**1. Introduction**

Optical communication systems require optoelectronic devices, such as sources, detectors and so on, and utilize fiber optics to transmit the light carrying the signals impressed by modulators. Optical fibers are affected by chromatic dispersion, losses, and nonlinearity. Dispersion control is, usually, achieved via fiber geometry and material composition. Losses limit the transmission distance in modern long haul fiber-optic communication systems, so in order to boost a weak signal, optical amplifiers have been developed. The basic idea, behind the fiber amplifier realisation, was to allow to the signal to remain in optical form throughout a link or network. Fiber amplifiers as repeaters offer a number of advantages, including the ability to change system data rates, or to simultaneously transmit multiple rates—all without the need to modify the transmission channel. The most important advantage is that signal power at multiple wavelengths can be simultaneously boosted by a single amplifier—a task that would require a separate electronic repeater for each wavelength. The state-of-the-art of optical amplifiers are erbium-doped fiber amplifiers (EDFAs), which are formed by doping the glass fiber host with erbium ions. In EDFAs, gain by stimulated emission at wavelengths in the vicinity of 1.55 μm is obtained, by optical pumping using light at either 1.48-μm or 0.98-μm wavelengths. EDFAs are lumped amplifier, this means that the signal, monotonically attenuated in the fiber span, is amplified at a point of the EDFA location to recover the original level, before entering the next fiber span. The distance between amplifiers is determined by the span loss, by the limit imposed from the power maximum value allowed in the fiber, without inducing nonlinear effects, and by the minimum acceptable power, avoiding a degradation of the signal-to-noise-ratio. EDFAs are compact and highly efficient devices with high gain and low noise [1].

Optical multiplexing allows to combine multiple optical signals into one to make full use of the immense bandwidth potential of an optical cannel. The basic idea is to divide the huge bandwidth of optical fiber into individual channels of lower bandwidth, so that multiple access with lower-speed electronics is achieved. The large gain bandwidth of EDFAs has enabled the realization of dense wavelength-division multiplexed systems, in which terabit/sec data rates have been demonstrated. EDFAs operate in the C-band (from 1530 to 1565 nm) while extended-band EDFAs can provide gain in the L-band (from 1565 to 1625 nm). However, the capacity of most fibers can be extended by opening up the S band (from 1480 to 1530 nm), and the S+ band (from 1430 to 1480 nm). In addition, we note that earlier transmission systems were deployed in the 1310-nm band, which can stretch between 1280 and 1340 nm, while there could be also a 1400-nm band, which is only useful in new fibers using special drying techniques to reduce the water peak absorption around 1390 nm. Definitely, the available communication range could span from 1270 to 1650 nm, corresponding to about 50 THz bandwidth, but the current strong limitation is that a great part of this range is inaccessible by EDFAs [1–4].

In optical communications, fiber nonlinearities are the basis of a number of devices such as amplifiers, switching and logic elements. The nonlinear effects in optical fiber can be divided in two class. The first one is due to the Kerr-effect, i.e., intensity dependence of refractive index of the medium, which manifests itself in three different effects: Self-Phase Modulation (SPM), Cross-Phase Modulation (CPM) and Four-Wave Mixing (FWM), depending upon the type of input signal. The second one is due to inelastic-scattering phenomenon, in which the optical field transfers part of its energy to the nonlinear medium. Such an inelastic-scattering phenomenon can induce stimulated effects such as Stimulated Brillouin-Scattering (SBS) and Stimulated Raman-Scattering (SRS). We note that each type of stimulated scattering process can be used as a source of gain in the fiber. In both phenomena, at high power level, the intensity of scattered light grows exponentially if the incident power exceeds a certain threshold value. Raman amplification is more useful, because of the relatively large frequency shift and the broader-gain bandwidth. The basic difference between them is that in the case of Brillouin the interaction is between the guided optical wave and low-frequency acoustic phonons, while in the case of Raman the interaction is between the guided wave and high-frequency optical phonons. Another fundamental difference is that SBS in optical fibers occurs only in the backward direction whereas SRS can occur in both directions [2–4].

In long-range high-datarate systems, involving high optical power levels and signals at multiple wavelengths, the consequences of fiber nonlinearities can be: (1) the generation of additional signal bandwidth within a given channel, (2) modifications of the phase and shape of pulses and can cause spectral broadening, which leads to increased dispersion, (3) the generation of light at other wavelengths at the expense of power in the original signal, and (4) crosstalk between signals at different wavelengths and polarizations. The first two arise from self-phase modulation, while the third and fourth effects arise from stimulated Raman or Brillouin scattering or four-wave mixing. These last can be used when generation or amplification of additional wavelengths is desired. We note that in order to obtain an error-free system performance, due to the finite sensitivity of the optical receiver, the signal should have a high-enough level. On the other hand, the nonlinear effects are proportional to the product of the signal power, P, and the transmission distance L. This means that by increasing the signal level, the distance and the transmission bit rate (speed), all the problems—dispersion, noise, and nonlinearity in the fiber are increased, too. It is worth noting that the signal distortions is an issue when the nonlinearity is involved, because it can couple a number of detrimental effects together, such as dispersion, noise, polarization mode dispersion, polarization-dependent loss/gain, etc. [2–4].

SRS is a nonlinear process, observed for the first time in 1962, which lies at the heart of fiber Raman amplifiers and fiber Raman lasers (FRAs and FRLs). Direct fiber optical amplification obtained by SRS, called Raman amplification, was demonstrated in the early 1970s. In Raman amplification, a power transfer from pump (s) to information carrying signals (usually described as probes) can occurs, if there is a sufficient pump power within the fiber. In principle, when optical pump energy is added along with signals in ordinary optical fibers, optical amplification can take place, providing low-noise, flat and wideband signal gain. The more attractive advantage is that FRAs, operating in a signal band outside the EDFA bands, could open new transmission windows in the future [2–4].

High-power fiber lasers have had a significant development, achieving output powers of multiple kilowatts from a single fiber. Due to its inherent material advantages, Ytterbium has been the primary rare-earth-doped gain medium, so fiber lasers are largely confined to its narrow emission wavelength region. It is worth noting that because of its Raman-shifted out, SRS is a workable method for generating coherent radiation at new frequencies. Taking advantages of the technology of high-power lasers, which serve as their input, FRLs can lead to conversion to wavelengths higher than the starting wavelength, using a series of Raman–Stokes shifts. In fiber-lasers, the tight spatial confinement of pump light, which is maintained over relatively large distances, significantly lowers threshold pump powers down to practical levels and enables continuous-wave operation. The main advantage of Raman laser is that essentially any laser wavelength can be achieved from the ultraviolet to the infrared with a suitable choice of the pump wavelength, providing that wavelengths are within the transparency region of the material and sufficiently high nonlinearity and/or optical intensity are reached. For this reason, currently, FRLs are the only wavelength scalable, high-power fiber laser technology that can span the wavelength spectrum [2–4].

Due to the extent of subject, a comprehensive review including mathematical [5–8] and physics aspects of SRS [9–14], architecture of fiber system [15–19] and laser systems would be impossible to realize. Therefore, the aim of our paper is to provide an overview of the field, emphasizing physical effects and working principles of fiber optics amplifier and laser based on SRS. This approach could have the advantage to provide a quick look of the state of art and to allow to scientists, who are new to Raman amplification, to go into the field. In addition, a comprehensive list of references is also provided for readers who wish to pursue any of the topics in more detail.

The paper is organized as follows. In next paragraph, for the sake of completeness an essential theoretical background is reported. In Section 3, Raman amplifiers are described, which are generally divided into two categories, namely, distributed and lumped amplifiers. Of course, the two classes can be combined to form a hybrid amplifier, where the discrete FRA can be replaced by an EDFA [20]. In Section 4, Raman lasers in fiber are introduced, which can be classified into two general categories. In the former, the wavelength is shifted by one Raman–Stokes shift, while, in the latter, called cascaded Raman laser, the wavelength is shifted by multiple Raman–Stokes shifts. In the last section, a brief introduction about Raman soliton laser is reported, too.
