After the FCC’s historic vote on July 14 establishing rules for microwave and mmWave broadband operations above 6 GHz in the US, much more attention has been paid to the mmWave region. The move effectively quadruples the amount of radio bandwidth that was available before to the mobile industry. Professor Ted Rappaport was the first to show that mmWaves were viable for cellular communications with his urban channel modeling tests done many years ago. Now he has done similar tests in a rural setting showing the viability of line-of-sight and non-line-of-sight transmission of millimeter mmWave communications. Prof Rappaport performed these tests at 73 GHz in rural Virginia.

NYU WIRELESS used this data to generate the first rural path loss model for mmWave frequencies, demonstrating pretty remarkable distances that can be achieved using mmWave (near theoretical limits).  The model will also reduce the cost and complexity of bringing Internet service to underserved rural areas. In the tests Professor Rappaport demonstrated that the approved 3GPP channel model standard (3GPP TR 38.900 [version 14]) has an error in its rural macro cellular model, making it not very accurate above 9.1 GHz. That 3GPP model was never verified with measurements as very few measurements have ever been made in these frequency ranges. Rappaport detailed how an errant model made its way into the current 3GPP standard, and then offered a proven model, based on his data, that can be used for rural deployments at all frequencies. The new model is a simpler, more reliable model based on measured data.

The industry is quickly moving to implement mmWave frequencies due to the vast bandwidths available. One popular frequency range seeming to catch on is the 28 GHz band. Qualcomm recently announced the Snapdragon X50 5G modem, making Qualcomm the first company to announce a commercial 5G modem chipset solution.  It is designed to support original equipment manufacturers that are building the next generation of cellular devices, as well as aid operators with early 5G trials and deployments. The Snapdragon X50 5G modem will initially support operation in the 28 GHz band.  It will employ MIMO antenna technology with adaptive beamforming that facilitates mobile broadband communications in non-line-of-sight environments.  With 800 MHz bandwidth support, the Snapdragon X50 5G modem is designed to support peak download speeds of up to 5 gigabits per second.

A few companies working on the mmWave RF phased arrays for 5G include ADI, Anokiwave and Peregrine. Anokiwave was the first to release a commercially available Ka-Band transceiver quad core IC for 5G communications markets. It operates at 27.5-30 GHz, supports 4 Tx/Rx radiating elements, and includes all requisite beam steering controls for 5-bit phase and gain control. The device operates in half duplex, enabling a single antenna to support both Tx and Rx operation. Anokiwave’s patent-pending IP blocks implemented in silicon technology enable low-cost hybrid beam forming for multi-antenna arrays with high energy efficiency. It also has gain compensation over temperature, temperature reporting, Tx power telemetry, and fast beam switching using eight on-chip beam weight storage registers.

The path loss of mmWave propagation is significant but still seems viable for commercial use due to beam steering and miniaturization advantages. 5G mmWave trials are happening quickly – it will be very interesting to see how fast companies can commercial mmWave technology and bring the cost down to a level that is practical for wide use. Mobile World Congress next year should be a watershed event for the release of this technology into the market.