Massive MIMO (Ma-MIMO) systems are widely expected to be a central component of future wireless networks. Using arrays of programmable radio elements, all Ma-MIMO systems attempt to shape RF energy and improve user throughput and overall system capacity by increasing SNIR and reusing spectrum between separated users. While the shape optimization process is often referred to as "beam-forming", most Ma-MIMO systems proposed to date for sub-6 GHz cellular bands do not actually form true beams, rather a large number of radio chains and digitizers are used to generate RF peaks and nulls based on user channel estimates. In this process, excess energy can spill into places outside the targeted users, causing unwanted interference at cell edges and wasting the available RF transmit power. Prior to onset of Ma-MIMO, the implementation of choice for active antennas was derived from military phased arrays. These classical phased arrays form true beams but suffer from extremely high costs even for limited surveillance applications. Their application to commercial wireless is further challenged by the complexity of industry use cases which require many simultaneous beams and maximum uplink and downlink transmission speeds over many users. Nevertheless, true beamforming is necessary in commercial mm-wave systems to be able to reach users given the large path loss at these frequencies. After reviewing the history and attributes of existing Ma-MIMO and phased array systems, a new Ma-MIMO solution will be outlined which starts by addressing a fundamental limitation that both existing systems have: the overhead and inaccuracies involved in either directly attempting to achieve RF coherency across the array or some digitally produced substitute. Combining an analog IC innovation - first described in a 2006 CICC poster as applied to VLSI clock skew alignment -- with additional radio architecture innovations, a hybrid Ma-MIMO system has been developed which builds MIMO processing on top of a fully synchronized array. Very well-defined beam shapes can then be produced for both TDD and FDD bands at much lower hardware complexity. Performance gains have been demonstrated in multiple live carrier trials, including clustered sectors which is a first for any Ma-MIMO system.
Dr. Mark Pinto is CEO and a co-founder of Blue Danube Systems, a venture capital backed startup developing intelligent wireless access solutions that deliver 5G experience on today's cellular networks and smartphones. Prior to Blue Danube, Dr. Pinto was an Executive Vice President at Applied Materials, where he served as CTO and led both the display and solar businesses. He began his career at Bell Laboratories in semiconductor research and later became CTO of the Lucent Microelectronics Group and General Manager of a network IC product division. Dr. Pinto received Bachelors' degrees from Rensselaer Polytechnic Institute and a Master's degree and a Ph.D. from Stanford University. He was included in Fast Company's 100 Most Creative People in Business in 2010 and has been named a Bell Labs Fellow, an IEEE Fellow and a member of the National Academy of Engineering. In addition, Dr. Pinto received the 2008 IEEE J.J. Ebers Award for contributions to semiconductor technology and the 2013 George E Pake Award for R&D management from the American Physical Society.