Massive MIMO architecture for 5G networks: Co-located, or distributed?

Massive MIMO has been identified as one of the promising disruptive air interface technologies to address the massive capacity requirement demanded by 5G wireless communications. Most of the existing works showed different benefits of massive MIMO system but for the co-located deployment scenario. In this paper, we are interested to find out extra benefits that may be brought by distributing the massive number of antennas to different levels. Specifically, we compare the performance of co-located and distributed deployment scenarios in the framework of area spectrum efficiency (ASE) and area energy efficiency (AEE) tradeoff. Closed-form expressions are derived to help discover important design insights. For instance, it is interesting to see the distribution of the massive antennas always helps to improve ASE, but not the case for AEE. Moreover, with the large number of transmit antennas available and the practical cost of channel estimation considered, collaboration between cells is not always welcomed in each deployment scenario and the positive use cases depend on system parameters such as the number of antennas and the way they are distributed.