Let's boot up a quantum computer. Far from being a simple push of a button, initializing a prototype quantum computer requires the precise tuning and calibration of many different parameters. Rather than switching transistors on and off, the controller of a quantum processor emits combinations of analog waveforms, each with a characteristic shape, frequency, and duration. These waveforms are used to either manipulate or read out the states of quantum bits (qubits), the basic units of quantum information in the processor. The analog nature of the system inherits many of the complexities of analog computing, including device parameter drift and offsets, system component tolerances and variabilities, and other well-known analog-circuit intricacies. Further, de livering these signals to the target qubits necessitates coordination between multiple low-noise and low-jitter instruments. In spite of these challenges, the quantum computing community has made tremendous progress toward useful quantum machines. We hope to provide an instructive introduction to the control, signal generation, and distribution principles currently used in small quantum systems that operate in the microwave frequency regime.
Let's boot up a quantum computer. Far from being a simple push of a button, initializing a prototype quantum computer requires the precise tuning and calibration of many different parameters. Rather than switching transistors on and off, the controller of a quantum processor emits combinations of analog waveforms, each with a characteristic shape, frequency, and duration. These waveforms are used to either manipulate or read out the states of quantum bits (qubits), the basic units of quantum information in the processor. The analog nature of the system inherits many of the complexities of analog computing, including device parameter drift and offsets, system component tolerances and variabilities, and other well-known analog-circuit intricacies. Further, de livering these signals to the target qubits necessitates coordination between multiple low-noise and low-jitter instruments. In spite of these challenges, the quantum computing community has made tremendous progress toward useful quantum machines. We hope to provide an instructive introduction to the control, signal generation, and distribution principles currently used in small quantum systems that operate in the microwave frequency regime.
Quantum computers have gained widespread interest as they can potentially solve problems that are intractable even for today’s supercomputers, such as the simulation of quantum systems as initially proposed by Feynman. To achieve such tremendous progress, a quantum computer relies on processing the information stored in quantum bits (qubits), the fundamental units of quantum computation, similar to the bits in a classical (i.e., non-quantum) processor. In general, a quantum processor comprising several qubits is connected through a quantum-to-classical interface consisting of classical electrical circuits for manipulating and reading out their quantum state. The accuracy of an operation on a qubit is characterized using the fidelity, which is 100 % in case of a perfect operation. In practice, a fidelity between 99 % and 99.9 % is typically achieved in today’s experiments with less than 100 qubits, limited by non-idealities in the control signals and by the implementation of the physical qubit itself. Since such qubit fidelities are too low to execute relevant quantum algorithms, quantum error correction schemes have been developed which rely on a much larger number of qubits to correct for possible errors, resulting in the need for quantum processors with thousands or millions of qubits. Solid-state qubits promise the largescale integration required to scale to millions of qubits, as they exploit the lithographic fabrication techniques borrowed from the semiconductor industry. However, state-of-theart solid-state qubits, such as transmons and single-electron spin qubits, must be typically cooled to cryogenic temperatures in a dilution refrigerator to exhibit quantum behavior. Because of the limited number of qubits (< 100) in state-of-theart solid-state quantum processors, the classical controller is currently implemented by general-purpose instruments or by tailor-made controllers operating at room temperature connected via several wires to the qubits in the cryogenic chamber. However, these systems must scale to support millions of qubits. While scaling up the underlying quantum processor is already exceptionally challenging, building the electronics required to interface such a large-scale processor is just as relevant and arduous.
Future error-corrected quantum computers will require analog/mixed-signal control systems capable of providing a million or more high-precision analog waveforms. In this article, we begin by introducing the reader to superconducting transmon qubit technology and then explain how the state of a quantum processor implemented in this technology is controlled. By example, we then show how one can implement one part of the quantum control system as a cryogenic integrated circuit and review experimental results from the characterization of such an IC. The article concludes with a discussion of analog/mixed-signal design challenges that must be overcome in order to build the system required to control a million-qubit quantum computer.
Current state of the art quantum computing experiments in the microwave regime use control pulses generated by modulating microwave tones with baseband signals generated by an arbitrary waveform generator (AWG). Recent advances in digital analog conversion technology have made it possible to directly synthesize arbitrary microwave pulses with sampling rates up to 92 gigasamples per second (GS/s). These new high bandwidth AWG's could dramatically simplify the classical control chain for quantum computing experiments, enabling more advanced pulse shaping and reducing the number of components that need to be carefully calibrated. Here we use a high speed AWG to study the viability of such a simplified scheme. We characterize the AWG and perform randomized benchmarking of a superconducting qubit, achieving average single qubit gate error rates below 5×10−4.
Quantum computing has made significant strides, with hardware platforms now advanced enough to tackle the challenges of scaling up to larger and more robust systems. Central to this progression is the interface between quantum processors, comprising isolated qubits, and the classical systems required for their control and readout. As these systems transition from few-qubit devices to fault-tolerant architectures, the complexity of this quantum-classical interaction poses significant engineering challenges. Field-Programmable Gate Arrays (FPGAs) have emerged as an adaptable and efficient solution, offering realtime control, low-latency operations, and seamless integration. This review delves into current FPGA-based platforms, including commercial systems, as well as open-source solutions developed by academic research groups. The architecture, features, and contributions of these platforms to addressing scalability and integration issues are thoroughly examined. Additionally, the review highlights innovations in cryogeniccompatible controllers and explores the potential of Cryo-FPGAs to reduce noise and improve fidelity in quantum systems. By providing a comprehensive analysis of current technologies and identifying opportunities for advancement, this review offers valuable insights for researchers and engineers working to develop scalable quantum computing solutions.
Quantum computing holds the promise to achieve unprecedented computation power and to solve problems today intractable. State-of-the-art quantum processors consist of arrays of quantum bits (qubits) operating at a very low base temperature, typically a few tens of mK, as shown in Fig. 15.5.1 The qubit states degrade naturally after a certain time, upon loss of quantum coherence. For proper operation, an error-correcting loop must be implemented by a classical controller, which, in addition of handling execution of a quantum algorithm, reads the qubit state and performs the required corrections. However, while few qubits (~10) in today's quantum processors can be easily connected to a room-temperature controller, it appears extremely challenging, if not impossible, to manage the thousands of qubits required in practical quantum algorithms [1].
During the past decade, quantum computing has grown from a field known mostly for generating scientific papers to one that is poised to reshape computing as we know it [1]. Major industrial research efforts in quantum computing are currently underway at many companies, including IBM [2], Microsoft [3], Google [4], [5], Alibaba [6], and Intel [7], to name a few. The European Union [8], Australia [9], China [10], Japan [11], Canada [12], Russia [13], and the United States [14] are each funding large national research initiatives focused on the quantum information sciences. And, recently, tens of start-up companies have emerged with goals ranging from the development of software for use on quantum computers [15] to the implementation of full-fledged quantum computers (e.g., Rigetti [16], ION-Q [17], Psi-Quantum [18], and so on). However, despite this rapid growth, because quantum computing as a field brings together many different disciplines, there is currently a shortage of engineers who understand both the engineering aspects (e.g., microwave design) and the quantum aspects required to build a quantum computer [19]. The aim of this article is to introduce microwave engineers to quantum computing and demonstrate how the microwave community's expertise could contribute to that field.