2. J Opt
1 3
than SOA [20]. The upstream modulation and downstream
amplification are combined in RSOA that provides higher
gain with less polarization dependency [3], lower injection
currents [3], lower noise figure [20], higher modulation
linearity, and lower sensitivity than the other amplifiers.
Our design is based on RSOA only. To the best of the
author’s knowledge, the design of ROM using RSOA has
not been published till now. The aforementioned beneficial
property of RSOA motivates us to design read-only mem-
ory (ROM), and also, this design can be reduced and inte-
grated into an optical scheme using the photonic band gap
(PBG) system [7, 21], so it can be used in nano-photonic
devices. This design is numerically simulated on the basis
of gain dynamics of RSOA in MATLAB (2018a) software.
The operating data rate which is the inverse of bit period
(T) = 1000/(data rate) = 8 ps @125 Gb/s, is considered
for this simulation.
Reflective semiconductor optical amplifier (RSOA)
functions as a switch
An anti-reflecting (AR) coating and a highly reflect-
ing coating (HR) are placed in the two facets of the
RSOA [22, 24]. RSOA has a high gain with low noise [20].
It is very popular in low-cost wave division multiplexing
(WDM) [23–25] and also used in passive optical network
(PON) [25, 26]. When probe (weak) and pump (strong)
signals are injected into RSOA, it saturates and gain is
where h(t) expressed as Eq. [27]
where Pin(t) is a input power and Es is a saturation energy
of the RSOA and g0 is the gain coefficient. The unsaturated
gain [3, 27] is defined as
where L and 𝛼d are the active length and internal loss
of RSOA. Using Eqs. 2 and 3, the following equation is
obtained
Here, Ec(t) and Es stand for the control pulse, and sat-
uration energies of the RSOA. Its schematic diagram is
given in Fig. 1. For input pump and probe signals, we take
a soliton pulse train as
(1)
G(t) = exp[2h(t)]
(2)
dh(t)
dt
=
g0L − h(t)
𝜏e
−
Pin(t)
Es
[exp h(t) − 1]
(3)
G0 = exp(Lg0 − L𝛼d)
(4)
h(t) = ln
[
1 −
(
1 −
1
G0
)
exp
(
−
Ec(t)
Es
)]
where soliton peak power is represented by,
All the values of constants are given in Table 2 and
represents the total control pulse energy. When pump and
probe signals are passed through the RSOA, its output signal
is defined as
When only probe signal is passed through the RSOA, its
output signal is defined as
The stability of the RSOA is also an important issue. If the
biasing electrical power fluctuates then the outputs of RSOA
may trouble. Therefore, since the proposed scheme is based
on RSOA, this scheme may also be in trouble. To maintain the
stability of the RSOA, a power source is needed that can flow
a constant current to the RSOA. The data rates that RSOA
can support in these applications are limited by the RSOAs
slow direct modulation speed, which in turn is constrained
by RSOAs finite modulation bandwidth as low as very few
GHz [28] according to evidence based on experimental meas-
urements [29–31]. A promising solution, which owing to its
clear concept, feasible implementation and passive nature has
gained wide popularity across the research community, is to
resort to frequency discrimination and equalization [28, 32].
(5)
P(t) =
n=N
∑
n=1
anA,BPsolitonsech2
(
1.763
(t − nT)
𝜏fwhm
)
Psoliton =
(
1.763
2𝜋
) Aveff 𝜆3
D
n2C𝜏2
fwhm
.
Ecp(t → ∞) = Psoliton × 𝜏fwhm = Ec
(6)
Pout = RG(t)Pin
(7)
Pout = RG0G0Pin
High
reflecve
coang
An reflecve coang
Filter
Pump Signal
Probe Signal
Output
Signal
Fig. 1 Schematic diagram of RSOA
3. J Opt
1 3
The operational principle of the RSOA‑based
2‑to‑4 line decoder
Here, 2-to-4 line decoder is designed using RSOAs. The
block diagram of decoder is given in Fig. 2. This design be
made up by five RSOAs, viz., RSOA1, RSOA2, RSOA3,
RSOA4, and RSOA5, given in Fig. 3. The A and B are the
input signals which are divided by the beam splitters in a
proper proportion to serve the purpose of pump and probe
signals. The outputs of all the RSOAs are obtained with the
help of a circulator and filters. Here, power of signals ‘A’ and
‘B’ is divided using beam splitters. The power of signal ‘A’
is divided into three parts: 45%, 45%, and10%. This 45% of
the power serves as the pump signal for RSOA1 and another
45%serves as the pump signal for RSOA3, while the remain-
ing10% serves as the probe signal for RSOA2. The power of
signal ‘B’ is divided into 70% and 30%. This 30% of power
is divided into three parts, each of 10% power, and these
three act as the probe signal power of RSOA1, RSOA2, and
RSOA4. 70% of the power of signal B’ serves as an RSOA2
pump signal. Now, the operation of a 2-to-4 line decoder is
explained, the following cases;
Case-1: When input signals A and B are applied as‘0’ then
probe and pump both the signals are absent for RSOA1,
RSOA2, RSOA3, and RSOA4. As these RSOAs are not
working so the outputs are ‘0’. In the case of RSOA5, the
pump signal is absent but a continuous probe signal (X) is
present so the output is obtained from RSOA5. Finally, the
outputs from DO1, DO3, DO4 are ‘0’ and for DO2, it is ‘1’.
2-to-4
line
DECODER
A
B
DO1
DO2
DO3
DO4
Fig. 2 Block diagram of 2-to-4 line decoder
V
O
A
A
B
DO1
DO2
DO3
DO4
RSOA4
RSOA1
RSOA2
RSOA3
RSOA5
X
B.S-4
B.S-3
B.S-2
B.S-1
B.S-5
B.S-7
B.S-6
B.S-8
Fig. 3 Diagram of 2-to-4 line decoder using RSOA
4. J Opt
1 3
Case-2: When input signals A and B are applied as‘0’ and
‘1’, respectively, then RSOA1, RSOA3 are in working con-
dition due to the absence of pump signal and presence of
probe signal. The output of RSOA3 acts as a pump signal
of RSOA4. Due to the presence of pump and probe signals,
RSOA4 does not yield output. The output of RSOA1 acts
as a pump signal of RSOA5 and its probe signal is also
present so the output of RSOA5 is ‘0’. Therefore, the final
output is obtained as ‘1’ at the DO1, whereas the other
outputs are zero.
Case-3: When input signals A and B are applied as‘1’
and ‘0’, respectively, then RSOA1, RSOA3 do not work
due to the absence of probe signal. The pump signals are
absent but the probe signals are present for the RSOA2 and
RSOA4. The output of RSOA3 acts as a pump signal of
RSOA4. Due to the presence of pump and probe signals,
RSOA5 does not yield the required output. Therefore, the
final output is obtained as ‘1’ at the DO2 and DO4, whereas
the other outputs are ‘0’.
Case-4: When input signals A and B are applied as‘1’ and
‘1’, respectively, then RSOA1, RSOA2, and RSOA3 do
not yield the output due to the presence of both probe and
pump signals. The output of RSOA3 acts as a pump signal
of RSOA4 but the pump signal is absent, so RSOA4 gives
the ‘1’ as the output.
Due to the presence of pump and probe signals, RSOA5
does not yield the output. So, the final output is obtained as ‘1’
at the DO4 terminal but the other outputs are ‘0’.
The proposed design of all‑optical read‑only
memory (ROM)
This proposed ROM has 2-bit address. This design using
decoder is given in Fig. 4. In each memory address, four non-
identical data can be sent. As an example, the ‘give’ word is
saved by the memory address and this is given in Table 1. The
following four cases are arising to store the word ‘give’.
Case-1: In this case, if we assign A = 0, and B = 0 then
the decoder output DO1 = 0, DO2 = 1, DO3 = 0, and
DO4 = 0 are obtained. The expression for 7th bit (MSB)
to 1st bit (LSB) for the ROM is given by
Utilizing the decoder outputs in the form of Eq. 8, we
obtained ‘1100111’ at the 7-bit outputs where 7th bit
(8)
7th Bit = DO1 + DO2 + DO3 + DO4
6th Bit = DO1 + DO2 + DO3 + DO4
5th Bit = DO2
4th Bit = DO1
3rd Bit = DO2 + DO3 + DO4
2nd Bit = DO2 + DO4
1st Bit = DO1 + DO3 + DO4
(MSB) is ‘1’ and 1st bit (LSB) is also ‘1’. These 7 bits
binary number represents the latter ‘g’ according to
ASCII code.
Case-2: When we assign A = 0, and B = 1, the decoder
outputs DO1 = 1, DO2 = 0, DO3 = 0, and DO4 = 0 are
obtained. Utilizing the decoder outputs in the form of
Eq. 8, we obtained ‘1101001’ at the 7-bit outputs where
7th bit (MSB) is ‘1’ and 1st bit (LSB) is also ‘1’. These
7 bits binary number represents the latter ‘i’ according
to ASCII code.
Case-3: In this case, if we assign A = 1, and B = 0 then
the decoder outputs DO1 = 0, DO2 = 0, DO3 = 1, and
DO4 = 0 are obtained. Utilizing the decoder outputs in
the form of Eq. 8, we obtained ‘1110110’ at the 7-bit
outputs where 7th bit (MSB) is ‘1’ and 1st bit (LSB) is
also ‘0’. These 7 bits binary number represents the latter
‘v’ according to ASCII code.
Case-4: In this case, if we assign A = 1, and B = 1 then
the decoder outputs DO1 = 0, DO2 = 0, DO3 = 0, and
DO4 = 1 are obtained. Utilizing the decoder outputs in
the form of Eq. 8, we obtained ‘1100101’ at the 7-bit
outputs where 7th bit (MSB) is ‘1’ and 1st bit (LSB) is
also ‘1’. These 7 bits binary number represents the latter
‘e’ according to ASCII code (Table 1).
Results and discussion
Here, for this model verification, two soliton pulse trains
are chosen as input signals A and B. For simulation,
RSOA’s parameters are taken as in Table 2. Signals (A
and B) are provided in Fig. 5. The simulated outputs of
the decoder’s bit patterns are provided in Fig. 6. Using this
decoder’s outputs signal, we have constructed read-only
memory, and these seven output bit patterns are provided
in Figs. 7, 8, and 9. The output quality factor or Q-value
is defined as the ratio of the difference between the aver-
age output power of ‘1’ and ‘0’ to the sum of the standard
deviation (sd1
, sd0
) of the output power of ‘1’ and ‘0’ [20,
27, 33]. It is denoted as
Table 1 2-bit address and ASCII code
Address (2 Bit) 7-bit binary (ASCII code) Character
(repre-
sented)
00 1100111 g
01 1101001 i
10 1110110 v
11 1100101 e
5. J Opt
1 3
The relative eye-opening (REOP) is represented by the
mathematical formula as [20, 27, 33]
The output extinction ratio (ER) is defined as the ratio
of the minimum values of the peak power of ‘1’ to the
(9)
Q =
P1
mean
− P0
mean
sd1
+ sd0
(10)
REOP =
P1
min
− P0
max
P1
min
maximum values of the peak power of ‘0’ [20, 27, 33]. It
is denoted as
The output contrast ratio (CR) is represented by the fol-
lowing formula
where P1
mean and P0
mean defined as the average value of the
pick power of ‘1’ and peak power of ‘0’, respectively. The
formula for the bit error rate (BER) is [20, 27, 33]
where Q represents the quality factor and erfc represents
the complementary error function. Simulated outputs data
are analysed with different parameters. We have evaluated
the quality factor (Q), relative-eye opening (REOP), extinc-
tion ratio (ER), contrast ratio (CR), and bit error rate (BER)
using Eqs. 9, 10, 11, 12, and 13, respectively. All the data
are provided in Table 3.
A comparative study of the proposed design with the pre-
vious designs is reported in Table 4
Physical realization of the proposed design
The C-band-based tunable diode laser can be used to physi-
cally realize the proposed model. The C-band wavelength is
in the range of 1535-1570 nm can be used for this operation
(11)
ER(dB) = 10 log
(
P1
min
P0
max
)
(12)
CR(dB) = 10 log
(
P1
mean
P0
mean
)
(13)
BER =
1
2
erfc
�
Q
√
2
�
=
exp
�
−Q2
2
�
Q
√
2𝜋
Fig. 4 Diagram of ROM using
decoder
2-to-4
line
DECODER
A
B
DO1
DO2
DO3
DO4
7th
Bit
(MSB)
6th
Bit 5th
Bit 4th
Bit 3rd
Bit 2nd
Bit 1st
Bit
(LSB)
Table 2 Parameter’s value of RSOA is taken for simulation [20]
Parameters Symbol Value
Injection current I 400 mA
Velocity of light c 3 × 108
m/s
Control pulse energy Ec 40 fJ
Active length L 150 µm
Active region depth d 250 nm
Width of the active region w 1.5 µm
Internal loss of the waveguide 𝛼D 2700 m−1
Nonlinear coefficient n2 2.6 × 10−20
m2
∕W
Dispersion constant D 1 ps/(nm km)
Carrier density at transparency Nt 1024
m−3
Fibre effective area Aeff 5 × 10−13
m2
Saturation energy Es 480 fJ
Wavelength of light 𝜆 1550 nm
Differential gain 𝛼N 3.3 × 10−20
m2
Confinement factor Γ 0.48
ASE factor Nsp 2
Optical bandwidth B0 3 nm
Bit period T 8 ps
6. J Opt
1 3
because RSOA works properly in this range. Losses occur
in the design due to beam splitter, mirror, beam combiner,
circulator, and RSOA. Losses due to RSOA can be subju-
gated by maintaining the proper electrical biasing of RSOA.
To minimize the loss, one can carefully select the optical
components. Since this proposed design describes the opera-
tion of a prototype dynamic memory unit in the all-optical
domain; therefore, it looks like a voluminous scheme. To
develop this scheme into a reduced and integrated optical
scheme, it can be realized in a photonic band gap (PBG)
system because RSOAs can be combined with PBG [7].
In the nano-photonic device, the use of the circulator and
beam splitter can be restricted. Therefore, optical delay due
to circulator and beam splitter can be avoided. This scheme
can perform in GHz range. The cost also is a considerable
matter for the proposed design converted into a reality. The
main cost is for tunable diode laser and the cost of RSOAs
is significant. Components of our proposed design are avail-
able in the market.
Fig. 5 Bit pattern of address
input signals (A, B)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
1
Power
(a.u.)
Address Input SIgnal A
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
1
Power
(a.u.) Address Input SIgnal B
7. J Opt
1 3
Fig. 6 Bit pattern of decoder
output signals (DO1, DO2,
DO3, and DO4)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Decoder Output Signal (DO1)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.) Decoder Output Signal (DO2)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Decoder Output Signal (DO3)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Decoder Output Signal (DO4)
8. J Opt
1 3
Fig. 7 Bit pattern of memory
output signals (7th Bit (MSB),
6th Bit, 5th Bit, and 4th Bit)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Memory Output Signal (7th
Bit)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.) Memory Output Signal (6th
Bit)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Memory Output Signal (5th
Bit)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Memory Output Signal (4th
Bit)
9. J Opt
1 3
Fig. 8 Bit pattern of memory
output signals (3rd Bit, 2nd Bit)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Memory Output Signal (3rd
Bit)
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.) Memory Output Signal (2nd
Bit)
Fig. 9 Bit pattern of memory
output signals (1st Bit (LSB))
0 10 20 30 40 50 60 70 80 90
Time (ps)
0
0.2
0.4
0.6
0.8
Power
(a.u.)
Memory Output Signal (1st
Bit)
Table 3 All the measured
values of output signals
Output signal REOP Q (dB) ER (dB) CR (dB) BER
7th Bit (MSB) 88.8941 19.554925 21.977013 34.448348 1.87 × 10−85
6th Bit 88.8941 19.554925 21.977013 34.448348 1.87 ×10−85
5th Bit 88.8941 30.576149 21.977013 45.410501 1.27 × 10−205
4th Bit 88.8941 60.269978 21.977013 59.616371 0
3rd Bit 88.8941 20.278074 21.977013 35.527520 1.0042 ×10−91
2nd Bit 88.8941 23.579139 21.977013 39.463159 3.1550 ×10−123
1st Bit (LSB) 88.8941 23.578859 21.977013 39.463757 3.1759 ×10−123
10. J Opt
1 3
Conclusion
Optical communication is a part of modern communication,
so scientists have a keen desire to design different types of
memory devices, in an all-optical domain. Here, we have
established an approach to design the ROM using RSOA.
For practical feasibility, we have simulated the proposed
design in MATLAB software using soliton pulses and meas-
ures different parameters like Q-value, extinction ratio (ER),
contrast ratio (CR), bit error rate (BER), etc., for all the
outputs. The author’s future intention is to design different
digital devices in the all-optical domain.
References
1. B. Ghosh, S. Hazra, N. Haldar, D. Roy, S.N. Patra, J. Swarnakar,
P.P. Sarkar, S. Mukhopadhyay, A novel approach to realize of
all optical frequency encoded dibit based XOR and XNOR logic
gates using optical switches with simulated verification. Opt.
Spectrosc. 124(3), 337 (2018)
2. K. Mukherjee, K. Majhi, A. Raja, A novel approach to all-optical
universal soliton logic gate NAND utilizing reflective semicon-
ductor optical amplifiers. J. Opt. 49(4), 516 (2020)
3. K. Mukherjee, K. Maji, A. Raja, M.K. Mandal, All-optical soliton
based universal logic NOR utilizing a single reflective semicon-
ductor optical amplifier (RSOA). Photon. Netw. Commun. 43(2),
101 (2022). https://doi.org/10.1007/s11107-021-00956-6
4. A. Kotb, K.E. Zoiros, W. Li, C. Guo, Theoretical investigation
of 120 Gb/s all-optical AND and OR logic gates using reflective
semiconductor optical amplifiers. Opt. Eng. 60(6), 066107 (2021)
5. A. Alquliah, A. Kotb, S.C. Singh, C. Guo, All-optical AND, NOR,
and XNOR logic gates using semiconductor optical amplifiers-
based Mach-Zehnder interferometer followed by a delayed inter-
ferometer. Optik 225, 165901 (2021)
6. S. Mitra, S. Mukhopadhyay, An all-optical scheme for implement-
ing a NAND logic by dibit representation of squeezed state of
light. J. Nonlinear Opt. Phys. Mater. 24(04), 1550048 (2015)
7. S. Saha, S. Mukhopadhyay, All optical frequency encoded quater-
nary memory unit using symmetric configuration of MZI-SOA.
Opt. Laser Technol. 131, 106386 (2020)
8. P.P. Sarkar, B. Ghosh, S.N. Patra, S. Mukhopadhyay, A new
scheme of an all optical frequency encoded dibit based latch with
its simulated result. J. Opt. Technol. 84(9), 631–634 (2017)
9. S. Saha, S. Dey, S. Mukhopadhyay, All optical wavelength
encoded 1-bit memory unit exploiting the nonlinear character of
asymmetric MZI-SOA switch, in International Conference on
Fibre Optics and Photonics (Optical Society of America, Th3A-
32, 2016)
10. A. Raja, K. Mukherjee, J.N. Roy, Design of dual semiconductor
optical amplifier structure based all-optical standard quaternary
inverter and quaternary clocked SR flip-flop. Opt. Quant. Elec-
tron. 54(1), 1–23 (2022)
11. K. Mukherjee, A.K. Meikap, D. Kumbhakar, Frequency
encoded all optical single bit memory unit using difference fre-
quency generation alone. Opt. Quant. Electron. 43(6), 101–107
(2012)
12. B. Chakraborty, S. Mukhopadhyay, A method of implementing
all-optical clocked flip-flop with phase encoded optical logic.
Optik 123(16), 1432–1435 (2012)
13. G.K. Bharti, J.K. Rakshit, Design of all-optical JK, SR and
T flip-flops using micro-ring resonator-based optical switch.
Photon Netw. Commun. 35(3), 381–391 (2018)
14. J.K. Rakshit, J.N. Roy, T. Chattopadhyay, A theoretical study of
all optical clocked D flip flop using single micro-ring resonator.
J. Comput. Electron. 13(1), 278–286 (2014)
15. D. Fitsios, C. Vagionas, G.T. Kanellos, A. Miliou, N. Pleros,
Memory speed analysis of an optical flip-flop employing a SOA-
MZI and a feedback loop. IEEE J. Quantum Electron. 49(2),
169–178 (2012)
16. L.Q. Guo, M.J. Connelly, A novel approach to all-optical wave-
length conversion by utilizing a reflective semiconductor optical
amplifier in a co-propagation scheme. Opt. Commun. 281(17),
4470–4473 (2008)
17. K. Mukherjee, All optical read only memory with frequency
encoded addressing technique. Optik 122(16), 1437–1440 (2011)
18. S. Kaur, R.S. Kaler, T.S. Kamal, All-optical binary full adder
using logic operations based on the nonlinear properties of a semi-
conductor optical amplifier. J. Opt. Soc. Korea 19(3), 222–227
(2015)
19. Y.J. Jung, N. Park, Y.M. Jhon, S. Lee, Design of all-optical read-
only memory. Appl. Opt. 48(31), G21–G27 (2009)
20. K. Maji, K. Mukherjee, A. Raja, J.N. Roy, Numerical simulations
of an all-optical parity generator and checker utilizing a reflective
semiconductor optical amplifier at 200 Gbps. J. Comput. Electron.
19(2), 800–814 (2020)
21. S. Bosu, B. Bhattacharjee, All-optical dibit-based Feynman
gate using reflective semiconductor optical amplifier with fre-
quency encoding scheme. J. Opt. (2022). https://doi.org/10.1007/
s12596-022-00875-3
22. A. Kotb, K.E. Zoiros, C. Guo, Performance investigation of 120
Gb/s all-optical logic XOR gate using dual-reflective semiconduc-
tor optical amplifier-based scheme. J. Comput. Electron. 17(4),
1640–1649 (2018)
23. A.S. Das, A.S. Patra, RSOA-based full-duplex WDM-PON for 20
Gbps transmission in two channels over a long-haul SMF using
external modulation scheme. J. Opt. Commun. 36(3), 231–235
(2015)
24. G.C. Mandal, R. Mukherjee, B. Das, A.S. Patra, A full-duplex
WDM hybrid fiber-wired/fiber-wireless/fiber-VLC/fiber-IVLC
transmission system based on a self-injection locked quantum
dash laser and a RSOA. Opt0 Commun. 427, 202–208 (2018)
Table 4 A comparative study
of the proposed design with the
previous designs
Design Main component Bit pattern
presented
Q-value ER, CR and BER
Ref. [19] SOA Yes Not calculated Not calculated
Ref. [17] SOA, ADM No Not calculated Not calculated
Ref. [7] SOA-MZI No Not calculated Not calculated
The proposed design RSOA Yes Calculated and it is
provided in Table 3
Calculated and these
are provided in
Table 3
11. J Opt
1 3
25. G.C. Mandal, R. Mukherjee, B. Das, A.S. Patra, Bidirectional
and simultaneous transmission of baseband and wireless signals
over RSOA based WDM radio-over-fiber passive optical network
using incoherent light injection technique. AEU Int. J. Electron.
Commun. 80, 193–198 (2017)
26. G.C. Mandal, R. Mukherjee, B. Das, A.S. Patra, Next-generation
bidirectional triple-play services using RSOA based WDM radio
on free-space optics PON. Opt. Commun. 411, 138–142 (2018)
27. T. Chattopadhyay, All-optical clocked delay flip-flop using a sin-
gle terahertz optical asymmetric demultiplexer-based switch: a
theoretical study. Appl. Opt. 49(28), 5226–5235 (2010)
28. Z.V. Rizou, K.E. Zoiros, T. Rampone, A. Sharaiha, Reflective
semiconductor optical amplifier direct modulation capability
enhancement using birefringent fiber loop. Appl. Sci. 10, 5328
(2020). https://doi.org/10.3390/app10155328
29. A.S. Das, A.S. Patra, Bidirectional transmission of 10 Gbit/s using
RSOA based WDM-PON and optical carrier suppression scheme.
J. Opt. Commun. 35(3), 239–243 (2014)
30. A.S. Das, P.K. Kuiri, A.S. Patra, A RSOA based full-duplex 80
channel CATV signal with 1.25 Gbps data-stream transmission
system using optical carrier suppression and injection-locked
FPLDs, in International Conference on Optics and Photonics,
vol. 9654, pp. 356–362 (2015)
31. K. Mallick, R. Mukherjee, B. Das, G.C. Mandal, A.S. Patra, Bidi-
rectional hybrid OFDM based wireless-over-fiber transport system
using reflective semiconductor amplifier and polarization multi-
plexing technique. AEU Int. J. Electron. Commun. 96, 260–266
(2018)
32. Z.V. Rizou, K.E. Zoiros, M.J. Connelly, Semiconductor optical
amplifier pattern effect suppression using optical notch filtering.
J. Eng. Sci. Technol. Rev 9(4), 198–201 (2016)
33. S. Bosu, B. Bhattacharjee, A novel design of frequency encoded
multiplexer and demultiplexer systems using reflected semicon-
ductor optical amplifier with simulative verification. J. Opt. 50(3),
361–370 (2021)
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.
Springer Nature or its licensor holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s);
author self-archiving of the accepted manuscript version of this article
is solely governed by the terms of such publishing agreement and
applicable law.