Adaptive Resource Allocation for Energy-Efficient Millimeter-Wave Massive MIMO Networks

Massive, green, soft and super-fast are the key attributes that will describe next-generation mobile networks (5G and beyond). Large millimeter-wave (mmWave) bandwidths, massive MIMO antenna arrays, ultra-dense small cells (UDN) and cloud radio access network (C-RAN), among others, will enable these future networks to deliver huge performance gains. Since the networks are anticipated to be green and the available spectrum will be more abundant, energy efficiency (EE) becomes a more critical design factor than spectral efficiency (SE). In the face of competing potentials and challenges brought about by the different enablers, efficient resource allocation (RA) schemes are important to optimize system performance. In this work, we propose a fully-adaptive RA for a dense, C-RAN-enabled mmWave massive MIMO network. We then compare its performance to a non-adaptive and two semi-adaptive RA schemes. The fully-adaptive scheme outperforms all the other RA schemes and shows promising potentials for the joint EE-SE optimization of future mobile networks. Our results show the optimal EE-SE points and the impact of the transmit power and the number of data streams on the EE and SE performance.