Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

IEEE International Midwest Symposium on Circuits and Systems (MWSCAS)

Reduction of signal-spontaneous noise in optical amplifiers

G. Du, Scott Roger Shepard, A. Wright

Published Nov 11, 2002• 0 citations
Publisher

Abstract

The beat between an amplified optical signal and the amplified spontaneous emission (ASE) noise is typically the primary limitation to the performance of optical fiber amplifiers. A filtering technique which might reduce this beat noise by 3dB creates a paradox since the noise figure's quantum limit of 3dB could then be reduced to 0dB implying noiseless amplification, which is impossible. The conjecture that this paradox might be resolved by different attenuations of the quantum and semiclassical descriptions of signal-spontaneous noise has been experimentally rejected . Herein, we present a variation of that experiment and telecommunication' applications of this filtering technique are discussed.

DOI
10.1109/mwscas.2000.952879

Related papers

A noiseless amplifier paradox

Scott Roger Shepard

Semiclassically, the signal-spontaneous beat noise in an optical amplifier can be reduced by 3 dB via a filtering technique. This creates a paradox since the noise figure's quantum limit of 3 dB could then be reduced to 0 dB, implying noiseless amplification even though noise is still being introduced by the amplifier. We discuss ways in which this paradox might be resolved and present experimental evidence that dispenses (up to the noise floor of our detectors) with one of these. A semiclassical treatment of the effects of a real (rather than ideal) filter is also presented and suggestions are made for improved (multi-mode) quantum models of this noise.

Journal of Optics B Quantum and Semiclassical Optics•May 24, 2002•1 citation

Noise in optical amplifiers

Gianfranco Cariolaro, Roberto Corvaja, Pierluigi Franco +2 more

This article presents an application of a recently published probablistic model for traveling wave optical amplification. Both optical signal and amplified spontaneous emission (ASE) noise are modeled as bunches of photons traveling along the amplifying region with size varying according to simple random rules. The results of the theory enable one to obtain a complete and accurate characterization of the signal and noise counting at the amplifier output. On the basis of the above statistics, we find a realistic evaluation of the limits imposed by both quantum and ASE noise on the performances of a receiver including an optical amplifier at a preamplifier stage.

Fiber & Integrated Optics•Jan 1, 1994•2 citations

Principles of light amplification

Stefano Bottacchi

Introduction Most of the lightwave transmission systems operating today benefit fully from extensive deployment of optical amplifiers (OA). These components perform the amplification of optical intensity directly in the optical domain, without resorting to any electrical-domain regeneration. Optical amplifiers increase the average number of output photons per unit time by several orders of magnitude compared with the incoming signal intensity. However, as with every amplifying process, optical amplification cannot be achieved without paying back for the unavoidable noise power added to the signal intensity. As we will see in detail in Chapter 6, the optical amplification process comes with optical noise generated by spontaneous emission inside the active medium. The intensity of the optical signal is represented by the flux of the signal photon population. Conversely, the photons generated by the spontaneous emission process inside the optical cavity represent the noise photon population. The fundamental phenomenon supporting optical amplification is stimulated emission. The basic property of stimulated emission, making the great difference from spontaneous emission and enabling optical amplification, is the coherent nature of the photon generation process. Accordingly, both signal and noise photons go through the same multiplication process by means of stimulated emission. Unless the optical cavity where the stimulated emission takes place (the active cavity) presents selective features for enhancing the signal photon population relative to the spontaneous one, there is no way to avoid noise photons going through stimulated emission, filling most of the cavity population. As a consequence, both photon populations increase, leading respectively to amplified signal intensity and to amplified spontaneous emission intensity. It should be apparent that every noise photon can create a new identical noise photon as a result of stimulated emission, effectively increasing the noise intensity in the active optical cavity. This is the well-known mechanism of amplified spontaneous emission (ASE) noise in optical amplification. Any improvement in the optical amplifier structure devoted to increasing stimulated over spontaneous emission will probably increase the optical signal-to-noise ratio (OSNR) at the cavity output. The spectrum of gain characteristic of the optical active medium is usually wide enough to support an efficient amplification of the broadband spontaneous emission spectrum, leading to the broadband ASE noise power spectral density.

Cambridge University Press eBooks•Sep 6, 2014•0 citations

Signal Enhancement by Optical Amplification in Autocorrelation Measurements

Michael Harris, R. Loudon, Terence J. Shepherd +1 more

The effect of optical amplification on the signal-to-noise ratio of autocorrelation measurements has been investigated experimentally. A light signal containing correlation information was amplified by an inverted population (phase-insensitive) optical amplifier. The data clearly show that the amplification may either degrade or enhance a subsequent autocorrelation measurement, depending on the experimental conditions. In particular, enhancement is possible when the detector quantum efficiency is low, or its noise is high. However, when all the signal photons are already being detected with no extra noise, then any optical amplification will inevitably lead to degradation of the data quality, due to addition of amplified spontaneous emission noise (ASE) from the amplifier itself.

Journal of Modern Optics•Apr 1, 1991•5 citations

Optimum noise performance of optical amplifiers

H. A. Haus

The concept of noise figure F and noise measure M applicable to radio frequency and microwave amplifiers is reviewed and extended to cover optical amplifiers. Two noise figures are defined in the optical regime: one gives the ASE output, while the other gives the ratio of signal-to-noise ratios at the input and output. In the quantum regime, the lower limit on these optimum values is set by a generalization of Heisenberg's uncertainty relation. A Raman back-pumped fiber link with post-amplification with an erbium-doped fiber amplifier is used to illustrate the use of noise measure.

IEEE Journal of Quantum Electronics•Jun 1, 2001•16 citations

Noise in an optical amplifier: formulation of a new semiclassical model

Silvano Donati, Guido Giuliani

After pointing out some questionable assumptions of the standard beating theory, we formulate a new semiclassical wave theory of the noise in optical amplifiers. The theory is simple yet rigorous and uses a few quantum statements in a classical signal framework. The amplifier is modeled as a 2/spl times/2 port device, and the amplified spontaneous emission and associated noise are shown to be just the amplified coherent state (or vacuum state) fluctuation of the field entering the idler input. The new theory can treat other closely related detection schemes as well, correctly supplying both mean signal and noise.

IEEE Journal of Quantum Electronics•Jan 1, 1997•41 citations
The beat between an amplified optical signal and the amplified spontaneous emission (ASE) noise is typically the primary limitation to the performance of optical fiber amplifiers. A filtering technique which might reduce this beat noise by 3dB creates a paradox since the noise figure's quantum limit of 3dB could then be reduced to 0dB implying noiseless amplification, which is impossible. The conjecture that this paradox might be resolved by different attenuations of the quantum and semiclassical descriptions of signal-spontaneous noise has been experimentally rejected . Herein, we present a variation of that experiment and telecommunication' applications of this filtering technique are discussed.
Semiclassically, the signal-spontaneous beat noise in an optical amplifier can be reduced by 3 dB via a filtering technique. This creates a paradox since the noise figure's quantum limit of 3 dB could then be reduced to 0 dB, implying noiseless amplification even though noise is still being introduced by the amplifier. We discuss ways in which this paradox might be resolved and present experimental evidence that dispenses (up to the noise floor of our detectors) with one of these. A semiclassical treatment of the effects of a real (rather than ideal) filter is also presented and suggestions are made for improved (multi-mode) quantum models of this noise.
This article presents an application of a recently published probablistic model for traveling wave optical amplification. Both optical signal and amplified spontaneous emission (ASE) noise are modeled as bunches of photons traveling along the amplifying region with size varying according to simple random rules. The results of the theory enable one to obtain a complete and accurate characterization of the signal and noise counting at the amplifier output. On the basis of the above statistics, we find a realistic evaluation of the limits imposed by both quantum and ASE noise on the performances of a receiver including an optical amplifier at a preamplifier stage.
Introduction Most of the lightwave transmission systems operating today benefit fully from extensive deployment of optical amplifiers (OA). These components perform the amplification of optical intensity directly in the optical domain, without resorting to any electrical-domain regeneration. Optical amplifiers increase the average number of output photons per unit time by several orders of magnitude compared with the incoming signal intensity. However, as with every amplifying process, optical amplification cannot be achieved without paying back for the unavoidable noise power added to the signal intensity. As we will see in detail in Chapter 6, the optical amplification process comes with optical noise generated by spontaneous emission inside the active medium. The intensity of the optical signal is represented by the flux of the signal photon population. Conversely, the photons generated by the spontaneous emission process inside the optical cavity represent the noise photon population. The fundamental phenomenon supporting optical amplification is stimulated emission. The basic property of stimulated emission, making the great difference from spontaneous emission and enabling optical amplification, is the coherent nature of the photon generation process. Accordingly, both signal and noise photons go through the same multiplication process by means of stimulated emission. Unless the optical cavity where the stimulated emission takes place (the active cavity) presents selective features for enhancing the signal photon population relative to the spontaneous one, there is no way to avoid noise photons going through stimulated emission, filling most of the cavity population. As a consequence, both photon populations increase, leading respectively to amplified signal intensity and to amplified spontaneous emission intensity. It should be apparent that every noise photon can create a new identical noise photon as a result of stimulated emission, effectively increasing the noise intensity in the active optical cavity. This is the well-known mechanism of amplified spontaneous emission (ASE) noise in optical amplification. Any improvement in the optical amplifier structure devoted to increasing stimulated over spontaneous emission will probably increase the optical signal-to-noise ratio (OSNR) at the cavity output. The spectrum of gain characteristic of the optical active medium is usually wide enough to support an efficient amplification of the broadband spontaneous emission spectrum, leading to the broadband ASE noise power spectral density.
The effect of optical amplification on the signal-to-noise ratio of autocorrelation measurements has been investigated experimentally. A light signal containing correlation information was amplified by an inverted population (phase-insensitive) optical amplifier. The data clearly show that the amplification may either degrade or enhance a subsequent autocorrelation measurement, depending on the experimental conditions. In particular, enhancement is possible when the detector quantum efficiency is low, or its noise is high. However, when all the signal photons are already being detected with no extra noise, then any optical amplification will inevitably lead to degradation of the data quality, due to addition of amplified spontaneous emission noise (ASE) from the amplifier itself.
The concept of noise figure F and noise measure M applicable to radio frequency and microwave amplifiers is reviewed and extended to cover optical amplifiers. Two noise figures are defined in the optical regime: one gives the ASE output, while the other gives the ratio of signal-to-noise ratios at the input and output. In the quantum regime, the lower limit on these optimum values is set by a generalization of Heisenberg's uncertainty relation. A Raman back-pumped fiber link with post-amplification with an erbium-doped fiber amplifier is used to illustrate the use of noise measure.
After pointing out some questionable assumptions of the standard beating theory, we formulate a new semiclassical wave theory of the noise in optical amplifiers. The theory is simple yet rigorous and uses a few quantum statements in a classical signal framework. The amplifier is modeled as a 2/spl times/2 port device, and the amplified spontaneous emission and associated noise are shown to be just the amplified coherent state (or vacuum state) fluctuation of the field entering the idler input. The new theory can treat other closely related detection schemes as well, correctly supplying both mean signal and noise.