, Minneapolis MN, Mountain View CA (2x), Palo Alto CA, Philadelphia PA (2x), Portland OR (2x), Raleigh NC, Rochelle IL, St. Cloud FL (2x), Santa Clara CA, Sunnyvale CA, and Tempe AZ (2x). In several cities we conducted the tests twice to detect changes in traffic and improvements in network service over time.

We discovered that, on the average, all of these networks have substantial performance and coverage. But the best Wi-Fi networks substantially outperformed the best of the cellular AND pre-WiMax networks.

Our test was an “apples-to-apples” comparison of performance (delay, uplink throughput, downlink throughput) and availability (percentage of tested locations with service within an advertised service area) for all of the major network technologies:

  1. ATT (Cingular), Sprint and Verizon cellular data networks;
  2. A number of metro WiFi networks using equipment by BelAir, SkyPilot,  Strix, Tropos; and
  3. Four of ClearWire’s pre-WiMax networks.

We tested outdoor coverage in an average of 20 locations per city, testing all networks with the same traffic load and in the same location and time. One of the important determinants of good performance is a good client modem, and indeed we tested a variety of these modems.

We took standard USB external modems for each of the cellular data networks, a higher power Wi-Fi modem (noting that current 802.11n modems appear to perform on par with these higher power clients), and a desktop directional CPE for the pre-WiMax ClearWire networks (no portable device was available at the time). We expected the WiMax modem (AC powered, directional antenna) to have the advantage in performance.

To our surprise, using similar client modems, averaged over good and bad networks, Wi-Fi networks delivered almost three times better performance than cellular networks and materially better performance than pre-WiMax networks – with similar levels of availability of service over the promised coverage area for all three network technologies.

If we look at the best and most recently deployed Wi-Fi networks, we see performance and availability superior to the best of the cellular data networks and pre-WiMAX networks – by a factor of 3!

The measured performance demonstrates that Wi-Fi networks materially outperform cellular data networks AND pre-WiMax networks – and do it over similar service area coverage, and likely lower deployment costs.

These network technologies are still evolving and we can expect improved performance and availability from cellular networks (with more extensive tower deployments and eventual migration to LTE from the CDMA and HSPA measured in these cities); from WiMax, as it moves from pre-WiMax OFDM systems measured here to standardized mobile WiMax; and from Wi-Fi as we begin to see the deployment of Wi-Fi metro networks based on IEEE 802.11n.


About the author

Ken Biba is a co-Founder and Chief Technical Officer of Novarum, an advisory firm specializing in wireless data networks. Ken has over 30 years experience in the network information systems industry combining a background of general management with a strong product and marketing focus in network systems and information security. Ken was an early engineer of the Internet in 1975. He has co-founded and managed four notable networking companies-Sytek, which was focused on cable TV-based local and metropolitan data networks, Agilis which delivered the first wireless handheld computers, Xircom, which pioneered local area networks for mobile computing, and Vivato, which was focused on scaling Wi-Fi infrastructure to cover campuses and metropolitan areas. Ken’s perspective as CEO, board member of public and private companies, and as a technologist brings unique insight to the business, market and technology of bringing useful wireless solutions to users. Ken has a Bachelor of Science in Physics (Magna Cum Laude, Tau Beta Pi) and a Master of Science in Computer Science from Case Western Reserve University.

– – – – – – – –

*Advertise now on Download our 2009 Media Kit and contact us.

*Join the MuniWireless / Muni WiMAX Linked In Group for networking and discussions.

*Sign up for the MuniWireless / WiMAX newsletter.

*Register now for the Muniwireless – Freedom to Connect Event, March 30-31, 2009 Washington DC


  1. Do you have similar figures for satellite service?

  2. I do … and it varies dramatically based on WHICH satellite service.

    But a typical residential satellite broadband will average about 700 msec delay, 120 kbps uplink and about 700 kbps downlink.

  3. Arnon Kohavi says

    Not a surprise. The cellular & WiMax industry has always focused in their marketing efforts on the downlink speed, which in some cases can come close to WiFi. As users increasingly send photos and videos from their mobile devices, the importance of uplink speed increases, which is where WiMax/4G will always be at a disadvantage. One can increase the power at the base station, but it is difficult to transmit a high speed return link from a small mobile device to a distant tower: Much easier to do so to a small box on the wall a few yards away. So WiFi, and in the future, Femtocell, is the way to go.

  4. That’s correct. Even cable and DSL providers focus on downlink speed. Are these people living in another century? Haven’t they used digital cameras and video cameras? People upload a lot of video files (see YouTube and similar sites). People do podcasts and upload their audio files. People post photos on sites like Flickr. I do.

    So talking only about downlink speeds is deliberately misleading. They don’t want to tell people about their paltry upload speeds because they don’t want people to know that they are living in the past and their service is SUB-PAR for the new world we live in.

    That is why in my articles, I do try to post UPLOAD speeds as well as download speeds.

  5. And not only uplink and downlink speeds .. the DELAY and delay variance are critical factors in whether some services – like VoIP and IPTV will work well.

    The surprise was actually not just on the uplink, but also on the downlink – where we still saw substantial performance advantages for Wi-Fi.

    This is in addition to the cost advantages. A large scale mobile WiMax deployment is probably going to require over $500k/square mile in infrastructure equipment – likely more. The best WiFi networks we measured were more on the order of $250k – and delivered substantially more performance.

    I suspect that new generation metro WiFi equipment based on 802.11n will substantially increase both the performance of WiFi and decrease the cost.

    The importance of the uplink has always been minimized by carriers. My home DSL and cable are capped at under 400 kbps as will Clear’s mobile WiMax speed.

    This uplink capacity limits means looooooong times for the uploading of photos or videos … and makes it effectively impossible to back up my home storage of family photos and videos on “cloud” storage.

    True “cloud” computing is very difficult until there is a more effective balance between uplink and downlink performance.

  6. Way to go Phil anf Ken on publishing a study that proves what most wireless experts and myself have been saying for years. WiFi and WiMax are complimentary services. Not competing ones!! Each has its own place in the market.
    WiFi is cheaper to deploy and has capacity of user advantages over its more rural based cousin. A lot of critics argue this is not the case but look at very large and populated cities and you will soon see that if wireless broadband grew quickly in popularity that the number of users per square mile could rapidly tax and oversubscribe the WiMax base station. More stations could be added but cost becomes a serious concern. Topography in the city is your design enemy due to buildings and other transient occlusions. WiMax handles these better but cannot completely overcome Line of Site issues and with the increased cell size leaves serious coverage holes and areas of less than peak performance. WiFi with its low cost access points, smaller cell size, and density of users per access point becomes the economical and technical performer in this environment. The coverage when well designed actually is better from an RF perspective. The smaller cells can be tailored to provide good coverage and throughputs to a larger percentage of the city.
    Now, let’s go rural, and the roles are reversed. In rural areas the cost of deploying 25-30 WiFi access points per square mile quickly becomes cost prohibitive when dealing with the much smaller density of users per square mile. In steps WiMax to save the day with extended range and sectorization. This drastically reduces the cost per square mile to deploy in areas with fewer users and thus no need for huge capacities of users per cell.
    When speed is the concern it comes down to the vendor and the design of the network. Well designed WiFi networks definitely outperform WiMax. Why analysts ask? It’s simple. WiFi is a much more mature technology. Huge players like Cisco, Aruba, Tropos, and Meraki have been competing with each other for years to build better, faster, and more resilient networks in the outdoor space. The results are better routing protocols, better radios, and better Radio Resource Management software suites.
    The long and short of the problem of broadband deployment in the US is simple. Money! Under the Obama Administration there should be available dollars to push the service out to underserved markets and strong ROI models like those I’ve published in the past need to be put into place in the larger metropolitan areas to offset the cost of deployments. I have heard from several sources in the popular media that municipal wireless broadband is dead. Nothing could be further from the truth. This is simply technology evolution at its best.

  7. The problem with WiFi is that everyone considers it a laptop technology. If it was deployed in both PTMP and municipal designs simultaneously, then it’s far more efficient and reliable. On the municipal side, our company has designs ranging from $3K to $9K per square mile. On the PTMP side, we have devloped techniques reducing the client side install to less than $200 per truckroll with equipment. There are better ideas yet to be deployed with WiFi.