At the beginning of any new standardization effort, one of the first and most critical steps is to measure the wireless channel at possible deployment frequencies. Researchers use these measurements to understand how wireless channels behave in a target environment. For example, channel measurements can not only show how signals propagate over free space, but also how the signals reflect-off of or are blocked by objects such as trees, buildings, cars and even people. Once the data has been captured, researchers create models used for simulations. The simulations estimate network performance in various scenarios and by varying system parameters researchers gain a better understanding of the challenges and are better able to explore design tradeoffs.
As the 5G standardization effort moves forward, mmWave channel measurement campaigns continue to be a focus for the entire industry. The challenge with mmWave spectrum is that it is not well understood and the deployment models may in fact be very different from traditional cellular networks below 6 GHz. Because of the directional nature of mmWave waveforms some models produced very narrow interpretations of the channel leading to a positive benefit in terms of MIMO (perhaps easily achieving rank 2 by exploiting polarization) but less optimistic in terms utilizing sidelobes and backscatter when the beam is occluded. We now know that the mmWave channels change very rapidly, and it is critically important to understand how the beams fluctuate over time. Time and specifically measurement time, is key.
Traditional channel sounders take measurements serially, meaning they essentially take a “snapshot” of the channel in one direction before moving to another angle and acquiring another snapshot to arrive at a 360-degree view. The time between snapshots varies with the equipment and if the channel changes in the time it takes to switch and setup for a new snapshot, valuable data may be lost. Additionally, traditional channel sounding systems take these snapshot measurements and then post process the data by piecing together the various views. Given the number of measurements required, it may take hours or even days before a holistic picture of the environment can be generated.
To combat this challenge, NI and AT&T collaborated on one of the world’s fastest mmWave channel sounders for 5G. Nicknamed internally within AT&T as the “Porcupine,” the state of the art system measures the channel from a 360-degree perspective in real-time. The “Porcupine” utilizes a unique antenna design developed by AT&T that deploys 16 quad-horn antennas yielding 64 elements configured in a 360-degree semi-sphere where data from each “quill” is captured and processed in real-time. No more waiting in between snapshots. Several PXIe FPGA modules process four independent signal streams in real-time using IP developed by NI in LabVIEW system design software. Instead of capturing and storing I/Q data in a singular direction and then stitching together the snapshots to construct a picture off line, the Porcupine delivers full channel impulse response in milliseconds. Most importantly, AT&T can get a picture of the mmWave channel within the coherence time – especially valuable when deploying and designing 5G networks.
5G continues to push the envelope demanding relentless innovation to meet the aggressive requirements set forth by the industry. Wireless researchers need new methods and tools to facilitate practical solutions to these daunting challenges. The “Porcupine” is an example of such an innovation and we look forward to seeing more of these kinds of breakthroughs on the road to 5G.