Code-division multiple-access multiuser demodulator by using quantum fluctuations

Abstract

We examine the average-case performance of a code-division multiple-access (CDMA) multiuser demodulator in which quantum fluctuations are utilized to demodulate the original message within the context of Bayesian inference. The quantum fluctuations are built into the system as a transverse field in the infinite-range Ising spin glass model. We evaluate the performance measurements by using statistical mechanics. We confirm that the CDMA multiuser modulator using quantum fluctuations achieve roughly the same performance as the conventional CDMA multiuser modulator through thermal fluctuations on average. We also find that the relationship between the quality of the original information retrieval and the amplitude of the transverse field is somehow a "universal feature" in typical probabilistic information processing, viz., in image restoration, error-correcting codes, and CDMA multiuser demodulation.

Modernizing quantum annealing using local searches

“…QAA-like protocols have also been successfully implemented in condensed matter systems [3]. While the quantum distribution obtained from the QAA will not generally be the thermal distribution one would find with zero transverse field, it has been demonstrated experimentally that a quantum annealer can be used for thermal sampling under some circumstances [10], it has also been demonstrated numerically that in some cases the transverse field can act as an effective proxy for finite temperature [11,12]. Recent work has also suggested that the dissipation from open quantum system effects in QA can lead to an improvement in performance over AQC [29].…”

“…It has been shown in [9] that for some machine learning tasks a quantum distribution at finite temperature is actually preferable to a Boltzmann distribution on a classical Ising model. For other tasks it has been shown that using quantum fluctuations as a proxy for thermal distributions is sub-optimal, but makes little difference in practice [11,12]. An approximate thermal sample can be obtained by first taking all of the data from the states output by algorithm 2 at a desired T eff (perhaps throwing the first few iterations away as warm up period), and then applying the metropolis rules at = T T eff to obtain a fair thermal sample within each of these local minima.…”

“…A complete list of all potential applications would be too long to give here. However applications have been studied in such diverse fields as finance [4], computer science [5], machine learning [6][7][8][9], communications [10][11][12][13], graph theory [14], and aeronautics [15], illustrating the importance of such algorithms to real world problems. While some of these applications rely on the ability of the QAA to perform optimization by finding the lowest energy state of a classical problem Hamiltonian, others such as [6][7][8][9][10]13], instead rely on the fact that open quantum systems effects allow for sampling of an approximate Boltzmann distribution.…”

Maximum-Entropy Inference with a Programmable Annealer

“…In other words, given prior information about the channel noise (and therefore about the likely distribution of the received set of couplers J′ ), maximum entropy inference can be used to obtain better decoding performance than the maximum likelihood approach. More recently the role of quantum fluctuations on inferential problems including decoding have been considered 15 16 17 . The current paper does not seek to determine the role of quantum fluctuations, in finding an optimal solution; instead it considers the extent to which thermal fluctuations (which will necessarily be present in any real implementation of a Sourlas decoder) may be exploited.…”

Infinite-range transverse field Ising models and quantum computation

“…The right panel shows Improvable regions (∆M > 0) in T -Γ (left) T0 = 1.0. Dashed lines indicate border at which ∆M = 0 holds for α = K/N = 2 (These were taken from ours[37]).…”