Early on while developing this idea, I realized that while anybody could create a system that delivers bandwidth, the retail price of bandwidth was going to probably go down, making a Capex recovery more difficult. I have realized now that I looked at it the wrong way and that there would never be enough bandwidth, regardless of the cost. I also felt back then that there weren’t enough apps out there to drive the need for more bandwidth so that was the first part of the equation that has to be solved. This problem started fixing itself with the advent of smart phones, YouTube, video conferencing, VoIP, etc… and bandwidth needs started increasing very quickly. More recently I have to personally thank Google for their new search engine and Direct TV with the NFL package. Just keep em’ coming boys.
Small businesses need better options than T-1 circuits. They are slow and expensive. DSL circuits are slightly faster down and almost slower up. Of course, if they could keep them working or keep the performance at the purchased rate, that would be even better (can you tell I had another Saturday morning wasted calling Qwest tech support for a circuit bouncing like a superball?). Cable is also starting to expand again. However, in any area that only has one local loop provider, let’s just say you aren’t getting any coupons in the mail.
If you really want to see the opposite of capitalism in action in this area, just visit Mexico and try to start a wireless internet business. You have to hand your business plan over to Telmex, all the details and technologies, and hope they approve of you opening up a company to compete with them. Yea, like that’s going to happen. It costs $1200-$2500 for a simple T-1 circuit down there. Imagine Cricket having to ask Verizon if they can open up a competing cell phone company. Mexico will always be playing catch up technically to the rest of the world until they allow open competition for small businesses. For those who want to tell me that Telmex is providing low-cost DSL service, try to make a VoIP International call over the service and amazingly, the quality is really bad. Wonder if that has anything to do with Telmex charging exhorbitant rates for international calling. If any country needed wireless options to compete with Telmex, Mexico does.
It’s also evident that the chokehold held by the local loop provides wasn’t going to change without a huge investment in fiber or cable. I still believe that fiber is the best technical way to deliver massive bandwidth. It is by far the best long term solution based on what I see in the future for wireless, but it’s not the most cost-effective unless you happen to be digging up the street for some reason and laying new conduit. The power companies had the best chance to break that monopoly but the Bandwidth over Powerlines idea was simply a bad idea. They had a better solution on the table and still do, but I don’t see any of them playing that card. They could have easily delivered 30Mbps to the home for a lot less money than Broadband over Powerline, but but I haven’t seen anyone deploy that idea, nor do I expect them to.
Investment in fiber is difficult to make for residential because of the long term ROI. That’s why most companies have abandoned it. The population density in the U.S along with the lack of new home construction and infrastructure mean that almost the only group willing to throw money at FTTH is the federal government since hey, it’s not their money and they need more votes. Whether it’s the most efficient use of our tax money is irrelevant. In the meantime, wireless costs have come way down while capacity has increased. Of course, cable companies haven’t been sitting on their hands and their capacity has increased simultaneously. DSL, well let’s just say that the DSL companies are still evaluating expanding into the Realty and Moving and Storage businesses. I really like the new marketing plan in Phoenix calling DSL “Heavy Duty or HD Internet.” I almost crashed my car I when I saw that billboard I was laughing so hard. I was thinking HD stood for Howdy Doody Internet after several more crashes of my service again this week.
Being the wireless technical gurus that we all are, we just simply say throw up access points everywhere which is what municipal wireless and mesh systems tried to do. This would cover about half the calls I have gotten from people interested in these types of projects. One minute of explanation concerning getting access rights and insurance along with pesky little details like ROI, and the conversation ends. As I pointed out in the last article on making money as a wireless ISP, as cheap as wireless is to deploy compared to fiber, it’s just as difficult to recoup the investment because it’s difficult to deliver a triple-play solution which has the highest revenue. It’s also logistically difficult, expensive, and a very long term project to deploy hundreds of vertical assets around a city, assuming you can even get access to them. Nowadays interference, as pointed out in Chapter 13, is also a way of life.
If you have read all the articles in this Tales from the Towers series, the blueprint is already there for alternative municipal deployments or my favorite phrase, Guerilla Wireless. The devil is in the details and implementation. We needed an inexpensive source of Internet backhaul (for as little as $1 per Mb). We need a backhaul infrastructure that can support up to 2 Gbps or more. That’s off the shelf today with all the licensed and unlicensed equipment out there and throughput ranging up to 4 Gbps or more. Our last mile equipment on the client side is less than $100. We also know that we can build systems capable of delivering last mile up to 30Mbps or more to residential or business systems with access points costing as little as $100.
Part of the formula in calculating ROI in an area depends on whether you are using point to multipoin (PTMP), open WiFi, or a hybrid model to provide service. It’s also important to decide if you are going to compete on price alone or on service levels. Going head to head with comparative pricing has been the battle that wireless hasn’t been able to fight. We can fight that now. However, you are going to have to make sure that your system is rock solid. In addition, if you can get $40 dollars or more, regardless of the deliverable bandwidth, your numbers start to look better. With all the people who are dropping cable TV and going to Internet TV only, if you can deliver the bandwidth, you can compete and be profitable. Keep the TV thing in mind as you calculate other revenue streams also. Also keep in mind the billing structure for your local power company. These are hints to maximize your revenue per bandwidth.
Now you are probably thinking that I’ve left out an important piece such as vertical assets. Again, I point out that I gave the blueprint for how to install systems in many environments. What I haven’t covered is how to get local assets, get the support of the community behind you, and do it for little or no monthly cost. You aren’t getting around the Capex part but if you can get a lot of your vertical assets and contribute to the community on many levels, that’s a huge advantage. WiFi is a game of inches (thank you Vince Lombardi) that has to be played at that level. It also has to be played at the political level. However, put the right players in the game to handle these, coupled with a solid technical team and the numbers work. Your team has to be fast, ambitious, and willing to take risks.
Let’s add a few more tools into our toolbox first. The biggest tool is using a single building supporting a 4 square mile area for less than $3000 in equipment. We know that large brick buildings block 2.4GHz and above pretty well (assume from the back of the antenna, not the front). Instead of putting antennas on the tops of buildings, let’s look at putting the radios on the side of the buildings. I can see property managers cringing already because they don’t want their buildings looking like NASA. However, our equipment is fairly small. For example, a Ubiquiti Nanostation 2M is about 11 inches tall and 3 inches wide. This access point supports up to 100 Mbps+ of real world throughput with a 20MHz wide channel using 802.11N. It’s basically the size of a brick. Paint it red (assuming red brick), hang it 30-100+ feet in the air and nobody would even see it. Throw three of them on a wall and there is 300 Mbps or more shooting one direction. It’s easy to hit a laptop with these at 1000 feet line of sight (LOS). This fits pretty much every school out there and since most of the high schools already rent their light poles for cellular phones, it’s not much of a stretch to get the building.
Now throw in the 5 GHz versions of the same radio on the same wall using four 20MHz channels in 5.8GHz, and you have 700 MHz of bandwidth shooting in one direction. The brick wall eliminates interference from behind the behind or on the sides. Expand this out to 4 walls and 2.8 Gbps of bandwidth is now available all over the surrounding area of the building. Assuming the building is 60 feet tall and the houses around the area are 30 feet tall with few trees above 40 feet, outdoor 5.8GHz CPEs could easily connect back up to 2-5 miles away with less than $150 in equipment. So for $3000 in radio equipment, you are delivering a massive amount of bandwidth from a central location.
If a student could get direct access to school computer assets at a high bandwidth rates, think about what other services the schools could offer. How does the idea of “Education Everywhere” sound? This was one of my earlier ideas to bring better assets to students and more control to their online experience. Also think in terms of low cost access, breaking the digital divide, and political capital. This idea has many legs.
Don’t stop at schools though. Any brick building with a height of 30 feet or more is a potential vertical asset. In some cases, you can make a deal to provide Internet for a fee inside the building to tenants. In other cases, you might have to give something away for free but you still get the asset. Compared to hundreds or thousands of dollars that cellular or data companies pay for vertical assets, this isn’t a bad idea. You just have to scale the cost to the potential market. In this game, you don’t get the advantage of amortizing losing areas across the big picture.
I leave the details of physically running cabling (through the wall or conduit outside the wall) and bolting against the wall to each individual installer. For example, if there is metal flashing on the roof, it might be easier to use a strap over the roof and paint the radio the same color as the flashing. If you are in the middle of the wall, it might be easier to drill through and bolt in from the back while running the cable through the same hole. There are a hundred ways to make it aesthetically pleasing, but there is no denying the performance capability. If you are really ambitious, you can build a metal box around each one of the radios to reduce noise and tighten up the beam patterns. Of course the installation, switches, and trying to figure out where to get 3 Gbps of back end bandwidth is still an issue. I would estimate a full deployment like this around $10,000 in Arizona.
Take this down to a smaller scale also. One of the complaints for residential deployment is that there may not be vertical assets in every area. Wrong. There are vertical assets in every area. Tell me that there isn’t at least one person every square mile that would trade free Internet for roof rights to their house. My guess is that there is 10 times that. Because the coverage zone is short, ½ mile every direction, it’s extremely cheap (less than $1000 installed) and quick to install this type of system. If you sell to 10 people per square mile, your monthly vertical asset cost is 1/10th the monthly revenue or $50. Even if you don’t trust the people in the house to pay the electrical bill, it’s not that expensive to add a meter. Who needs lights when every house on the planet is a potential vertical asset?
On the house, think chimney with some type of metallic shield if you want the signal isolation. Four access points with 400 Mbps of bandwidth up there cost less than $300 for 360 degrees of short range coverage. If it’s a smaller area, even a single omni-directional antenna can deliver 40 Mbps. If there area is covered with trees, chimneys aren’t high enough, or the Home Owners’ association is run by Ghengis Kahn, then use your imagination. In areas where you aren’t sure that your AP site is going to be stable, use 1×1 802.11N radios with omni-directional antennas on the surrounding client areas so that if you have to move the AP, you don’t have to reposition a bunch of client radios.
Plan for your system to be dynamic and budget for that. Wireline companies generally don’t have that issue. However, the ability to be dynamic and spontaneous is also the advantage to WiFi. As long as you plan on that type of environment instead of being surprised, then this model become significantly more successful.
All of these ideas and techniques are to provide the tools to make WiFi competitive. However, these are still just the technical ideas without getting into the financial details. I wouldn’t propose them if I didn’t know they could be financially successful. If your focus is 100% technical, this doesn’t work. You will absolutely need someone with business and marketing experience to make this happen. I keep learning about new ideas that can generate more revenue daily if the infrastructure is in place, so don’t limit yourself to simply being a bandwidth providers. Associations and strategic partnerships are just as important to the creation of a growing system.
With the economy in a slump, there is always a place for someone to come in and do something cheaper, better, and more efficiently. Southwest isn’t the leader in the industry because they made the best martinis in First Class. They use their people more efficiently and deliver what the client wants. WiFi has the same option as demonstrated by Triad Wireless and other companies. Let’s get it moving people.