The general problem of performance advantage obtainable by the use of nonclassical transmitted states over classical ones is considered. Attention is focused on the situation where system loss is significant and additive Gaussian noise may be present at the receiver. Under the assumption that the total received state is classical, rigorous output density operator representations and their trace distance bounds are developed for classical and nonclassical transmitted states. For applications with high loss in all modes, a practical no-go theorem is enunciated that rules out the possibility of significant advantage of nonclassical over classical states. The recent work on quantum illumination is discussed as an example of our no-go approach.
The general problem of performance advantage obtainable by the use of nonclassical transmitted states over classical ones is considered. Attention is focused on the situation where system loss is significant and additive Gaussian noise may be present at the receiver. Under the assumption that the total received state is classical, rigorous output density operator representations and their trace distance bounds are developed for classical and nonclassical transmitted states. For applications with high loss in all modes, a practical no-go theorem is enunciated that rules out the possibility of significant advantage of nonclassical over classical states. The recent work on quantum illumination is discussed as an example of our no-go approach.
Quantum information investigates the possibility of enhancing our ability to process and transmit information by directly exploiting quantum mechanical laws. When searching for improvement opportunities, one typically starts by assessing the range of outcomes classically attainable, and then investigates to what extent control over the quantum features of the system could be helpful, as well as the best performance that could be achieved. In this thesis we provide examples of these aspects, in linear optics, quantum metrology, and quantum communication. We start by providing a criterion able to certify whether the outcome of a linear optical evolution cannot be explained by the classical wave-like theory of light. We do so by identifying a tight lower bound on the amount of correlations that could be detected among output intensities, when classical electrodynamics theory is used to describe the fields. Rather than simply detecting nonclassicality, we then focus on its quantification. In particular, we consider the characterisation of the amount of squeezing encoded on selected quantum probes by an unknown external device, without prior information on the direction of application. We identify the single-mode Gaussian probes leading to the largest average precision in noiseless and noisy conditions, and discuss the advantages arising from the use of correlated two-mode probes. Finally, we improve current bounds on the ultimate performance attainable in a quantum communication scenario. Specifically, we bound the number of maximally entangled qubits, or private bits, shared by two parties after a communication protocol over a quantum network, without restrictions on their classical communication. As in previous investigations, our approach is based on the evaluation of the maximum amount of entanglement that could be generated by the channels in the network, but it includes the possibility of changing entanglement measure on a channel-by-channel basis. Examples where this is advantageous are discussed.
Despite various approaches to encode nonclassical light, most existing formulations either suffer from limited practicality or rely on optimization procedures that are computationally demanding. Here, we develop a framework for quantifying optical nonclassicality without invoking any optimization, using generic classicalization channels, p-norms and linear functionals generated by displaced rotationally-invariant operators. However, for a subset of such measures, involving s-ordered distributions and a Gaussian classicalization channel, we analytically establish a no-go theorem demonstrating that no such universal measure can exist. We further substantiate our theoretical result through explicit examples involving both Gaussian and non-Gaussian states, followed by a generic criterion for a no-go theorem, whose the existence remains to be established.
The classical information capacity of channels that are subject to quantum Gaussian noise is studied. Recent work has established the capacity of the pure-loss channel, as well as bounds on and a conjecture for the capacity of the lossy channel with isotropic-Gaussian excess noise. This work is applied to the pure-loss free-space channel that uses multiple Hermite-Gaussian (HG) or Laguerre-Gaussian (LG) spatial modes to communicate between soft-aperture transmit and receive pupils, and to the lossy channel with anisotropic (colored) Gaussian noise.
The principles are elaborated which underlie the applications of general nonclassical states to communication and measurement systems. Relevant classical communication concepts are reviewed. Communication and measurement processes are compared. The possible advantages of nonclassical states in classical information transfer are assessed. The significance of novel quantum amplifiers and duplicators in communication is emphasized. A general approach is developed for determining the ultimate accuracy limit in quantum measurement systems. It is found that bandwidth or mode number is a most important parameter and ultrahigh precision measurement is possible in systems with a fixed energy but many modes. The problem of the standard quantum limit in monitoring the position of a free mass is also addressed.
Quantum networks are rapidly being developed using the noisy quantum devices available today.As quantum networks scale, noise will lead to significant challenges in quantum network characterization, design, and automation, challenges that classical methods may be ill equipped to tackle.Moreover, the advantage and value of quantum networks is not well understood in the presence of noise, making it difficult to justify the cost of quantum network development for real-world applications.In this dissertation, we define operational nonclassicality as a quantifier of quantum advantage in general multipoint communication networks and describe a procedure for deriving operational tests of nonclassicality in general communication networks.Then, we develop a quantum-hardware-compatible variational optimization framework for optimizing quantum networks to exhibit nonclassicality.In a wide range of communication network topologies, including nonsignaling networks, multiaccess networks, broadcast networks, and interference networks, we derive operational tests that witness nonclassicality.We then use our variational framework to optimize various quantum resource configurations for maximal performance against these operational tests of nonclassicality.In all communication network topologies, we find examples where quantum resources lead to observable violations of the nonclassicality witnesses, implying that quantum resources provide a strict advantage over classical resources.Furthermore, we investigate how the presence of noise diminishes these advantages, and by extension, the value of quantum resources.Finally, we demonstrate that our variational optimization techniques can be deployed on quantum hardware and applied well beyond the scope of finding nonclassical quantum behaviors.In conclusion, we find that nearly all quantum resource configurations in communication networks can provide operational advantage as witnessed by operational tests of nonclassicality.These nonclassical network behaviors show novel ways that quantum physics defies the classical assumptions of locality, causality, and realism, but nonclassicality can also be used to test and certify quantum resources in communication networks.Furthermore, nonclassicality can also provide advantages in information security and distributed computing.We assert that variational quantum optimization techniques are well-suited to design and automation tasks in quantum networks.The advantages of these methods are that they are hardware agnostic, do not require full network characterization, and can optimize quantum systems against their inherent and unknown noise models.Thus, we introduce variational quantum networking as an engineering paradigm for designing and automating noisy quantum networks.In many ways, variational quantum networking circumvents the challenges of characterizing, designing, and automating noisy quantum networks.
Quantum resources, such as entanglement or quantum communication, offer significant communication advantages in information processing. We develop an operational framework for realizing these communication advantages in resource-constrained quantum networks. The framework computes linear bounds on the input/output probabilities of classical networks with limited communication and globally shared randomness. Since the violation of these classical bounds witnesses nonclassicality, a measurable communication advantage, the framework maximizes the violation of the classical bound using variational quantum optimization methods tailored to the communication network and quantum resources. This operational framework for nonclassicality can be scaled on quantum computers or deployed in the field to optimize noisy quantum networks for communication advantages. Applying this framework, we investigate the nonclassicality of communication networks that are assisted by quantum resources. We find that entanglement between communication-constrained parties is sufficient for nonclassicality to be found, whereas in networks with multiple senders, quantum communication with no entanglement-assistance is sufficient for nonclassicality to be found. As a result, entanglement is necessary for nonclassicality when a single sender broadcasts to multiple receivers.