| Popis: |
The massive multiple-input/multiple-output (MIMO) technique has been gaining momentum lately as a potential key enabler of the network spectral efficiency (and hence throughput) increase expected from the fifth generation (5G) ofcellularnetwork technology. This technique consists in equipping the base stations (BSs) with arrays having a large number, typically from tens to few hundreds, of antenna elements. This, in theory, guarantees the possibility of serving larger numbers of concurrent transmissions to or from the BSs than the multiplexing techniques used in previous generations of cellular technology, without compromising the throughput of each transmission. By exploiting the high spatial multiplexing gain that can be realized through massive MIMO systems, 5G networks could provide the expected boost to the overall throughput with respect to previous generations. On the one hand, having more antenna elements in the BS array increases the spatial resolution of both outgoing and incoming signals. On the other hand, it diminishes the effect of the additive white gaussian noise (AWGN) at the receiver. These features lead to several advantages both in terms of access mechanisms and associated signal processing: little intra-cell and multicell interference leakage, optimality of simple linear precoding/detection schemes, capability of simultaneously serving a multitude of competing wireless connections in each cell area without compromising their individual throughput, just to name a few. However, this comes at the cost of the acquisition of accurate uplink/downlink channel state information (CSI) at the BS. This operation is challenging in many respects, and has a non-negligible impact in terms of both algorithms implemented at the physical/access/network layer and practical hardware design. It is worth noting that the extent of this impact strongly depends on the frequency bands over which the transmissions are performed. In this regard, it is a common belief that signals transmitted in future 5G networks will likely span a wide spectrum of frequencies.In other words, their radio carrier wavelengths will range from decimeters, as in legacy cellular networks, to millimeters, as in millimeter wave (mmWave) communications. Evident peculiarities will characterize such signals, and future massive MIMO systems will have to leverage them in effective ways in order to achieve the promised network throughput increase. This chapter starts from these observations to present a discussion on the potential and the challenges associated to the deployment of massive MIMO systems for 5G networks. |
