How Much Spectrum is too Much in Millimeter Wave Wireless Access

Great increase in wireless access rates might be attainable using the large amount of spectrum available in the millimeter wave (mmWave) band. However, higher propagation losses inherent in these frequencies must be addressed, especially at ranges beyond 100 m and in non-line-of-sight (NLOS) settings. In contrast to the interference limited legacy cellular systems, where using more bandwidth is favorable, to use wider bandwidth for mmWave channels in noise limited settings may be ineffective or even counterproductive when accounting for channel estimation penalty. In this paper, we quantify the maximum beneficial bandwidth for mmWave transmission in some typical deployment scenarios where pilot-based channel estimation penalty is considered assuming a minimum mean square error channel estimator at the receiver. We find that, under I.I.D. block fading model with coherence time T
_c
and coherence bandwidth B
_c
, for transmitters and receivers equipped with a single antenna, the optimal (rate maximizing) signal-to-noise-ratio is a constant that only depends on the product B
_c
T
_c
, which measures the channel coherence and equals the average number of orthogonal symbols per each independent channel coefficient. That is, for fixed channel coherence B
_c
T
_c
, the optimal bandwidth scales linearly with the received signal power. Under 3GPP urban micro NLOS path loss model with coherence time T
_c
= 5 ms and coherence bandwidth B
_c
= 10 MHz, using 52-dBm equivalent isotropic radiated power (EIRP) at the transmitter and 11-dBi antenna gain at the receiver, the maximum beneficial bandwidth at 28 (resp. 39) GHz is less than 1 GHz at a distance beyond 210 (resp. 170) m with maximum throughput about 200 Mbps, and less than 100 MHz beyond 400 (resp. 310) m with maximum throughput about 20 Mbps. At EIRP of 85 dBm, corresponding to the FCC limit of 75 dBm per 100 MHz, 1 Gbps rate can be delivered using 1-GHz bandwidth up to 860 (resp. 680) m.