Open Access
Add to BibSonomy
Abstract
We propose a novel coherent analog radio over fiber (A-RoF) scheme to realize the generation, separation, and detection of four-independent mm-wave signals with the same carrier frequency on a single-wavelength for 5th generation (5G) mobile communication, and no digital signal processing (DSP) algorithms are required in remote antenna unit (RAU). In baseband unit (BBU), four-independent mm-wave signals are modulated on the two orthogonal polarization states of a single wavelength based on a dual-polarization IQ modulator using the dual single-sideband (SSB) modulation and polarization division multiplexing (PDM) technique. In RAU, a novel carrier polarization rotation module based on the self-polarization stabilization technique is proposed, and thus the four-independent mm-wave signals can be detected by self-coherent detection. Besides, the power fading effect induced by the chromatic dispersion could be overcome thanks to the optical SSB modulation, contributing to the increased coverage. By these means, no DSP algorithms are required in RAU, and the latency of signal processing could be significantly reduced. The experimental results show our proposed scheme could support 38.4 Gbps 16-ary quadrature amplitude modulation (16QAM) signals at 14 GHz over 30 km standard single-mode fiber (SSMF) transmission without DSP, satisfying 3rd Generation Partnership Project (3GPP) requirements. Besides, the measured error vector magnitude (EVM) value of 800 MBaud 16QAM signals at 28 GHz over 50 km SSMF transmission is 12.99%. This research provides a potential solution for the 5G mobile fronthaul.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The combination of analog radio over fiber (A-RoF) technique and millimeter-wave communication (mm-wave) shows a promising solution for the mobile fronthaul network (MFN) in the 5th generation (5G) mobile communication by considering the spectral efficiency, latency, transmission capacity, infrastructure, and cost [1,2]. The major technical challenge related to the A-RoF-based MFN is high signal-to-noise-and-distortion ratio requirements [3] especially in the case of higher carrier frequency, larger transmission bandwidth, and higher modulation format order.
To address this problem, digital coherent detection technology for RoF transmission systems is introduced [4]. Digital coherent detection technology can not only make full use of the multiple dimensions of amplitude, phase, and polarization to effectively improve the system efficiency, but also enhance the receiver sensitivity by utilizing a high-power local oscillator (LO) laser source. More importantly, it can effectively compensate for the fiber dispersion, link linear and nonlinear distortions through optical field reconstruction and digital signal processing (DSP) algorithm, which greatly improves the transmission performance [5]. However, the compensation of the carrier frequency offset and phase noise (PN) introduced by the transmitter and LO lasers require exhaustive DSP resources [6]. Besides, analog to digital converter (ADC) and digital to analog converter (DAC) are indispensable for DSP modules, further increasing the system cost. It is well-known that DSP-free is very helpful to design a remote antenna unit (RAU) with the advantages of low cost, low latency, and simple infrastructure. DSP-free RAU can not only avoid the use of the ADC and DAC and reduce energy consumption but also directly transmit mobile signals as analog waveforms without digitizing which is an essential process in the common public radio interface (CPRI) protocol [7]. In addition, the RAU without DSP is helpful to meet the requirement of ultra-reliable and low latency communications (uRLLC) which is one of the three typical application scenarios of the 5G mobile communication networks [8]. Recently, the self-coherent detection technique in which the LO and signal are transmitted together from the same optical source and the frequency offset and carrier phase noise are efficiently eliminated [9] is considered as a potential solution in the design of DSP-free RAU.
On the other hand, to enhance system capacity, the polarization division multiplexing (PDM) technique which could double the spectral efficiency (SE) attracts much attention [10–15]. The PDM technique is also helpful for massive multiple-input multiple-output (MIMO) applications, which is a key technology for the radio frequency (RF) interface of 5G mobile communications [8]. But the crosstalk between two polarization states in the PDM system would degrade the transmission performance. Therefore, the effective polarization demultiplexing technique is curial. Various methods have been proposed to realize polarization demultiplexing [11–15]. However, polarization feedback tracking [11] or MIMO signal processing [12] are not cost-effective for RAU in A-RoF-based MFN. In [13,14], a polarization-tracking-free PDM-RoF mechanism is proposed, but two optical carriers are needed in the transmitter to modulate two independent signals on two orthogonal polarization states. A self-polarization diversity mechanism in a PDM direct detection RoF system is also proposed in [15], and this method could only transmit two independent SSB signals on two orthogonal polarization states. Half of the optical spectrum is wasted when the optical SSB modulation is used. A dual-SSB modulation proposed in [16] could double the SE compared with the SSB modulation. Understandably, the combination of PDM technique and optical dual-SSB modulation on a single-wavelength could provide four times SE compared with optical SSB modulation without PDM technique, but this scheme will bring new challenges to the effective separation and detection of four-independent RF signals when no polarization feedback tracking technique and MIMO signal processing could be used at the DSP-free RAU.
In this paper, we propose a coherent A-RoF scheme for four-independent mm-wave signals transmission at the same carrier frequency and on a single wavelength, and no DSP algorithms are required in RAU. In this scheme, four-independent mm-wave signals are modulated on the two orthogonal polarization states of single-wavelength based on a dual-polarization IQ modulator (DP-IQMZM) enabled by dual-SSB modulation and PDM technique in BBU. A novel carrier polarization rotation (CPR) module based on the self-polarization stabilization technique [17] is proposed and used in RAU to transform the optical carrier to its orthogonal polarization state, and the polarization interference caused by the change of external environment is avoided. Therefore, the four-independent mm-wave signals can be detected by self-coherent detection based on a single polarized carrier. In addition, the power fading effect induced by the chromatic dispersion is overcome thanks to the optical SSB modulation. In the proof-of-concept experiment, the proposed scheme could support 38.4 Gbps 16QAM signal at 14 GHz over 30 km standard single-mode fiber (SSMF) transmission without DSP algorithm, satisfying 3GPP requirements. In our performance verification experiment, the measured error vector magnitude (EVM) value of 800 MBaud 16QAM signals at 28 GHz over 50 km SSMF transmission is 12.99%, which is close to 3GPP requirements.
2. Principle
2.1 Combination of optical dual-SSB modulation and the PDM technique
Figure 1(a) shows the proposed scheme to generate the optical dual-SSB mm-wave signal on single polarization of an optical carrier based on a single-polarization IQMZM. The two-independent baseband 16QAM signals (${S_1}(t)$ and ${S_2}(t)$) are up-converted to the mm-wave band by multiplying $\textrm{exp} ( + j2\pi {f_s}t)$ and $\textrm{exp} ( - j2\pi {f_s}t)$, respectively. Here, we use upper-sideband (USB) and lower-sideband (LSB) to denote ${S_1}(t)\textrm{exp} ( + j2\pi {f_s}t)$ and ${S_2}(t)\textrm{exp} ( - j2\pi {f_s}t)$, respectively. ${f_s}$ is the frequency of the electrical carrier. The real and imaginary parts of the summation of USB and LSB signals are used to drive the IQMZM. The optical IQMZM operates at its linear bias point (π, π, π/2), and thus an optical dual-SSB signal with the suppressed carrier is obtained at the output of IQMZM.
Fig. 1. The principle of the optical dual-SSB mm-wave signals generation (a), the principle of the modulation for four-independent mm-wave signals on two orthogonal polarization states of single-wavelength (b).
Therefore, by using a DP-IQMZM, four-independent mm-wave signals with the same carrier frequency could be modulated on one single-wavelength, and the optical spectra diagram is plotted in Fig. 1(b). XI and XQ in Fig. 1(b) are the real and imaginary parts of the summation of S1 and S2, respectively. Similarly, YI and YQ in Fig. 1(b) are the real and imaginary parts of the summation of S3 and S4, respectively.
2.2 Principle of the optically polarization demultiplexing
Figure 2 shows the schematic diagram of the generation, transmission, separation, and detection of four-independent mm-wave signals. For self-coherent detection, an external cavity laser (ECL) is divided into two parts using a 1×2 optical power splitter 1 (OPS1). With the help of the modulation scheme shown in Fig. 1, four-independent mm-wave signals can be loaded onto one part of the optical source with the suppressed carrier using a DP-IQMZM. Another part of the optical source is used as LO, and it is aligned to the X-pol of the DP-IQMZM by adjusting a polarization controller (PC) before the optical coupler (OC). The optical spectra before and after OC are shown in the inset (a-c) of Fig. 2.
Fig. 2. The schematic diagram of the proposed DSP-free remote antenna unit in coherent radio over fiber mobile fronthaul.
To realize the optical polarization demultiplexing after 30 km SSMF propagation, a CPR module is proposed based on the self-polarization stabilization technique, and its operation principle is shown in Fig. 3. This CPR module contains an optical circulator (OCIR), a delay line interferometer (DI), a ${45^ \circ }$ Faraday rotator mirror (FRM), and a Faraday mirror (FM). The optical spectra at Port 1 of this OCIR are shown in the inset (a) of Fig. 3. The green and blue curves represent the optical response curves of channel 1 (Ch 1) and channel 2 (Ch 2) of the used DI, respectively. By tuning the free spectral range (FSR) of DI, the optical carrier and four-independent mm-wave signals could be separated effectively. The inset (b) and (c) of Fig. 3 show the output optical spectra of DI. By using a ${45^ \circ }$ FRM, the polarization state of the optical signal at node (d) is rotated by ${90^ \circ }$ compared with the polarization state of the optical signal at node (b), and the optical spectra in these two nodes are shown in the inset (b) and (d) of Fig. 3, respectively. The polarization state of the optical signal at node (e) remains unchanged compared with the polarization state of the optical signal at node (c) after FM, and the optical spectra in these two nodes are shown in the inset (c) and (e) of Fig. 3, respectively. After the DI and OCIR, the polarization state of the optical carrier is transformed to its orthogonal polarization state compared with the optical signal at Port 1 of the OCIR, and the polarization of the signals remains unchanged. The inset (f) of Fig. 3 shows the optical spectra at Port 3 of OCIR. It can be seen in the inset (f) of Fig. 3 that if the optical carrier and the modulated signals are not effectively separated, the residual signal on X-pol would superimpose on the signals on Y-pol after polarization transformation, resulting in performance degradation. Fortunately, the optical carrier and the modulated signals are experienced two separations in our proposed CPR module, which is helpful to separate the optical carrier and the modulated signals as much as possible. By this means, polarization feedback tracking [11] and MIMO signal processing [12] are no longer needed for polarization demultiplexing.
Fig. 3. The principle of the proposed CPR module using the self-polarization stabilization technique.
Generally, the polarization state of the optical signal is extremely sensitive to mechanical vibration or temperature. To cope with this problem, the evolution process of the polarization state of optical signals is represented by the Jones matrix. Based on the analysis in our previous work [17], the Jones matrix from node (b) to node (d) can be expressed as
2.3 Detection of four-independent mm-wave signals
After optical bandpass filter (OBPF), these four-independent mm-wave signals can be detected by four photodiodes (PD) respectively. Take the signal S1 as an example, the square-law detection can be expressed as
3. Experimental setup and results
3.1 Proof-of-concept experimental setup and results
To verify the principle of the proposed scheme, a proof-of-concept experimental setup is built based on Fig. 4. At the transmitter, the generated two dual-SSB signals in the digital domain are loaded into a four-channel arbitrary waveform generator (AWG, Keysight M8195A) with a 3-dB bandwidth of 25 GHz operating at 64 GSa/s. These electrical signals with an optimized peak-to-peak voltage (Vpp) of 200 mV from the AWG are boosted by a four-channel electrical amplifier (EA, Centellax OA3MHQM4) with 24 dB gain. Therefore, the Vpp of driving signals after the used four-channel EA are about 3.17 V. Subsequently, a DP-IQMZM (FTM7977HQA) is applied to produce two optical dual-SSB on two orthogonal polarization states. The 3 dB bandwidth and half-wave voltage of the DP-IQMZM are 23 GHz and 3.5 V, respectively. A 14.2 dBm continuous wave (CW) from an external cavity laser (ECL) with a wavelength of 1550.01 nm is split into two parts by optical power splitter 1 (OPS1). One part is injected into the DP-IQMZM, and an erbium-doped fiber amplifier (EDFA, Amonics-C-DWDM-23-B-FA) is used to compensate for the insertion loss of DP-IQMZM. In our experiment, both two IQMZM in the DP-IQMZM operate at its linear bias point (π, π, π/2), and thus two optical dual-SSB signals without the central optical carrier on two polarizations are generated at the output of DP-IQMZM. The other part is utilized as the optical carrier, and this optical carrier is aligned to the X-pol of DP-IQMZM by adjusting a polarization controller (PC). A 50 m SSMF is applied to match the time delay of the used DP-IQMZM and EDFA. An alternative solution is to generate a central optical carrier on the X-pol by biasing the IQMZM on X-pol a little bit above the null point [19], and the IQMZM on Y-pol still operates at its linear bias point. By this means, the OPS1, PC, 50 m SSMF, and polarization alignment operation in our experiment are no longer required, contributing to a simpler system structure.
Fig. 4. The experimental setup of the proposed scheme. (i) The measured optical power values at the corresponding points of the experimental system. (ii)-(iii) The optical spectra at the corresponding nodes. (iv)-(v)The electrical spectra of the received electrical signal.
After 30 km SSMF propagation, the optical signal is divided into two parts. One part is amplified by another EDFA, and an optical bandpass filter (OBPF) is applied to remove the unwanted optical sideband. A variable optical attenuator (VOA) is used to sweep the received optical power (ROP). Subsequently, the optical SSB signal on X-pol is detected by a PD with a 3-dB bandwidth of 40 GHz. Another part passes through our proposed CPR module, and thus the polarization state of the optical carrier is rotated from X-pol to Y-pol. Therefore, the optical dual-SSB signal on Y-pol can be detected in the same way as the signal on X-pol. Finally, the detected electrical signal is captured by a digital sampling oscilloscope (DSO, LeCroy LabMaster 10-36Zi-A) with a 3-dB bandwidth of 36 GHz operating at 80 GSa/s. The captured digital signal is processed offline. After frequency down-conversion, matched filter, and timing recovery, the EVM and BER performance is counting. No chromatic dispersion compensation, frequency offset compensation, carrier phase recovery, polarization feedback tracking, MIMO signal processing, and equalization algorithms are used before the EVM and BER counting in our experiment. Note that, in the real scenario the RF signal after PD detection is amplified by an RF amplifier and then directed fed into the antenna. That means the analog-to-digital conversion, frequency down-conversion, matched filter, and timing recovery are no longer needed at the RAU. The inset (i) of Fig. 4 shows the measured optical power values at the corresponding points of the experimental system. The inset (ii) and (iii) of Fig. 4 show the optical spectra at the corresponding nodes in the experiment. The electrical spectra of the received signal when the carrier frequency is 14 GHz are shown in the inset (iv) of Fig. 4. The SSBI and the desired signal can be seen in the electrical spectra, and the DC term is blocked by the used AC-coupled PD. The inset (v) of Fig. 4 shows the electrical spectra of the desired signal at the frequency range of 13 GHz-15 GHz.
The carrier-to-signal power ratio (CSPR) which is defined as the ratio of carrier power to signal power is a critical parameter, and the tuning of the CSPR value is achieved by adjusting the output power of the EDFA behind the DP-IQMZM. Figure 5 shows the measured EVM performance versus the CSPR for 800 MBaud 16QAM signal with the carrier frequency of 14 GHz at the USB after optical back-to-back (OB2B) and 30 km SSMF transmission. And the ROP value is set as −3 dBm in this test. It can be seen that the measured EVM performance for all cases is less than 12.5% (the EVM value specified in the 3GPP standard for 16QAM) when the CSPR value is in the range of 5–14 dB, and the EVM performances are close when the CSPR value is 8 dB and 10 dB. The optimum CSPR value of the USB signal on Y-pol after optical B2B transmission is 10 dB, whereas the optimum CSPR value of the other three cases is 8 dB. This difference could be attributed to the drift of the optical response curve of the used DI and OBPF. Considering the transmission performance and system stability, we have fixed the CSPR value at 10 dB in the rest experiments.
Fig. 5. The EVM performance versus the CSPR for 800 Mbuad 16QAM signals with the carrier frequency of 14 GHz at the USB.
Figures 6(a) and 6(b) present the measured BER performances in the terms of the ROP for 800 MBaud 16QAM signals with the carrier frequency of 14 GHz on X-pol and Y-pol after OB2B and 30 km SSMF transmission, respectively. It can be seen in Fig. 6(a) that when the ROP is higher than −16 dBm, the BER performances of all four cases are better than the 7% overhead hard-decision forward error correction (HD-FEC) threshold of 3.8×10−3. Compared with the signals on Y-pol, the signals on X-pol have negligible 0.5 dB power gain which is induced by our proposed CPR module. This indicates that our proposed CPR module has little impact on transmission performance, and it is further verified that the proposed scheme can realize the separation and detection of four independent signals without DSP at the receiver. The measured BER performances after OB2B and 30 km SSMF transmission are very close, and it is attributed to that the power fading effect induced by the chromatic dispersion is avoided thanks to the optical SSB modulation at the transmitter. It can be seen in Fig. 6 that the BER performances of the LSB and USB signals on X-pol after OB2B are roughly the same. The same phenomenon could be observed for the LSB and USB signals on Y-pol after OB2B and 30 km SSMF transmission. Only the LSB signal on X-pol after optical 30 km SSMF transmission has about 1 dB power penalty compared to the USB signal on X-pol after optical 30 km SSMF transmission. And this difference could be attributed to the drift of the optical response curve of the OBPF. Therefore, it is acceptable that only the performance of the signal on USB is analyzed in the rest experiments.
Fig. 6. The BER performance versus ROP of 800 MBaud 16QAM signals with the carrier frequency of 14 GHz on (a) X-pol and (b) Y-pol.
Figure 7(a) shows the EVM performance versus different 16QAM signal baud rates with the carrier frequency of 14 GHz after 30 km SSMF transmission. The ROP is fixed at −3 dBm, and both the signals on X-pol and Y-pol are measured in this test. As the signal baud rate increases, the measured EVM performance gradually deteriorates. This could be attributed to the inter-symbol interference (ISI) induced by the uneven frequency response of the used experimental system. When the signal baud rate is 3.2 GHz, the measured EVM value of 16QAM signal with the carrier frequency of 14 GHz on X-pol and Y-pol is 11.4% and 12.7%, respectively. This EVM performance difference between these two polarization states could be explained by the uneven optical response curve of the used DI which can be seen in Figs. 8(a) and 8(b). However, when the signal bandwidth is 2.4 GHz, the measured EVM value of these two cases could both achieve lower than 12.5%. In other others, our proposed scheme could support 38.4 Gbps (2.4×4×4 = 38.4) 16QAM signals with the carrier frequency of 14 GHz over 30 km SSMF transmission satisfying 3GPP requirements for 5G systems without DSP.
Fig. 7. The EVM performance versus different 16QAM signal baud rates with the carrier frequency of 14 GHz after 30 km SSMF transmission (a). The EVM performance of 800 MBaud 16QAM signals with different carrier frequencies after 30 km SSMF transmission (b).
Fig. 8. The optical response curves of the used DI and the optical spectra at the corresponding nodes in Fig. 3 when the electrical carrier frequency is 8 GHz (a) and 14 GHz (b).
Figure 7(b) shows the measured EVM performance of 800 MBaud 16QAM signal with different electrical carrier frequencies after 30 km SSMF transmission. The EVM performance difference between the signals on X-pol and Y-pol could be negligible when the electrical carrier frequency is higher than 14 GHz. However, the EVM performance of the signal on Y-pol is much worse than the signal on X-pol when the electrical carrier frequency is 8 GHz. This could be attributed to the crosstalk of the signals on X-pol to the signals on Y-pol induced by the limited extinction ratio (ER) of the used DI. The ER and FSR of the used DI are positively correlated, and the FSR is twice the electrical carrier frequency. Figures 8(a) and 8(b) show the optical response curves of the used DI and the optical spectra at the corresponding nodes in Fig. 3 when the electrical carrier frequency is 8 GHz and 14 GHz, respectively. The ER of the used DI is 13.3 dB and 17.5 dB in these two cases, respectively. In other words, the optical carrier and the modulated signals cannot be separated effectively when the ER of the DI is small, and the residual signal on X-pol would superimpose on the signals on Y-pol after polarization transformation, resulting in performance degradation. Therefore, the crosstalk of the signals on X-pol to the signals on Y-pol when the carrier frequency is 8 GHz is larger than the cases when the electrical carrier frequency is higher than 14 GHz. It should be noted that our proposed scheme operates at the mm-wave frequency band in 5G mobile communication, and thus the polarization crosstalk induced by the limited ER of the DI has little influence on the transmission performance. The reason why the achieved EVM performance at 20 GHz is worse than that at 23 GHz could be attributed to the worse system response at 20 GHz.
3.2 Performance verification experimental setup and results
To verify the feasibility of the proposed scheme for four-independent mm-wave (carrier frequency of 28 GHz) signals transmission, we built up another experimental setup shown in Fig. 4. The AWG used in Section 3.1 is replaced by the AWG (Keysight M8196A) with a 3-dB bandwidth of 32 GHz operating at 92 GSa/s in this experiment. The four-channel EA and DP-IQMZM used in Section 3.1 are replaced by a 64 Gbaud coherent driver modulator CDM (NeoPhotonics, HB-CDM CLASS 40 SN: S190220B), which is co-packaged with a linear, quad-channel, and differential 64 GBaud driver and an Indium Phosphide based Mach-Zehnder quadrature modulator chip. All the other experimental devices are the same as the experimental system in Section 3.1.
In Fig. 6, we have got the conclusion that the signal transmission performance on the USB and the LSB are very close. And the transmission performance difference between the signals on X-pol and Y-pol could be negligible when the carrier frequency is higher than 14 GHz as shown in Fig. 7(b). Therefore, only the transmission performance of USB signals on X-pol or Y-pol is measured in this test. We have tested the EVM performance in four cases including single-polarization single-band (SPSB) case, single-polarization dual-band (SPDB) case, dual-polarization cross-band (DPXB) case, and dual-polarization dual-band (DPDB) case. Note that, SPSB means only the USB signal on X-pol is transmitted in the test. SPDB means the USB signal and the LSB signal on X-pol are transmitted, and only the EVM performance of the USB signal on X-pol is measured in the test. DPXB means the USB signal on X-pol and the LSB signal on Y-pol are transmitted simultaneously, and only the EVM performance of the USB signal on X-pol is measured. DPDB means four-independent mm-wave signals at the USB band and LSB band on X-pol and Y-pol are transmitted simultaneously, and only the EVM performance of the USB signal on Y-pol is measured in the test. Figure 9(a) shows the measured EVM performance of 800 MBaud 16QAM signals with different electrical carrier frequencies over 50 km SSMF transmission in four cases. The constellations of the received signals are plotted in the insets of Fig. 9(a), and the optical spectra of these four cases can be seen in Fig. 9(b). It can be seen that SPSB has the best EVM performances because no sideband crosstalk and polarization crosstalk exist in this case and the EVM performance of SPDB is worse than that of DPXB. Note that, the EVM performances of the USB signal on X-pol are measured for SPSB, SPDB, and DPXB cases. Therefore, we could conclude that the crosstalk between USB and LSB induced by the limited ER of the used DP-IQMZM [20] is higher than the crosstalk between two polarization states induced by the limited polarization ER of this DP-IQMZM. The measured EVM performance of 800 MBaud 16QAM signal with the carrier frequency of 28 GHz in the DPDB case is 12.99%, which is much better than the achieved EVM performance of 800 MBaud 16QAM signals with the carrier frequency of 23 GHz in Section 3.1. This could be attributed to the limited system bandwidth in Section 3.1. However, it is proven that our proposed scheme for four-independent mm-wave signals transmission with the same electrical carrier frequency on single-wavelength can be realized without any DSP algorithms in the receiver.
Fig. 9. The measured EVM performance of 800 MBaud 16QAM signals with different electrical carrier frequencies over 50 km SSMF (a), and the optical spectra of four test cases (b).
4. Conclusions
We have experimentally demonstrated a four-independent mm-wave signal generation, transmission, separation, and detection for coherent A-RoF in 5G mobile communication, and DSP-free RAU could be achieved in the proposed scheme. In BBU, four-independent mm-wave signals are modulated on two polarization states of single-wavelength using a DP-IQMZM enabled by dual-SSB modulation and PDM technique. In RAU, a novel carrier polarization rotation module based on self-polarization stabilization technique is designed for optically polarization demultiplexing, and thus four-independent mm-wave signals can be detected by self-coherent detection. Besides, the power fading effect induced by the chromatic dispersion could be overcome thanks to the optical SSB modulation in the transmitter. In this way, no DSP algorithms are required in RAU, contributing to low-latency, simple structure, and low-cost RAU. In our experiment, the measured EVM value of 800 MBaud 16QAM signals with the carrier frequency of 28 GHz over 50 km SSMF transmission could be achieved to 12.99%. These results show that the proposed scheme is a protentional solution for the 5G mobile fronthaul network.
Funding
National Key Research and Development Program of China (2018YFB1800903); National Natural Science Foundation of China (61675083); Fundamental Research Funds for the Central Universities (2019kfyXMBZ033).
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. A. Delmade, C. Browning, T. Verolet, J. Poette, A. Farhang, H. H. Elwan, R. D. Koilpillai, G. Aubin, F. Lelarge, A. Ramdane, D. Venkitesh, and L. P. Barry, “Optical heterodyne analog radio-over-fiber link for millimeter-wave wireless systems,” J. Lightwave Technol. 39(2), 465–474 (2021). [CrossRef]
2. J. P. Santacruz, S. Rommel, U. Johannsen, A. Jurado-Navas, and I. T. Monroy, “Analysis and compensation of phase noise in Mm-wave OFDM ARoF systems for beyond 5G,” J. Lightwave Technol. 39(6), 1602–1610 (2021). [CrossRef]
3. H. Kim, “RoF-based optical fronthaul technology for 5G and beyond,” in Optical Fiber Communication Conference (OFC) 2018, OSA Technical Digest (Optical Society of America, 2018), paper Tu3J.1.
4. K. Kitayama, A. Maruta, and Y. Yoshida, “Digital coherent technology for optical fiber and radio-over-fiber transmission systems,” J. Lightwave Technol. 32(20), 3411–3420 (2014). [CrossRef]
5. K. Kikuchi, “Fundamentals of coherent optical fiber communications,” J. Lightwave Technol. 34(1), 157–179 (2016). [CrossRef]
6. B. Schrenk, “The EML as analogue radio-over-fiber transceiver—A coherent homodyne approach,” J. Lightwave Technol. 37(12), 2866–2872 (2019). [CrossRef]
7. A. Pizzinat, P. Chanclou, F. Saliou, and T. Diallo, “Things you should know about fronthaul,” J. Lightwave Technol. 33(5), 1077–1083 (2015). [CrossRef]
8. M. Shafi, A. F. Molisch, P. J. Smith, T. Haustein, P. Zhu, P. De Silva, F. Tufvesson, A. Benjebbour, and G. Wunder, “5G: A tutorial overview of standards, trials, challenges, deployment, and practice,” IEEE J. Select. Areas Commun. 35(6), 1201–1221 (2017). [CrossRef]
9. X. Chen, S. Chandrasekhar, and P. Winzer, “Self-coherent systems for short reach transmission,” in 2018 European Conference on Optical Communication (ECOC) (IEEE, 2018), pp. 1–3.
10. L. Tao, Z. Dong, J. Yu, N. Chi, J. Zhang, X. Li, Y. Shao, and G. K. Chang, “Experimental demonstration of 48-Gb/s PDM-QPSK radio-over-fiber system over 40-GHz mm-wave MIMO wireless transmission,” IEEE Photonics Technol. Lett. 24(24), 2276–2279 (2012). [CrossRef]
11. B. Koch, R. Noé, D. Sandel, V. Mirvoda, J. Omar, and K. Puntsri, “20-Gb/s PDM-RZ-DPSK transmission with 40 krad/s endless optical polarization tracking,” IEEE Photonics Technol. Lett. 25(9), 798–801 (2013). [CrossRef]
12. H. Huang, C. Sun, C. Lin, C. Wei, W. Zeng, H. Chang, B. Shih, and A. Ng’oma, “Direct-detection PDM-OFDM RoF system for 60-GHz wireless MIMO transmission without polarization tracking,” in Optical Fiber Communication Conference (OFC) 2015, OSA Technical Digest (Optical Society of America, 2015), paper W3F.2.
13. S. Liu, J. Yan, C. Tseng, and K. Feng, “Polarization-tracking-free IFoF mobile fronthaul with adaptively modulated PDM multiband DDO-OFDM,” IEEE Photonics Technol. Lett. 29(14), 1211–1214 (2017). [CrossRef]
14. J. Yan, M. Xu, H. Huang, M. Liao, K. Feng, and G. Chang, “A PDM based spectral aggregation and cell densification for 5G point-to-multipoint mobile fronthaul with a polarization-tracking-free RAU design,” in Optical Fiber Communication Conference (OFC) 2018, OSA Technical Digest (Optical Society of America, 2018), paper Tu3J.7.
15. Y. Chen, C. Tseng, S. Liu, and K. Feng, “PDM DDO-OFDM with self-polarization diversity for the backhaul of radio-over-fiber system,” in Optical Fiber Communication Conference (OFC) 2016 OSA Technical Digest (Optical Society of America, 2016), paper Th4A.2.
16. X. Pan, X. Liu, H. Zhang, K. Wang, Y. Zhang, D. Ran, X. Wang, and C. Wang, “Independent dual single-sideband vector millimeter-wave signal generation by one single I/Q modulator,” Opt. Express 27(14), 19906–19914 (2019). [CrossRef]
17. L. Huang, Q. Yu, L. Deng, S. Fu, M. Tang, M. Cheng, M. Zhang, M. Choi, D. Chang, G. K. P. Lei, and D. Liu, “Novel dual-loop optoelectronic oscillator based on self-polarization-stabilization technique,” Opt. Express 25(18), 21993–22003 (2017). [CrossRef]
18. K. Y. Cho, K. Tanaka, T. Sano, S. P. Jung, J. H. Chang, Y. Takushima, A. Agata, Y. Horiuchi, M. Suzuki, and Y. C. Chung, “Long-reach coherent WDM PON employing self-polarization-stabilization technique,” J. Lightwave Technol. 29(4), 456–462 (2011). [CrossRef]
19. Z. Li, M. S. Erkılınç, K. Shi, E. Sillekens, L. Galdino, B. C. Thomsen, P. Bayvel, and R. I. Killey, “SSBI mitigation and the Kramers-Kronig scheme in single-sideband direct-detection transmission with receiver-based electronic dispersion compensation,” J. Lightwave Technol. 35(10), 1887–1893 (2017). [CrossRef]
20. T. Kawanishi, T. Sakamoto, M. Tsuchiya, and M. Izutsu, “160GHz two-tone lightwave generation using high extinction-ratio optical modulation,” in Proc. LEOS 2006-IEEE Lasers Electro-Opt. Soc., Annu. Meeting Conf., 2006, pp. 195–196.
