Smartphones and other condense positioning devices use satellite systems like GNSS to provide positioning. Nevertheless, to be most efficient, GNSS needs a clear line of sight to increase satellites, and even at its peak performance, it still only provides meter-level accuracy, which is simply not accurate enough for industrial and indoor IoT applications. On the other hand, 5G millimeter wave (mmWave), claims senior director of engineering at Qualcomm, Kiran Mukkavilli can, in some cases, achieve centimeter-level precision.
“Cellular compliments GNSS really well, and if you look at cellular deployments […] they have all the ingredients to get the positioning to work,” Mukkavilli said. “In looking at the kind of deployments that we have. We have the macro deployments in FR1, sub-7 GHz and we have our FR2 millimeter wave [mmWave] deployments, and then we also have some of our indoor deployments.”
Each one of these deployments, he detailed, have special characteristics that help in positioning. Massive MIMO deployments in the sub-7 GHz band, for instance, have large antenna arrays, which allow for precise angle domain estimation for wide-area positioning. On the other hand, indoor deployments are commonly quite dense to allows high capacity communications. Denser mmWave deployment means more bandwidth, and more bandwidth means better positioning precision. mmWave, though, has both the benefit of density and lots of spectrum, according to Mukkavilli.
“From a positioning point of view, a lot of bandwidth means very fine and precise time resolution. Small cell deployments also typically means we get denser nodes […] which gives us very fine beams. So putting [together] the bandwidth and the fine beams that we get out of these dense arrays […] we can easily get to the centimeter-kind of accuracy with mmWave,” he said, adding that as a result, mmWave is going to be “a very important component” to Qualcomm’s overall positioning solutions portfolio.
Nevertheless, because precise time determination demands a lot of bandwidth for accuracy, applications involving low-power and low-complexity devices like IoT devices, which use smaller bands such as 20 MHz or even 5 MHz, present a positioning challenge. Put simply, time-based precise positioning techniques are not well suited for narrowband IoT devices.
“With small bandwidth, the time resolution is not as what we get with 100 MHz or 400 MHz and that’s where we can make use of […] spatial resolution,” Mukkavilli detailed, offering an angle-based positioning technique called DL-AoD, or downlink angle-of-departure positioning, as an option.
DL-AoD is a device-based positioning technique in which an IoT device calculates each beam RSRP (Reference Signal Received Power) on a network, using the relative differences between them to evaluate the angle of departure (AoD). Then, these scaled differences provide horizontal angle and promotion information via pattern matching.
In this way, DL-AoD delivers high-accurate positioning for 5G IoT devices with restricted bandwidth. In fact, according to Qualcomm’s demo, using an angle-based technique attained at least a 60 per cent development over the median precision accuracy in the 5 MHz bandwidth.
“Put all of this together, what cellular can do for positioning is complement and deliver the less than a meter kind of accuracy most places and even centimeter level positioning […] for industrial IoT and indoor precise positioning,” summarized Mukkavilli.
Watch an OTA presentation of 5G mmWave positioning using a 400 megahertz channel in the 28 GHz band and narrowband positioning with DL-AoD.