## Abstract

An ideal linear optical amplifier increases the amplitude of an incoming optical signal by a constant gain factor and induces a linear phase shift. In a fundamental paper from 1962 [1], Heffener derived, using the uncertanty principle of quantum mechanics, that it is not possible to construct a linear noiseless amplifier. This implies that the intrinsic noise source (amplified spontaneous emission) induces random perturbations on the phase of the incoming optical signal. The fundamental question that arises is on the minimum variance of the induced phase fluctuation and the corresponding spectral broadening. We demonstrate that the quantum limit on the minimum phase variance derived by Heffener in [1] can be surpassed by performing

a statistically optimum optical phase measurement. We propose a new limit, valid for low to moderate signal-to-noise ratios and a practical receiver architecture to approach it. Using the proposed receiver, we experimentally demonstrate a significant reduction of the amplifier noise impact. These new insights may play a significant role for the design of high–power narrow–linewidth laser sources, transmission of optical signals and optical systems used for metrology.

a statistically optimum optical phase measurement. We propose a new limit, valid for low to moderate signal-to-noise ratios and a practical receiver architecture to approach it. Using the proposed receiver, we experimentally demonstrate a significant reduction of the amplifier noise impact. These new insights may play a significant role for the design of high–power narrow–linewidth laser sources, transmission of optical signals and optical systems used for metrology.

Original language | English |
---|---|

Journal | Optica |

Number of pages | 8 |

ISSN | 2334-2536 |

Publication status | Accepted/In press - 2021 |