Another year is upon us and the Internet of Things (IoT) has already begun to take center stage. Last month at CES 2015, companies gave a sneak peak at what we [as consumers] have to look forward to in the New Year.
Even though devices now have the ability to communicate with each other, they aren’t worth anything more to the consumer if they don’t solve a problem. Is it really important that my microwave can talk to my oven? Well, if what is in my microwave only takes three minutes to cook and the oven still has another fifteen minutes remaining, it might be worth it to have my oven notify my microwave when it also has only three minutes remaining.
But it seems like companies are more focused on connecting devices to the internet than they are with whether or not they should be connecting those devices and whether or not the consumers want and need it.
Moving Too Fast
Before companies go connecting every device in our lives, there are a few things that need to be considered beforehand. To begin with, both businesses and consumers alike need to begin accepting the paradigm shift that is inevitably upon us. With sensors being embedded in more and more of our everyday devices, we need to accept the fact that we will generate (unexpectedly and even unknowingly) massive amounts more data as we go through our day to day lives. This data will allow others to know more than they ever could know about us before, but the tradeoff could be [in my opinion] well worth it.
Second, with massive amounts more of data being generated in our daily lives, questions about security and privacy need to be addressed in the forefront. How do we keep hackers from remotely unlocking our doors or stealing sensitive health data from our fitness trackers? If all of our data is being pushed into “the cloud”, who has the final decision of who owns that data and who else has access to our data? Maybe you want to share your health information with your doctor but not your insurance company? The same could be true for your “smart-connected car” where you might want your driving behavior to be shared with your mechanic but not with your automotive insurance provider. Who’s to say what third parties can see that data?
Lastly, networks such as our home wireless access points and internet connections as well as the overall internet infrastructure need to be capable of supporting huge network traffic demands that IoT will bring. Many IoT companies believe that the wireless routers we have in our homes today are perfect for acting as gateways between IoT devices and the internet. Although that sounds good in theory, I [personally] along with many other industry specialists are not sure it is such a good idea. Most Wi-Fi routers claim to support up to 255 connected devices. However, connecting that many devices to a single Wi-Fi access point is not recommended as it will require more processing power than what most routers are capable of. Plus, many IoT futurists foresee typical households having more than 255 connected devices within the next few years. Current households already have between five and fifteen devices. This includes computers, tablets, cellphones, gaming consoles, etc… Once we begin having our appliances, furniture, doors, windows, utility outlets, faucets, toilets, etc… connected (not to mention devices we haven’t even conceived of connecting yet), this number will drastically increase.
Machine learning will also carry a much more important role in IoT than it has in the past. To account for things like network latency, many devices will need to have their own machine learning capabilities so that they can understand context faster and better, even in offline mode such as when their connection to “the cloud” is unavailable.
Do we really need all of the “things” in our lives to act autonomously? For example, I have a device located under my house that can detect water leaks and report back to me if one of my home’s pipes were to freeze and burst. If that were to happen, I wouldn’t want the water leak sensor to report back to the sensor’s manufacturer that I need a new pipe and they automatically ship one to me. With that said, I probably wouldn’t mind [see paradigm shift above] if the sensor was capable of locating the cheapest replacement part and shipping it to my house instead. If the former were to happen, the device manufacturer could stick me with whatever price they wanted, causing me to pay a premium when there are better prices for the same exact part. Even though I don’t (currently) want my water leak sensor to automatically spend my money for me, I do have it setup so that when a water leak is detected my house will automatically shut off the water flow to that particular part of the house via an inline solenoid valve.
At CES, president and CEO of Samsung Electronics (Boo-Keun Yoon) announced that by 2017, all Samsung products will be IoT-capable. Over the last few years, we have already started seeing internet-connected televisions, entertainment systems, and even appliances.
Before IoT can truly be successful, we need to begin developing and adopting open standards. Finishing out 2014, it appeared as though MQTT is gaining a lot of traction with manufacturers and software developers. Organizations such as the AllSeen Alliance have made great traction at formulating open standards as well. With that said, many companies are still convinced that their own technologies should be the standard and aren’t willing to consider alternatives, especially from their competition. As an example, many companies are still convinced that Z-wave is the way to go due to its openness while others are preaching for ZigBee due to its ability to scale much larger than Z-wave. There are also other debates continuing such as the argument between Bluetooth and WiFi, also Bluetooth and Z-wave. From what I gathered at CES, the number of Z-wave followers is slowing due to the increasing popularity of Bluetooth Low Energy (BTLE 4.0). Likewise, the number of ZigBee followers is also slowing in favor of the much more open and already widely adopted TCP/IP. This is due in part that the ZigBee Alliance board has repeatedly refused to make their license compatible with GPL even though Bluetooth had already changed their license to make it compatible with GPL.
Gateways and Connectivity
Devices called “hubs” are becoming more and more popular among home and office IoT automation. Hubs are intended to be language-translating-traffic-cops that sit between all of your devices and translate messages between those devices so that they can all speak the same language. Due to the low-power expectation of most devices, hubs play a huge role in allowing your devices to continue running on a tiny amount of power while still providing interconnectivity between different manufacturer devices, even while the power is out and their internet connection is unavailable.
While many companies are betting the bank on centrally-dedicated hubs, others are pushing toward the ever-so-common smart phone to be used as a hub. Even though the smart phone has the horse power to make a good hub, it has the limitation that your house or office cannot function efficiently once you leave the house – taking your smart phone with you.
Another, not so new, trend that showed its face at CES was the “connected car”. Not only do consumers expect their homes to be “connected” and “smarter”, but they also expect their cars to be the same. After all, we spend a great deal of time in our cars commuting to and from work. So, why not have our cars interconnected with our homes and the rest of our lives?
Many automobile manufacturers are already including devices within the cars that can notify their owners’ homes of their comings and goings. By knowing where the car is and the direction it is moving, the house (and office) can adjust itself accordingly. For example, when my car is on its way out of my neighborhood (preferably with me in it), my car will notify my house that I am leaving for a while and that the house should enable the security alarm, turn off all interior & exterior lights, turn off the thermostat, and lock the doors if not already done so. Not only does this provide more security for my house, it also saves on electricity usage.
Getting it Right
A semi-decent example of IoT done right at CES was a Wi-Fi connected ceiling fan that was controlled by a Nest thermostat. Instead of running up your electricity bill by turning on your main AC unit to keep cool, Nest can detect hot spots within specific rooms and turn on much less-expensive-to-run ceiling fans.
As sensors get smaller and cheaper, we can expect to see them embedded into objects never thought of before. For example, if you were to crack open your smartphone, you would find sensors that measure temperature, humidity, sound, proximity, and ambient light as well as an accelerometer, gyroscope, magnetometer, and a barometer. The costs of all of these sensors add up to around $5.00. With sensors that cheap, they can and will be embedded into more and more devices, all of which will be capable of gathering massive amounts of data. One thing is for sure, we won’t see what “big data” really is until IoT is in its full swing.
PayPal will open in a new tab.