We propose and analyze a technique of an optical carrier transmitting two RF signals using optical carrier suppression. A single optical Mach-Zehnder modulator is used for both optical carrier suppression and signal modulation, and optical carrier suppression modulation is also used for frequency conversion of RF signals. This work shows that in contrary to the case of an optical carrier transmitting a single RF signal with optical carrier suppression where stronger optical carrier suppression improves the upconverted RF signal, weaker optical carrier suppression is preferred for an optical carrier transmitting two RF signals due to nonlinear distortion because the nonlinear distortion is reduced by using weaker optical carrier suppression. We find that the usable range of optical carrier suppression ratio is from 10 to 18 dB for RF signal upconverted to 20 GHz and beyond, and the best optical carrier suppression ratio is around 10 dB. We verify the concept and analysis with experiment. In experiment, we used two RFs at 6 and 18 GHz transmitting two 750 Mb/s signals. The experiment for the first time demonstrated that an optical carrier can transmit two RF signals using optical carrier suppression and showed that upconverted RF signals are degraded by nonlinear distortion, particularly for upconverted RF signal at 12 GHz, i.e. the RF signal at the lower frequency.
© 2007 Optical Society of America
The growth of information technology demands improvements in communication system capacity, bandwidth, security, mobility, and flexibility. Currently deployed communication systems do not offer these features simultaneously. The currently developing 3G wireless has transmitted date rate of up to 144 Kb/s for high mobility traffic, 384 Kb/s for low-mobility traffic, and 2 Mb/s in good conditions. However, there are two main limitations with 3G. One is the difficult extension to very high data rate such as 100 Mb/s with code division multiplexing access. The other is the difficulty of providing a full range of multi-rate services. Therefore, the future 4G wireless with features of high data rate and open network architecture is desired to satisfy the increasing demand for broadband wireless access. The key objectives of 4G are to provide reliable transmission of data rate from 100 Mb/s for high mobility applications to 1 Gb/s for low mobility applications. Optical fiber technology can provide tremendous bandwidth, but it does not support user mobility or flexible system reconfiguration. Wireless communication systems using traditional RF and microwave frequencies can provide user mobility, but they do not support high data rates and security. More importantly, radio frequency bands are going to be used up soon except for unlicensed millimeter-wave (mm-wave) band (26–75 GHz). Therefore mm-wave is one possible frequency band for 4G wireless communication. When we use mm-wave band for 4G wireless communications, we have three transmission choices: mm-wave over air, mm-wave over coaxial cable and mm-wave over fiber. Mm-wave over air transmission is limited to tens of meter due to high transmission loss, thus it is impossible to transport wireless signals by use of mm-wave over air. Mm-wave over coaxial cable transmission is limited by the coaxial cable bandwidth and high cost as well as high transmission loss. However, it is well-known that optical fiber has much broad bandwidth and low loss as well as very low cost. As a consequence, it has been believed that mm-wave over fiber for transporting 4G wireless signals is cost-effective technology and has large impact on future potential information technology and wireless communications. Moreover, fiber based broad-band wireless access infrastructure enables transparent delivery of mm-wave radio signals to remote antenna sites. The mm-wave radio-over-fiber (RoF) access systems can be seamlessly integrated with wavelength division multiplexing (WDM) passive optical network infrastructure and the allocation of channel spacing in the future. Therefore, many research works of using fiber to distribute mm-wave radio signals for realizing high speed and capacity wireless access networks have been done [1–23].
One technical challenge of mm-wave over fiber system design is to cost-effectively generate and deliver high quality mm-wave signals to remote antenna sites while maintaining the link simplicity, such as photonic frequency upconversion technique for mm-wave generation. To photonically obtain mm-waves for RoF downlink systems, a number of techniques in mm-wave generation for RF signals have been proposed, such as optical carrier suppression [2–5, 12, 16] using either optical Mach-Zehnder modulator (MZM) or optical phase modulator with optical notching filter, optical frequency multiplication using an optical MZM or optical phase modulator [7, 9, 23], optical remote heterodyne upconversion [14, 22], etc. It has been found that optical carrier suppression technique is one of the cost-effective solutions in photonic frequency upconversion, which usually requires two MZMs, one for frequency upconversion and the other for signal modulation [3, 5] (note this technique cannot be used for an optical carrier transmitting multiple RF signals). In order to simplify the system, it is desirable that one optical modulator is used for both frequency upconversion and signal modulation. Just recently, it was demonstrated that both frequency upconversion and signal modulation can be obtained with one MZM . On the other hand, it is much desired that one optical carrier, i.e. one wavelength, transports multiple optical subcarriers or RF signals, to further leverage the cost. It has been shown that one optical carrier can transport multiple RF signals using optical subcarrier modulation such as single sideband modulation [13, 17] (note photonic frequency upconversion using optical carrier suppression is not obtained). When the optical carrier suppression with one MZM is used for both frequency upconversion and signal modulation, the case of one optical carrier transporting one RF signal has been only demonstrated so far . In other words, an optical carrier transmitting multiple RF signals with optical carrier suppression technique for both frequency upconversion and signal modulation in a single MZM has not been investigated. As a matter of fact, the case of an optical carrier transmitting two or more RF signals is expected to be quite different from the case of an optical carrier transmitting a single RF signal using optical carrier suppression or optical subcarrier modulation. This is due to the fact that radio over fiber systems may not be limited by nonlinear distortion, if an optical carrier transmits a single RF signal. However, it is very likely that radio over fiber systems are limited by nonlinear distortion such as intermodulation distortion when an optical carrier transmits two or more RF signals. When optical carrier suppression modulation is used with multiple RF signals, both optical modulation index, which determines the optical carrier suppression, and frequency allocation will determine the impact of nonlinear distortion, and thus the quality of upconverted RF signals.
In this paper, we first numerically investigate the limitations of RoF downlinks using optical carrier suppression for both frequency upconversion and signal modulation for the technique of an optical carrier transmitting two RF signals. It will be shown that both optical carrier suppression and RF allocation are crucial and quite different from the case of an optical carrier transmitting one RF signal. Then, we experimentally verify the concept.
2. Proposed frequency upconversion of two RF signals using optical carrier suppression
The principle of proposed RoF downlink with an optical carrier transmitting two RF signals using optical carrier suppression in a single MZM is illustrated schematically in Fig. 1. Two different baseband signals are electrically up-converted to radio signals at f 1 and f 2 (f 2>f 1). The converted two RF signals are combined through a RF power combiner. The MZM is biased at the minimum transmission to realize optical carrier suppression modulation. The MZM is driven by the combined two RF signals and modulates the incident optical carrier, thus four first-order optical subcarriers are generated at -f 2, -f 1, f 1 and f 2 with respect to the optical carrier, and the optical carrier is suppressed as shown in Fig. 1 inset (a). The four optical subcarriers transmit down a fiber from a central station to a remote antenna site. After photodetection, two upconverted RF signals at frequencies of 2f 1 and 2f 2 are generated at least due to beating between the four optical subcarriers. The other beats such as at 2f 2-f 1, 2f 1-f 2, and f1+f2 etc. are not shown in Fig. 1 inset (e). This is called electrical approach. The other method is to first separate the four optical subcarriers into two groups before photodetection by optical filtering as shown in Fig. 1 insets (b) and (c), then photodetection generates two RF signals at 2f 1 and 2f 2, as shown in Fig. 1 inset (d). This is referred to optical approach. Intuitively, RoF downlink system may experience more nonlinear distortion by using the electrical approach compared to the optical approach. In order to illustrate the impact of two RF signals carried by one optical carrier compared to one RF signal using optical carrier suppression, we consider RFs at 10 and 15 GHz and each RF carries a 2.5 Gb/s signal. The Mach-Zehnder modulator, optical extinction ratio of 35 dB, is biased at minimum transmission, and driven by RF signals with modulation index of VRF/Vπ=1, VRF -modulation voltage and Vπ -the modulator π -phase shift voltage. In the optical approach, we used two cascaded fiber Bragg gratings (FBG) with an optical circulator and the two FBGs, 3-dB bandwidth of 5 GHz for each FBG, and one allocated at +15 GHz and the other allocated at -15 GHz, reflected the optical subcarriers at ±15 GHz and transmitted the optical subcarriers at ±10 GHz. Thus input, transmitted and reflected optical spectra of the optical filter are shown in Fig. 2. There may be overlapped nonlinear distortions which are not visible in the optical spectra in Fig. 2. Correspondingly, eye diagrams for using the electrical approach are shown in Fig. 3 and the optical approach in Fig. 4, compared to the case of an optical carrier transmitting a single RF signal at 10 GHz. It is seen from both Figs. 3 and 4 that nonlinear distortion does involve in the generation of upconverted RF signals, and the performance of the RoF downlink becomes worse when an optical carrier transports two RF signals compared to one RF signal transmitted. Figures 3 and 4 simply show that there are nonlinear distortions which are overlapped and/or very close to the optical subcarriers. It is quite different that an optical carrier transmits one and two RF signals using optical carrier suppression. In addition, note the RF signal at the lower frequency is more degraded by distortion and residual optical carrier than that at the higher frequency as seen in Figs. 3 and 4. Therefore, we only focus on the analysis of the RF signal at the lower frequency. To obtain eye diagrams, we used an electrical bandpass filter (6th-order Bessel) centered at 20 and 30 GHz with a bandwidth of 3.5 GHz, an electrical mixer with a local oscillator at 20 and 30 GHz and a low-bandpass filter with a bandwidth of 2.5 GHz for each RF signal.
3. Numerical analysis by simulation
When optical carrier suppression modulation is used for RF signal frequency upconversion and signal modulation simultaneously, it is found that two important parameters must be considered and investigated, i.e. optical carrier suppression ratio and the frequency allocation. Particularly, the optical carrier suppression ratio plays a significant role when two RF signals are carried by one optical carrier with optical carrier suppression. We suppose that each RF signal carries a 2.5 Gb/s signal as mentioned above. Also we assume that a standard single-mode fiber with group velocity dispersion (GVD) of 17 ps/nm/km and loss of 0.2 dB/km is used. We used VPI TransmissionMaker 7.01 to model the RoF downlink system and carry out our investigation and analysis. The other parameters are the same as above if not stated.
3.1 Impact of optical carrier suppression ratio
We first investigate the impact of optical carrier suppression ratio. The optical carrier suppression ratio is defined as the power of the first-order optical sidebands, i.e. optical subcarrier signals, divided by the power of the suppressed optical carrier. The optical carrier suppression ratio can be altered with the modulation driving voltage of the MZM. We use simulated Q factor to characterize the frequency up-converted RF signals. To this end, we have assumed that the launched optical power into fiber is assumed 3 dBm, and optical receiver thermal noise is 10-12 A/(Hz)1/2. The optical receiver is limited by thermal noise and nonlinear distortion in sensitivity. As an example, we consider f 1=10 GHz and f 2=15 GHz as shown in Fig. 1 without loss of generality. Thus, the two upconverted RF signals will be allocated to 20 and 30 GHz at remote antenna sites. For easy comparison, we also consider a case that an optical carrier transports one RF signal at 10 GHz, and thus 20-GHz RF signal at remote antenna sites is upconverted by optical carrier suppression and photodetection. Simulated Q factor versus fiber length with respect to optical carrier suppression ratio is shown in Fig. 4 for the two cases: an optical carrier transmitting a single RF signal at 10 GHz and two RF signals at 10 and 15 GHz. Note that for transporting a single RF signal using optical carrier suppression as in Fig. 5(a), the maximum fiber distance approximately follows the walk-off length LB=TB/(D2f 1), where TB is the bit period of data rate, D is the fiber chromatic dispersion and 2f 1=20 GHz is the frequency interval of the two optical subcarriers. It is worth noting that a higher optical carrier suppression ratio improves Q factor for this case. In other words, a higher optical carrier suppression ratio is preferable for the case of a single RF signal carried by optical carrier suppression modulation. Correspondingly, for an optical carrier transmitting two RF signals, Fig. 5(b) shows the Q-factor with fiber distance for upconverted RF signal at 20 GHz using the electrical approach (up-converted RF signal at 30 GHz is not shown here). When an optical carrier delivers two RF signals simultaneously using optical carrier suppression, it is obvious as shown in Fig. 5(b) that the Q-factor is significantly reduced due to serious nonlinear distortion compared to Fig. 5(a), and also the performance is degraded with the increase of optical carrier suppression ratio, which is contrary to Fig. 5(a). In other words, a weaker optical carrier suppression ratio is preferred rather than a higher optical carrier suppression ratio for this case. This suggests that frequency upconversion with optical carrier suppression used for carrying one RF signal and two RF signals is quite different in the requirement of the optical carrier suppression ratio. The physical origin lies in that stronger optical carrier suppression will transfer more optical power from the optical carrier to the optical sidebands, thus more RF power will be obtained after photodetection, but also more high-order optical subcarriers are induced. Therefore, due to multiple optical subcarriers including first- and higher orders, more nonlinear distortion is introduced which limits the maximum optical carrier suppression ratio. Thus, the balance between the RF signal power and nonlinear distortion has to be correctly set. It is found that using the optical approach a similar behavior as in Fig. 5(b) is obtained, even though optical bandpass filtering reduces nonlinear distortion to some extent.
3.2 Impact of frequency allocation
Because optimum optical carrier suppression ratio may be dependent on frequency allocation, as an example a frequency difference of 5 GHz between the two RFs which drive the optical MZM is considered to investigate the impact of the frequency allocation on optical carrier suppression ratio. The optimum optical carrier suppression ratio with the relation of the lower RF, i.e. f 1, is shown in Fig. 6. Figure 6 gives the optimum optical carrier suppression ratio range, within which fiber reach is reduced by maximum 20% of the longest fiber distance, and the best optical carrier suppression ratio is indicated by a dot on the line. It is clear that as the lower RF become smaller, i.e. optical subcarriers at ± f 1 are getting closer to the optical carrier, the optimum optical carrier suppression ratio range is reduced and a larger optical carrier suppression ratio is preferred. This is because as the optical subcarriers approach the optical carrier, upconverted RF signal will be more degraded by the baseband induced distortion. Therefore, stronger optical carrier suppression is preferred. However, when RF signal frequency is increased to 10 GHz and beyond (thus upconverted RF of 20 GHz and beyond), optimum optical carrier suppression ratio range is increased and ranged from 10 to 18 dB, and also note that the best optical carrier suppression ratio is around 10 dB. This suggests that for the upconversion of RF signals to mm-wave band from 26 to 75 GHz using optical carrier suppression, an optical carrier suppression ratio of 10 dB around is the best when an optical carrier transmits two RF signals. It is expected that when an optical carrier transmits three or more RF signals with optical carrier suppression, a further lower optical carrier suppression ratio may be used because much more nonlinear distortion will be introduced.
Now we investigate the impact of frequency difference. For this case, we keep an optical carrier suppression ratio of 10 dB fixed. We use eye opening penalty, i.e. eye closure, of the upconverted RF signals to characterize the frequency upconverted RF signals. We alter the lower RF f 1 for a given RF difference Δf=f 2-f 1 to analyze the impact of frequency allocation. As an example we consider two cases: Δf=f 2-f 1=5 and 10 GHz (the minimum Δf is limited by data rate). Simulated eye opening penalty of upconverted RF signal at 2f 1 versus the lower RF f 1 is shown in Fig. 7(a) for the electrical approach and Fig. 7(b) for the optical approach. It is found that frequency difference Δf plays a critical role in the upconverted RF signals for f 1<10 GHz, particularly for the electrical approach. Figure 7 shows that the eye-opening penalty quickly decreases as f 1 increases. Particularly for f 1<Δf, both electrical and optical approaches result in a significant increase of eye opening penalty due to nonlinear distortion, and residual optical carrier for the lower RF signal. Figure 7(b) clearly shows that the optical approach leads in a better eye opening than the electrical approach for f 1<10 GHz. This can be easily understood that the two groups of optical subcarriers in the optical domain are separated before the optical receiver, and thus less nonlinear distortion is produced. On the other hand, it is seen that the frequency difference of 5 and 10 GHz does not lead to significant difference in performance when f 1 is 10 GHz and beyond as shown in Fig. 7. This suggests that the upconverted RF signal is improved significantly if both optical carrier suppression ratio of 10 dB around and the lower RF f 1 of beyond 10 GHz are used. Consequently, our analysis shows that the frequency upconversion of multiple RF signals to mm-wave band (26–75 GHz) with optical carrier suppression modulation may not depend on frequency allocation and only optical carrier suppression ratio will play a significant role.
4. Experimental setup and concept proof
The experimental setup for one optical carrier transmitting two RF signals using optical carrier suppression is shown in Fig. 8 to verify the above concept. Due to limited resources we only verify the optical approach with f 1=6 GHz and f 2=18 GHz. The upconverted RF signals will be thus at 12 and 36 GHz at remote antenna sites. A CW lightwave was generated by a distributed feedback laser-diode (DFB-LD) with a wavelength of 1540.2 nm and modulated via a LiNbO3 (LN)-MZM driven by the combined two RF signals. The two RF signals were obtained by using two individual electrical mixers to up-convert two individual data signals at 750 Mb/s to 6 and 18 GHz. The two 750 Mb/s signals were generated from two different pattern generators with a pattern length of 231-1. The LN-MZM was biased at the minimum transmission point to realize optical carrier suppression modulation. The optical spectrum after the LN-MZM is inserted in Fig. 8 inset (i). We can see that the optical carrier suppression ratio is around 15 dB, and the high-order sidebands are 25 dB lower than the four first-order optical sidebands. Then the all optical subcarrier signals were transmitted over 20 km single-mode fiber before they were separated by optical filtering. We used a 25/50 GHz optical interleaver to separate the two groups of optical subcarriers. After the optical separation, we used another 0.3-nm tunable optical filter to further suppress higher-order sidebands. Then the separated optical subcarriers were pre-amplified by an erbium doped fiber amplifier (EDFA) with a small-signal gain of 30 dB and filtered by an optical filter with a bandwidth of 1 nm before photodetection. The optical spectra of the two groups of the optical subcarriers are shown in Fig. 8 insets (ii) and (iii). It is seen that the residual optical carrier accompanies the optical subcarriers at ±6GHz as shown in Fig. 8 inset (ii), and the optical carrier is suppressed in the optical spectrum as shown in Fig. 8 (iii). This is very similar to optical filtering in the simulation as shown in Fig. 2. The upconverted RF signals are generated with a 60-GHz PIN photodiode. In the receiver, converted electrical signals at 12 GHz and 36 GHz were amplified by electrical amplifiers with a bandwidth of 5 GHz centered at 12 GHz and a bandwidth of 10 GHz centered at 40 GHz, respectively. Two electrical local oscillator signals at 12 GHz and 36 GHz were used to down-convert the two upconverted RF signals to baseband. Eventually, the down-converted 750 Mbit/s signals were detected by an optical oscilloscope and a bit error rate (BER) tester.
Figure 9 shows eye diagrams after transmission over 20 km single-mode fiber before electrical down-conversion and after down-conversion for the two RF signals at 20 and 36 GHz, respectively. The eye diagrams were measured by using an oscilloscope with a 50-GHz bandwidth. Obviously eye diagrams at both 12 GHz and 36 GHz are distorted seriously by nonlinear distortion. This is partially because an optical carrier suppression of 15 dB is used, and such a ratio has induced serious nonlinear distortion. Particularly note that the measured eye diagram shows that the RF signal at 12 GHz experiences more nonlinear distortion, and thus the RF signal at 12 GHz is much worse than that at 36 GHz, which well agrees with the above numerical prediction. Correspondingly, measured BER versus optical receiver power is shown in Fig. 10. It is seen that no BER floor is observed even though nonlinear distortion degrades RF signals seriously.
We have proposed and analyzed a technique of an optical carrier transmitting two RF signals using optical carrier suppression modulation, where optical carrier suppression modulation is used for both signal modulation and frequency upconversion. It is found that comparing to the case of an optical carrier transmitting a single RF signal where stronger optical carrier suppression ratio results in a better performance, for an optical carrier transmitting two RF signals, on the contrary much weaker optical carrier suppression is required in order to obtain less-distorted or better upconverted RF signals. Moreover, we found that when an optical carrier transmits two RF signals using optical carrier suppression modulation for both frequency upconversion and signal modulation, the lower frequency of RF signals is limited to 10 GHz and beyond. If beyond 10 GHz, the optical carrier suppression ratio ranged only from 10 to 18 dB can be used, and the best optical carrier suppression ratio is around 10 dB. This implies that using optical carrier suppression with two or more RF signals can be used for upconversion to mm-wave (26–75 GHz) only if optical carrier suppression is carefully chosen. The experiment successfully verified the proposed technique of an optical carrier transmitting two RF signals using optical carrier suppression. In the experiment, two upconverted RF signals at 12 and 36 GHz with data rate of 750 Mb/s after transmission over 20 km fiber are seriously distorted by nonlinear distortion, but no BER floor is induced by nonlinear distortion. Moreover, the experiment clearly shows that upconverted RF signal at 12 GHz is worse than that at 36 GHz, which agrees well with the above prediction by simulation.
References and links
1. M. Bakaul, A. Nirmalathas, C. Lim, D. Novak, and R. Waterhouse, “Investigation of performance enhancement of WDM optical interfaces for millimeter-wave fiber-radio networks,” IEEE Photon. Technol. Lett. 19, 843–845 (2007). [CrossRef]
2. K. Wang, X. Zheng, H. Zhang, and Y. Guo, “A radio-over-fiber downstream link employing carrier-suppressed modulation scheme to regenerate and transmit vector signals,” IEEE Photon. Technol. Lett. 19, 1365–1367 (2007). [CrossRef]
3. J. Yu, Z. Jia, T. Wang, and G. Chang, “Centralized lightwave radio-over-fiber system with photonic frequency quadrupling for high-frequency millimeter-wave generation,” IEEE Photon. Technol. Lett. 19, 1499–1501 (2007). [CrossRef]
4. J. Yu, Z. Jia, T. Wang, and G. K. Chang, “A novel radio-over-fiber configuration using optical phase modulator to generate an optical mm-wave and centralized lightwave for uplink connection,” IEEE Photon. Technol. Lett. 19, 140–142 (2007). [CrossRef]
5. L. Chen, Y. Shao, X. Lei, H. Wen, and S. Wen, “A novel radio-over-fiber system with wavelength reuse for upstream data connection,” IEEE Photon. Technol. Lett. 19, 387–389 (2007). [CrossRef]
6. C. Lin, J. Chen, P. Peng, C. Peng, W. Peng, B. Chiou, and S. Chi, “Hybrid optical access network integrating fiber-to-the-home and radio-over-fiber systems,” IEEE Photon. Technol. Lett. 19, 610–612 (2007). [CrossRef]
7. M. Garcia Larrode, A. Koonen, J. Vegas Olmos, and E. Verdurmen, “Microwave signal generation and transmission based on optical frequency multiplication with a polarization interferometer,” J. Lightwave Technol. 25, 1372–1378 (2007). [CrossRef]
8. H. Lu, S. Tzeng, Y. Chuang, Y. Chi, and C. Liao, “Bidirectional radio-over-DWDM transport systems based on injection-locked VCSELs and optoelectronic feedback techniques,” IEEE Photon. Technol. Lett. 19, 315–317 (2007). [CrossRef]
9. G. Qi, J. Yao, J. Seregelyi, S. Paquet, C. Belisle, X. Zhang, K. Wu, and R. Kashyap, “Phase-noise analysis of optically generated millimeter-wave signals with external optical modulation techniques,” J. Lightwave Technol. 24, 4861–4875 (2006). [CrossRef]
10. T. Cho and K. Kim, “Effect of third-order intermodulation on radio-over-fiber systems by a dual-electrode Mach-Zehnder modulator with ODSB and OSSB signals,” J. Lightwave Technol. 24, 2052–2058 (2006). [CrossRef]
11. M. Bakaul, A. Nirmalathas, C. Lim, D. Novak, and R. Waterhouse, “Simultaneous multiplexing and demultiplexing of wavelength-interleaved channels in DWDM millimeter-wave fiber-radio networks,” J. Lightwave Technol. 24, 3341–3352 (2006). [CrossRef]
12. C. Lin, W. Peng, P. Peng, J. Chen, C. Peng, B. Chiou, and S. Chi, “Simultaneous generation of baseband and radio signals using only one single-electrode Mach-Zehnder modulator with enhanced linearity,” IEEE Photon. Technol. Lett. 18, 2481–2483 (2006). [CrossRef]
13. M. Bakaul, A. Nirmalathas, C. Lim, D. Novak, and R. Waterhouse, “Hybrid multiplexing of multiband optical access technologies towards an integrated DWDM network,” IEEE Photon. Technol. Lett. 18, 2311–2313 (2006). [CrossRef]
14. X. Zhang, B. Liu, J. Yao, K. Wu, and R. Kashyap, “A novel millimeter-wave-band radio-over-fiber system with dense wavelength-division multiplexing bus architecture,” IEEE Trans. Microwave Theory Tech. 54, 929–937 (2006). [CrossRef]
15. A. Kaszubowska, L. Hu, and L. Barry, “Remote downconversion with wavelength reuse for the radio/fiber uplink connection,” IEEE Photon. Technol. Lett. 18, 562–564 (2006). [CrossRef]
16. L. Chen, H. Wen, and S. Wen, “A radio-over-fiber system with a novel scheme for millimeter-wave generation and wavelength reuse for up-link connection,” IEEE Photon. Technol. Lett. 18, 2056–2058 (2006). [CrossRef]
17. C. Wu and X. Zhang, “Impact of nonlinear distortion in radio over fiber systems with single-sideband and tandem single-sideband subcarrier modulations,” J. Lightwave Technol. 24, 2076–2090 (2006). [CrossRef]
18. B. Masella and X. Zhang, “A novel single wavelength balanced system for radio over fiber links,” IEEE Photon. Technol. Lett. 18, 301–303 (2006). [CrossRef]
19. K. Wu, J. Yao, X. Zhang, and R. Kashyap, “Millimeter-wave photonic techniques for broadband communication and sensor applications,” Proceedings of IEEE LEOS annual meeting 2006, Montreal, pp.270–271.
20. K. Ikeda, T. Kuri, and K. Kitayama,” Simultaneous three-band modulation and fiber-optic transmission of 2.5-Gb/s baseband, microwave-, and 60-GHz-band signals on a single wavelength,” J. Lightwave Technol. 21, 3194–3202 (2003). [CrossRef]
21. M. Attygalle, C. Lim, and A. Nirmalathas, “Dispersion-tolerant multiple WDM channel millimeter-wave signal generation using a single monolithic mode-locked semiconductor laser,” J. Lightwave Technol. 23, 295–303 (2005). [CrossRef]
22. T. Nakasyotani, H. Toda, T. Kuri, and K. Kitayama, “Wavelength-division-multiplexed Millimeter-waveband radio-on-fiber system using a supercontinuum light source,” J. Lightwave Technol. 24, 404–410 (2006). [CrossRef]
23. M. Mohamed, B. Hraimel, X. Zhang, and K. Wu, “Efficient photonic generation of millimeter-waves using optical frequency multiplication in radio over fiber systems,” Proceedings of IEEE Topic meeting on Microwave Photonics 2007, paper Th.-4.20, Victoria, Canada.