Accurate Approximation of System Capacity in MISO Broadcast Channel: A Massive-Antenna Perspective

Abstract

The massive-antenna is widely considered as one of key technologies in the 5th generation (5G) networks and beyond. To approach the system capacity with the massive-antenna, it is highly demanded for fundamental research on <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i</i> ) how to obtain the system capacity of massive-antenna based broadcast channel (BC) with simple closed-form expressions, and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ii</i> ) how to approach the system capacity efficiently. To address such two problems, this paper focuses on the massive-multiple-input single-output (mMISO) BC transmission, which is a common scenario in cellular networks. With regarding the massive-antenna, we derive a precise approximation of the system capacity for all possible power splits, and its expression is as simple as the formula of Shannon capacity. It is proved that the derived approximation is a lower-bound of the system capacity and asymptotically approaches the system capacity with respect to the number of transmit antennas. Thus, the derived approximation is valuable in the massive-antenna cases. Based on the approximation, we show how to jointly employ the massive-antenna and superposition coding to approach the system capacity with a low computational complexity. The simulations validate the correctness of the derived outcomes in mMISO-BC. Consequently, the aforementioned two complicated problems are effectively circumvented in this paper.