Antonio Puglielli

TL;DR: The Sparse CNN (SCNN) accelerator as discussed by the authors employs a dataflow that enables maintaining the sparse weights and activations in a compressed encoding, which eliminates unnecessary data transfers and reduces storage requirements.

TL;DR: The Sparse CNN (SCNN) accelerator architecture is introduced, which improves performance and energy efficiency by exploiting thezero-valued weights that stem from network pruning during training and zero-valued activations that arise from the common ReLU operator.

TL;DR: The paper discusses both RF frequencies below 10 GHz, where fully digital techniques are preferred, and operation at millimeter (mm)-wave bands where a combination of digital and analog techniques are needed to keep cost and power low.

TL;DR: An active transmitter (TX) cancellation scheme enabling integration of the antenna interface for frequency division duplex systems is presented and Propagation of the TX phase noise in the RX band is analyzed and shown to be feed-forward cancelled in the proposed system.

TL;DR: This work proposes an array architecture based on a common module which serves a small number of antennas with RF transceivers, data converters, and several support functions, and demonstrates that by using this approach, the maximum backhaul datarate scales as the number of users rather than thenumber of antennas.