Open Access
Add to BibSonomy
Abstract
In this paper, we propose a high spectral efficiency modulation scheme based on joint interaction of orthogonal compressed chirp division multiplexing (OCCDM) and power superimposed code (PSC) under the intensity modulation and direct detection (IM/DD) system. OCCDM is a novel orthogonal chirp division multiplexing technology featuring spectral compression through the implementation of processing similar to a discrete Fourier transform, enhancing the spectral efficiency (SE) through bandwidth savings without loss of orthogonality of each chirp. Meanwhile, PSC technology enables multiple code words being transmitted superimposed on the same chirp. This technique involves allocating varying power levels to different users, thereby distinguishing them, increasing the transmission's net bit rate and substantially boosting the SE. The transmission has been performed experimentally using a 2 km 7-core fiber span. The impact of the above-mentioned technologies on the bit error rate (BER) performance is assessed in the power, frequency, and joint domain. The BER and enhancements in the SE can be balanced when the spectral bandwidth compression factor (α) and power distribution ratio are equal to 0.9 and 4, respectively. The observed outcome leads to the transmission's SE increase to more than double the baseline value, at 2.22 times. Based on the above analysis, we believe this structure is expected to become a potential for developing next-generation PON.
© 2024 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
The rapid growth in the data traffic demands over optical access networks exerts significant pressure on the available spectral resources [1]. 6 G technology in the future will continue to bring higher speed, more reliability, lower latency, and broader coverage of communications services through improved utilization of spectrum resources. Passive optical networks (PONs) are considered to be the ideal solution for next-generation large-scale optical access networks, and the development of new services such as augmented reality and meta-universe has accelerated the strong demand for PON systems with high spectral efficiency (SE) and high transmission capacity [2,3]. Nowadays, the primary emphasis on improving the performance of PON transmission SE is mainly centered on expanding the available optical spectrum bands [4,5]. However, there has been a scarcity of all-encompassing enhancement strategies put forward concerning the realm of frequency and power domains. Besides, IM/DD systems rely heavily on modulated light intensity modulation without the use of phase information. Coherent access enables complex digital signal processing (DSP) technology in the fiber-optic systems with high SE [6]. Compared with this, light intensity modulation provides only a limited modulation depth, limiting the amount of information that can be transmitted over a limited spectral range [7,8]. Thus it is imperative to enhance the performance of IM/DD system transmissions using a high spectral efficiency scheme.
Power superimposed code (PSC) technology, also known as a kind of non-orthogonal multiple access, is not only a key technology for 5 G, but also the core of next-generation PONs [9]. Traditional PSC is based on the power allocation multiplexing, which can be combined with the OCDM technology [10]. It assigns power values with large differences to different users on one resource unit corresponding to the same subcarrier and the same OCDM symbol, which means the time-frequency domain of the signals among multiple users is overlapped and divided by the power domain to achieve multiple access [11,12]. PSC introduces serial interference in the form of non-orthogonal transmissions at the transmitter by allocating the users’ transmitting power and demodulates the signals through serial interference cancellation technology at the receiver. Considering that this technology helps to improve the system SE from the perspective of power domain, bringing about high cellular edge throughput, slack channel feedback and low transmission latency [12–14], PSC is suitable for application in the circumstance of PONs with tight spectrum resources [15,16].
Spectrally efficient frequency division multiplexing (SEFDM) technology has been regarded as a significant method to enhance the SE of OFDM and OCDM signals since its birth [17]. Nevertheless, SEFDM technology causes non-orthogonality of OFDM subcarriers by compressing the spacing between subcarriers [18,19], consequently raising several issues. On one hand, it is challenging to generate SEFDM real-valued signals through the intensity modulation and direct detection (IM/DD) system, on the other hand, since SEFDM technology actively introduces ICI to the channel state information (CSI), it poses a number of challenges for channel estimation and equalization in the demodulation process. Although various measures have been used to improve the above situation, they inevitably increase the complexity of the system computation [20]. Therefore, we designed an OCCDM modulation scheme based on the principle of SEFDM. Conventional OCDM modulation uses chirp spread spectrum (CSS) method to generate a series of orthogonal waveforms named chirps, which do not interfere with each other as subcarriers in OFDM. However, due to different signal-to-noise ratio (SNR) between different frequency subcarriers, OFDM can lead to frequency selective fading in the state of bandwidth limitation and optical waveguide dispersion, which reduces the signal transmission quality [21,22]. Then, OCDM mitigates the situation of systematic impairments during the signal transmission process [7,23], showing great resistance to frequency selective fading. The inversed discrete Fresnel transform (IDFnT) operation experienced by OCDM can be realized only by phase rotation [6] on the basis of OFDM's inversed discrete Fourier transform (IDFT). Besides, the OCDM-PSC technology reduces the bit error rate (BER) of the signal transmission in contrast to the OFDM-PSC technology [24]. OCCDM modulation scheme is based on OCDM and undergoes a discrete Fourier transform (DFT)-like precoding process before IDFnT, so that the generated chirps of OCCDM signals can still maintain orthogonality in the frequency domain. Therefore, OCCDM not only effectively improves the SE, but also has high compatibility with other technologies. The complexity of channel estimation and equalization is greatly simplified as well [25]. To minimize the impact of the introduced ICI during signal demodulation, we use a standardized cascaded binary-phase-shift-keying iterative detection (SCBID) algorithm with lower complexity to reduce the effect of ICI [26]. Accordingly, the system can effectively address the scarcity of PON spectrum resources.
In this paper, we proposed a high spectral efficiency modulation scheme based on joint interaction orthogonal compressed chirp division multiplexing and power superimposed code designed in an IM/DD system for PON systems. Three sets of the experiments are carried out to verify the impacts of OCCDM, PSC, and their combined influence in a system utilizing a 2 km, 7-core fiber transmission. The experimental results demonstrate that the addition of OCCDM and PSC technologies respectively increases the spectrum utilization while having a less negative impact on the BER of OCDM-PSC and OCCDM systems. Simultaneously, the BER performance of the OCCDM-PSC system under the combination of the two technologies performs very well.
2. Principle
2.1 OCCDM and PSC technologies under the IM/DD system
OCDM technology under the traditional digital communication system will complete the modulation through IDFnT, the OCCDM signals will firstly go through a DFT-like pre-coding process, and then modulated by IDFnT to generate the final OCCDM baseband signal. The generation process of this signal can be expressed as Eq. (1):
Since DFnT does not have complex conjugate symmetry like DFT, the modulated OCCDM signals are complex-valued. However, the IM/DD system can only transmit real-valued signals. Therefore, it is necessary to convert the complex-valued signal into a real-valued signal containing complex information through the digital up-conversion (DUC) operation [27]. The DUC of the OCCDM signals will be represented as Eq. (6):
The power domain enhancement of spectral efficiency proposed in this paper is a PSC technology based on user-allocated power multiplexing. Frequency domain signals of different powers are allocated to multiple users and then the signals are multiplexed by subsequent coded superposition. The PSC signal can be described as Eq. (7):
where pi and xi are the power and signal assigned to the ith user. In this experiment, we take the value of n to be 2 and carry out the power division for two users. At the same time, we employ the parameter power division ratio (PDR) to represent the quotient of power p 1 and p 2 (p 1 > p 2), which is perceived as one of the parameter of signal quality in the experiments. In this paper, two 4QAM signals are superimposed, whose constellation points are only standardized for 16QAM when PDR is 4. The constellation points are not equally spaced in the coordinate system in the rest of the PDR cases. Hence, a standardization process of calibrating the constellation point positions for subsequent demodulation is in demand. The non-standardized 16QAM modulation is only a linear variation from the standardized 16QAM modulation, thus, only the effect of the coefficients of this linear variation need to be eliminated to obtain a standardized 16QAM modulation format.In this paper, a high SE signal transmission scheme is proposed based on OCCDM and PSC techniques in frequency and power domains respectively. The schematic is displayed in Fig. 1. It can be seen that the number of chirps has decreased in the frequency domain and the users has multiplied in the power domain, leading to improvement of spectral efficiency.
2.2 SCBID algorithm for minimizing the ICI and demodulation
Prior to signals recovery, the complex conjugate transpose of the matrix of Wsqu is employed, which is multiplied with the autocorrelation matrix $W_{squ}^H$, explained by the given equation:
For signal compression recovery at the receiver, it is imperative to multiply this conjugate transpose matrix on the left side of the received signal matrix to recover the number of signal subcarriers to N. ICI elimination and signal demodulation will be performed by the SCBID algorithm.
The SCBID algorithm adopted in this paper uses the IQ-independent BPSK soft decision strategy to reduce the impact brought by ICI, which solves the issue of ICI-compensated distortion brought by the previous ID algorithm of M-QAM where one side of the real part of the imaginary part satisfies the judgment threshold and the other does not satisfy the judgment threshold, increasing the judging accuracy [26]. For the 16-QAM modulation format involved in this paper, the signal can be regarded as two PAM4 symbols after IQ separation. Figure 2 demonstrates the maximum likelihood function of the PAM4 signal reception without ICI. As can be seen from the figure, in the absence of ICI influence, the transmitter has a higher probability of being misjudged as 1 when sending a symbol of 3 and a very small probability of being misjudged as -1 and -3. Similarly, the probability of other symbols being misjudged as neighboring symbol values is greater than the probability of symbol values further away. If ICI is introduced, the modulated symbols of each carrier will be affected by the power of other modulated symbols. The larger the modulation format is, the more high-power modulation symbols we will get. Consequently, a certain symbol will be susceptible to the interference of other high-power modulation symbols, increasing the probability of being misjudged as the value of symbols farther away from itself. It raises a majority of judgment errors in the initial stage of the judgment iteration, alleviating the effectiveness of the subsequent ICI distortion compensation.
The conventional CBID algorithm divides the M-QAM symbols ($\sqrt M $=2 k, k ${\in} N^\ast $) of the square constellation map into $\sqrt M $-1 stages for ICI distortion compensation. One 4QAM symbol is detected in each stage, then, the 4QAM symbols of each stage are superimposed to determine the final detected M-QAM. Nevertheless, owing to the difference in PDR after encoding through PSC, the signal is not exactly the standard M-QAM modulation format type after superposition. Therefore demodulation using the SCBID algorithm involves an operation of destandardization opposite to that used for modulation. It is done by removing the effect of linearly varying coefficients during normalization at the transmitter.
For the 16QAM selected in this paper, the constellation point will be decomposed into one of the two BPSK signal values of -1 and +1 at each judgment stage. At the end of the judgment, the results of the three judgments will be summed up to get the complete PAM4 signal. Each detection stage of SCBID is a process of ICI distortion compensation and the ICI interference will be decreasing with the increase of the number of detections. The detection procedure is shown in Fig. 3.
The pseudocode of the SCBID algorithm for 16QAM is demonstrated in Algorithm 1. Where C is the autocorrelation matrix mentioned above, E is the N × N identity matrix, a is the linear coefficient of destandardization and the initial signal matrix input to SCBID is S 1. The formula in line 6, Algorithm 1 represents the judgment threshold equation. The parameter qth is assigned a value of 3 for the first stage of the iterative process and sequentially assigned values of 2 and 1 for the subsequent iterations, where v denotes the total number of iterations in the stage and m is the number of iterations in the current location. As shown in Fig. 3, each iteration is a bi-directional approximation process from ± qth to 0, whose threshold will keep approaching 0. Points outside the threshold domain will be judged as -1 or 1, points inside waiting to be judged. At the end of each stage of the adjudication process, a 4QAM symbol is obtained. The power of the previous adjudication signal is subtracted from the total signal, compensating for the ICI distortion caused by the earlier 4QAM. After that, the next stage of adjudication is performed. The three stages result in three 4QAM symbols, SS 1, SS 2 and SS 3, which are then superimposed to give us the original 16QAM constellation. The final result S’ can be used for demodulation of x 2 by removing the linear coefficient.
Algorithm 1. Standardized cascaded binary-phase-shift-keying interative detection
2.3 Proposed encrypted scheme
The high SE OCCDM-PSC signal transceiver procedure is displayed in Fig. 4. The two bundles of bitstreams undergo a series-parallel transform S/P respectively. After constellation diagram mapping of 4QAM modulation, two sets of 4QAM signals are superimposed to generate a set of 16QAM-like signals xPDM. When the signals go through the operation of standardization, DFT-like compression is performed. The processed signals begin to undergo an IDFnT transform to become OCCDM signals. After the insertion of CP and parallel to serial conversion P/S process, up-sampling and DUC processing will be operated to generate the signal Tx, which is to be transmitted over the fiber channel. We perform inversed operations at the receiver to obtain the signals to eliminate ICI. These signals are required to achieve channel estimation by comparing the transmitted and received training sequences (TS). The process can be expressed as (9):
where R and X are the frequency domain signals at the receiver and transmitter, TSre and TStr denote the training sequences for receiving and transmitting, and H is the frequency domain channel response. The SCBID algorithm computes the signal equalized by the TS training sequence, and ICI's influence is greatly weakened. According to the demodulation principle of PSC signal, the signal R can be directly used for signal x 1 demapping. Since the SCBID demodulation algorithm is related in this paper, it will instantly perform the constellation point position judgment operation on the overall signal. The two branches of signals do not need to carry out channel estimation individually so that the signal x 2 can be directly calculated by Eq. (7) and then carry out the demapping operation.3. Experimental setup and results
We performed three sets of validation experiments to thoroughly evaluate the impact of PSC and OCCDM on the transmission BER performance of high SE OCCDM-PSC signals. These experiments investigated key parameters, such as the PDR in PSC and the α in OCCDM signals. We varied PDR and α individually during these experiments to examine their separate effects. Furthermore, the joint impact is also conducted to investigate the influence of these technologies across the frequency domain, power domain, and codomain, respectively. The system diagram of the validation experiments is shown in Fig. 5.
Fig. 5. IM/DD verification experimental system for OCCDM-PSC signals (DSP: digital signal processing; AWG: arbitrary waveform generator; EA: electrical amplifier; MZM: Mach-Zehnder modulator; EDFA: erbium-doped fiber amplifier; MCF: multi-core fiber; PS: power splitter; VOA: variable optical attenuator; PD: photodiode; MSO: mixed signal oscilloscope).
At the OLT side, we use an offline DSP to generate real-valued OCCDM-PSC signals. The two bitstreams shown in Fig. 3 are each mapped into a 4QAM symbol, which is then loaded into 128 chirps by power splitter 1(PS1). Meanwhile, the signal chirps are compressed with a compression factor of α. The offline data modulated by the DSP is loaded onto the AWG to perform the analog-to-digital conversion to generate the corresponding electrical waveform, which is amplified by the EA to drive a Mach-Zehnder modulator (MZM) biased at the quadrature point to implement intensity modulation. A laser source with a linewidth of less than 100 kHz was used to generate a 1550 nm optical carrier wave into the MZM in order to modulate the optical signals, which an EDFA will amplify. At the ODN side, the signals will pass through 1:8 PS2, dividing the signals into 7 equal parts and fans into the corresponding cores of the 7-core fiber (MCF) cores. After 2 km transmission, the transmitted 7-core signal is demultiplexed spatially into a single-mode fiber through a fan-out device. At the ONU side, we employ a variable optical attenuator (VOA) to adjust the value of the accepted optical power we need in real-time. A photodiode (PD) is used for photoelectric signal conversion detection. Ultimately, the analog-to-digital conversion (ADC) by a mixed signal oscilloscope (MSO, TekMSO73304DX) whose sampling rate is 50 GSa/s, is used to receive the waveform, with an off-line DSP for demodulation.
Figure 6 displays the curves of two branches of OCCDM-PSC signals after 2 km transmission in 7 fiber cores. The modulation format for both x 1 and x 2 is 4QAM in the following tests. As mentioned before, three sets of experiments in this paper allocate more power to x 1 than to x 2 and the x 1 signal is less influenced by channel noise than x 2. Therefore, it can be seen in the figure that the transmission BER performance of the x 1 signal is better than that of x 2. From the longitudinal range, with the ascending received optical power, the BER of the two signals transmitted in the various fiber cores gradually descended. In the horizontal range, the BER curves for different cores at different power levels are found to be more horizontal lines. When BER = 1 × 10−3, the difference in received optical power between the best core and the worst core for transmitting signal x 1 and x 2 is 0.68 dB and 0.49 dB, reflecting the stable transmission effect of the 7 cores. The whole transmission keeps good uniformity. By utilizing 7-core fiber, high bandwidth, data transfer rate and system reliability is gained in the signal transmission [28]. In the subsequent experiments, we choose core 6 as a response of medium transmission performance of this system.
Fig. 6. BER performance curves of two branches of OCCDM-PSC signals in 7-core fiber (MCF). (a) x 1 (b) x 2.
In the initial set of experiments, the impact of the PSC technology on the OCCDM signal transmission performance is verified in the power domain. As mentioned previously, PDR is the key parameter in the PSC. When PDR = 4, it can be observed that the transmitted signal constellation points after PDM superposition are equally spaced standard 16QAM modulation. In this experiment, the compression factor α of the OCCDM signal is fixed to 0.90, and the impact of PSC on the signal transmission performance of this system architecture is determined in five cases, namely PDR = 2, 3, 4, 5, and 6 respectively. At the same time, a group of no-PSC technology tests are also set up to compare the impact on the performance. At this time, the sampling rate of AWG is set to 10GSa/s in all five sets of tests. In the no-PSC comparison test, the effective transmission data volume is halved due to the absence of power division multiplexing. Thus, to ensure the consistency of the net bit rate, the AWG sampling rate is set to 20GSa/s, and the modulation format is set to 4QAM under this test.
Figure 7 displays the BER performance curves of two branches of signals transmission with different PDR when α=0.90. We mark the hard-decision forward error correction (HD-FEC) threshold value and soft-decision forward error correction (SD-FEC) threshold value, which is 3.8 × 10−3 and 2.4 × 10−2 respectively. The optimal transmission BER performance of the PSC technology test set did not produce a significant degradation compared to the signaling performance of the No-PSC. It reflects that the use of PSC did not bring much negative impact on OCCDM signaling. The figure shows that when PDR = 4, the signal transmission BER of the two branches is significantly lower than the values of other PDR values at each received optical power. Signal x 1 reaches HD-FEC and SD-FEC at the received optical power values of about -18.3 dBm and -19.8 dBm. Signal x 2 reaches HD-FEC and SD-FEC at received optical power values of about −17.3 dBm and −18.8 dBm. This is due to the fact that the distribution of the constellation points is not equidistant when the PDR is not equal to 4. Therefore, after standardization, the positions of the constellation points are not strictly distributed at the points of the judgment conditions of 1 and 3, leading to increase of the probability of error in the judgment. Then, the larger the difference between the value of PDR with 4, the value of transmission BER will increase accordingly. The determination accuracy of signal x 1 is determined by the distance of the constellation point stack within each of the four quadrants. As depicted in Fig. 8., when PDR > 4, the constellation points are stacked farther apart. The interference of the x 1 signal determined during the determination process is smaller in contrast to that when PDR < 4, resulting in an increase in the accuracy. Under the same received optical power (ROP), the transmission BER of signal x 1 at PDR = 5 and 6 is lower than that at PDR = 2 and 3. The determination accuracy of x 2 signals will also be determined by the distance of the four points within each quadrant. Combining the above factors, the transmission BER of signal x 2 at PDRs of 3 and 5 is less than that exhibited by signal x 1 compared to the difference at 2 and 6. It can be seen that the closer the PDR is to 4, the better the BER performance of signal x 2 will perform. Meanwhile, Fig. 9 demonstrates the BER difference curves of two branches of signals transmission with different PDR when α=0.90. The BER difference between the two branches of signals at PDR of 5 and 6 is larger than the difference when 2 and 3, which is affected by the constellation point distribution demonstrated above.
Fig. 7. BER performance curves of two branches of signals transmission with different PDR when α=0.90 (a) x 1 (b) x 2.
Fig. 8. Distribution of constellation points with different PDR (a) PDR = 4 (b) PDR = 6 (c) PDR = 2.
Fig. 9. BER difference curves of two branches of signals transmission with different PDR when α=0.90.
In the second set of experiments, we verified the impact of the OCCDM technology on the signaling performance of the OCDM-PSC architecture in the frequency domain. Likewise, α is a key parameter in the OCCDM technology. A smaller α leads to a large ICI interference introduced by the compression, and vice versa, without a better spectral performance improvement. Therefore, in this experiment, the PDR of the OCDM-PSC architecture is fixed to 4. The effects of OCCDM on the signal transmission performance of this system architecture are determined for five cases of compression coefficient α=0.86, 0.88, 0.90, 0.92, and 0.94 respectively, while a set of uncompressed state of the OCCDM technology (α=1) test is set for the comparison of the performance impact. The amount of transmitted bit data will vary due to different levels of spectral compression. To ensure the consistency of the net bit rate, The sampling rate for all five sets of tests in this experiment was set to 10 GSa/s.
Figure 10 depicts the BER performance curves of two branches of signals transmission with compression factor at PDR = 4. It can be seen that the transmission BER performance of the signals x 1 and x 2 decreases as the compression coefficients become smaller and smaller while the ICI interference between the introduced subcarriers becomes larger and larger. The compression processed signals do not produce a large degradation in transmission performance compared to α = 1, reflecting that the OCCDM technology does not bring much negative impact on the OCDM-PSC transmission architecture. When α=0.92 and 0.94, the performance curves have a higher degree of approximation to the uncompressed performance curves, and the received optical power values to reach HD-FEC and SD-FEC are similar. When α = 0.86 and 0.88, a significant degradation in performance is produced as seen by the curve direction. In contrast, it balances the compression degree and BER performance when α=0.90, showing a better performance in this test.
Fig. 10. BER performance curves of two branches of signals transmission with compression factor when PDR = 4 (a) x 1 (b) x 2.
Fig. 11. BER heatmap diagram of two branches of signals transmission with different compression factor and PDR. (a) x 1 (b) x 2.
Figure 11 shows the BER heatmap diagram of two branches of signals transmission with different compression factor and PDR. In order to facilitate a clear and systematic representation of the BER variation rule, we use a variety of colors to divide the BER values into intervals, adopting star and triangle patterns to indicate the locations where the HD-FEC and SD-FEC thresholds are reached. We divide the BER into five classes based on their magnitude as shown in the legend and indicate them with color blocks. Within each class, we also utilize the shade of the color to indicate the change in BER under that class, forming a heat map. For this experiment, we set the gradient difference of PDR to 0.2 and the gradient difference of α to 0.01. In the figure, we can see that a little change in the value of PDR will not affect the BER performance too much, but a little change in the value of α will have a large impact on the ultimate BER. In this paper, the number of subcarriers is set to 128, and a 0.01 change in the α value only makes a difference of about 1 subcarrier number. Both branches show large changes in overall BER performance at α = 0.89 and 0.90. We believe that this is due to the fact that there is a judgment stage threshold condition in the SCBID algorithm during the process of constellation point location judgment. When the interference of ICI exceeds a certain threshold, the judgment accuracy of the SCBID algorithm shows a large gap before and after that threshold.
Notably, from the HD-FEC and SD-FEC thresholds, it can be seen that the BER of x 1 and x 2 have a large reduction at α = 0.90 compared to α < 0.90, while there is not a large gap in BER performance compared to α > 0.90. And it can be found that although the BER performance is better at α > 0.90, it reaches the HD-FEC and SD-FEC thresholds under similar ROP conditions. Therefore, it can be judged that α = 0.90 is the best compression factor value for this experimental condition. Although the subtle PDR changes do not have a large impact on the experimental results, under each compression factor condition, it can still be seen that the overall BER performance is smooth and good at PDR = 4, so PDR = 4 is the best value of the power distribution ratio under this experimental condition.
It can be concluded that α=0.90 and PDR = 4 have the best performance for high spectrum OCCDM-PSC signal transmission. The formula for calculating the spectral efficiency is shown below:
where R denotes the effective information rate transmitted by the system and B denotes the communication channel bandwidth. The use of PSC technology can increase R by a factor of 2 compared to no use of itself. By using a compression factor of α=0.90, the bandwidth of the communication channel can be reduced to 0.90. At last, it can be increased to 2.22 times of the original one, thus, a higher information transmission rate can be obtained.4. Conclusion
In this paper, we propose a high spectral efficiency modulation scheme based on OCCDM and PSC, which improves the spectral efficiency of the system transmission to 2.22 times of the original one. It not only effectively improves the spectral efficiency and even the signal transmission rate from the perspective of frequency and power domains, but also enhances the frequency selective fading resistance of the system. In the experiments, the signals are transmitted in a 2 km 7-core fiber, and the impacts of PSC and OCCDM technologies on the system in the power and frequency domains, as well as the performance impacts in the codomain are verified. It is demonstrated that the incorporation of PSC and OCCDM technology brings few BER performance effects to the original system, but some SE enhancement can be obtained. In addition, in the subsequent experiments, we also find that the BER performance and SE enhancement can be balanced when the spectral compression factor α=0.90 and the power distribution ratio PDR = 4 for PSC. Based on the above analysis, we believe that this system architecture is expected to be a potential for the development of next-generation PON.
Funding
National Key Research and Development Program of China (2021YFB2800904); National Natural Science Foundation of China (62225503, 62275127, 62171227, 62205151, U22B2009); Jiangsu Provincial Key Research and Development Program (BE2022079, BE2022055-2); Natural Science Research of Jiangsu Higher Education Institutions of China (22KJB510031); Startup Foundation for Introducing Talent of Nanjing University of Information Science and Technology; Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX23_1353).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. Sun-Young Jung, Chang-Hun Kim, Sang-Min Jung, et al., “Optical pulse division multiplexing-based OBI reduction for single wavelength uplink multiple access in IM/DD OFDMA-PON,” Opt. Express 24(25), 29198–29208 (2016). [CrossRef]
2. Yiming Ma, Jianxin Ren, Bo Liu, et al., “NOMA security scheme based on constellation camouflage and selective mapping,” Opt. Lett. 48(15), 4101–4104 (2023). [CrossRef]
3. Xiaogeng Xu, Enbo Zhou, Gordon Ning Liu, et al., “Advanced modulation formats for 400-Gbps short-reach optical inter-connection,” Opt. Express 23(1), 492–500 (2015). [CrossRef]
4. Qiuyan Yao, Hui Yang, Bowen Bao, et al., “SNR re-verification-based routing, band, modulation, and spectrum assignment in hybrid C-C + L optical networks,” J. Lightwave Technol. 40(11), 3456–3469 (2022). [CrossRef]
5. Victor Lopez, Benyuan Zhu, Daniela Moniz, et al., “Optimized design and challenges for C&L band optical line systems,” J. Lightwave Technol. 38(5), 1080–1091 (2020). [CrossRef]
6. Xing Ouyang and Jian Zhao, “Orthogonal chirp division multiplexing for coherent optical fiber communications,” J. Lightwave Technol. 34(18), 4376–4386 (2016). [CrossRef]
7. A. A. Khansa, Y. Yin, G. Gui, et al., “Power-domain NOMA or NOMA-2000?” 2019 25th Asia-Pacific Conference on Communications (APCC), pp. 336–341 (2019).
8. Wenchao Xia, Bo Liu, Jianxin Ren, et al., “High-security transmission scheme of secure key generation and distribution based on polling-permutation encryption,” J. Lightwave Technol. 42(1), 149–157 (2024). [CrossRef]
9. Shuaidong Chen, Bo Liu, Jianxin Ren, et al., “Power sparse code division non-orthogonal multiple access scheme for next-generation flexible optical access,” J. Lightwave Technol. 40(22), 7236–7245 (2022). [CrossRef]
10. Shuaidong Chen, Jianxin Ren, Bo Liu, et al., “Performance improvement of non-orthogonal multiple access with a 3D constellation and a 2D IFFT modulator,” Opt. Express 31(5), 7425–7439 (2023). [CrossRef]
11. Jin Shi, Yang Hong, Rui Deng, et al., “Demonstration of real-time software reconfigurable dynamic power-and-subcarrier allocation scheme for OFDM-NOMA-based multi-user visible light communications,” J. Lightwave Technol. 37(17), 4401–4409 (2019). [CrossRef]
12. Smaranika Swain, Srikrishna Bhashyam, Ravinder D. Koilpillai, et al., “Digital power division multiplexed DD-OFDM using fundamental mode transmission in few-mode fiber,” Opt. Express 28(12), 17809–17819 (2020). [CrossRef]
13. Feng Lu, Mu Xu, Lin Cheng, et al., “Power-division non-orthogonal multiple access (NOMA) in flexible optical access with synchronized downlink/asynchronous uplink,” J. Lightwave Technol. 35(19), 4145–4152 (2017). [CrossRef]
14. Yongcan Han, Jianxin Ren, Bo Liu, et al., “Highly secure non-orthogonal multiple access based on key accompanying transmission in training sequence,” Optics Express 32(2), 1979–1997 (2024). [CrossRef]
15. Zabih Ghassemlooy, Xuan Tang, Zhang Lin, et al., “Experimental demonstration of an noma-pon with single carrier transmission,” Optics Communications 396, 66–70 (2017). [CrossRef]
16. Izzat Darwazeh, Tongyang Xu, Tao Gui, et al., “Optical SEFDM system; bandwidth saving using non-orthogonal sub-carriers,” IEEE Photonics Technol. Lett. 26(4), 352–355 (2014). [CrossRef]
17. Wei-Ren Peng, Bo Zhang, Kai-Ming Feng, et al., “Spectrally efficient direct-detected OFDM transmission incorporating a tunable frequency gap and an iterative detection techniques,” J. Lightwave Technol. 27(24), 5723–5735 (2009). [CrossRef]
18. Matthias Schuster, Sebastian Randel, Christian Alexander Bunge, et al., “Spectrally efficient compatible single-sideband modulation for OFDM transmission with direct detection,” IEEE Photon. Technol. Lett. 20(9), 670–672 (2008). [CrossRef]
19. W. Ozan, H. Ghannam, T. Xu, et al., “Experimental evaluation of channel estimation and equalisation in non-orthogonal FDM systems,” 2018 11th International Symposium on Communication Systems, Networks & Digital Signal Processing (CSNDSP), pp. 1–6 (2018).
20. Zhenhua Feng, Ming Tang, Songnian Fu, et al., “Performance-enhanced direct detection optical OFDM transmission with CAZAC equalization,” IEEE Photonics Technol. Lett. 27(14), 1507–1510 (2015). [CrossRef]
21. Zhenhua Feng, Qiong Wu, Ming Tang, et al., “Dispersion-tolerant DDO-OFDM system and simplified adaptive modulation scheme using CAZAC precoding,” J. Lightwave Technol. 34(11), 2743–2751 (2016). [CrossRef]
22. Xing Ouyang and Jian Zhao, “Orthogonal chirp division multiplexing,” IEEE Trans. Commun. 64(9), 3946–3957 (2016). [CrossRef]
23. Xing Ouyang, Giuseppe Talli, Mark Power, et al., “Orthogonal chirp-division multiplexing for IM/DD-based short-reach systems,” Opt. Express 27(16), 23620–23632 (2019). [CrossRef]
24. Yibin Wan, Bo Liu, Jianxin Ren, et al., “Performance-enhanced optical non-orthogonal multiple access enabled by orthogonal chirp division multiplexing,” J. Lightwave Technol. 40(16), 5440–5449 (2022). [CrossRef]
25. Zhouyi Hu and Chun-Kit Chan, “Non-orthogonal matrix precoding based faster-than-nyquist signaling over optical wireless communications,” 2020 Optical Fiber Communications Conference and Exhibition (OFC), pp. 1–3 (2020).
26. Jun Huang, Qi Sui, Li Zhaohui, et al., “Experimental demonstration of 16-QAM DD-SEFDM with cascaded BPSK iterative detection,” IEEE Photonics J. 8(3), 1–9 (2016). [CrossRef]
27. Xing Ouyang, Giuseppe Talli, Mark Power, et al., “Iterative block decision feedback equalization for IM/DD-Based OCDM to compensate chromatic-dispersion-induced power fading,” J. Lightwave Technol. 37(17), 4349–4358 (2019). [CrossRef]
28. Yiming Ma, Bo Liu, Jianxin Ren, et al., “A Coalesce security system of PDM and SDM based on a flexible configuration of multi-channel keys,” J. Lightwave Technol. 41(5), 1364–1374 (2023). [CrossRef]
