Problems with hardware cost and power consumption, however, arise from deploying large antenna arrays coupled with power-demanding ADCs. To overcome these challenges, hybrid analog-digital processing architectures that attempt to reduce the burden of fully digital processing, and receivers with low-resolution ADCs have attracted the most interest in recent years.
To take advantage of the two considered architectures, WNCG Professor Brian L. Evans, WNCG student Mr. Jinseok Choi, and Futurewei Baseband SOC CTO Dr. Alan Gatherer propose a hybrid massive-MIMO architecture with resolution-adaptive ADCs for mmWave communications. They investigate the ADC bit-allocation problem to minimize the quantization distortion of received signals by leveraging the flexibility of ADC resolutions. A derived closed-form bit-allocation solution reveals that the optimal number of ADC bits increases logarithmically proportional to the RF chain's SNR raised to the 1/3 power given the ADC power constraint. The proposed algorithm that utilizes the solution outperform the conventional fixed ADCs in terms of both the sum rate and energy efficiency in the mmWave communication environment.
This research was supported by Futurewei Technologies.
J. Choi, B. L. Evans and A. Gatherer, "Resolution-Adaptive Hybrid MIMO Architectures for Millimeter Wave Communications", IEEE Transactions on Signal Processing, vol. 65, no. 23, pp. 6201-6216, Dec. 2017, DOI 10.1109/TSP.2017.2745440.
While it is always possible for handset manufacturers to introduce new receiver circuitry for mmWave bands in 5G, the price of such a handset will be more expensive and the battery will deplete faster when having to measure two bands simultaneously.
WNCG Professor Brian L. Evans and PhD student Mr. Faris Mismar have introduced a new category of partially blind algorithms for inter-radio access technology band switching between LTE-Advanced and 5G. The band switch is partially blind because it makes a decision to switch to mmWave bands based on the current sub-6 GHz LTE-Advanced channel measurements. The decision comes from the probability of band switching success from an extreme gradient boosting classifier that has been trained using simultaneous LTE-Advanced nd mmWave channel measurements. If the classifier finds that the switch to mmWave is likely to succeed, it will allow the band switching measurements to start by allowing the LTE-Advanced base station to configure an occasion for the handset to measure the mmWave channel. This occasion is known as the measurement gap. If not, the base station instructs the handset to remain in the LTE-Advanced service area. This way, the measurement gap is not opened, and the handset continues to receive scheduled data from the better performing technology.
F. B. Mismar and B. L. Evans, "Partially Blind Handovers for mmWave New Radio Aided by Sub-6 GHz LTE Signaling", Proc. IEEE International Conference on Communications Workshop on Evolutional Tech. & Ecosystems for 5G Phase II, May 20-24, 2018, Kansas City, MO, USA.
F. B. Mismar, A. AlAmmouri, A. Alkhateeb, J. G. Andrews, and B. L. Evans, "Deep Learning Predictive Band Switching in Wireless Networks", IEEE Transactions on Wireless Communications, vol. 20, no. 1, Jan. 2021, pp. 96-109, DOI 10.1109/TWC.2020.3023397.
Mail comments about this page to bevans@ece.utexas.edu.