The main goal of designing a multistage fiber amplifier is to achieve higher performance for long-haul optical transmission systems. In this paper, multistage triple-pass (TP) TDFAs are represented. Three types of TP-TDFA are modeled, optimized, and compared in the S-band for next-generation fiber optical communication systems. First, to examine the performance of these 3 types of TP-TDFAs, signal power, pump power, thulium-doped fiber (TDF) length, and thulium ion concentration were optimized in the input signal at the 1469 nm wavelength, which gives the highest gain value in the S-band, respectively. Second, the input signals in the 1444–1499 nm band gap were amplified by using these specified optimized values in all designs. Gain and noise figure values of all three types were obtained and analyzed by plotting separately for easy comparison. At the end of the study, all results were evaluated and all types were compared to determine the best TP-TDFA.
Research Article
Modelling and Optimizations of Triple-Pass TDFAs for Next-Generation Fiber Optical Communication Systems
https://doi.org/10.21203/rs.3.rs-3101449/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
thulium-doped fiber amplifier (TDFA)
multi-stage
triple-pass
optimization
In recent years, communication traffic has grown tremendously with high bandwidth applications such as voice-over IP, video conferencing, online gaming, social networks, and high-definition video streaming. While these new-generation technologies, the usage of which is increasing, require an increase in transmission speed and capacity, they also necessitate a change in the communication infrastructure. Fiber optic communication systems and optical amplifiers have a great share in meeting all these requirements [2–5]. The studies carried out to date are on Erbium Doped Fiber Amplifier (EDFA) with high efficiency in the C and L bands [6–10]. Since the C and L bands may be insufficient, optical amplifiers that can operate effectively in the S-band have been researched. Thulium-doped fiber amplifier (TDFA) and fiber Raman amplifier (FRA) are optical amplifiers with high efficiency in the S-band [11]. Due to some limitations of FRA, such as high pump power and multipump usage, TDFAs are seen as a better alternative. [12–14]. To increase the gain of the TDFA, in some studies, the amplifier parameters were changed [15–20], while in others, different pumping configurations were used to excite all thulium ions [21–26]. In addition, in some research, the gain has been increased by adding a mirror or an optical circulator to the basic TDFA design with double-pass configurations [27–31]. In some of them, these designs are called double-pass; in others, they are called double-stage [30–31].
Multistage fiber amplifiers have fundamentally changed the architecture of long–haul optical communication systems and networks. Although studies of multipass optical amplifiers are mostly focused on double-pass amplifiers, there are also studies of triple-pass (two-stage triple-pass and three-stage triple-pass) EDFAs [32–34]. Akhter and Ibrahimy modeled all possible EDFA configurations, examined their characteristics, and compared them with each other [32, 33]. In their study, S.W. Harun et al. used and compared two types of two-stage triple-pass EDFAs [34]. On the other hand, considering the S-band, no study has been done on the TP-TDFA other than the research we have done before [30].
In this paper, multistage TP-TDFA structures are used, optimized, and compared for next-generation fiber optical communication systems. There are three types of TP-TDFAs. The performance of each structure was analyzed with signal power, pump power, TDF length, and thulium ion concentration. All these parameters have been optimized, and the results have been examined. At the end of the study, all types were compared to determine the best TP-TDFA.
2.1. Simulation setups
All types of TP-TDFA designs are shown in Fig. 1. The first type, composed of one-stage single-pass TDFA and one-stage double-pass TDFA, is connected serially to obtain triple-pass TDFAs. Figure 1a shows the schematic diagram of Type 1 TP-TDFA. The output signal of the first TDFA is the input signal for the second TDFA. In addition to TDFs, two pump couplers (PCs), an optical circulator (OC), a power splitter (PS), and a mirror are used in Type 1 and Type 2 TP-TDFA designs. A PC is used to combine the pump laser signal and input signal into the TDF. A circulator (CIR) is used to ensure separation between the input and output signals. A PS is used to split the power from two or more ports, and a mirror is used to repass the amplified signal through the TDF.
In Type 2 TP-TDFA, there is a single-pass TDF and a double-pass TDF, as in Type 1. In addition, in Type 2, the signal goes through the double pass TDF and then the single pass TDF, whereas in Type 1, it is vice versa. As a result, the output signal power of the double-pass TDF is the input signal power of the single-pass TDF, as shown in Fig. 1b. As in Type 1, in this setup, the pump power is divided into two by PS.
Figure 1c shows the Type 3 TP-TDFA design. It consists of a serial combination of three single-pass TDFs. The output signal power of the first TDF is the input signal power of the second TDF. The output signal power of the second TDF is the input signal power of the third TDF. Since the signal passing through the TDFs in total is amplified three times, it is called TP-TDFA. Different from the others in this setup, the pump power is divided into three by PS.
3.2. Optimization steps
The cases to be tried in this study are listed in Table 1. In the first case, each design is simulated separately by feeding at different input powers changed from − 40 to 0 dBm, and the most efficient input power is determined.
Table 1. Input powers, pump powers, TDF lengths, and TDF ion concentration of all designs used in the simulations.
In the second case, using the input power determined in the first case, each design is separately simulated by increasing 250 mW each time at pump powers between 250 and 3500 mW. Gain and noise figure spectra are created using the obtained values, and the most efficient pump power is determined.
In case 3, using the values determined in the first and second cases, the TDF 1 lengths of Type 1 and Type 2 are optimized separately. Then, using the optimum TDF 1 length, these are simulated by increasing 0.5 m each time at TDF 2 length between 3 and 7 m. Gain and noise figure spectra are created using the obtained values, and the most efficient TDF 2 length is determined. Then, again, using the values determined in the first and second cases, the TDF 1 length of Type 3 is optimized. Using the optimum TDF 1 length, it is simulated by increasing 0.5 m each time at a TDF 2 length between 3 and 7 m. After that, the most efficient TDF 2 length is determined. Then, using the optimum TDF 1 and TDF 2 lengths, it is simulated by increasing 0.5 m each time at a TDF 3 length between 3 and 7 m. Finally, gain and noise figure spectra are generated using the obtained values, and the most efficient TDF 3 length is determined.
In case 4, the ion concentration of TDF 1 is called ion concentration 1, the ion concentration of TDF 2 is called ion concentration 2, and the ion concentration of TDF 3 is called ion concentration 3. Using the values determined in the first, second, and third cases, ion concentrations 1 of Type 1 and Type 2 are optimized separately. Then, using the optimum ion concentration 1, Type 1 and Type 2 were simulated by increasing 2.5e + 24 m-3 each time at ion concentration 2 between 5e + 24 and 30e + 24 m-3. Gain and noise figure spectra of ion concentration 2 are produced, and the optimum value is determined. Then, again, using the values determined in the first and second cases, ion concentration 1 of Type 3 is optimized. Second, using the optimum ion concentration 1, it is simulated by increasing 5e + 24 m-3 each time at ion concentration 2 between 5e + 24 and 30e + 24 m-3. After that, the most efficient ion concentration 2 is determined. Then, using the optimum ion concentrations 1 and 2, it is simulated by increasing 5e + 24 m-3 each time at ion concentration 3 between 5e + 24 and 30e + 24 m-3. Finally, gain and noise figure spectra are generated using the obtained values, and the optimum ion concentration 3 is determined. The parameters used in all simulations are listed in Table 2.
In the last and fifth cases, using all values determined in the previous cases, all designs are simulated, and gain and noise figure spectra are produced using the obtained values in the S bands.
Table 2. Parameters used in the Simulation
In TDFAs, the gain depends on the ion density and reverse deposition of thulium ions at different energy levels [17]. The energy band diagrams of the thulium and Tm emission cross-section spectra are shown in Fig. 2. The energy levels 0, 1, 2, 3, and 5 are named 3H6, 3F4, 3H5, 3H4, and 1G4, respectively.
The rate equation of the different energy levels of thulium was proposed by P. Peterka et al. as follows [35]:
\(\frac{{d{N_2}}}{{dt}}={N_0}({W_{01}}+{W_{02}}) - {N_1}({W_{10}}+{W_{13}}+{W_{14}}+A_{1}^{{nr}}+A_{{10}}^{r})\) \(+{N_3}({W_{31}}+{W_{32}}+A_{3}^{{nr}}+A_{{32}}^{r}+A_{{31}}^{r})+{N_5}(A_{{51}}^{r}+A_{{52}}^{r})\) (1)
\(\frac{{d{N_3}}}{{dt}}={N_0}({W_{03}})+{N_1}({W_{13}}+{W_{14}}) - {N_5}(A_{5}^{{nr}}+A_{{52}}^{r}+A_{{53}}^{r})\) \(- {N_3}({W_{35}}+{W_{32}}+{W_{31}}+{W_{30}}+A_{3}^{{nr}}+\sum\limits_{{j=0}}^{2} {A_{{3j}}^{r}} )\) (2)
$$\frac{{d{N_5}}}{{dt}}={N_0}({W_{05}})+{N_3}{W_{35}}+{N_5}({W_{50}}+A_{5}^{{nr}}+\sum\limits_{{j=0}}^{4} {A_{{5j}}^{r}} )$$
3
$${N_t}={N_0}+{N_1}+{N_3}+{N_5}$$
4
where the N0, N1, N3, and N5 values represent the population densities of the 3H6, 3F4, 3H4, and 1G4 levels, respectively. Ni and Wij variables are functions of r, φ, z positions.
Gain and noise figure values were obtained by simulating all TP-TDFA designs separately. These values are shown in two separate graphs as gain and noise figures for easy comparison.
In the first case, the input signal at 1469 nm wavelength, pumps at 1050 nm wavelength with 1000 mW power, 5 m length, and 20e+ 24 ion concentration for each TDF were applied to all designs. Each design was simulated separately by feeding at different input powers from 0 dBm to -40 dBm. Figure 3a shows the spectrum of the gain according to the input power, and Fig. 3b shows the spectrum of the noise figure according to the input power. According to Fig. 3a, -30 dBm was determined to be the most efficient input power. While the noise figure values of Types 1 and 2 are close to each other, Type 3 is half of the others.
In the second case, using − 30 dBm, which is determined in the first case, each design was separately simulated by increasing 250 mW each time at pump powers between 250 and 3500 mW. Figure 4a shows the spectrum of the gain according to the wavelength, and Fig. 4b shows the spectrum of the noise figure according to the wavelength. Figure 4a shows that the most efficient pump power was 1750 mW for Types 1 and 2, while it was 2750 mW for Type 3. Since the pump power is divided into three, it does not reach saturation until higher values.
When the noise figure spectra in Fig. 4b are considered, it is seen that the noise figures of Type 2 are higher. This increase is attributed to the higher counterpropagating ASE at the input part of the amplifier.
In case 3, using 1750 mW, which was determined as the optimum pump power for Type 1 and Type 2 in the second case, TDF 1 of Type 1 and Type 2 were optimized separately by using the software, and it was 5.2 m for both of them. Then, using TDF 1: 5.2 m, Type 1 and Type 2 were simulated by increasing 0.5 m each time at TDF 2 lengths between 3 and 7 m. Figure 5a shows the spectrum of the gain according to the TDF 2 length, and Fig. 5b shows the spectrum of the noise figure according to the TDF 2 length. Figure 5a shows that 5.5 m is the optimum value of TDF 2 length for Type 1 and 5 m is the optimum value of TDF 2 for Type 2. These values continued to be used in the next case. In addition, it is seen in Fig. 5b that the noise figure values of these types are close to each other.
Then, again, using 2750 mW, which was determined as the optimum pump power for Type 3 in the second case, TDF 1 of Type 3 was optimized by using the software, and it was 6.3 m. Using TDF 1: 6.3 m, it was simulated by increasing 0.5 m each time at TDF 2 length between 3 and 7 m. Figure 6a shows the spectrum of the gain according to the TDF length, and Fig. 6b shows the spectrum of the noise figure according to the TDF length. As shown in Fig. 6a, the most efficient TDF 2 length is 6 m. Then, using the optimum TDF 1: 6.3 m and TDF 2: 6 m lengths, it was simulated by increasing 0.5 m each time at TDF 3 length between 3 and 7 m. Finally, the obtained values were added in Fig. 6. The most efficient TDF 3 length is 3.5 m. Additionally, since the third TDF is close to saturation, the noise figure values are lower.
In case 4, using TDF 1: 5.2 m and TDF 2: 5.5 m, which were determined to be the optimum lengths for Type 1 and Type 2 in the third case, ion concentration 1 of Type 1 and Type 2 was optimized separately and was 18e+ 24 m− 3 for both. Then, using ion concentration 1: 18e+ 24 m− 3, Type 1 and Type 2 were simulated by increasing 2.5e+ 24 m− 3 each time at ion concentration 2 between 5e+ 24 and 30e+ 24 m− 3. Figure 7a and 7b shows the gain and noise figure spectra according to ion concentration 2. Figure 7a shows that 20e+ 24 m− 3 is the optimum value of ion concentration 2 for Type 1 and 22.5e+ 24 m− 3 is the optimum value of ion concentration 2 for Type 2. These values continued to be used in the next case.
Then, again, using TDF1: 6.3 m, TDF 2: 6 m, and TDF 3: 3.5 m, which were determined to be the optimum TDF lengths for Type 3 in the third case, ion concentration 1 of Type 3 was optimized by using the software and was 16e+ 24 m-3. Using ion concentration 1: 16e+ 24 m-3, it was simulated by increasing 2.5e+ 24 m-3 each time at ion concentration 2 between 5e+ 24 and 30e+ 24 m-3.
Gain and noise figure graphs in Fig. 8 were created using the obtained values. As shown in Fig. 8a, 22.5e+ 24 m-3 is the optimum value of ion concentration 2 for this type. Then, using the optimum ion concentration 1: 16e+ 24 m-3 and ion concentration 2: 22.5e+ 24 m-3, it was simulated by increasing 2.5e+ 24 m-3 each time at ion concentration 3 between 5e+ 24 and 30e+ 24 m-3. Finally, the obtained values were added in Fig. 8. The most efficient ion concentration 3 is 17.5e+ 24 m-3.
Additionally, since the third TDF is close to saturation, it is seen that the noise figure values are slightly lower. The noise figure is 3.45 dB at the 17.5e+ 24 m-3 ion concentration. When noise figures of other ion concentrations are taken into account, this value is an average value.
All optimization studies thus far have been made for the input signal at 1469 nm wavelength, where the highest gain is obtained in the S-band. The values obtained for this signal listed in Table 3 were applied to the input signals in the 1444–1499 nm wavelength range (S-band).
Table 3. Optimized values were determined in the all cases.
The pump power values of Type 1 and Type 2 are the same, and the TDF lengths and ion concentrations are quite close to each other. TDF 2 of Type 1 is slightly longer, while the ion concentration 2 of Type 2 is slightly higher. In Type 3, the optimum fiber lengths are longer. For easier comparison, the obtained gain and noise graphs of all types are shown in Fig. 9.
As shown in Fig. 9a, the gain values of the models optimized for the signal in the 1469 nm wavelength are 44.64 dB for Type 1, 45.67 dB for Type 2 and 43.95 dB for Type 3. While the highest gain and noise figure values belong to Type 2, the lowest gain and noise figure values belong to Type 3. The gain values of Type 2 are higher than those of Type 1. As a result, the gain value of the signal amplified with a single pass after a double pass is higher compared to the gain of a double-pass amplified signal used after a single pass.
The highest noise figure values occurred in the Type-2 design. This increase is attributed to the higher counterpropagating ASE at the input part of the amplifier. This reduces the population inversion at the input part of TDF and afterward increases the noise figure [29].
Three types of TP-TDFA have been presented in this research. The input power, pump power, TDF length, and ion concentration were optimized for all three types. Then, a performance comparison was made over the gain and noise figure values in the S-band for all types using the determined optimized values. -30 dBm was determined as the most efficient input power for all types. The most efficient pump power was 1750 mW for Types 1 and 2, while it was 2750 mW for Type 3. Since the pump power is divided into three, it does not reach saturation until higher values. While the optimum TDF 1 is 5.2 m for both Type 1 and Type 2, the TDF 2 is 5.5 m for Type 1 and 5 m for Type 2. The TDF values of Type 3 are 6, 3.6, and 3.5, respectively. While the optimum ion concentration 1 is 18e+ 24 m-3 for both Type 1 and Type 2, the optimum ion concentration 2 is 20e+ 24 m-3 for Type 1 and 22e+ 24 m-3 for Type 2. The ion concentration values of Type 3 are 16e+ 24 m-3, 22.5e+ 24 m-3 and 17.5e+ 24 m-3.
Examining the gain performance of all optimized TP-TDFA types, it can be concluded that Type 2 is the best. Nevertheless, its noise figure values are higher than those of other types because of the higher counterpropagating ASE at the input part of the amplifier and seem to be a disadvantage. On the other hand, considering the noise figure graphs in Fig. 9b, Type 3 has less than half of the noise figure values of the other types and seems to be the best choice for systems requiring low noise. If a comparison is made by looking at the results obtained, two-stage triple-pass models (Type 1 & 2) are more useful and economical than the three-stage triple-pass model (Type 3). Using these two types, whose gain values are high and close to each other, can increase the repeater distance in long-haul next-generation fiber optical communication systems.
Ethical Approval
Not applicable.
Competing interests
Both authors declare that they have no financial interests.
Authors' contributions
Ayten Dincer stimulated all designs, drew the figures and created the tables. Ayten Dincer and Murat Yücel modelled all designs, wrote and reviewed the main manuscript text together.
Funding
There is not any Funding.
Availability of data and materials
Optisystem Software 30-day Evaluations is used.
- Vargo D, Zhu L, Benwell B and Yan Z(2021) Digital technology use during COVID-19 pandemic: a rapid review. Human Beh and Emerg Tech 3:13-24. https://doi.org/10.1002/hbe2.242
- SunY, Srivastava AK, Zhou J, and Sulhoff JW(1999) Optical fiber amplifiers for WDM optical networks. Bell Labs Tech J 4:187-206. https://doi.org/10.1002/bltj.2153
- Singh S, Singh A, Kaler RS (2013) Performance evaluation of EDFA, RAMAN and SOA optical amplifier for WDM systems. Optik 124:95-101. https://doi.org/10.1016/j.ijleo.2011.11.043
- Ding X, Wu G, Zuo F, Chen J (2019) Bidirectional optical amplifier for time transfer using bidirectional WDM transmission. Optoelect Letters 15:401–405. https://doi.org/10.1007/s11801-019-9025-1
- Malik D, Kaushik G, Wason A (2018) Performance evaluation of hybrid optical amplifiers in WDM system. J of Opt 47: 396–404. https://doi.org/10.1007/s12596-018-0470-1
- Yücel M, Göktas HH, Akkaya G (2012) Three stage L-band EDFA optimization. 2012 20th Signal Processing and Communication Applications 1-4. https://doi.org/10.1109/SIU.2012.6204451
- Yucel M, Aslan Z(2013) The noise figure and gain improvement of double-pass C-band EDFA. Mic and Opt Tech Let 55:2525-2528. https://doi.org/10.1002/mop.27854
- Bebawi JA, Kandas I, El-Osairy MA, Aly MH(2018) A comprehensive study on EDFA characteristics: temperature Impact. Applied Science 8:1640. https://doi.org/10.3390/app8091640
- Anwar NM, El-Sahn Z, El-Badawy EA, Aly MH(2020) A few mode EDFA with different pumping schemes: performance evaluation. Opt and QuanT Elect 52:109. https://doi.org/10.1007/s11082-020-2226-9
- Akcesme O, Yucel M, Burunkaya M (2021) The design and implementation of a software based gain control for EDFAs used in long-haul optical networks. Opt 239. https://doi.org/10.1016/j.ijleo.2021.166850.
- Kani J, Jinno M(1999) Wideband and flat-gain optical amplification from 1460 to 1510 nm by serial combination of a thulium-doped flouride fiber amplifier and fiber Raman amplifier. Elect Let 35:1004-100. https://doi.org/10.1049/el:19990708
- Zhu S, Pidishety S, Feng Y, Demas J, Sidharthan R et al(2018) Multimode-pumped Raman amplification of a higher order mode in a large mode area fiber. Opt Exp 26:23295-23304. https://doi.org/10.1364/OE.26.023295
- Husein AHM, El-Nahal FI(2013) Optimizing thulium doped fiber amplifier (TDFA) gain and noise figure for S-band 16x10 Gb/s WDM systems. Opt 124:4052-4057. https://doi.org/10.1016/j.ijleo.2012.12.057
- Li Z, Heidt AM, Daniel JMO et al (2013) Thulium-doped fiber amplifier for optical communications at 2 µm. Opt Exp 21:9289-9297. https://doi.org/10.1364/OE.21.009289
- Aozasa S, Masuda H, Shimizu M(2006) S-band Thulium-doped fiber amplifier employing high Thulium concentration doping technique. J of Lightwave Tech 24:3842-3848. https://doi.org/10.1109/JLT.2006.881480
- El-Nahal FI, Hakeim A, Husein N(2012) Thulium doped fiber amplifier (TDFA) for S-band WDM systems. Open J of Appl Sciences 2. https://doi.org/10.4236/ojapps.2012.24B002
- Singh R, Singh ML(2016) Evaluating the effect of doping concentration and doping radius on the gain of silica-based Thulium doped fiber amplifier. 2016 Inter Conf on Recent Adv and Innov in Eng 1-5. https://doi.org/10.1109/ICRAIE.2016.7939469
- Kaur I, Gupta N (2014) Performance analysis of TDFA using 1050 nm pumping power for 1479 nm-1555 nm Range. Inter J of Com and Digital Sys 3. https://doi.org/10.12785/ijcds/030306
- Dissanayakel KPW, Emami SD et al(2013) Design and performance of an S-band Thulium-doped modified silica fiber amplifier. 2013 IEEE 4th Inter Conf on Phot (ICP) 288-290. https://doi.org/10.1109/ICP.2013.6687141
- Yücel M, Dincer A(2020) Gain and noise figure analysis of single and two stage Thulium-doped fiber amplifiers in S-band. Duzce Uni J of Sci and Tech 8:2062-2072. https://doi.org/10.29130/dubited.730502
- Singh R, Singh ML, Kaur B(2012) A novel triple pump 1050 nm+1400 nm+800 nm pumping scheme for Thulium doped fiber amplifier. Opt 123:1815-1816. https://doi.org/10.1016/j.ijleo.2011.11.102
- Yam SH, Akasaka Y, Kubota Y, Inoue H, Parameswaran K(2005) Novel pumping schemes for flouride-based Thulium-doped fiber amplifier at 690 and 1050 nm (or 1400 nm). IEEE Phot Tech Let 17:1001-1003. https://doi.org/10.1109/LPT.2005.846749
- Roy F, Bayart D, Sauze A, Baniel P(2001) Noise and gain band management of Thulium-doped fiber amplifier with dual-wavelength pumping schemes. IEEE Phot Tech Let 13: 788-790. https://doi.org/10.1109/68.935804
- Gomes ASL, Carvalho MT, Sundheimer ML, Bastos-Filho CJA et al (2003) Characterization of efficient dual-wavelength (1050+800 nm) pumping scheme for thulium-doped fiber amplifiers. IEEE Phot Tech Let 15:200-202. https://doi.org/10.1109/LPT.2002.806847
- Singh R, Singh ML (2017) Performance evaluation of S-band Thulium-doped silica fiber amplifier employing multiple pumping schemes. Opt 140:565-570. https://doi.org/10.1016/j.ijleo.2017.04.098
- Jamalus MSK, Yusoff NM, Sulaiman AH(2019) Simulation of dual stage Thulium-doped amplifier using pump power distribution technique. Indonesian J of Elect Eng and Com Science 15:1203-1211. https://doi.org/10.11591/ijeecs.v15.i3.pp1203-1211
- Emami SD, Harun SW, Abdul-Rashid HA, Daud SA, Ahmad H(2009) Optimization of the 1050 nm pump power and fiber length in single-pass and double-pass Thulium doped fiber amplifiers. Prog in Electr Research B 14:431-448. https://doi.org/10.2528/PIERB09022503
- Bastos-Filho CJA, Martins-Filho CF (2003) 38 dB from a double-pass single-pump Thulium doped fiber amplifier. The 2003 SBMO/IEEE MTT-S Inter Mic and Optoelect Conf 125-128. https://doi.org/10.1109/IMOC.2003.1244844
- Emami SD, Harun SW, Abdul-Rashid HA, Daud SA, Ahmad H, Abd-Rahman F, Ghani ZA(2010), “A theorical study of double-pass Thulium-doped fiber amplifiers. Opt 121:1257-1262. https://doi.org/10.1016/j.ijleo.2009.01.017
- Yucel M, Dincer A(2022) The gain and noise figure comparison of single-pass, double-pass, and dual-stage triple-pass Thulium-doped fiber amplifiers in S-band with Thulium-doped fiber optimization. Mic and OpT Tech Letters 64:1863- 1870. https://doi.org/10.1002/mop.33364
- Mohammed D, Hammadi A(2021) Performance analysis on double-pass thulium-doped fiber amplifier for 16-channel WDM system at S-band. J of Opt Com https://doi.org/10.1515/joc-2020-0313
- Akhter F, Ibrahimy M, Naji A, Siddiquei H(2012) Modeling and characterization of all possible triple pass EDFA configurations. Inter J of the Physical Sci 7. https://doi.org/10.5897/IJPS12.172
- Akhter F, Naji AW, Ibrahimy MI, Siddiquei HR(2012) Characterization of triple pass Erbium-doped fiber amplifiers. 2012 Int Conf on Com and Commun Eng (ICCCE) 462-466. https://doi.org/10.1109/ICCCE.2012.6271230.
- Naji AW, Ibrahim HFH, Ahmed B, Harun SW, Mahdi MA(2010) Theoretical analysis of triple-pass Erbium-doped fiber amplifier. Int Conf on Com and Commun Eng (ICCCE'10) 1-4. https://doi.org/10.1109/ICCCE.2010.5556844
- Peterka P, Faure B, Blanc W, Karasek M, Dussardier B(2004) Theoretical modeling of S-band Thulium-doped silica fibre amplifiers. Opt and QuanT Elect 36:201–212. https://doi.org/10.1023/B:OQEL.0000015640.82309.7d
No competing interests reported.