Session K67: Quantum networks modular architecture and distribution

Quantum Data Center: Theories and Applications

We propose the Quantum Data Center (QDC), an architecture combining Quantum Random Access Memory (QRAM) and quantum networks. We give a precise definition of QDC, and discuss its possible realizations and extensions. We discuss applications of QDC in quantum computation, quantum communication, and quantum sensing, with a primary focus on QDC for $T$-gate resources, QDC for multi-party private quantum communication, and QDC for distributed sensing through data compression. We show that QDC will provide efficient, private, and fast services as a future version of data centers.The talk is based on https://arxiv.org/abs/2207.14336.   

Quantum-Accelerated Distributed Algorithms for Approximate Steiner Trees and Directed Minimum Spanning Trees

We present two algorithms in the Quantum CONGEST-CLIQUE model of distributed computation that succeed with high probability; one for producing an approximately optimal Steiner Tree, and one for producing an exact spanning arborescence of minimum weight, the analog of a Minimum Spanning Tree in a directed graph, each of which uses O~(n^(1/4)) rounds of communication and O~(n^(9/4)) messages, achieving a lower round and message complexity than any known algorithms in the classical CONGEST-CLIQUE model.

The CONGEST distributed computational model allows limited-sized messages to be transmitted within a network described by a communication graph of size n in a series of rounds to address a computational problem. The size limitation for such messages is O(log(n)) bits at each edge of the communication graph per round. The communication graph in the CONGEST-CLIQUE model is fully connected. In the Quantum CONGEST-CLIQUE model, at most O(log(n)) classical and quantum bits (qubits) can be communicated across each edge of the communication graph per round.

At a high level, we achieve these results by combining classical algorithms with fast quantum subroutines. These speedups further contribute to understanding what problems can be solved more efficiently when we allow quantum communication in this CONGEST-CLIQUE model of distributed computation.    

Optimal Control Policies for Distributed Quantum Computing with Quantum Walks

Distributed quantum computing (DQC) is a key application of quantum networks. It enables interconnected quantum computers to virtually implement a powerful quantum machine that uses entanglement to solve problems that cannot be addressed by individual computers alone. One of the key capabilities that quantum networks must support to DQC is the execution of quantum controlled gates among qubits residing in geographically separated quantum computers. In order to perform such control, we describe a quantum walk network control plane protocol that captures the logic operations needed for the execution of remotely controlled quantum gates. The quantum walk protocol provides multiple ways to distribute a given quantum circuit in a network. The multitude of valid distribution schemes for the same circuit naturally leads to the definition of a combinatorial optimization problem that captures the optimal way to use the network for circuit execution. In this context, we define an integer programming formulation to compute network policies to implement a given quantum circuit minimizing network resource utilization, i.e the number of channel uses necessary to implement the circuit. The formulation determines both the assignment of qubits described by a logical description of the circuit tophysical qubits in the network, and how to route quantum control information with quantum walks.   

Synchronization of FNAL and ANL for quantum communications

We distribute clock synchronization on a three-node quantum network between Fermi and Argonne National Laboratories, connected via 59km of optical fiber. Quantum photon pairs are generated at the telecommunications C-band whereas the classical optical clock signal is generated in the O- and L-bands and coexists with the quantum photons in the fiber. The photon pairs and clock signals are directed to nodes through a standard telecommunication switch, with the clock signals providing picosecond-scale timing resolution to synchronize the photon pairs. We study the properties of the Raman scattering for the clock signal in the optical fiber used in the network, determining that we have a Raman scattering coefficient of 20.8 ± 0.3 x 10-10 for the L-band and 4.6 ± 0.1 x 10-10 for the O-band. We measure the coincidence-to-accidental ratio (CAR) of the photon pairs with and without the coexisting clock signal to determine the amount of Raman scattering occurring from the clock signal into the quantum frequency channel. We notice a reduction in the CAR from 51 ± 2 to 5.3 ± 0.4 when the quantum light coexists with the O-band clock pulses. Even with this reduction, the CAR is suitable for quantum networks.   

Adaptive, Continuous Entanglement Generation for Quantum Networks

Quantum networks rely on entanglement between qubits at distant nodes to transmit information; however, creation of these links is not dependent on the information to be transmitted. Researchers have thus explored schemes for continuous generation of entanglement, where network nodes may generate entanglement links before receiving user requests. In this work, we present a scheme for continuous generation of entanglement utilizing an adaptive method to better tailor generated entanglement links to incoming requests and reduce latency. This method selects links to generate randomly according to some distribution and updates this distribution upon receiving requests. We first construct a simple simulator to test the scheme and derive parameter spaces where latency improvement is likely. For these spaces, we observe improvements in latency versus networks with no pre-generated entanglement and networks with a static distribution for pre-generated links. We then move our simulation to the more realistic Simulator of Quantum Network Communication (SeQUeNCe) and observe the performance of our scheme for varying hardware parameters, request structures, and network topologies.   

Towards a Global Satellite Quantum Network: Optimized Entanglement Distribution in a Dynamic Satellite System

Quantum networks will undoubtedly play crucial role in advancing quantum technologies including quantum communication, distributed quantum computing, and quantum metrology. Currently a nascent topic, the fundamental understanding of how to properly and efficiently simulate usable quantum networks will have a profound impact on how to design and implement this technology on a global scale. Reaching high entanglement distribution rates over long distances through fiber or atmosphere is difficult due to exponential photon loss, and can be ameliorated by incorporating satellite links into the quantum communication network. Within this project, an optimized theoretical framework for a global network of satellites is developed.

The goal of this project is to simulate an entanglement distribution scheme via a quantum global network, in which a quantum repeater architecture is used. All relevant sources of noise and loss, as well as the capabilities of quantum memories, are accounted for in a dynamic satellite constellation. The effects various parameters have on the distribution rate and fidelity of the system are analyzed. Ultimately, a modular approach is provided for simulating a global network for various capabilities, paving the way for a global quantum internet.    

Experiments on an entanglement-based quantum network in the lab

Quantum networks promise the ability to distribute quantum entangled states over large distances for the purposes of quantum communication, distributed and blind quantum computation, or quantum sensing. Recently, we have realized the first multi-node quantum network in the lab, combining remote quantum photonic links with small quantum processors containing a diamond NV center communication qubit and a carbon-13 memory qubit[1]. This network can serve as a testbed for control stack development and for exploring quantum network protocols. As an example, quantum teleportation between two non-neighboring stationary nodes was recently demonstrated[2].

In this talk, we will report on the exploration of more network protocols using NV centers in diamond that are important for the upscaling of quantum networks. We will discuss some of the underlying physical layer and control layer challenges, and our approach to solving these. This work will build up on the detailed study on factors that affect both the rate and fidelity of the quantum-entangled state between two distant qubits[3].   

Quantum Networks Status and Prospects

 Quantum networks promise accelerated development of a wide array of applications in several fields including secure communication, computing, quantum simulations, metrology, sensing etc. While quantum networks are still in their nascent stages of development, the different stages and the corresponding known applications of each of the stages are clearly identified, setting up a clear path for a large-scale quantum internet.  Here we report our progress towards each of the stages including distribution of weak coherent states, generation and distribution of entangled states using fibers and free space as quantum channels and finally interfacing with hybrid species that are able to hold and process the quantum information. We present some of the key milestones that are achieved in this process including high clock rate time-bin quantum key distribution, high rate time-bin entanglement generation and distribution using ultra low loss fiber, low jitter synchronization of distant national labs, FNAL and ANL, and development of SNSPDs with high count rates and low jitter. We also discuss the development of different architectures needed for each of the above mentioned stages for a successful deployment in the field. 

    

Modeling Short-range Quantum Networks for Scaling Superconducting Quantum Computation

A core challenge for superconducting quantum computers is to scale up the number of qubits without increasing noise or cross-talk. Distributing a quantum computer across nearby, small qubit arrays, known as chiplets, addresses several relevant challenges. We propose chiplet architectures connected over microwave links with potential to exceed monolithic performance on near-term hardware. We model and compare architectures in a way that bridges the physical and network layers. We find evidence that short-range microwave networks may yield overall lower-noise operations despite higher noise figures at links. Chiplet topologies, latencies, and bandwidths may also compete reasonably with monolithic analogs, especially in applications that map naturally to distributed architectures. In the long term, short-range networks may underlie quantum computers just as local area networks underlie classical datacenters and supercomputers. Understanding these local networks requires quantum-based models, differing from classical expectations..

 

Based on work with Kaitlin N. Smith (ColdQuanta), Poolad Imany (National Institute of Standards and Technology; University of Colorado Boulder), Kevin L. Silverman (National Institute of Standards and Technology) & Frederic T. Chong (Dept. of Computer Science, University of Chicago).   

Scalable surface-code quantum error correction based on cavity-QED network

The exploration of an efficient and scalable architecture of fault-tolerant quantum computing (FTQC) is among the most important keys to the demonstration of useful quantum computing. In this presentation, we propose a scalable and practical architecture with a cavity network, each cavity including neutral atoms as data qubits. In our architecture, all atomic qubits are connectable each other via intra cavity fields and the interconnection of cavities by fibers or waveguides, which is flexibly reconfigurable with optical switches. The flexibility in the connection of qubits is advantageous in efficiently constructing a logical qubit. We consider several specific structures of cavity network and analyze the error model in each structure along with a discussion of the difficulties of implementing it. We then numerically estimate the logical error probability and threshold values under the circuit-level noise model. Through these analyses, we also evaluate the efficacy of a heralded syndrome measurement relying on the detection of photon loss events.   

Exploring the Potential of Quantum Network Simulators to Guide the Development of Quantum Network Tomography

Just like classical networks, quantum networks will require tools to diagnose problems and monitor performance. In classical computer networks, the quality of a given network is assessed using network tomography techniques, however, no equivalent has been developed for a quantum network. In order to address this need, we investigate how to utilize quantum network simulators as a virtual quantum network laboratory for developing quantum network tomography protocols. Quantum network simulators allow complete network customization, enabling us to alter parameters of the network such as error rates and network topologies in order to identify characteristics of network errors. We will discuss how to identify and characterize types of error such as coherent Pauli errors within a network through protocols developed on simulated testbeds.   

An Investigation of the Potential of Quantum Game Theory to control Routing in Quantum Networks

With increasing progress in quantum information science, the need for a quantum network allowing  communication between quantum systems is obvious. In a functional quantum network, users need to be able to reliably send information to each other without loss of information. When quantum information is sent at the same time on the same network channel,  there is the possibility of interference, thus resulting in loss of information. In classical networks, game theory has successfully been applied to mitigate routing congestion due to its ability to find optimal strategies to increase successful outcomes in game play. We will discuss how quantum game theory, the fusion of game theory with quantum mechanics, can minimize congestion and optimize sending information inside a quantum network. More specifically, we present results showing  how  the superposition and entanglement aspects of quantum mechanics can be used to increase the efficiency of routing inside a quantum network in realistic scenarios.