A significant hurdle towards realization of practical and scalable quantum computing is to protect the quantum states from inherent noises during the computation. In physical implementation of quantum circuits, a long-distance interaction between two qubits is undesirable since, it can be interpreted as a noise. Therefore, multiple quantum technologies and quantum error correcting codes strongly require the interacting qubits to be arranged in a nearest neighbor (NN) fashion. The current literature on converting a given quantum circuit to an NN-arranged one mainly considered chained qubit topologies or Linear Nearest Neighbor (LNN) topology. However, practical quantum circuit realizations, such as Nuclear Magnetic Resonance (NMR), may not have an LNN topology. To address this gap, we consider an arbitrary qubit topology. We present an Integer Linear Programming (ILP) formulation for achieving minimal logical depth while guaranteeing the nearest neighbor arrangement between the interacting qubits. We substantiate our claim with studies on diverse network topologies and prominent quantum circuit benchmarks.
This paper considers multiple-input multiple-output (MIMO) relay communication in multi-cellular (interference) systems in which MIMO source-destination pairs communicate simultaneously. It is assumed that due to severe attenuation and/or shadowing effects, communication links can be established only with the aid of a relay node. The aim is to minimize the maximal mean-square-error (MSE) among all the receiving nodes under constrained source and relay transmit powers. Both one- and two-way amplify-and-forward (AF) relaying mechanisms are considered. Since the exactly optimal solution for this practically appealing problem is intractable, we first propose optimizing the source, relay, and receiver matrices in an alternating fashion. Then we contrive a simplified semidefinite programming (SDP) solution based on the error covariance matrix decomposition technique, avoiding the high complexity of the iterative process. Numerical results reveal the effectiveness of the proposed schemes.
Mar 28 2017 cs.ET
In his seminal paper, Chua presented a fundamental physical claim by introducing the memristor, "The missing circuit element". The memristor equations originally supposed to represent a passive circuit element because, with active circuitry, arbitrary elements can be realized without limitations. Therefore, if the memristor equations do not guarantee that the circuit element can be realized by a passive system, the fundamental physics claim about the memristor as "missing circuit element" loses all its weight. Recent work of Chua indicates that certain type of memristor features may imply an active device. To make the question more physical, we incorporate thermodynamics into the study of this question. By using the Second Law of Thermodynamics and thermal noise, we prove that the memristor model represents an active device for an infinitely large number of cases of memristor functions M; particularly, any power function with an odd power exponent, including the situation of linear M. This situation implies rectifier features and, according to the Brillouin paradox, driving a passive circuit element of rectifying feature with thermal noise would allow the construction of perpetual motion machines. The memristor equations require an active device at infinitely many situations.