arrow left

Photonic Qubits

Photonic Qubits

Photonic Quantum Computing (PQC) uses an individual particle of light, known as a photon, as a qubit, the fundamental unit of quantum information. Several properties make this modality attractive, including:

  • Quantum information can be encoded in multiple ways, including in the polarization of the photons (horizontal or vertical) or in their paths of travel (path encoding).
  • Because of limited environmental interactions, photons are naturally resilient to certain types of noise sources that plague other qubit modalities.
  • Photons can travel long distances along existing fiber optic cables, connecting quantum computers at the speed of light over future quantum networks.
  • Photons can maintain their coherence long enough to be useful for both computation and global communication.

As with all modalities, however, photonics is not without its challenges. The two primary challenges are:

  • Photons can be lost in numbers great enough to require quantum error correction (QEC) in order for the measurement results to be useful.
  • Low Density Parity Check (LDPC) quantum error correction codes (QECC) are often cited in regard to photonic quantum computing.
  • Although touted as operating at room temperature, certain components of photonic quantum computers require bulky cryogenic equipment.

For more information, a diginomica article titled “Is phototonic quantum computing the answer to commercial quantum use? Maybe” delves a little deeper into what photonic qubits are, what their advantages are, and how cryogenics is involved. There are also some comparisons of photonic qubits vs superconducting qubits and trapped ion qubits.

What are Photonic Qubits

Several paradigms of photonic qubits are currently being researched. However, three in particular are arguably the most common. These include squeezed states, dual-rail encoding, and qumodes.

  • Squeezed states trade off Heisenberg uncertainty, decreasing uncertainty for some phases at the expense of increasing uncertainty for other phases.
  • Dual-rail encoding uses two paths, aka “modes,” one designated 0 and one designated 1, and the paths the photons take determine the measurement outcomes.
  • Qumodes use integrated optics and waveguides and rely on quantum tunneling to allow photons to pass from one waveguide to another.

Entangled photonic qubits typically have one of two potential sources. Photons can be created as entangled pairs before computation commences. Photons can also become entangled during computation through the use of interferometry.

Quantum Gates and Photonic Qubits

In comparing photonic qubits vs superconducting qubits and trapped ion qubits, gate operations are executed quite differently. The latter receive pulses of microwaves or lasers, respectively, and they interact with each other directly. Superconducting qubits are frozen in place, whereas ions are mostly stationary, but they move around for their interactions.  

Though the former can interact through interferometry, they mostly interact with linear optical components and photodetectors. As opposed to passively receiving manipulation, photons actively go get manipulated. They cannot be held in place, so they physically move through their circuits at the speed of light.

Photonic operations include:

  • Interferometry can perform important operations such as entangling operations and comparing quantum states.
  • Photonic Integrated Circuits (PIC) can not only include interferometers, but also the beam splitters and phase shifters that implement the quantum algorithms.
  • Photodetectors can distinguish the number of photons that reach them, which provides an additional way to encode quantum information.

The photonic quantum computing landscape continues to change. A Stanford University team, for example, has proposed a paradigm in which photons in a loop are controlled by a single atom. You can read more about this one in an article titled “Stanford engineers propose a simpler design for quantum computers.” But, again, this is just one example, and it is not alone in being a proposed alternative to current implementations.

No items found.