Search

A key performance metric in blockchains is the latency between when a transaction is broadcast and when it is confirmed (the so-called, confirmation latency). While improvements in consensus techniques can lead to lower confirmation latency, a fundamental lower bound on confirmation latency is the propagation latency of messages through the underlying peer-to-peer (p2p) network (inBitcoin, the propagation latency is several tens of seconds). The de facto p2p protocol used by Bitcoin and other blockchains is based on random connectivity: each node connects to a random subset of nodes. The induced p2p network topology can be highly suboptimal since it neglects geographical distance, differences in bandwidth, hash-power and computational abilities across peers. We present Perigee, a decentralized algorithm that automatically learns an efficient p2p topology tuned to the aforementioned network heterogeneities, purely based on peers' interactions with their neighbors. Motivated by the literature on the multi-armed bandit problem, Perigee optimally balances the tradeoff between retaining connections to known well-connected neighbors, and exploring new connections to previously-unseen neighbors. Experimental evaluations show that Perigee reduces the latency to broadcast by $33\%$. Lastly Perigee is simple, computationally lightweight, adversary-resistant, and compatible with the selfish interests of peers, making it an attractive p2p protocol for blockchains., Comment: Accepted at ACM PODC 2020

This paper presents a 1-bit digital transmitter system with voltage-mode class-D digital power amplifiers (DPA) and an FPGA-based 1-bit modulator for the first time. The DPA employs quadruple cascode configuration with 45 nm SOI CMOS process to increase an operation voltage. The DPA exhibits output swing voltage over 3.2Vpp for 20Gbps RF-pulses under 4V with power consumption of 264mW. The 1-bit digital transmitter system with a digital transmitter IC integrating four DPAs and 4-way transformer-based power combiner convers 2.1-3.8GHz with output power of more than 100mW.

This paper presents a novel band-limited digital predistortion (BL-DPD) with band-switching feedback architecture and update procedure for linearization of a wideband 5G mmWave power amplifier (PA). In the proposed BL-DPD, inside and outside of original input signal band in PA output are acquired alternately. This is done while keeping ADC bandwidth same as original input signal bandwidth, thus, avoiding increase in cost and energy consumption. It is demonstrated that the proposed BL-DPD achieves ACLR of −43.6/-42 dBc and EVM < 3% for Watt-class PA operating on a 256 QAM modulated 26 GHz OFDM signal with bandwidth of 800 MHz.

A novel optical multicast tree design algorithm is proposed for elastic optical path networks; it considers signal regeneration and modulation format adaptation. The algorithm minimizes not only the number of costly regenerators needed, but also the total frequency slots necessary by optimizing the modulation formats applied. By formulating the problem as two-stage integer linear programming on an auxiliary graph, the proposed algorithm successfully calculates optimal trees on large networks within several tens of seconds.