IoE: Dead or Alive on Wireless
When we examine the average wireless client on a network today, we see a few options. There is the smartphone, the tablet, or the laptop. Even the Apple TV and Roku are often based off of similar technology and chipsets. All of these devices connect to a wide range of services, often consuming large amounts of bandwidth: we stream music while we browse; we video chat with friends, family, and coworkers across the globe; catch up on our favorite shows or sports teams. Occasionally, we do all of these things at the same time.
Our current wireless networks are built to handle this type of traffic. With 802.11AC, it is clear that we will be ready for anything that our standard client will encounter in the near future. But what happens when the standard client model is broken? The increasing shift to an Internet of Everything (IoE) forces us to face this question about the future of wireless clients.
There are lots IoE devices already on the market, and the next couple of years will see developments not that dissimilar from those during the “internet revolution” of the late 90’s. Finally, the average user will have the access and the ability to afford the smart homes we have been dreaming of for the past thirty-plus years. We already have devices on the market that can switch every light in our home on and off. Anything that is plugged into an outlet is also ours to command: our thermostats can be tweaked from the other side of the world, our plants can call out to us when they need water, and we can control our sprinkler systems with a couple of finger presses on our smart device.
Over the next few years, as this trend continues, major appliances will acquire this technology. Forget to turn the oven off or to start the dishwasher? Take care of that from the airport terminal. Going to be gone on vacation for a week? Your hot water heater is taking a vacation as well, and your power or gas bill thanks you. After you return from vacation, your health tracker can help you re-acclimate by turning on your coffee maker and adjusting the thermostat just before you awake. After all, who can sleep when your whole house smells like coffee? This is the IoE. This is the value, the business case, and “ahhh” factor.
But this isn’t an article about how cool IoE is. This is about wireless, and how we might kill IoE before it is old enough to know what coffee is. Going through the IoE devices that I listed above, you will see a trend. These devices are all basic users of the wireless network. You won’t be streaming music or video from your coffee maker, thermostat, or light switch. These devices sip data, with small transactions sent or received from a controlling server. These devices won’t utilize multiple data streams, will never saturate 150 mbps, nor suffer excessively from a few lost packets.
These devices will only suffer when they can’t access the network. This is rarely a problem inside the common living areas of a home, because coverage is already provided by a device that is built for highly demanding clients. But what happens when the device leaves the house, as with moisture sensors, irrigation control, outside lighting, a garage with an IoE connected car? The Internet of Everything truly includes everything, which means it leaves the traditional connected spaces in our homes, offices, and industrial spaces.
Let’s move our discussion out of the consumer space, and towards the commercial. During the 2013 Cisco Live Keynote, John Chambers and Jim Grubb discussed the connected farm. Their discussion outlines using irrigation control and moisture sensors to ensure that a corn field maintains optimum conditions. The benefit is two-fold: reduced water consumption and increased output, since the areas that need water get it, and the areas that don’t need any more water stay dry, preventing root rot, fungus, and mold.
Here again, let’s consider the wireless network required to support these features. According to the latest farm census, the average farm size in the US is 440 acres. That is 19 million square feet of required coverage. What is the cost of covering that area wirelessly? Remember to include the cost of power for each AP spread across the fields. The argument for IoE on the farm is very compelling, until the price tag starts at $500,000.
If we, as engineers and enthusiasts, are going to support the transition to IoE, we need a few things. First, we need the ability to cover larger areas. This is a power problem. The current 2.4Ghz FCC power maximum of 1 watt needs to be raised. While there are cases that can legally output more than 1 watt, our farm example isn’t one of them. Lifting the power maximum to 2 watts for outside areas would allow for higher dbi antennas to be used, thus extending the range which an AP could effectively cover, while also increasing the ability of an AP to “hear” weak signals. The farmer spends less on equipment and infrastructure and the ROI starts to make sense.
Next, manufactures need to design proper antenna systems for their WiFi supplicants. While size can be a limiting factor in some devices, antenna design must be considered as part of the development cycle. This was a lesson that Apple taught the world with the design (and later trouble) with the iPhone 4G.
Finally, as 2.4Ghz and 5Ghz are further consumed by channel bonding and higher density, we need more available frequencies. 60Ghz is available for higher speed networks with small cell sizes. For this use, we need spectrum that is lower than 2.4Ghz. Lower frequencies would better penetrate walls, and would provide better coverage for the same power output. We could then solve the issue of the homeowner who can’t understand why his backyard or workshop is off limits to IoE, and we could further reduce the number of AP’s the farmer needs to deploy for the same coverage.
The user experience with IoE will live or die on wireless networks. If it succeeds, it won’t be without some challenges in the wireless space. The 2008 spectrum auction proved how valuable spectrum can be. While it may be difficult for governments to pass up the billions for which that spectrum can be sold, there is a real need for more spectrum to be added to the public, unlicensed space.