The paper presents for the first time an experimental demonstration of RIN noise transfer dampening at low frequencies in random distributed feedback ultralong Raman fibre lasers based on conventional telecommunication fibres. In addition, a comprehensive theoretical description of the phenomenon is presented and a predictive model is provided, identifying the general requirements for the improvement of the system through the reduction of RIN transfer.
What is a random laser? Let's first remember what a conventional laser is. Inside a confined space, a laser medium, which can be a vapor, a solid or a dye, is “pumped” so that its atoms reach an excited state. When atoms fall to a lower energy level, they emit photons. By bouncing between the two mirrors in a cavity, this emission is amplified in a coherent way, producing a beam of light of a single wavelength.
However, in so-called 'random lasers' the light bounces off a large number of floating particles that scatter the light, rather than between two mirrors, and there is no light confinement mechanism.
Unlike conventional lasers where the gain occurs in the successive reflections that light makes between the mirrors, in a random laser the gain is obtained by multiple scattering of light in a disordered medium. When characterizing a conventional laser, light scattering is often viewed as a detrimental factor for the laser process to occur. On the other hand, in a disordered medium with gain, the scattering of light happens to have a positive effect on the emission of laser radiation. Multiple light scattering causes an increase in the time the light spends inside the sample, thus increasing the time in which the light is amplified.
The concept of laser light generation in random active media without cavities was introduced by Letokhov in 1966-1968, initially in the context of interstellar radiation studies.
Random lasers have attracted a great deal of attention thanks to their unique characteristics, including their relative simplicity and the lack of need for a defined cavity structure. At the same time, its messy nature has often become an obstacle to its systematic design and implementation. A notable exception to this rule is provided by the Raman Raman Distributed Feedback Fiber (RDFL) lasers, first proposed in 2010 as a particular type of ultra-long fiber laser.
In fiber optic random lasers we do not have bounces in floating particles, as the medium is not a gas or a liquid. The dispersive centers are made up of small fluctuations in density randomly distributed along the length of the fiber, and the rebound (feedback) mechanism is known as Rayleigh scattering, the same mechanism responsible for the blue color of our sky. Rayleigh scattering is caused by particles smaller than the wavelength of light.
Unlike other types of random lasers, the use of conventional fiber optics as a gain and transport medium allows high efficiency, narrow bandwidth, and inherently directional output. Thanks to this, multiple applications have been made possible both in detection and in communications.
In a fiber Raman laser, amplification occurs from the absorption of energy from an intense light source at shorter wavelengths by the signal. This light source is known as "pumping." Apart from the own intensity fluctuations that are generated in the laser, the own fluctuations that the pumping source has are also transferred to the signal. This relative intensity noise transfer (RIN) limits laser quality and performance. In earlier work, we theoretically predicted that in some particular configurations the maximum RIN transfer in RDFL that typically occurs at low modulation frequencies could be damped, leading to an abnormal RIN transfer profile.
In this work we have experimentally demonstrated and theoretically described this phenomenon for the first time, showing how to achieve this noise reduction through the interaction between the pumps and the different components of the signal. In addition, we have identified the conditions to obtain a reduction of RIN transfer in random laser configurations based on conventional telecommunication fiber, paving the way for its direct application in amplification schemes with application both in sensors and in telecommunications.
The work is a collaboration between the Institute of Optics, the University of Bordeaux, the department of electronics and telecommunications (DET) of Turin, the Public University of Navarra (UPNA), the Institute of Smart Cities (ISC) of Navarra and the University of Alcalá de Henares (UAH)
Link to article