One of the technologies that is going to be needed to make the Internet of Things work better is MIMO. MIMO stands for multiple-input, multiple output and refers to using an array of antennas to communicate instead of a single antenna. MIMO technology can apply to different kinds of wireless including WiFi and cellular.
MIMO has been around for a few years and the latest high performance WiFi routers include the first generation MIMO technology. These wireless routers include multiple antennas that work together and the purpose for the antennas is to establish separate wireless routes to different devices.
When done smartly, MIMO dynamically sets up a different wireless path to a given device, so there would be a separate wireless path to your cell phone, your TV and your speaker system. The current MIMO routers can only establish a few separate paths at a time. So if you have more than a few wireless devices running at the same time (which many of us now do), then there is also a general broadcast signal that can be picked up by any device within range.
As you can imagine, establishing separate paths and doing it well can be a challenge. Some devices like cell phones and tablets are mobile within the environment and the router has to keep track of where each device is at. Done well the router will determine the right amount of power and bandwidth to give to each device.
But fast forward a few years when you also have a host of IoT devices in your home. Today in my house we often are running seven WiFi devices, but add to this an array of smart appliances, smoke detectors, security cameras, medical monitors and various toys and it’s easy to see that the normal home router could get overwhelmed in a hurry.
Scientists are already working on more sophisticated MIMO devices so that they can understand the challenges of handling large numbers of multiple devices simultaneously. Scientists at Rice University have constructed an array of 96 MIMO antennas that is letting them a look into our future. They have named their array Argos and it is giving them a tool for exploring the ways to process and integrate inputs and outputs from many sources. They are calling their application Mammoth MIMO.
Mammoth MIMO antenna arrays are more efficient than a bunch of single antennas. The large array that Rice is studying can do a whole lot more than connect to 96 devices and they are claiming that the multiplicative efficiency appears to make the large array as much as ten times more efficient than using a host of individual routers.
That kind of efficiency is going to be necessary in the future in two circumstances. First, this technology could be used immediately in crowded environments. We are all aware of how hard it is to get a cell phone signal when there are a lot of people together in a convention center or stadium. Mammoth MIMO could enable many more connections.
But the more widespread use will be in a world where the normal home or business is filled with scores of IoT devices all wanting to make connections to the network. Without improved MIMO this is not going to be possible.
Massive MIMO is going to require massive processing power to make sense of the huge inflow of simultaneous signals. That will require more computational and data storage locally just to process and make sense of IoT data. I have several friends who work in the field of artificial intelligence and they think their technology is going to be needed to help make sense of the massive data flood that will flow out of IoT.
Actually, this is a great example of an astro-physical technology development and its other uses…
The multiple antenna array is, I believe, first used by space satellites that fly beyond our solar system. Antennae reception on Earth is received by an array of antennae, which is then combined by the computer into a single, clearer signal…
MIMO is traditional antenna arrays on steroids. Satellites are very low power and so they use multiple antenna arrays in order to be able to send more data back down to earth on each orbit. But MIMO not only sends multiple signals, but it dynamically figures out where to send them and how strong the signal should all at the same time. And it can do this while also dynamically receiving data from inputs. From an engineering perspective it’s a challenge to always get it right.