High speed 1.

6 Tbps hybrid TWDM-OFDM PON using 2D

bidirectional symmetric spectral/spatial-single weight zero cross-
correlation
SingSH

Abstract: The hybrid time and wavelength division multiplexing passive optical network (TWDM-PON) with a
1:128 split ratio and 16- quadrature amplitude modulation orthogonal frequency division multiplexing (QAM-
OFDM) modulation system using optical code division multiple access (OCDMA) code results in high
transmission security, channel capacity and robustness against interference. Thus in this paper, a bidirectional
symmetric 1.6 Tbps hybrid TWDM-OFDM PON system using two-dimensional spectral/spatial-single weight
zero cross-correlation (2D-S/S SWZCC) code supporting a large number of users for long-reach as well as high-
speed applications has been proposed and analysed considering fiber impairments and noise. Numerically
calculated receiver sensitivity (RS) and splitter power budget (SPB) are -36dBm and 23dB for downstream as
well as RS and SPB for upstream transmission of -33dBm and 10dB respectively showing the supremacy of
using 2D-S/S SWZCC code in the system than other OCDMA codes. Also, the simulation results show that the
proposed system using 2D-SS SWZCC code provides a faithful 120km transmission distance as well as a high
transmission rate of 16×15/20Gbps with a maximum received power of -32.8dBm supporting more than 190
simultaneous users. The designed PON system offers the best performance over other conventional PONs. The
number of users can be increased up to 250 with an optical signal-to-noise ratio of 110 dB and -30 dBm
received power.

Keywords: optical code division multiple access (OCDMA); orthogonal frequency division multiplexing
(OFDM); passive optical network (PON); time and wavelength division multiplexing (TWDM); single weight
zero cross-correlation (SWZCC)

1. Introduction
In the recent few years, network operators have encountered the exponential growth of data traffic carried over
their networks and they still confronted the increasing bandwidth demand, bulldozed by higher-bandwidth
consuming services and applications. For the past thirty years, fiber-to-the-x (FTTx) systems have been
recommended as the bonanza of the broadband access technology. In access networks, the technology hurdles to
fiber deployment are steadily evacuated by permeate advancement in technology. The recent situations of the
broadband access networks are anticipated to change considerably across the following years. Moreover, new
services and applications like high-speed Ethernet and mobile backhaul/front-haul cause demand for the
extended fiber architectures. Hence, by 2035, certainly FTTx will become the prominent broadband technology,
making the approach toward all-optical infrastructure. For now, a passive optical network (PON), a point-to-
multipoint technology is composed to play a significant role in realizing FTTx networks in urban areas with a
cost-effective and eco-friendly deployment. PON exceeds wireless access solutions concerning the least energy
consumption per bit for broadband access and it also offers lower cost, longer lifespan, less attenuation loss,
large capacity, and a highly developed network [1].
In the last couple of years, the full-service access network (FSAN) group has successfully developed a
new standard i.e. time and wavelength division multiplexing (TWDM) PON on basis of next-generation PON
stage 2 (NG-PON2), in order to obtain the definite end goal of fulfilling a flexible network operation in the
incoming future access network and preserving the multiple service capability and high capacity transmission
advantages [1]. Several TWDM-PON has been realized to 40km transmission distances at 40Gbps data rate in
the previous work. Since the growth of end users’ demand, the primary target of future-based high-capacity
PON systems is to offer a high transmission rate, long reach, and a greater number of users. To achieve TWDM
architecture at ≥100Gbps data rate, various solutions for downstream (D/S) as well as upstream (U/S)
modulation formats such as on-off keying (OOK)[2–5], four-level pulse amplitude modulation (4-PAM) [6,7]
and duobinary [8,9] have been considered. But these modulation formats are not feasible because of the
presence of fiber dispersion, chromatic dispersion (CD), and expensive 40GHz trans-receiver. For the future
≥100Gbps TWDM-PON, to acquire a cost-effective higher transmission rate, the high spectral efficiency-based
quadrature amplitude modulation incorporated orthogonal frequency division multiplexing (QAM-OFDM)
modulation would be mostly utilized [10,11]. Moreover, generally, the recent PON based on data distribution
can be categorized into three major categories: a) time division multiplexing (TDM) based utilizing single
carrier modulation e.g. gigabit PON (GPON) and ethernet PON (EPON), where both GPON and EPON are the
mature and most deployed PON architectures; b) frequency division multiplexing (FDM) based which is an
extendible solution for TDM PON system and has the capability to transmit data at distinct data rates on each
subcarrier with high spectral efficiency; c) time and frequency division multiplexing utilizing OFDM scheme.
OFDM has been regarded as a promising candidate for future high-bandwidth PON technology. It is a
multicarrier transmission technology that transfers high-speed information by splitting it into several parallel
low-speed orthogonal channels. Thus, OFDM has the maximum spectral efficiency and permits dynamic
bandwidth allocation within several applications both in frequency and time domain dimensions. Besides this,
the digital signal processing (DSP)-based OFDM-PON can offer highly suitable features like resistance to CD,
non-linear distortion, and polarization mode dispersion (PMD) [12].
Further, the optical code division multiple access (OCDMA) technique is mostly utilized multiplexing
technique in optical communication systems that permit various transmitters to send data synchronically over a
single communication channel. In OCDMA, the users share the corresponding physical resources by utilizing
the spread-spectrum technique means a coding method where a specific code is allocated to each transmitter. It
is a significant candidate for NG-PONs networks. But OCDMA also undergoes some shortcomings like inter-
symbol-interference (ISI) which causes multiple access interference (MAI), reduced spectral efficiency, and low
transmission rate as the number of users increases. OFDM has been identified as an effective scheme for
information transmission in multipath, multiuser, and fading channels. Therefore, using the OFDM method with
OCDMA can enable better spectrum, and increased data rate as well as generates higher no. of subcarriers
employing lost cost optical components using QAM modulation [13]. Here, a hybrid TWDM-OFDM PON
system using OCDMA for high-speed long-reach applications has been proposed and analyzed.
1.1 Related work
Hybrid TWDM-PON and OFDM system has been reported recently to develop better receiver sensitivity (RS),
transmission distance as well as data rate for different end users considering fiber impairments. Yufeng Shao
proposed experimentally a 16QAM-OFDM radio over fiber (RoF) system utilizing superior over Nyquist pulse
shaping (BTN-PS) and discrete multi-tone (DMT) modulation technique to lessen the peak-to-average power
ratio (PAPR) as well as to strengthen the RS. The results demonstrated that the proposed system offers a faithful
42km and 0.4m over fiber and wireless channel respectively at a 5Gbps D/S data rate [14]. Cheng Ju et al.
introduced an OFDM-PON U/S system based on various modulation format conversion schemes. The
experimental results show that the proposed system offers 10Gbps and 20Gbps data rates for binary phase-shift
keying (BPSK) to quadrature phase-shift keying (QPSK) and QPSK to 16-QAM conversion techniques
respectively [12]. Chien-Hung Yeh et al. proposed a symmetric ≥67Gbps per wavelength OFDM-intensity
modulation and direct detection (IMDD) based wavelength division multiplexing (WDM) PON network to
minimize the Rayleigh backscattering noise. The results show that the fiber length of 20km can be achieved
with minimum CD in fiber and radio frequency (RF) fading [15]. Reginaldo B. Nunes et al. proposed a medium
access control (MAC) layer protocol for a bandwidth scalable OFDM-PON system over 20km fiber length for
32 optical network units (ONUs) at a data rate of 33Gbps per wavelength [19]. C.H. Yeh et al. proposed and
investigated an asymmetric 115/10Gbps TWDM-PON using QAM-OFDM modulation at 11.25Gbps. The
results show that the proposed system provides a 20km transmission range at a threshold bit error rate (BER) of
3.8×10-3[10]. Bangjiang Lin et al. proposed and experimentally demonstrated a 4×25/10 Gbps TWDM-PON
system using double-sideband (DSB) OFDM D/S and OOK U/S modulations. The results indicate that the
proposed system offers a power budget (PB) of 35.5dB over a 20km fiber range [4]. Iwa Kartiwa et al.
experimentally demonstrated a WDM-OFDM-PON over a 20km fiber link at 20Gbps using a reflective
semiconductor optical amplifier (RSOA) in uplink transmission successfully [16]. Cheng Ju et al. proposed a
16QAM-OFDM system at 40Gbps over a 100km fiber range using a rectangular filter [17]. Xiaosong Fu et al.
realized the OFDM-PON using a modified-discrete Fourier transform (DFT) encryption method at 10Gbps over
a 20km transmission range [18]. Lognquanchen et al. realized a WDM-PON using OFDM U/S and inverse-duo
binary-return to zero (RZ) D/S schemes. The results show that the centralized-light source proposed system
offers a 50km fiber range at a symmetric 10Gbps data rate [8]. Chang-Hun Kin et al. proposed an OFDM-PON
using an all-optic virtual private network (AOVPN) over 20km range at 6Gbsp data rate [12]. Hui yang et al.
proposed an OFDM/offset QAM (OQAM) PON using direct detection (DD) scheme over 80km fiber and
20Gbps data rate [19]. Siyuan Liang et al. demonstrated imaginary interference (IM)/DD-OFDM/OQAM-PON
using an artificial neural network (ANN)-based channel estimation algorithm at 20Gbsp over a transmission
distance of 110km successfully [20]. Yaoquiang Xiao et al. demonstrated a 16QAM-OFDM modulation-based
WDM-RoF-PON over a 40km fiber range at a 10Gbps transmission rate [21]. Amber sultan et al. proposed and
demonstrated an OFDM-PON using chaotic distribution QAM symbols at a data rate of 9.4Gbps and the
transmission range of 25km [22]. Avnnet Kaur et al. demonstrated a TWDM-OFDM PON using a re-
modulation technique for U/S. The results show that the transmission range of 60km with a 4×25/10Gbps data
rate can be obtained in the proposed system [23]. Pallab K. Chaudhary et al. proposed and analyzed a 10Gbps
full-duplex cross-wavelength reuse WDM-PON using spectral shaped 16QAM-OFDM signals. The results
demonstrated that transmission distance of 25km can be obtained with OFDM signals at acceptable RS against
the CD and Rayleigh backscattering noises [24]. XuweiXue et al. demonstrated a single-sideband modulation
based OFDM-RoF-PON system 2.5Gbps transmission rate. Results indicate that the proposed system offer
transmission range of 40km below forward error correction (FEC) limit of 10 -3 with error vector magnitude
(EVM) less than 0.1 [25]. Rong Tang et al. experimentally demonstrated an OFDM-PON on the basis of fourth
dimensional (4D) joint encryption and cascading chaotic model. Results indicate that the system can offer 25km
transmission length at 16.48Gbps data rate [26]. Vishal Sharma et al. demonstrated an OFDM-dense wavelength
division multiplexing (DWDM)-EPON system over 27km transmission distance with split ratio of 16 at
2.5Gbps data rate [27]. Chang-Hun Kim et al. proposed a multi-band OFDM-PON utilizing raised-cosine
window over 20km transmission range. The results show that the system shows optimum performance by
minimising side lobes at 10Gbps data rate [28]. Cheng Ju et al. demonstrated a DD based 16QAM-OFDM-PON
system for U/S transmission over 100km fiber length at 20Gbps data rate successfully [29]. Yufeng Shao et al.
demonstrated a WDM-RoF system employing D/S OFDM-continuous phase modulation (CPM) and U/S OOK
modulation techniques. The results show that OFDM-CPM 10Gbps signals can be transmitted over 25km fiber
length faithfully [30]. Paulo Almeida et al. investigated the performance of a WDM-OFDM-PON using optical
single sideband (OSSB) modulation over 130km distance [31]. Jing Zhang et al. introduced a digital coherent
superposition (DCS)-OFDM-PON at 40Gbps over 25km distance[32]. A.O. Aldhaibani et al. proposed a
flexible cross-correlation (FCC) based OCDMA system using OFDM modulation. Results demonstrated that the
system can provide 2.5Gbps transmission rate for 100 users at 10-9 BER [13].
1.2 Motivation and Contribution
From the comprehensive study, it is noticed that the prior researchers mostly focused on individual TWDM-
PON and OFDM modulation methods for the communication systems. In these works, except a couple of
[13,33], researchers don’t consider OFDM and TWDM-PON using two-dimension (2D) spectral/special (S/S)
SWZCC code for NG-PON2 based systems. In short, the hybrid TWDM-OFDM PON system using 2D-
SWZCC code is a superb system as it has the potential to transfer high-speed continuous coded information
signals considering fiber non-linearities and impairments. The development of the proposed system allows
attractive advantages of high transmission rate, bandwidth, support a huge number of users, cost-effectiveness,
flexibility, scalability, and ease of upgradation to NG-PON3.
The four significant contributions of this proposed work are as below:
1. Propose a symmetric 1.6Tbps hybrid TWDM-OFDM PON using 2D S/S SWZCC code with a
1:128 split ratio.
2. Mathematically analyse system performance in respect of BER, EVM, receiver noise, RS, PB, and
splitter ratio for downlink/uplink channels. Also, to analyse the system performance considering
inter-symbol interference (ISI), CD, four-wave mixing (FWM), and different OCDMA codes.
3. Analyse system performance for varying fiber lengths, number of users, and transmission rate for
downlink/uplink channels through simulation.
4. Proposed work comparison w.r.t. previous recent literature work for distant codes and systems
designs.
This work is organized as: Section 2 presents the system design. Section 3 defines system numerical analysis.
Section 4 defines the results with analysis. Lastly, a conclusion is drawn in Section 5. The structure of the paper
is illustrated in Fig. 1.
 Fig. 1 Structure of the proposed work in paper

2. System design
Fig. 2 depicts the system design of a full-duplex and symmetric 16×10Gbps hybrid TWDM-OFDM PON
utilizing 2D S/S SWZCC code. This system is designed and analysed using OptiSystem v.18 software by
considering a number of samples=32768 and symbol rate=2.5×109 symbols/sec. In this system, sixteen pairs of
D/S wavelengths (λDN=1596) and U/S wavelengths (λUP) i.e. 1596nm to 1602nm and 1527nm to 1533nm having
ITU-T channel spacing of 0.4nm are utilized. TWDM-PON employs a 16-QAM OFDM modulation technique
for both U/S and D/S transmission.
 OLT/CO
Transmitter 1 dn

CW Laser (5)
OFDM Modulation LP Cosine Electrical Gain
MZM
(1) (3) (4) Roll-off

Power Combiner
Modulator
QAM Filter
CW PRBS OFDM
FBG1
Laser Generator
Sequence
Modulator
Generator LP Cosine (6)
Roll-off MZM
(2) Filter Electrical Gain Modulator (7)

Ideal Mux
Transmitter 16dn

Receiver 1up

BER Buffer
OFDM Demodulation

Ideal Demux
Analyzer Selector (8)

Receiver 16up

(9) SMFDownstream

SMF Upstream
ONUs Receiver 1dn

OFDM Demodulation FBG1+

PIN
(11) PIN PD
Photodetector
X X (12) DSP
Coupler φ Coupler
FBG1+
PIN
Electrical (14)
Phase Amplifier
Shift PIN PD
Photodetector
Local OFDM
Oscillator Decision
Demodulator
FBG1+
PIN
(10) PIN PD
Photodetector
X X (13)
Coupler QAM
Coupler
FBG1+ Sequence
PIN Electrical
PIN PD Detector
Photodetector Amplifier
(15)

BER
(16) Analyzer
Ideal Demux

Receiver 16dn
Bidirectional Splitter

Transmitter 1up

OFDM Modulation Time switch CW

Ideal Mux

section Laser

Transmitter 16up

(a)
 (1) (2) (3) (4) (5) (6)

(12) (11) (10) (9) (8) (7)

(13) (14) (15) (16)

(b)
Bessel Filter

CD Compensation
Resampling

QI Compensation

Adaptive Equalizer

FOE CPE

FOE CPE
Q

(c)
Fig. 2 Schematic of (a) symmetric 16×10Gbps hybrid TWDM-OFDM PON using 2D S/S SWZCC code, (b)
spectra, constellation diagrams, eye diagram of D/S signals, and (c) DSP unit in conjunction with constellation
diagrams
2.1 Downstream section
At the central office (CO), sixteen pairs of D/S wavelengths (λ DN) ranging from 1596nm to 1602nm are
emanated by utilizing sixteen continuous waves (CW) lasers (input power, P in=10dBm, and
linewidth=0.15MHz). As shown in Fig. 2(a), in each D/S transmitter section, the binary data bits from the
pseudo-random bit sequence generator at a 10Gbps data rate are fed into a 16-QAM (4-bits/symbol) sequence
generator. Then, the 16-QAM output signal is modulated through an OFDM modulator possessing 512
subcarriers, 1024 fast Fourier transform points along with 64 cyclic prefixes [34,35] to generate in-phase ( I )
means and quadrature (Q ) components of the OFDM signal to move forward in I /Q optical modulator. After
filtering these I /Q signals through a couple of low pass cosine roll-off filters (roll-off factor=0.2), I / Q optical
modulator components is used to modulate the incoming CW laser signal with fiber Bragg grating (FBG1)
component by a pair of Mach-Zehnder modulators (MZMs) to be spectrally encoded in 2D-SWZCC code as
given in Table 1.

Table 1 2D-SWZCC code design with 3×3 cardinality [36,37]

Y 0= [ 1 0 0 0 ] Y 0= [ 0 0 1 0 ] Y 0= [ 0 0 0 1 ]
 [] [ ] [ ] [ ]
1 1 0 0 0 0 0 1 0 0 0 0 1
T 0 0 0 0 0 0 0 0 0 0 0 0 0
X 0=
0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0

[] [ ] [ ] [ ]
0 0 0 0 0 0 0 0 0 0 0 0 0
X =0
T
1
0 0 0 0 0 0 0 0 0 0 0 0
1 1 0 0 0 0 0 1 0 0 0 0 1
0 0 0 0 0 0 0 0 0 0 0 0 0

[] [ ] [ ] [ ]
0 0 0 0 0 0 0 0 0 0 0 0 0
T 0 0 0 0 0 0 0 0 0 0 0 0 0
X2 =
0 0 0 0 0 0 0 1 0 0 0 0 0
1 1 0 0 0 0 0 0 0 0 0 0 1

Two pairs of electrical gain components with MZMs (extinction ratio=60dB) and a 2×1 power combiner (PC)
are also utilized for spectrally encoded OFDM signal [38]. Like so, sixteen D/STWDM-OFDM signals are
multiplexed through an ideal multiplexer (MUX). It is forwarded towards a single mode fiber (SMF)as well as a
1×2 bidirectional power splitter (PS). An optical amplifier (gain, G=13dB and noise figure, NF=4dB) is also
used over fiber link to improve the RS considering fiber non-linearities, dispersion, and attenuation.
For coherent orthogonal reception, the incoming coded signal from an ideal demultiplexer (DEMUX)
is transferred to the two X couplers pairs followed by local oscillators as well as phase shift (phase
difference=900) components, which are used. Further, a combination of two pairs of PIN photodetectors (PD)
along with FBG1 component having the same grating number accompanied by a pair of electrical subtractors, a
couple of electrical amplifiers (=20dB), and then OFDM demodulator is utilized to obtain the desired electrical
signals at ONU side. Single polarized DSP module, decision module, and QAM sequence decoder are used
consecutively to compensate nonlinearities losses, to the normalized amplitude of electrical I and Q channel to
16-QAM grid and to demodulate the coherent OFDM signal respectively. At the receiver, the performance
parameters of the OFDM demodulator should be matched with a transmitter to restore the QAM symbols at the
end user side as shown in Fig. 2(a). Quality (Q)-factor and eye diagrams are measured in the BER analyzer
[34,39]. Fig. 2(b) shows the spectra, constellation diagrams, and BER curves of D/S signals.
Again, the single polarized DSP module shown in Fig. 2(c) is used to recover the decoded OFDM
demodulated signal. In the DSP module, different components are used such as the Bessel filter (samples per
symbol=8×samples/bit and bandwidth=0.75×symbol rate), re-sampling (rate=4samples/symbol), I /Q
compensation, CD, adaptive equalizer, frequency offset estimation (FOE) and carrier phase estimation (CPE)
[40].Mathematically, the received signal at the input of the DSP module is expressed as [40]:
j ( 2 π f o kT +φ k ) (1)
R ( k ) =D ( k ) . e +G(k )

where D ( k ) is data symbol, f o is offset carrier frequency, T is symbol period and φ k carrier phase respectively;
G(k ) mean zero-mean Gaussian random variable. Further, the signal in the fourth power is defined as [40]:
R4 ( k )= A . e ( o
4 j 2 π f kT +φ k ) (2)
+ P(k )
where A depicts constant amplitude and P ( k ) means noise process having mean equal to zero. Here, Frequency
offset can be measured based on spectral density of signal as [40]:

1 (3)
R4 ( k )= arg { max ⁡[|J (f o )|] }
4
where
n−1
1 (4)
|J (f o )|= n ∑ R4 ( k ) . e
− j ( 2 π f o kT )

k=0
where n means the block length. In the 2D complex the decision component normalizes the desired space
2
|d k, b| to the closest constellation point plane as follows[40]:
2 (5)
|d k, b| =|J k e− j φ −X k , b|
2 b
 −jφ
where X k ,b depicts the decision J k e b. Finally, after decoding the received sequences for single polarization
followed by a parallel to serial converted to get the output bits.
2.2 Upstream section
For uplink transmission, 16×10Gbps OFDM signals are generated at sixteen uplink wavelengths (λUP). The U/P
data is transmitted from the different ONUs and then detected at the CO side. Like D/S transmission, the U/S
transmission 2D-SWZCC coded OFDM signals are generated and passed through two cascaded dynamic selects
Y components having different switching time (T s 1=0 to0.71×10-06 s and T s 2=0.10×10-06 to 0.81×10-06 s) and
time slots (0 to 15s) [41]. For U/S signal reception at the CO side, each λUP is detected. Buffer selector is
employed to choose the recent simulation iteration. Fig. 3 indicates the simulation diagram of the hybrid
TWDM-OFDM PON system using 2D S/S SWZCC code in OptiSystem simulation software. Table 2 shows the
various components’ parameters used in the proposed design.

Fig. 3 Simulative model of hybrid TWDM-OFDM PON using 2D S/S SWZCC code

Table 2 Proposed work parameters [42,43]

Parameter Value
Fiber Distance 10-150km
Input power 10-15dBm (DN) and 0dBm (UP)
Attenuation 0.2dB/km
Temperature 300 K
Dispersion slope 0.075ps/nm2/k
Reference wavelength 1550nm
No. of channels 16
Split ratio 128
Data rate 2.5-20Gbps
Modulation type 16-QAM
Dispersion 17ps/nm/km

3. Numerical analysis
In this section, the system incorporating 2D-SWZCC code is numerically analysed w.r.t. BER, EVM, receiver
noise, RS, PB, and splitter ratio for downlink/uplink channels. Also, the impact of ISI, CD, FWM, and different
OCDMA codes on the system design is calculated.

3.1 Code design with BER analysis

The 2D-SWZCC code, generated from 1D-SWZCC is represented by a N × M matrix, where N and M mean
code lengths for spectral and spatial components respectively. The 2D-SWZCC code is generated by utilizing
two code sequences of 1D-SWZCC evidenced byY and X for corresponding spectral and spatial components
having code length of Y with cardinalities K 1and X with cardinalities K 2 are: N=K 1 and N=K 2. The basic
equation of 2D-SWZCC code comprising zero cross-correlation feature is represented as [37]:
A =X Y
T (6)
h, g g h

where h=0,1,2 , … , K 1−1 and g=0,1,2, … , K 2−1.WhileY h= [Y 0 , Y 1 ,Y 2 , … , Y N −1 ] and

X g=[ X 0 , X , X 2 , … , X M −1 ] represent two code sequences of 1D-SWZCC. Here, K 1 × K 2 present the code
size and length of 2D-SWZC code. Further, the cross-correlation of Ad and Ah , g can be defined as [37]:

{
N −1 M −1
w w =1 for h=0∩ g=0 (7)
R ( h , g ) = ∑ ∑ bi , j b(i+h )( j +g) = 1 2
d (d)

i=0 j=0 0 else

b
where d ∈ {0,1,2,3 },b i, j and (i+h)( j +g) express the elements of Ad and Ah , g matrix respectively. The 2D-
SWZCC code for cardinality equal to 3 ( K 1=K 2=3 ¿ is shown in Table 1.
Again, to analyse the 2D-SWZCC code following assumptions are considered [37]:

1. The broadband light source spectrum is flat over f c −

[ f
2
f
]
, f c + , where f c and f represent the central
2
frequency and optical bandwidth respectively.
2. The input light source is idealistic un-polarized and per power spectral component has an equal spectral
width.
3. Each bit stream is identically and synchronized from the respective user.
Again, the signal-to-noise ratio (SNR) expression is given as:
I2 (8)
SNR= 2
⟨ itn ⟩
where i tn and I mean total photocurrent noise current and PD current respectively. Basically, it contributes three
major types of noises i.e. shot ( i s ) , thermal ( i t ) and phase-induced intensity noise (PIIN, i p) as below [44]:
⟨ i2tn ⟩ =⟨ i2s ⟩ + ⟨ i2t ⟩+ ⟨ i2PIIN ⟩ (9)
To evaluate the SNR, the power spectral density (PSD) of the signal received at the receiver side is presented as
[37]:
Pr K N−1 M −1
S ( f )= ∑ d (w) ∑ ∑ b(0i , )j b i , j ( w ) ∏ ( f , i ) (10)
f w−1 i=0 j=0
where ∏ ( f , i ), the broadband source i th spectral element is given as:

{[ ][
∏ ( f , i )= u f −f c − 2fN (−N +2 i ) − f −f c−u 2fN (−N +2 i+ 2 ) =u Nf ]} [ ]
(11)

where P r represents the effective received power and respectively. Again, i tn from equation (9) can be
expressed as:

[ ] [ ]
2 2
2 e Br R P r 4 B r K b T n 2 B r R Pr (12)
⟨ i ⟩= N + R + B r R ∆ fN
2
tn
L
where e , Br , K b , R Land T n represent electron charge, electric bandwidth, Boltzmann’s constant, load
resistance, and absolute temperature respectively [44–46]. The SNR can be calculated from equations (8), (9),
and (12) with N=K 1 w 1 and K= K 1 K 2 as:
 [ ]
2
R Pr
M
SNR=

[ ]
2 2 (13)
1 Br R P r K 2 4B K T eB P K
w1 + r b n + r r 2
2 ∆ fK RL K
Also, in the WDM analyzer, optical power, SNR, optical SNR (OSNR), noise, and wavelength for all WDM
channels at visualize port are detected, calculated, and displayed as [47,48]:

SNR ( dB )=10 log 10

[ P s ( mW )
Pn ( mW ) ] =Ps ( dBm )−Pn ( dBm ) (14)

where Ps and Pn present the corresponding signal and noise power. Besides this, the resulting OSNR in WDM
analyzer within the chosen resolution bandwidth ( RB=2∗Sybmol rate¿ for the analyzer, is measured as
follows [47,48]:

OSNR ( dB )=10 log 10

[ Ps ( mW )
Pn 0.1 ( mW ) ] =Ps ( dBm )−Pn 0.1 ( dBm ) (15)

where Ps and Pn 0.1 present the signal power within RB for the analyzer and noise power within a 0.1nm
bandwidth. Now, considering Gaussian noise with standard deviations σ 0 and σ 1 the BER is [47][48]:
U V (16)
Pe = P + P
U +V e 0 U + V e 1
whereU and V present the number of samples for logical 0 and logical 1 respectively. Pe 0 and Pe 1 are the
probabilities of symbols 0 and 1 respectively, given below [48]:
1
Pe 0= erfc
2
S−μ0
√2 σ0 ( ) (17)

and
1
Pe 1= erfc
2
μ1−S
√2 σ 1 ( ) (18)

where μ0 , μ1, σ 0and σ 1 represent average values and standard deviations of sampled values at 0 and 1
respectively, and S means the threshold value. Also, the Q-factor is calculated as[48]:
|μ1 −μ 0| (19)
Q=
σ 1+ σ 0
The eye height, eye amplitude, and eye closure are measured in equations (20), (21), and (22) respectively
as[48]:
E H =( μ1 −3 σ 1 )−( μ 0+ 3 σ 0 ) (20)

E A =μ1−μ 0 (21)
and
Ec =min ( V 1 )−max ⁡(V 0 ) (22)

wheremin ( V 1 ) and max ⁡(V 0 ) mean the minimum value of the amplitude for the marks and maximum value
for the amplitude of the space respectively. Hence, BER in terms of SNR can be defined as [44][48]:
1
BER= erfc
2
where erfc (.) means complementary error function.
SNR
8 √ (23)

3.2 Symbol error rate (SER) as well as error vector magnitude (EVM)
SER is basically to test for modulation systems of higher order in the system. After the decision when received
signals are compared with the original transmitted symbols to count the symbol errors as SER ( Pe ¿ as [49]:
Errors (24)
Pe =
S ymbolSequenceLength−PilotSymbols−TrainingSymbols
Further, for SER calculation, symbol error is given as[49]:
 [∑ ( )]
x x
Qzl (25)
1
Pe =∑ P m [ z ] erfc
z−1 l=1 ,l ≠ z 2 √2
where x is the number of received symbols, Pm [ z ] is the probability of occurrence of symbol z and Q zl is
given as [49]:
d zl (26)
Q zl =
σ zl +σ lz

where,

√ 2
d zl = [ I av ,Q av ] z + [ I av , Qav ]l
2 (27)

√ 2
σ zl = [ σ ( I )] z + [ σ (Q) ] z
2 (28)

and
√ 2
σ lz = [ σ ( I) ] l + [ σ (Q) ] l
2 (29)

where I av and Q av mean the average values of all received I amplitudes and Q amplitudes respectively.
Standard deviations of amplitudes I or Q are given as [49]:
σ ( I)= I −I √( 2
av ) (30)

and
√
σ (Q)= ( Q−Qav )
2 (31)

Moreover, EVM is calculated as below [49]:

( )
x 2 2
d yl
∑ x (32)
l=1 ,l ≠ y
EVM = x 2
d
∑ x
y

l=1 ,l ≠ y

where d y is the reference distance of region y . In order words, after performing a decision on each symbol in
the decision block, EVM% can be measured as [49]:

EVM %=
√|S −|S || ×100 %
s
2
D
2

(33)
|S D|
where Ss , (… .) and |S D| are symbol sequence, mean value, and decision of Ss respectively. In terms of EVM,
the BER for the 16-QAM modulation format is calculated as [50]:

3
BER16−QAM = erfc
8 (√ 1
×
1
2 EVM 2 ) (34)

3.3 Receiver sensitivity (RS)

RS at the receiver side is defined as the optical signals strengths with lowest acceptable values at the receiver
side. It can be evaluated as[51]:
RS=RS ( dB )− { ILtotal + PP } >−32.5 dBm (35)
where ILtotal and PP represent total insertion loss as well as power penalty respectively. Also, the insertion loss
for downlink/uplink channels i.e. ILdn and ILup are defined as [51]:
ILdn (dB)=IL ( MZM + MUX +Circulator+ DEMUX ) (36)

and
ILup (dB)=IL ( Dynamic Y switch+ MUX +Circulator ) (37)

Thus, the calculated RSdn and RSup for downlink/uplink channels transmission respectively, over 10km in
16×10/10Gbps TWDM-PON is given as [51]:
ILdn (dB)=3+2+1+ 3=9 dB (38)

RSdn =−25− {9+ 2 }=−36>−32.5 dBm (39)

ILup (dB)=3+2+1=6 dB (40)

RSup=−25− {6+ 2 }=−33>−32.5 dBm (41)

Table 3 Measured performance parameters for D/S and U/S TWDM-OFDM PON for varied transmission
distance
Transmission DN UP
Distance (km) 10 40 70 100 130 10 40 70 100 130
Transmitted power (dBm) 10 10 10 10 10 0 0 0 0 0
Received power(dBm) -20 -24 -28 -33 -38 -16 -20 -23 -26 -30
RS(dBm) -36 -40 -44 -49 -54 -33 -37 -40 -44 -47
OSNR(dB) 74 69 64 59 54 89 84 79 74 69
log(BER) -24 -16 -12 -4 1 -29 -24 -17 -11 -4
Q-Factor 11.5 9.5 8.5 3.5 0 16.5 12.5 10.5 8.5 3.5

Table 3 summarizes the system performance of TWDM-OFDM PON at 1596nm for downlink and 1527nm for
uplink channels at 10/10Gbsp data rate per channel at -9 log(BER) [36,51].

3.4 Power budget (PB) and splitter ratio evaluations

PB for the system can be evaluated as follows[51]:
A ( dB ) =αL+ α s n s+ α c nc (42)

where α is fiber attenuation, α s is splices’ attenuation and α c represent connectors’ attenuation. L is the fiber
length, n s present no. of splices and n c represent no. of connectors.
Thus, A ( dB ) of the system without splices loss is evaluated as [51]:
A ( dB ) =(0.2)50+ 10=20 dB (43)

Also, PB is calculated as below [51]:

PB ( dB )=Transmitted Power−RS (44)

For D/S and U/P the PB is given below [51]:

PB dn ( dB )=10−(−36 )=46 dB (45)

and
PB up ( dB )=0−(−33 )=33 dB (46)

Also, insertion loss produced due to splitter because ofn splitters, number of splitters is measured as [51]:
 PL splitter =10 log ⁡( 1/n splitters ) (47)

PL splitter =10 log ( 1/128 )=−21 dB (48)

Thus splitter power budget ( SPB) with Link Margin, Lm ( dB )=3 dB is given as below [51]:
SPB=PB− A total−Lm (49)

For downlink/uplink channels, the SPB dn and SPB up are given in equations (49) and (50) respectively, as [51]:
SPB dn=PB dn− Atotal−Lm =46−20−3=23 dB (50)

and
SPB up =PBup−A total−Lm =33−20−3=10 dB (51)

3.5 Effect of ISI

ISI is the signal distortion where one symbol interferes with subsequent symbols. In the TWDM-PON system,
ISI is produced due to the overlapping of adjacent bit intervals pulses which causes pulse spreading ( p) as
[52]:
p=LD λ (52)

where λ (=0.4nm) is the pulse spectral line width with dispersion coefficient ( D ¿. For symmetric TWDM-PON,
the λ for varied fiber distance from 10km to 130km is given below in Table 4.

Table 4 Calculated pulse spreading in the hybrid TWDM-OFDM PON system for different fiber lengths
l (km) λ (ps)
10 22
40 38
70 44
100 55
130 66
Here, as it is observed that as distance (L) increases, the value of λ increases correspondingly results in ISI
accordingly.

3.6 Impact of CD
CD is defined as a linear effect which cumulates on an optical fiber channel with link distance and causes
restriction in bandwidth. In the proposed system at the reference wavelength of 1550 nm, the dispersion( D Fiber
)at specific wavelength ( λ ¿ is presented as below[52]:
D Fiber =L[D+ c ( λ−1550 ) ] (53)

where c is a slope coefficient of different wavelengths of the fiber. The calculated CD over 100km fiber at
symmetric 1.6Tbps transmission rate for D/S (1596nm) and U/S (1527nm) wavelengths is 1120ps/nm and
990ps/nm respectively. Thus, the maximum acceptable distance for D/S and U/S wavelengths under CD having
tolerance range of 1176ps/nm and 18817ps/nm respectively is 100km D/S and >100km U/Sin the proposed
system[43,53,54].

3.7 Impact of FWM

FWM is defined as interference of at least three distinct wavelengths in the optical fiber link. This effect
generates the unwanted wavelengths by correlations with the main channel. The generated unwanted FWM
wavelengths ( λ FWM ) for q number of wavelength links in the TWDM-PON system is presented as[55]:
2
q
λ FWM = (54)
2(q−1)
 Thus, the hybrid TWDM-OFDM PON using 2D-SWZCC code generates approximately eight FWM
wavelengths. Besides this, with an increase in transmission rate, the FWM side lobes too increase and hence
result in degradation of the system performance.

3.8 Impact of OCDMA codes

Table 5 indicates the compassions of the hybrid TWDM-OFDM PON system utilizing seven different OCDMA
codes w.r.t. gain (G ¿, noise figure ( NF ¿, P i, output power, Po input OSNR, and output OSNR for distinct
D/S and uplink wavelengths. The G and NF are defined as [56,57]:

G=10 log 10 ( )
Po
Pi
dB=Po ( dB ) −Pi (dB) (55)

and

NF =10 log 10
( )
SNR i
SNR o
dB=SNR i ( dB )−SNR o ( dB) (56)

where SNRi and SNRo mean input and output SNR respectively. The tabulated values in Table 5 reveal that the
2D-SWZCC code performs best in the proposed system over 10km fiber for both D/S and U/S wavelengths.

Table 5 Hybrid TWDM-OFDM PON system performance utilizing distinct codes over 10km distance for 100
users at 10/10Gbps data rate for downlink/uplink wavelengths
D/S (1596nm) U/S (1527nm)
Input Outpu Input
OCDMA Noise Input Output Noise Input Output Output
Gain powe t Gain powe
Code figure OSNR OSNR figure OSNR power OSNR
(dB) r power (dB) r
(dB) (dB) (dB) (dB) (dB) (dB) (dB)
(dB) (dB) (dB)
2D-
-1.5 2 10 108 5 105 -1 1.5 0 108 -5 105
SWZCC

MD -3 3 10 107 5 105 -2 2 0 107 -5 104

Hadamar
10 0
d -3 2.5 107 5 105 -3 2 107 -5 105

SZCC 10 0
-2 2 106 4 104 -2 2 107 -5 105
ZCC -2 2 10 106 4 104 -2 2 0 107 -5 105

FCC 10 105 0 105

-2 2 107 5 -2 2 106 -4
MQC -2 2 10 101 2 99 -2 2 0 101 -1 100

4. Results and discussions

To verify the performance of the proposed system, in this proposed work similar parameters (in Table 2) are
utilized as accepted in [43,53,54]. The system performance is measured by concerning BER, EVM%, received
optical power (ROP), OSNR, number of users, and SNR.
 35 With 2D-SWZCC
D/S 1596nm with 2D-SWZCC code
U/S 1527nm with 2D-SWZCC code
30 D/S 1596nm without 2D-SWZCC code
U/S 1527nm without 2D-SWZCC code
(b)(c)
25 D/S, L: 10km 100km
BER: 10-27 10-9
EVM (%)

Q-Factor: 12 6
Without 2D-SWZCC
20
15%EVM

10%EVM
12.5% EVM
15 4%EVM
6%EVM

(d) (e)
U/S, L: 10km 120km
BER: 10-24 10-4
Q-Factor: 10 4
10
10 30 50 70 90 110 130 150
Fiber Length [km]
Fig. 4(a) EVM (%) vs. fiber length for downlink/uplink channels 2D-SWZCC code and without it, (b) to (e) eye
diagrams for D/S and U/S channels for different fiber distance

Fig. 4(a) illustrates the EVM (%) versus fiber distance of the proposed system at 10dBm (for D/S) and 0dBm
(for U/S) input power ( Pi ¿at symmetric 10Gbps per channel transmission link with and without 2D-SWZCC
code for both D/S (1596nm) and uplink (1527nm) transmission channels. EVM is generally used to measure the
vector signal performance and can be obtained from the constellation diagram by the equation (33) [58]. It is
depicted that as fiber range increases, EVM (%) too increases due to the occurrence in the inflated error level
over fiber link. Without 2D-SWZCC code, EVM (%) for the downlink channel and the uplink channel is
measured as 17% and 15%over 10km fiber length, whereas over 150km the measured EVM (%) is 34% and
32% for 1596nm and 1527nm channels respectively. Again, with the 2D-SWZCC code, the measured EVM (%)
over 10km and 150km for DN/UP channels are 14%/12% and 20%/18% respectively. It is also noted that the
U/S channel shows better performance than D/S channel using the 2D-SWZCC code than without the 2D-
SWZCC code. For D/S and uplink channels, the increment in EVM (%) is 4% and 6% respectively over 10km,
while it is 10% and 15% for downlink and uplink channels over 150km corresponding with and without 2D-
SWZCC code. Thus, the faithful fiber distance with a 12.5% EVM limit (for 16-QAM) is 100km (at 1596nm)
and 120km (at 1527nm) with 2D-SWZCC code. This is because 2D-SWZCC code required lower bandwidth
and zero-cross correlation permits cardinality with single weight which will enhance the system capacity [13].
Also, Fig. 4(b) to 4(e) show the eye diagrams for both D/S and U/S channels at error-free performance
FEC limit of 3.8×10-3 BER. Respective values of log (BER) and Q-factor are too stated under the eye diagrams.
The eye diagrams at a 10km fiber distance for DN and UP channels are widely open with very little distortion.
While for both D/S and U/S channels when eye lips turn wider, the height of the eye becomes narrower as fiber
length increases from 10 to 150km. For D/S and uplink channels using SWZCC code, the eye diagrams continue
to maintain open even if the optical fiber range reaches 100km and 120km respectively, which indicates that the
system performance is acceptable [58].
 Downstream

Upstream

Upstream

Downstream

Fig. 5 Measured BER performance of proposed PON for D/S and U/S transmission; Insets: corresponding eye
patterns

Fig. 5 depicts the measured BER performance of the proposed TWDM-OFDM PON system for both
downstream and upstream transmission over 50, 100, and 150km fiber range. It is depicted that the BER value
increases as received optical power increases and hence system performance diminishes. Also, the system
performance decreases as transmission ranges increase from 50 to 150km. For upstream transmission, the
received power obtained of the system over 50, 100, and 150km distance is -16, -15, and -12dBm respectively,
at BER threshold limit of 10-3. While for downstream transmission, -8, -6, and -5dBm received power is
obtained over 50, 100, and 150km distance respectively, at the BER threshold. Therefore, upstream transmission
shows better performance than downstream over 50km followed by 100 and 150km. The obtained eye patterns
at -8dBm received power over 50km distance for D/S and U/S transmission strength in the above statements.
 Downstream

Upstream

Downstream Upstream

(a)

Downstream

Upstream

Downstream Upstream

(b)
Fig. 6 Measured BER performance comparisons of hybrid TWDM-OFDM PON with other systems for (a)
16QAM and (b) 32QAM modulation format for D/S and U/S transmission; Inserts: corresponding constellation
diagrams of hybrid TWDM-OFDM PON at -12dBm received power

Fig. 6 depicts the measured BER analysis comparisons of hybrid TWDM-OFDM PON, TWDM-PON, and
GPON systems for both D/S and U/S transmission over a 50km distance. Fig. 6(a) and 6(b) indicate the
proposed system comparisons with other systems for 16QAM and 32QAM modulation formats respectively.
However, the 16QAM modulation-based proposed system offers better performance than the 32QAM
modulation-based system. Fig. 6(a) indicates that hybrid 16QAM- based TWDM-OFDM PON offers better
performance than TWDM-PON and GPON systems. Received optical power of >-24 and -12dBm for hybrid
TWDM-OFDM PON, >-24 and -9dBm for TWDM-PON and -24 and 4dBm for GPON for U/S and D/S
 transmission at BER threshold. Again, Fig. 6(b) depicts that hybrid 32QAM based TWDM-OFDM PON offers
better performance than TWDM-PON and GPON systems. The received optical power of >-24 and -8dBm for
hybrid TWDM-OFDM PON, -23 and -4dBm for TWDM-PON, and -16 and 0dBm for GPON for U/S and D/S
transmission at BER threshold. It is due to the reason that the system incorporating a lower modulation format
offers better noise immunity than a higher one. The constellation diagram at -12dBm received power for both
downstream and upstream transmissions indicate the superiority of the proposed system over other existing
systems.

Table 6 Performance analysis of TWDM-OFDM-PON using 2D-SWZCC code for varying fiber length
for D/S and U/S transmission
D/S U/S
Fiber
<12.5% EVM =12.5% EVM >12.5% EVM <12.5% EVM =12.5% EVM >12.5% EVM
range
SNR RP (in SNR RP (in SNR RP (in SNR RP (in SNR RP (in SNR RP (in
(km)
(in dB) dBm) (in dB) dBm) (in dB) dBm) (in dB) dBm) (in dB) dBm) (in dB) dBm)
30 34 4 32 2 29 2 36 5 34 3 33 3
70 31 1 28 -2 26 -2 34 3 30 1 28 0
110 30 -3 25 -7 23 -7 32 -1 28 -3 24 -3
150 30 -6 26 -11 24 -11 32 -5 28 -9 25 -10

The above-mentioned results in Table 6 summarized that the proposed PON system is most significant for both
D/S and U/S transmission in terms of SNR and ROP for varied fiber distances under acceptable EVM.

0
FEC limit @ 3.8×10-3 BER D/S 1596nm
U/S 1527nm
-10

D/S, ROP:-8dBm -32dBm

-20
log(BER)

BER: 10-45 10-2

Q-Factor: 16 2
-30 (b)(c)

-40

-50 U/S, ROP:-8dB -32dBm

BER: 10-50 10-4
Q-Factor: 18 4
(d) (e)
-33 -30 -27 -24 -21 -18 -15 -12 -9 -6
Received optical power [dBm]
(a)
Fig. 7(a) log(BER) versus ROP of downlink/uplink channels, (b) to (e) eye diagrams for downlink/uplink
channels at -8 and -32dBm ROP

The BER performance of symmetric 1.6Tbps hybrid TWDM-OFDM PON using 2D-SWZCC code for 1596nm
D/S and 1527 uplink channels, is shown in Fig. 7(a) as log (BER) versus ROP over a 10km distance. It is
observed that 1596nm D/S and 1527nm U/S channel wavelengths offer error-free transmission performance
with ROP ranges from -8 to -32.5dBm and -8 to -32.8dBm respectively. Uplink channel offers better
performance than D/S channel with 0.3dBm advantage ROP due to less ISI and fiber nonlinearities. Also, the
receiver sensitivities are detected at -32.5 and -32.8 dBm for downlink and uplink channels respectively
resulting in a 0.3dB power penalty over a 10km fiber range. Thus, the effective received power for hybrid
TWDM-OFDM PON system using 2D-SWZCC code is optimum in a manner that it is the minimum received
power that can be achieved at photo detector at FEC limit (BER of 3.8×10-3). Also, Fig. 7(b) to 7(e) show the
eye diagrams for both D/S and U/S channels at error-free performance FEC limit. The eye curves with large eye
opening at ROP of -8dBm proved that the proposed system has superb transmission performance with not many
errors like attenuation, dispersion, ISI, FWM, and receiver noise as compared to eye diagrams at -32dBm ROP
 [40]. Respective values of log (BER) and Q-factor are too stated under the eye diagrams. The comparative
performance analysis of downlink/uplink transmission of hybrid TWDM-OFDM-PON using 2D-SWZCC code
for varying fiber length is tabulated in Table 7.

Table 7 Comparison of hybrid TWDM-OFDM-PON for D/S and U/S transmission

D/S U/S
10km 50km 100km 10km 50km 100km
Log(BER) OSNR RP OSNR RP OSNR RP OSNR RP OSNR RP OSNR RP
(dB) (dBm) (dB) (dBm) (dB) (dBm) (dB) (dBm) (dB) (dBm) (dB) (dBm)
-20 -39 -24 -32 -28 -30 -34 -43 -27 -34 -31 -31 -36
-16 -36 -25 -28 -29 -24 -35 -39 -28 -30 -32 -25 -37
-12 -30 -26 -25 -30 -20 -36 -36 -29 -27 -33 -23 -38
-9 -19 -27 -23 -31 -18 -37 -26 -30 -25 -34 -21 -39

The above-mentioned results in Table 7 summarized that the proposed PON system is most desirable in both
downlink/uplink transmission in terms of minimum log (BER), OSNR, and ROP till 100km fiber length.

(a)
 (b)
Fig. 8 log (BER) performance for varied number of users for (a) D/S and (b) U/S channels

Fig. 8(a) and 8(b) illustrate the BER performance of symmetric 1.6Tbps hybrid TWDM-OFDM PON using
different OCDMA codes over a 10km transmission distance as a function of different numbers of active users
for D/S and U/S transmission respectively. Fig. 8(a) it is depicted that 2D-SWZCC code can enhance the system
performance with minimum log (BER) values less than 3.8×10 -3 for 190 users for the D/S 1596nm channel.
While OCDMA codes i.e. Hadamard, multi-diagonal (MD), shift zero cross-correlation (SZCC), modified new
ZCC (MNZCC), flexible cross-correlation (FCC) as well as a modified quadratic congruence (MQC) degrade
the system performance in terms of increasing BER from 10-26 to 10-2 for 190 simultaneous users. The highest
number of end users supported by the proposed system using 2D-SWZCC code is 190 followed by less than 190
by MD, MQC, SZCC, FCC, MNZCC, and Hadamard for D/S transmission. Also, as depicted from Fig. 8(b),
2D-SWZCC code upgrade the proposed system for U/S channel 1527nm with BER values less than 10 -3 for 190
users. Likewise, considering 190 users for MD, Hadamard, SZCC, MNZCC, FCC, and MQC codes decreases
the system performance from 10-29 to 10-19. Therefore, more than 190 users can be supported by the system using
2D-SWZCC code followed by MD, MQC, SZCC, FCC, MNZCC, and Hadamard for uplink transmission.

Table 8 Comparative performance analysis of hybrid TWDM-OFDM-PON using different OCDMA codes for
D/S and U/S users
D/S U/S
No. of users
10 90 190 10 90 190
SNR RP ( in SNR RP ( in SNR RP ( in SNR RP ( in SNR RP ( in SNR RP ( in
Code
(in dB) dBm) (in dB) dBm) (in dB) dBm) (in dB) dBm) (in dB) dBm) (in dB) dBm)
2D-SWZCC 29 4 28 -4 20 -11 31 6 29 -1 25 -8
MD 25 0 27 -7 21 -16 28 1 28 -4 22 -13
MQC 22 -19 25 -27 21 -38 24 -17 26 -23 20 -33
SZCC 24 -8 23 -15 15 -26 26 -6 25 -12 18 -21
FCC 25 -16 24 -23 19 -34 26 -14 25 -20 22 -29
MNZCC 23 -12 24 -19 13 -31 27 -11 25 -16 14 -25
Hadamard 17 -4 16 -11 08 -21 19 -3 18 -8 10 -17

The above-mentioned results in Table 8 summarized that the proposed PON system using 2D-SWZCC code is
most desirable in both downlink/uplink transmission in terms of minimum SNR and ROP for a varied number of
users.
 -30
 
-25
 
-20
 
-15
 
-10
 
-5
 
0

(a)

-30
 
-25
 
-20
 
-15
 
-10
 
-5
 
0

(b)
Fig. 9 No. of users and ROP versus OSNR for (a) D/S and (b) U/S channels

Fig. 9(a) and 9(b) depict the number of users and OSNR with a resolution of a 0.1nm as a function of ROP for
symmetric 1.6Tbps hybrid TWDM-OFDM PON using 2D-SWZCC code over a 10km fiber link at launch power
of 15dBm. It is noted that as the no. of meanwhile users increases, the OSNR value decreases and ROP
decreases causing degradation in system performance for D/Sand U/P channels. It is the reason that at high ROP
and high OSNR, users can be supported by the TWDM-OFDM receiver. The maximum OSNR and ROP at the
minimum acceptable 3.8×10−3 BER under 7% FEC for D/S is 90dB and -30dBm respectively for 250 users as
shown in Fig. 9(a). Also, the U/S channel presents better performance than the D/S channel with a maximum
obtained OSNR and ROP values for 250 users is 110dB and -30dBm respectively under the FEC limit as shown
in Fig. 9(b).
As the network capacity is measured by the maximum number of end users multiplied by the data rate
per user. Fig. 10(a) and 10(b) show the DN and UP output graphs of the four no. of users respectively, in the
form of received spectra, eye diagrams, and constellation diagrams for different data rates.

Data rate
Spectrum Eye diagram Constellation diagram
(Gbps)

2.5

10

15
 20

(a)

Data rate
Spectrum Eye diagram Constellation diagram
(Gbps)

2.5

10
 15

20

(b)
Fig. 10 Performance analysis of hybrid TWDM-OFDM-PON using 2D-SWZCC code for (a) D/S and (b) U/S
transmission for different data rate

From Fig. 10(a) and 10(b) it is noted that when the data rate increases from 2.5Gbps to 20Gbps, the power in the
received power decreases from -85 to -95dBm for D/S and -75 to -85dBmfor U/P with respective eye-opening
to closing and clear to distorted constellation diagrams under FEC level. These constellation diagrams present
the received power for 16-QAM OFDM symbols which are transmitted over a 120km fiber range. The clear
scatter-plot is obtained from 2.5 to 15Gbps and 2.5 to 20Gbps for the downlink and uplink channels respectively
at the BER limit of 3.8×10-3, which is undistorted and well detected, signifying little deterioration because of the
transmission of fiber considering real environment conditions. This is also because of increase in bandwidth
spectrum due to increase in data rate and thus proposed system can attain 15Gbps (for D/S) and 20Gbps (for
U/S) for four users [13,59]. Tables 9 as well as 10 depict the comparisons of the proposed system with respect to
(w.r.t.) latest existing work for various codes and system designs respectively.

Table 9 Analysis of the hybrid TWDM-OFDM PON system utilizing 2D-SWZCC code with other codes
Security
Code (Low/ Code Number of Cross- Length
Reference OCDMA code
Dimension(D) High) Weight Subscribers correlation of code

Dynamic cyclic shift

[60] 1 Low 4 30 ≤1 30
(DCS)
Modified double
[61] 1 Low 16 9 1 30
weight (MDW)
[62] Hadamard 1 Low 16 30 ≤1 32
Enhanced double
[63] 1 Low 4 10 ≤1 30
weight (EDW)
Not Not
[33] 2D-HC 2 High 1 920
discussed discussed
Proposed
2D-SWZCC 2 High 1 250 0 57
system

From Table 9, it is depicted that, the proposed system using 2D-SWZCC code outperforms other codes in terms
of high security, less code weight, and the moderate number of subscribers, negligible cross-correlation except
for the code length. Although the code length is more it also further enhances the security of the system.
 Table 10 Comparisons of TWDM-OFDM-PON system with existing works
Ref. & No. of Modulation Code Symmetric/ Split Fiber length Data rate Obtained
System
Year channels format used Asymmetric ratio (km) (Gbps) BER
WDM-
[21], RoF-PON Not
16-QAM Not used Symmetric 3 40 10 10-4
2013 with discussed
OFDM
WDM
[8], Not
PON using QPSK Not used Symmetric 2 50 10 10-3
2013 discussed
OFDM
WDM-
[16], Not 32-/16-/8-/ Not Not
OFDM- Not used 20 20 (UP) 2×10-3
2013 discussed 4-QAM discussed discussed
PON
[27], Hybrid
2014 OFDM- Not
2 OSSB Not used 16 27 1.25 10-9
DWDM- discussed
EPON
TWDM-
Hybrid PON: 15
[10], TWDM- Not OFDM:
4 16-QAM Not used Asymmetric 20 3.8×10-3
2014 OFDMA- discussed 11.25
PON (D/S)
10 (U/S)
Hybrid
[4], TWDM- Not
4 16-QAM Not used Asymmetric 20 100/40 3.8×10-3
2015 OFDMA- discussed
PON
OCDMA Not Not
[13], Not Not
with QAM FCC discussed discussed 2.5 10-9
2015 discussed discussed
OFDM
Hybrid
[34], TWDM-
4 4-QAM Not used Asymmetric 2 60 25/10 10-12
2017 OFDMA-
PON
[64], OFDMA- Not
16-QAM Not used Symmetric 32 20 33 3.8×10-3
2017 PON discussed
[12], OFDMA- Not Not
2 16-QAM Not used 100 20 (U/S) 3.8×10-3
2017 PON discussed discussed
[33], Not Not
LR-PON 16 Not used 2D-HC 190 40 10-9
2019 discussed discussed
[51], TWDM- Not used Not used
8 Symmetric 640 60 80 10-9
2021 PON
Hybrid
TWDM-
OFDM-
Proposed 2D-
PON 16 16-QAM Symmetric 128 120 320 3.8×10-3
work SWZCC
system
with
OCDMA

From the comparative analysis of the proposed symmetric TWDM-OFDM PON system w.r.t. recent previous
work in Table 10 it is clear that the proposed work describes best performance than other system designs in
terms of a higher number of channels (=16), modulation format (16-QAM), code (2D-SWZCC), split ratio
(=128), fiber length (=120) and data rate (=20Gbps per channel) at respective BER values. In the previous
literature work, except [33] uses code with a maximum fiber range of 190km at a maximum 40Gbps data rate,
and shows limited performance than the proposed TWDM-OFDM PON system. Fig. 11 shows the flowchart for
designing of hybrid TWDM-OFDM-PON using 2D S/S-SWZCC code [65], [66].
 Fig. 11 Flowchart for designing of hybrid TWDM-OFDM-PON using 2D S/S SWZCC code

5. Conclusion
NG-PON2-based hybrid TWDM-OFDM-PON is a point-to-multipoint optical high-speed access network,
adapting full existing consequence access network architecture as well as a multi-wavelength network utilizing
16-QAM OFDM modulation technique. A full-duplex symmetric 1.6Tbps hybrid TWDM-16QAM-OFDM-
PON using 2D-SWZCC code with a 1:128 split ratio has been demonstrated and analysed to enable high
bandwidth access network and to meet the forthcoming high-speed application as well as services such as fifth
generation (5G), artificial intelligence (AI), internet of things (IoT), etc. Based on the numerical calculations,
the hybrid system provides optimal performance in terms of good receiver sensitivity of -36dBm and -33dBm
for D/S and U/S transmission respectively, under the impact of fiber dispersion, attenuation, ISI, FWM, and
receiver noise. Also, the system using 2D-SWZCC code offers better performance in terms of adequate splitter
power budget of 23dB for downlink and 10dB for uplink transmission with a high gain of -1dB, low noise figure
of 1.5dB and high OSNR of 105dB than MD, MQC, SZCC, FCC, MNZCC, and Hadamard codes. The hybrid
TWDM-16QAM-OFDM-PON offers the best performance than other existing networks based on TWDM and
GPON. Moreover, the simulation results show that the proposed system provides a 100km D/S and 120km U/S
transmission distance, a high data rate of 15/20Gbps for every sixteen channels, minimum received power of -
32.8dBm at an input power of 10dBm supporting 190 users. At a high input power of 15dBm, 250 simultaneous
users can be handled by the TWDM-OFDM PON using SWZCC code with OSNR of 110dB, -30dBm received
power and clear constellation along with eye diagrams successfully. Additionally, a comparison of the proposed
work with existing recent past work shows the supremacy of the proposed system w.r.t. different OCDMA
codes as well as system designs.

Declarations:

Availability of Data and Materials: Data is provided under the request.

Conflict of Interest: The author SingSH declares that he has no conflicts of interest, and agree e to publish this
version of this paper under academic ethics.
Funding: This research received no specific grant from any funding agency in the public, commercial, or not-
for-profit sectors.

References

[1] V. Kachhatiya, S. Prince, Performance analysis and optimization of wavelength routed (WR) and
wavelength selected (WS) hybrid optical distributed network (ODN) for next generation passive optical
network stage 2 (NG-PON2), Opt. Laser Technol. 106 (2018) 335–347.
https://doi.org/10.1016/j.optlastec.2018.04.021.
[2] J. Wang, C. Lu, S. Li, Z. Xu, 100 m/500 Mbps underwater optical wireless communication using an
NRZ-OOK modulated 520 nm laser diode, Opt. Express. 27 (2019) 12171.
https://doi.org/10.1364/oe.27.012171.
[3] N. Sotiropoulos, T. Koonen, H. De Waardt, D8PSK / OOK Bidirectional Transmission over a TDM-
PON, 14th Int. Conf. Transparent Opt. Networks. (2012) 6–9.
[4] B. Lin, Y. Li, S. Zhang, X. Tang, Asymmetrical TWDM-PON with 4 × 25-Gb/s downstream DSB
OFDM and 4 × 10-Gb/s upstream OOK modulations, Opt. Fiber Technol. 26 (2015) 206–210.
https://doi.org/10.1016/j.yofte.2015.09.007.
[5] Y. Chen, J. Li, P. Zhu, J. Zhu, Y. Tian, Z. Wu, H. Peng, Y. Xu, J. Chen, Y. He, Z. Chen, Demonstration
of flexible multicasting and aggregation functionality for TWDM-PON, Opt. Commun. 392 (2017) 202–
207. https://doi.org/10.1016/j.optcom.2017.02.006.
[6] X. Tang, J. Zhou, M. Guo, J. Qi, F. Hu, Y. Qiao, Y. Lu, 40-Gb/s PAM4 with low-complexity equalizers
for next-generation PON systems, Opt. Fiber Technol. 40 (2018) 108–113.
https://doi.org/10.1016/j.yofte.2017.11.015.
[7] C.Y. Li, X.H. Huang, H.H. Lu, Y.C. Huang, Q.P. Huang, S.C. Tu, A WDM PAM4 FSO-UWOC
Integrated System with a Channel Capacity of 100 Gb/s, J. Light. Technol. 38 (2020) 1766–1776.
https://doi.org/10.1109/JLT.2019.2960525.
[8] L. Chen, Y. Qiao, Y. Ji, A cost-saving scheme of centralized-light-source WDM PON utilizing inverse-
duobinary-RZ downstream and OFDM upstream, Opt. Commun. 294 (2013) 123–128.
https://doi.org/10.1016/j.optcom.2012.12.073.
[9] D.T. Van Veen, V.E. Houtsma, Proposals for Cost-Effectively Upgrading Passive Optical Networks to a
25G Line Rate, J. Light. Technol. 35 (2017) 1180–1187. https://doi.org/10.1109/JLT.2017.2665166.
[10] C.H. Yeh, C.W. Chow, H.Y. Chen, Y.L. Liu, 115 Gbit/s downstream and 10 Gbit/s upstream TWDM-
PON together with 11.25 Gbit/s wireless signal utilizing OFDM-QAM modulation, Opt. Fiber Technol.
20 (2014) 84–89. https://doi.org/10.1016/j.yofte.2013.12.004.
[11] B. Lin, Y. Li, S. Zhang, X. Tang, Asymmetrical TWDM-PON with 4 × 25-Gb/s downstream DSB
OFDM and 4 × 10-Gb/s upstream OOK modulations, Opt. Fiber Technol. 26 (2015) 206–210.
https://doi.org/10.1016/j.yofte.2015.09.007.
[12] C. Ju, N. Liu, Z. Zhang, X. Chen, A flexible optical OFDMA-PON upstream scheme based on
modulation format conversion technique, Opt. Laser Technol. 90 (2017) 237–241.
https://doi.org/10.1016/j.optlastec.2016.11.017.
[13] A.O. Aldhaibani, S.A. Aljunid, M.S. Anuar, A.R. Arief, C.B.M. Rashidi, H.Y. Ahmed, Performance of
OCDMA system using OFDM technique based on flexible cross correlation code, Opt. Fiber Technol.
22 (2015) 256–258. https://doi.org/10.1109/ICP.2014.7002372.
[14] Y. Shao, PAPR reduction and receiver sensitivity improvement in 16QAM-OFDM RoF system using
DMT modulation and BTN-PS technique, Opt. Laser Technol. 77 (2016) 59–63.
https://doi.org/10.1016/j.optlastec.2015.08.013.
[15] C.H. Yeh, C.W. Chow, L.Y. Wei, Symmetric >67 Gbps OFDM-IMDD based WDM access network for
mitigating Rayleigh backscattering interference noise, Opt. Commun. 454 (2020) 124504.
https://doi.org/10.1016/j.optcom.2019.124504.
[16] I. Kartiwa, S.M. Jung, M.K. Hong, S.K. Han, 20 Gb/s WDM-OFDM-PON over 20-km single fiber
uplink transmission using optical millimeter-wave signal seeding with rate adaptive bit-power loading,
Opt. Fiber Technol. 19 (2013) 231–235. https://doi.org/10.1016/j.yofte.2013.02.001.
[17] C. Ju, P. Yang, X. Chen, Z. Zhang, N. Liu, 40 Gbps 100-km SSMF VSB-IMDD OFDM transmission
experiment based on FBG filter, Opt. Commun. 329 (2014) 184–189.
https://doi.org/10.1016/j.optcom.2014.05.020.
[18] X. Fu, M. Bi, X. Zhou, G. Yang, Q. Li, Z. Zhou, X. Yang, A chaotic modified-DFT encryption scheme
for physical layer security and PAPR reduction in OFDM-PON, Opt. Fiber Technol. 42 (2018) 126–
131. https://doi.org/10.1016/j.yofte.2018.03.004.
[19] H. Yang, X. Zhang, A. Yi, R. Wang, B. Lin, H. Xing, B. Sha, A modified convolutional neural network-
based signal demodulation method for direct detection OFDM/OQAM-PON, Opt. Commun. 489 (2021)
126843. https://doi.org/10.1016/j.optcom.2021.126843.
[20] S. Liang, F. Zhao, F. Zhao, Y. Huang, C. Wang, ANN-based channel estimation algorithm of IM/DD-
OFDM/OQAM-PON systems with mobile fronthaul network in 5G, Opt. Fiber Technol. 59 (2020)
102310. https://doi.org/10.1016/j.yofte.2020.102310.
[21] Y. Xiao, J. Yu, A novel WDM-ROF-PON architecture based on 16QAM-OFDM modulation for
bidirectional accessnetworks, Opt. Commun. 295 (2013) 99–103.
https://doi.org/10.1016/j.optcom.2013.01.036.
[22] A. Sultan, X. Yang, A.A.E. Hajomer, S.B. Hussain, W. Hu, Chaotic distribution of QAM symbols for
secure OFDM signal transmission, Opt. Fiber Technol. 47 (2019) 61–65.
https://doi.org/10.1016/j.yofte.2018.11.022.
[23] A. Kaur, B. Kaur, K. Singh, Design and performance analysis of bidirectional TWDM-PON employing
QAM-OFDM for downstream and re-modulation for upstream, Opt. - Int. J. Light Electron Opt. 134
(2017) 287–294. https://doi.org/10.1016/j.ijleo.2017.01.009.
[24] P.K. Choudhury, Enhanced noise tolerance for 10 Gb/s Bi-directional cross-wavelength reuse colorless
WDM-PON by using spectrally shaped OFDM signals, Opt. Fiber Technol. 42 (2018) 6–10.
https://doi.org/10.1016/j.yofte.2018.02.011.
[25] X. Xue, W. Ji, Z. Kang, K. Huang, X. Li, High-efficiency optical coupling single-sideband modulation
for OFDM-RoF-PON systems, Opt. Commun. 356 (2015) 500–509.
https://doi.org/10.1016/j.optcom.2015.08.035.
[26] R. Tang, B. Liu, Y. Mao, R. Ullah, J. Ren, X. Xu, J. Zhao, M. Li, S. Chen, Y. Han, High security
OFDM-PON based on an iterative cascading chaotic model and 4-D joint encryption, Opt. Commun.
495 (2021) 127055. https://doi.org/10.1016/j.optcom.2021.127055.
[27] V. Sharma, S. Sharma, Hybrid multiplexed OFDM-DWDM-EPON access nutwork, Optik (Stuttg). 125
(2014) 6404–6406. https://doi.org/10.1016/j.ijleo.2014.06.123.
[28] C.H. Kim, S.M. Jung, S.Y. Jung, S.M. Kang, S.K. Han, Inter-band interference suppression in multi-
band OFDM-PON uplink transmission using window shaping, Opt. Fiber Technol. 27 (2016) 35–40.
https://doi.org/10.1016/j.yofte.2015.12.002.
[29] C. Ju, N. Liu, X. Chen, Iteration SSII cancellation in DD-OFDM PON upstream scheme, Opt. Commun.
364 (2016) 9–12. https://doi.org/10.1016/j.optcom.2015.11.014.
[30] Y. Shao, N. Chi, J. Fan, Y. Wang, W. Fang, Novel centralized-light-source WDM-RoF system with
OFDM-CPM downstream and OOK upstream, Opt. Commun. 285 (2012) 3437–3440.
https://doi.org/10.1016/j.optcom.2012.03.024.
[31] P. Almeida, H. Silva, Power optimized OSSB modulation to support multi-band OFDM services along
hybrid long-reach WDM-PONs, Opt. Fiber Technol. 23 (2015) 129–136.
https://doi.org/10.1016/j.yofte.2015.03.005.
[32] J. Zhang, H. Yang, D. Zhao, K. Qiu, Proposal of DCS-OFDM-PON upstream transmission with
intensity modulator and collective self-coherent detection, Opt. Fiber Technol. 30 (2016) 12–16.
https://doi.org/10.1016/j.yofte.2016.01.009.
[33] H. Mrabet, I. Dayoub, S. Haxha, R. Attia, Performance analysis of 2D-OCDMA system in long-reach
passive optical network, Opt. Laser Technol. 117 (2019) 64–72.
https://doi.org/10.1016/j.optlastec.2019.04.017.
[34] A. Kaur, B. Kaur, K. Singh, Design and performance analysis of bidirectional TWDM-PON employing
QAM-OFDM for downstream and re-modulation for upstream, Optik (Stuttg). 134 (2017) 287–294.
https://doi.org/10.1016/j.ijleo.2017.01.009.
[35] A. Kaur, M.L. Singh, A. Sheetal, Simulative analysis of co-existing 2.5 G/10 G asymmetric XG-PON
system using RZ and NRZ data formats, Optik (Stuttg). 125 (2014) 3637–3640.
https://doi.org/10.1016/j.ijleo.2014.01.074.
[36] A. Cherifi, N. Jellali, M. Najjar, S.A. Aljunid, B.S. Bouazza, Development of a novel two-dimensional-
SWZCC – Code for spectral/spatial optical CDMA system, Opt. Laser Technol. 109 (2019) 233–240.
[37] A. Cherifi, N. Jellali, M. Najjar, S.A. Aljunid, B.S. Bouazza, Development of a novel two-dimensional-
SWZCC – Code for spectral/spatial optical CDMA system, Opt. Laser Technol. 109 (2019) 233–240.
https://doi.org/10.1016/j.optlastec.2018.07.078.
[38] V. Kachhatiya, S. Prince, Four-fold increase in users of time-wavelength division multiplexing
(TWDM) passive optical network (PON) by delayed optical amplitude modulation (AM) upstream, Opt.
Fiber Technol. 32 (2016) 71–81.
[39] V. Kachhatiya, S. Prince, Downstream performance analysis and optimization of the next generation
passive optical network stage 2 (NG-PON2), Opt. Laser Technol. 104 (2018) 90–102.
https://doi.org/10.1016/j.optlastec.2018.02.007.
[40] H. Yang, J. Li, B. Lin, Y. Wan, Y. Guo, L. Zhu, L. Li, Y. He, Z. Chen, DSP-based evolution from
conventional TDM-PON to TDM-OFDM-PON, J. Light. Technol. 31 (2013) 3035–3041.
https://doi.org/10.1109/jlt.2013.2271909.
[41] M. Kumari, R. Sharma, A. Sheetal, Comparative Analysis of High Speed 20 / 20 Gbps for Long-Reach
NG-PON2, J. Opt. Commun. 0 (2019) 1–14.
[42] M. Kumari, R. Sharma, A. Sheetal, Performance analysis of high speed backward compatible TWDM-
PON with hybrid WDM–OCDMA PON using different OCDMA codes, Opt. Quantum Electron. 52
(2020). https://doi.org/10.1007/s11082-020-02597-x.
[43] M. Kumari, R. Sharma, A. Sheetal, A hybrid next-generation passive optical network and visible light
communication for future hospital applications, Optik (Stuttg). 242 (2021) 166978.
https://doi.org/10.1016/j.ijleo.2021.166978.
[44] M.S. Anuar, S.A. Aljunid, N.M. Saad, S.M. Hamzah, New design of spectral amplitude coding in
OCDMA with zero cross-correlation, Opt. Commun. 282 (2009) 2659–2664.
https://doi.org/10.1016/j.optcom.2009.03.079.
[45] T. Hussein, S.A. Aljunid, H. Adnan, R.A. Ahmad, N.M. Saad, Optical Fiber Technology Development
of a new code family based on SAC-OCDMA system with large cardinality for OCDMA network, Opt.
Fiber Technol. 17 (2011) 273–280. https://doi.org/10.1016/j.yofte.2011.04.002.
[46] S. Panda, Effect of SHIFTZCC codes for Optical CDMA system, World Sci. News. 67 (2017) 365–389.
[47] W. Moench, E. Loecklin, Measurement of Optical Signal to Noise Ratio in Coherent Systems using
Polarization Multiplexed Transmission, in: Opt. Fiber Commun. Conf., 2017: pp. 1–4.
[48] Y. Qi, Z. Wang, J. He, Q. Li, Design and performance analysis of 2 x 10 GB/S bidirectional green
WDM-PON based on circulator, Proc. - 2011 4th IEEE Int. Conf. Broadband Netw. Multimed. Technol.
IC-BNMT 2011. (2011) 56–60. https://doi.org/10.1109/ICBNMT.2011.6155895.
[49] A. Sheetal, H. Singh, 5 × 10 Gbps WDM-CAP-PON based on frequency comb using OFDM with blue
LD, Opt. Quantum Electron. 50 (2018) 1–14. https://doi.org/10.1007/s11082-018-1703-x.
[50] D.-N. Nguyen, J. Bohata, J. Spacil, D. Dousek, M. Komanec, S. Zvanovec, Z. Ghassemlooy, B. Ortega,
M-QAM transmission over hybrid microwave photonic links at the K-band, Opt. Express. 27 (2019)
33745. https://doi.org/10.1364/oe.27.033745.
[51] R.A. Pagare, S. Kumar, A. Mishra, Design and analysis of hybrid optical distribution network for worst-
case scenario of E2-class symmetric coexistence 80 Gbps TWDM NG-PON2 architecture for FTTX
access networks, Optik (Stuttg). 228 (2021) 166168. https://doi.org/10.1016/j.ijleo.2020.166168.
[52] B.G.P. Agrawal, G. Agrawal, Applications of Nonlinear Fiber Optics, 2010.
[53] M. Kumari, A. Sheetal, R. Sharma, Performance analysis of symmetrical and bidirectional 40 Gbps
TWDM-PON employing m-QAM-OFDM modulation with multi-color LDs based VLC system, Opt.
Quantum Electron. 53 (2021) 1–29. https://doi.org/10.1007/s11082-021-03108-2.
[54] M. Kumari, A. Sheetal, R. Sharma, Performance Analysis of a Full-Duplex TWDM-PON Using OFDM
Modulation with Red LED Visible Light Communication System, Wirel. Pers. Commun. 119 (2021)
2539–2559. https://doi.org/10.1007/s11277-021-08343-0.
[55] C. Ghosh, V. Priye, Suppression of four-wave mixing in a 22 × 10 Gbps dense wavelength division
multiplexed system by linearly chirped fiber Bragg gratings, Opt. Quantum Electron. 51 (2019) 1–14.
[56] U.C. De Moura, J.R.F. Oliveira, J.C.R.F. Oliveira, A.C. Cesar, EDFA adaptive gain control effect
analysis over an amplifier cascade in a DWDM optical system, in: SBMO/IEEE MTT-S Int. Microw.
Optoelectron. Conf. Proc., 2013. https://doi.org/10.1109/IMOC.2013.6646469.
[57] I. Lavrinovica, J. Porins, Noise Figure Analysis of EDFA with different pumping configurations in 40
Gbit/s 8 channel DWDM transmission system, in: Proc. - 2015 Adv. Wirel. Opt. Commun. RTUWO
2015, 2015.
[58] J. Ma, M. Zhou, Y. Zhan, H. Liang, C. Yu, A novel ROF link scheme with frequency quadrupling
optical millimeter-wave carrying dual-stream of 10 Gbit/s 16-QAM signals, Opt. Laser Technol. 46
(2013) 81–87. https://doi.org/10.1016/j.optlastec.2012.04.034.
[59] R.Q. Shaddad, A.B. Mohammad, A.M. Al-Hetar, Spectral efficient hybrid wireless optical broadband
access network (WOBAN) based on transmission of wireless MIMO OFDM signals over WDM PON,
 Opt. Commun. 285 (2012) 4059–4067. https://doi.org/10.1016/j.optcom.2012.06.002.
[60] T.H. Abd, S.A. Aljunid, H.A. Fadhil, A new technique for reduction the phase induced intensity noise in
SAC-OCDMA systems, J. Opt. Commun. 32 (2011) 263–267.
[61] S.A.A. El-Mottaleb, H.A. Fayed, N.E. Ismail, M.H. Aly, M.R.M. Rizk, MDW and EDW/DDW codes
with AND subtraction/single photodiode detection for high performance hybrid SAC-OCDMA/OFDM
system, Opt. Quantum Electron. 52 (2020). https://doi.org/10.1007/s11082-020-02357-x.
[62] S. Samanta, G.K. Maity, S. Mukhopadhyay, All-optical Walsh-Hadamard code Generation using, in:
2019 Devices Integr. Circuit, 2019: pp. 515–518.
[63] S. Kumawat, M. Ravi Kumar, Generalized optical code construction for enhanced and Modified Double
Weight like codes without mapping for SAC-OCDMA systems, Opt. Fiber Technol. 30 (2016) 72–80.
https://doi.org/10.1016/j.yofte.2016.03.004.
[64] R.B. Nunes, J.M.R. Bacalhau, J.A.L. Silva, M.E.V. Segatto, A MAC layer protocol for a bandwidth
scalable OFDMA PON architecture, Comput. Commun. 105 (2017) 145–156.
https://doi.org/10.1016/j.comcom.2017.01.016.
[65] Elsayed, E. E., Alharbi, A. G., Singh, M., & Grover, A. (2022). Investigations on wavelength-division
multiplexed fibre/FSO PON system employing DPPM scheme. Optical and Quantum Electronics, 54(6).
https://doi.org/10.1007/s11082-022-03717-5
[66] Elsayed, E.E., Kakati, D., Singh, M. et al. Design and analysis of a dense wavelength-division
multiplexed integrated PON-FSO system using modified OOK/DPPM modulation schemes over
atmospheric turbulences. Opt Quant Electron 54, 768 (2022). https://doi.org/10.1007/s11082-022-04142-4