Open Access
Add to BibSonomy
Abstract
An ultra simple and low cost method to detect low-power RF signal is proposed and experimentally demonstrated based on optoelectronic feedback DFB semiconductor laser. To our knowledge, according to public reports, this is the simplest photonics-assisted method which avoids using high sensitive optical modulators and narrow bandwidth optical filters. The RF signal, which matches the oscillation mode at the relaxation oscillation peak of the DFB laser, is amplified based on optoelectronic feedback. The RF signal from 1 to 4.5 GHz can be detected by adjusting the frequency of relaxation oscillation which is related to the laser bias current. The system provides a maximum gain of 15 dB for the low-power RF signal. The sensitivity of the system can reach up to as high as -97 dBm. Considering the real application of the detection system, the properties like dynamic range, resistance to large signals and performance for detecting modulated RF signal are also investigated.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Detecting low-power radio frequency (RF) signal is very important in many applications, such as modern radar, wireless communication, radio astronomy, channelization and electronic warfare systems [1–5]. How to detect the weak signal in the cluttered environment is the key problem to be solved in these applications. It is significant not only to determine the frequency of the signal, but also amplify and demodulate the signal with data in signal detection and analysis systems. In electronic domain, many approaches have been proposed to detect and amplify low-power RF signal commonly based on narrowband filters and amplifiers [6]. However, it is difficult to realize RF signal detection of large instantaneous bandwidth and wide frequency coverage by only electronic methods [7]. The photonics-assisted method provides a potential way to detect the RF signal with the advantages of broad frequency coverage, small size, low cost, instantaneous bandwidth and avoiding the electromagnetic interference [8]. Many photonic techniques have been proposed to detect RF signal over a broad bandwidth, such as dispersion element [9], fine tunable optical filters [10], and microwave channelization [11]. However, the sensitivity of these systems is not high enough.
The multimode optoelectronic oscillator (OEO) provides an effective method to detect low-power RF signal. Differ from single mode OEO which can generate high-frequency and low phase noise microwave signal [12–14], the multimode OEO, which works just below the single mode oscillation threshold gain, provides gain for the RF signal which aligns with one oscillation mode, while gives loss to that outside the oscillation mode. Many approaches are proposed to realize RF signal detection based on multimode OEO [15–17]. The wideband low-power RF signal can be detected using stimulated Brillouin scattering-based OEO [15]. The bandwidth can be detected from 1 to 17 GHz which reflects the advantage of optical detection, while the sensitivity is just -72 dBm. The structure is a little bit complex by using the pump laser and electronic amplifier, the high nonlinear fiber will increase the instability of the system. In [16], multimode OEOs without electrical bandpass filters inside the OEO cavity has been proposed to detect low power RF signals based on the injection locking. A gain of 8 dB and a detection sensitivity of -83 dBm for RF signals ranging from 1 to 6 GHz are demonstrated. The tunable OEO based on a phase-shifted fiber Bragg grating (PS-FBG) is demonstrated for RF signal detection and amplification [17]. The detection bandwidth is from 1.5 to 5 GHz with the power as low as -91 dBm. However, the optical modulator with high sensitivity is indispensable which is ensured that the low-power RF signal can be modulated to the optical carrier in above systems. Other active/passive components will increase system cost and structure complexity.
In this paper, an ultra simple and low cost method to detect low-power RF signal is proposed and experimentally demonstrated based on optoelectronic feedback distributed feedback (DFB) semiconductor laser. To our knowledge, according to public reports, the system is the simplest photonics-assisted method which avoids using high sensitive optical modulators, narrow bandwidth optical filters or electrical amplifiers. The RF signal, which matches the oscillation mode at the relaxation oscillation peak of the DFB laser, is amplified based on optoelectronic feedback. The tunable of the system can be realized by adjusting the frequency of relaxation oscillation which is related to the laser bias current. So, the frequency from 1 to 4.5 GHz can be detected. The system provides a maximum gain of 15 dB for the low-power RF signal. The sensitivity of the system can reach up to as high as -97 dBm. Considering the real application of the detection system, the properties such as dynamic range, resistance to large signals and performance for detecting modulated RF signal are also investigated.
2. Principle and experimental setup
Figure 1 illustrates the experimental diagram of proposed RF signal detection. An optical carrier generated from the DFB laser (DWT0029 LD15DM-SP-LC) is sent to the erbium-doped fiber amplifier (EDFA, EDFA-BA15-APC). The threshold of the DFB laser is 8 mA, the center wavelength is 1549.23 nm which can be controlled by temperature. The amplified optical signal is convert to electrical signal by the photodetector (PD, TELEDYNE LECROY OE6250G-M) with the bandwidth of 40 GHz and the responsivity of 0.85 A/W. The electrical signal is divided equally into two parts by the 5:5 electrical splitter (ES, KEYTAR MODEL 6005265 0.5-26.5 GHz). One part is sent to the electrical spectrum analyzer (ESA, ROHDE&SCHWARZ FSW-SIGNAL&ANALYZER) for monitoring the low RF signal. Another 50% part is sent to an electrical coupler (EC, KEYTAR MODEL 6005265 0.5-26.5 GHz) with 5:5 coupling ratio. The low RF signal generated from the vector signal generator (ROHDE & SCHWARZ SMBV100A) is sent to the EC and then directly modulated to the DFB laser. The EDFA is used to compensate the loss of the optoelectronic loop which is consist of the photoelectric and electro-optic conversion loss, transmission loss and connection loss. The optical output power after the EDFA is adjusted below the max input power of the PD and above the minimum power needed to overcome the loss. There are only three active devices including DFB laser, EDFA and PD, which makes the system ultra simple. The optical modulator with extremely high sensitivity and fine optical filters are also avoided using which makes the system lower cost.
Fig. 1. Schematic diagram of the proposed RF signal detection (EDFA: erbium-doped fiber amplifier, PD: photodetector, ES: electrical splitter, EC: electrical coupler, OSA: optical spectrum analyzer, ESA: electrical spectrum analyzer).
The realization of weak RF signal detection benefits from the gain provided by the optoelectronic feedback in the relaxation oscillation of the DFB laser. The DFB laser generates an optical carrier with the relaxation oscillation frequency (fm) as Fig. 2(a) shows. As we know, the frequency located at fm will achieve the most effective modulation efficiency when a semiconductor laser is directly modulated by an RF signal. It can be clearly seen from the frequency response of a semiconductor laser in Fig. 2(b), where the fm is determined. Before the loop is closed, the output RF signal is lower than the input signal because of the photoelectric and electro-optic conversion loss as Fig. 2(c) shows. After the loop is closed, adjusting the gain to make the system work at the multi mode oscillation state in Fig. 2(d). The loop gain is controlled just below oscillation threshold by adjusting the pump current of the EDFA which is different from single mode OEO. So, the oscillation modes are almost equal to each other. Accordingly, the frequency located at fm which matches with the oscillation mode will acquire the largest gain in the loop as Fig. 2(e) shows. During the RF detection, the system is sensitive to the low power RF signal which is equal to one oscillation mode because it provides gain for the signal. While, the system provides loss to the RF signal which is not matching with the oscillation modes in Fig. 2(f).
Fig. 2. Illustration of the RF signal detection: (a) optical carrier from the DFB laser before modulated. (b) the modulation response of the laser with a relaxation oscillation at fm. (c) the output RF signal before the loop is closed. (d) multi mode oscillation state after the loop closed around fm when there is no input RF signal. (e) the RF signal is amplified when it matches with the oscillation mode under optoelectronic feedback.(f) the RF signal gets loss when it mismatches with the oscillation mode.
3. Experimental results
An experiment is carried out based on the configuration in Fig. 1 shows. At first, to investigate the selectable amplification of the system, the bias current of the laser is set at 10.3 mA whose threshold current is 8 mA. Adjust the gain of the EDFA to make the system at the multi-mode oscillation state. The electrical spectrum is the inset in Fig. 3(a). The oscillation modes are not flat because the modulation response at the relaxation oscillation is not flat, the mode at the peak gets more gain like Fig. 2(c) shows. Two RF signals of f1=2.6134 GHz and f2=2.6171 GHz with same power of -75 dBm are injected to the system. As we can see from Fig. 3(a), the span of ESA is 1 GHz, the f1 which matches one oscillation mode is amplified and detected, while the f2 which mismatches with the mode is drowned in the noise floor. Here, the oscillation modes are identified as noise. If the measured power of RF signal is below the power of oscillation modes (-81.14 dBm), it can not be detected. Set the span at 25 MHz, the RF signals can be seen clearly as Fig. 3(b) shows. The power of f1 is -59.8 dBm which acquires the gain of 15.2 dB compared with RF signal source. The power of f2 is -80.64 dBm which gets loss of 5.64 dB. The oscillation mode spacing is about 7.4 MHz, which is mainly determined by the loop length, including the pigtails of the optoelectronic devices and the erbium-doped fiber in the EDFA.
The amplification property for RF signals at different frequencies is measured. According to the principle, the RF signal at the relaxation oscillation peak will get the max gain, so, the tunable of the detection system can be acquired by adjusting the relaxation oscillation which can be controlled by the bias current of the laser. In the experiment, the bias current of the laser is tuned from 8.4 mA to 12.9 mA to make sure the RF signal at different frequencies acquire the max gain. The power of RF signal source is fixed at -75 dBm. The detected RF signal at the frequency from 1 to 4.5 GHz is shown in Fig. 4. During the detection range, the RF signal will acquire gain and the power is above -75 dBm. The RF signal gets amplified about 15 dB from 1 GHz to 3 GHz benefitted from the optoelectronic oscillation which is 5 dB more than the system in Ref. [17]. It should be noted that, the RF signal power will leak out from the EC and ES to the ESA. The isolations between the two input/output ports of EC/ES are all 18 dB. So, the total isolation from RF signal source to ESA is 36 dB. Set the power of RF signal at -75 dBm, the signal power detected on the ESA is -117 dBm. The leakage of the RF signal will decrease the detected signal power because less power is injected to the DFB laser. So, the amplification performance is a bit degraded.
While the gain rolls down from 3 GHz. To investigate the limitation of the system’s detection bandwidth, the influence factor like transmission response (S21) and reflection response (S11) of the DFB laser are measured. As we can see from Fig. 5, by tuning the bias current from 8.4 mA to 13.8 mA, the peak of the relaxation oscillation varies from 1 to 5 GHz. The inset figure is the reflection response of the laser. The gain at lower frequency is as large as 15 dB because the bandwidth of the transmission response is narrow at the low frequency, the low-frequency oscillation modes will get full gain. While, as the improvement of the relaxation oscillation, the gain provided by the system will be distributed to all oscillation modes before the relaxation oscillation although the oscillation mode at the peak gets more gain than others. So, the gain of the RF signal at high frequency decreases relatively. This question can be solved by increasing the gain provided by the EDFA. However, the max input power of the PD used in the experiment is no more than 8 dBm which limits the output power of the EDFA. So, employing another PD with higher maximum input power will increase the gain for high frequency RF signal and improve the detection bandwidth further. On the other hand, the reflection response of the DFB laser at high frequency is higher than that at low frequency, which means the power of high frequency RF signal injected into the system is lower. So, employing the DFB laser with high frequency packaged will improve the gain at high frequency too. The maximum detection range is determined by the modulation frequency of the DFB laser based on the scheme. The solution to greatly improve the detection range is to improve the modulation frequency of the laser, such as optical injection, which is taken into account in the next work.
Fig. 5. Modulation response of the DFB laser when the bias current varies from 8.4 mA to 13.8 mA, the inset is S11 response.
Another important parameter of the detection system is the sensitivity which is defined as the lowest RF power the system can detect. The RF signal will be amplified when its frequency equals with one oscillation mode. Decreasing the power of the RF source until the detected RF signal is drown under the oscillation mode, it can not be detected. Therefore, the sensitivity depends on the noise floor of the oscillation modes. Figure 6 shows the electrical spectrum of the detected RF signal when the input power varies from -80 dBm to -97 dBm. It can be seen that the noise floor is -81.14 dBm. The detected RF signal can be distinguished with the oscillation mode until the input power reduced to -97 dBm. So, the sensitivity of the detection system is -97 dBm which is 6 dB better than that in Ref. [17]. The gain dynamic range can be explained as the difference between the lowest detectable power and the highest power when the gain to the detected RF signal tends to be saturated. The output power with different input power from -97 dBm to -9 dBm is measured, and the results are shown in the inset of Fig. 6. It can be seen that as the input power is low, the gain to the injected signals is constant. When the input power reaches -9 dBm, the gain to the injected signal decreases to 0 dB. Accordingly, the gain dynamic range of the system is 88 dB for the signal of 2.6134 GHz.
Considering that the system performance is related to the modulation response of the DFB laser, which is different at different frequencies as Fig. 5 shows. The sensitivity and gain dynamic range are measured at 1.5012, 2.9777 and 4.4991 GHz which are shown in Fig. 7. The sensitivity remains almost unchanged although the gain at high frequency decreases 10 dB compared with that at low frequency. Because the sensitivity is determined by the noise floor which at the same time decreases as the gain decreases. The gain for the signal at 1.5012, 2.9777 and 4.4991 GHz reduces to 0 dB when the input power reaches to -18, -10 and -10 dBm. That’s to say, the dynamic range at 1.5012 GHz is smaller. As we can see from Fig. 5, the bandwidth of the modulation response at 1.5 GHz is narrower than that at higher frequency. So, the number of the oscillation modes is fewer. Within the bandwidth, the other oscillation modes are more sensitive to the big input signal which will grab the system gain easier. Therefore, the detected signal will get less gain resulting in the reduced dynamic range.
Fig. 7. Electrical spectrum of the detected signal at different frequency with different power. (a) 1.5012 GHz, (b) 2.9777 GHz, (c) 4.499 GHz.
Next, we research how the system will operate when injecting interference signal at the next mode with high power to the low-power detected signal. When a large interference signal is injection to the system, it may dominate the gain provided by the system and the detected signal will get less gain until 0 dB. In the experiment, the RF signal source is set to 1.5012, 2.6134, 2.9777 and 4.4991 GHz with the fixed power of -75 dBm respectively, the corresponding interference signal which is at the next oscillation mode is 1.5086, 2.6208, 2.9851 and 4.5065 GHz. Increasing the power of interference signal from -40 dBm, the detected RF signals are measured like Fig. 8 shows. The RF signal remains unchanged until the interference power increases to -36, -32, -28 and -24 dBm respectively. On the other hand, the detected signal will get no gain when the interference signal power increases to -17, -10, -10, -9 dBm respectively, because the power of detected signal will decrease as the interference signal becomes bigger. All in all, the anti-interference ability to big signals of the system for higher frequency becomes stronger.
Fig. 8. Electrical spectrum of the detected signals at (a) 1.5012, (b) 2.6134, (c) 2.9777 and (d) 4.4991 GHz when injecting interference signal at the next mode with different power.
Finally, to investigate the real application for signal detection, the modulated signal is generated by the VSG are injected to the system. In the experiment, the quadrature phase shift keying (QPSK) modulated signal is utilized for its advantage of high spectral and modulation efficiency. The RF signal with different QPSK modulation rates of 2.4, 4.8, 10 and 20 ksps, whose frequency is 2.6314 GHz and power is fixed at -75 dBm, is injected to the input of the EC. The detected signal is measured by the VSA module in the ESA and the corresponding results are plotted in Fig. 9. As we can see, the constellation of the signal is clear. The error vector magnitudes (EVMs) are 6.54%, 7.87%,11.54% and 15.87%, and the corresponding signal-to-noise ratio (SNR) are 23.69, 21.18, 18.75 and 15.99 dB respectively.
Fig. 9. QPSK constellation with different modulation rates at the power of -75 dBm with the center frequency of 2.6134 GHz.
The sensitivity of the system for the RF signal is researched above, then, that for the modulated signal is also need to investigate. The bit error rate (BER) can directly reflect the performance of the detection system, which can be calculated by the measured SNR [18]. So, the 4.8 ksps QPSK signal with different power from -72 dBm to -92 dBm is injected to the system. The EVM and SNR are measured and plotted in Fig. 10(a). With the increasing of input power of the modulation signal, the SNR increases almost linearly from 8.02 dB to 28.07 dB, the EVM improves from 39.74% to 3.95% correspondingly. The calculated BER, which is plotted in Fig. 10(b), rises with the input power decreasing.
Fig. 10. The SNR, EVM (a) and calculated BER (b) for the modulation signal with the data rate of 4.8 ksps at different power.
For the synchronous voice signal reception in the satellite phones, the BER is required to be below 10−3. It can be seen the BER of the detected signal is 1.3*10−4 when the input power is -89 dBm. In addition, the maximum allowed BER for the asynchronous data signal, such as fax and instruction detection is 10−6. Thus, the detection sensitivity is -86 dBm with the BER of 3.25*10−7. The measured constellations at the input power of -89 and -86 dBm are shown as the insets in Fig. 10(b), from which it can be seen that the data points can be distinguished well. The detection sensitivity is better than that in Ref. [17].
4. Discussion and conclusion
To conclude, in this paper, we have proposed and experimentally demonstrated an ultra simple and low-cost method to detect low-power RF signal. To our knowledge, this is the simplest photonics-assisted method according to public reports. The system is based on optoelectronic feedback direct modulate the DFB laser which works at the multimode oscillation state. The gain provided by the EDFA compensates the loss of the feedback loop and just lower than the threshold gain for single mode oscillation. The RF signal obtains the largest gain which aligns with the oscillation mode at the relaxation oscillation peak. The system can detect RF signal from 1 to 4.5 GHz by tuning the bias current of the laser to adjusting the frequency of the relaxation oscillation peak. It provides the maximum gain of 15 dB for RF signal and the sensitivity is as high as -97 dBm. The detected signal of 2.6134 GHz is amplified until the input power reaches -9 dBm which means the dynamic range is 88 dB. The ability of resisting to large signal is investigated by injecting a large signal at the next oscillation mode. The detected small signal will still acquire gain until the large signal is over -10 dBm. As to the stability of the system, although the system works at multi-mode oscillation, the gain in the loop is below the threshold gain for single mode oscillation. So, no oscillation mode can exceed others, the mode competition is very small. No mode hopping will happen and the system is stable. The performance of system for detecting modulated RF signal is also investigated. The proposed system has broad applications for low-power RF signal detection.
Funding
National Natural Science Foundation of China (61974165, 62071487); National Outstanding Youth Science Fund Project of National Natural Science Foundation of China (61901480).
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. P. Ghelfi, F. Laghezza, F. Scotti, G. Serafino, A. Capria, S. Pinna, D. Onori, C. Porzi, M. Scaffardi, A. alacarne, V. Vercesi, E. Lazzeri, F. Berizzi, and A. Bogoni, “A fully photonics-based coherent radar system,” Nature 507(7492), 341–345 (2014). [CrossRef]
2. J. Raza, A.-J. Boonstra, and A. J. van der Veen, “Spatial filtering of RF interference in radio astronomy,” IEEE Signal Process. Lett. 9(2), 64–67 (2002). [CrossRef]
3. C. Wang, F. Haider, X. Gao, X. You, Y. Yang, D. Yuan, H. M. Aggoune, H. Haas, S. Fletcher, and E. Hepsaydir, “Cellular architecture and key technologies for 5G wireless communication networks,” IEEE Commun. Mag. 52(2), 122–130 (2014). [CrossRef]
4. G. W. Anderson, D. C. Webb, A. E. Spezio, and J. N. Lee, “Advanced channelization for RF, microwave, and millimeter wave applications,” Proc. IEEE 79(3), 355–388 (1991). [CrossRef]
5. T. S. Rappaport, S. Sun S, R. Mayzus, H. Zhao, Y. Azar, K. Wang, G. N. Wong, J. K. Schulz, M. Samimi, and F. Gutierrez, “Millimeter wave mobile communications for 5G cellular: It will work!” IEEE Access 1, 335–349 (2013). [CrossRef]
6. D. Cabric, S.M. Mishra, and R.W. Brodersen, “Implementation issues in spectrum sensing for cognitive radios,” in Conference on Signals, Systems and Computers, IEEE Pacific Grove, Ca., USA, 2004, pp. 772–776.
7. X. Zou, B. Lu, W. Pan, L. Yan, A. Stohr, and J. Yao, “Photonics for microwave measurements,” Laser Photonics Rev. 10(5), 711–734 (2016). [CrossRef]
8. J. Yao, “Microwave photonics,” J. Lightwave Technol. 27(3), 314–335 (2009). [CrossRef]
9. L. V. T. Nguyen, “Microwave photonic technique for frequency measurement of simultaneous signals,” IEEE Photonics Technol. Lett. 21(10), 642–644 (2009). [CrossRef]
10. M. S. Rasras, D. M. Gill, S. S. Patel, K.-Y. Tu, Y.-K. Chen, A. E. White, A. T. S. Pomerene, D. N. Carothers, M. J. Grove, D. K. Sparacin, J. Michel, M. A. Beals, and L. C. Kimerling, “Demonstration of a fourth-order pole-zero optical filter integrated using CMOS Processes,” J. Lightwave Technol. 25(1), 87–92 (2007). [CrossRef]
11. X. Xie, Y. Dai, K. Xu, J. Niu, R. Wang, L. Yan, Y. Ji, and J. Lin, “Digital joint compensation of IMD3 and XMD in broadband channelized RF photonic link,” Opt. Express 20(23), 25636–25643 (2012). [CrossRef]
12. L. Maleki, “The optoelectronic oscillator,” Nat. Photonics 5(12), 728–730 (2011). [CrossRef]
13. X. Xie, C. Zhang, T. Sun, P. Guo, X. Zhu, L. Zhu, W. Hu, and Z. Chen, “Wideband tunable optoelectronic oscillator based on a phase modulator and a tunable optical filter,” Opt. Lett. 38(5), 655–657 (2013). [CrossRef]
14. X. Zhang, J. Zheng, T. Pu, Y. Zhang, Y. Shi, J. Li, Y. Li, H. Zhu, and X. Chen, “Simple frequency-tunable optoelectronic oscillator using integrated multi-section distributed feedback semiconductor laser,” Opt. Express 27(5), 7036–7046 (2019). [CrossRef]
15. G. Wang, T. Hao, W. Li, N. Zhu, and M. Li, “Detection of wideband low-power RF signals using a stimulated Brillouin scattering-based optoelectronic oscillator,” Opt. Commun. 439(15), 83–136 (2019). [CrossRef]
16. P. S. Devgan, M. W. Pruessner, V. J. Urick, and K. J. Williams, “Detecting low-power RF signals using a multimode optoelectronic oscillator and integrated optical filter,” IEEE Photonics Technol. Lett. 22(3), 152–154 (2010). [CrossRef]
17. Y. Shao, X. Han, M. Li, and M. Zhao, “RF signal detection by a tunable optoelectronic oscillator based on a PS-FBG,” Opt. Lett. 43(6), 1199–1202 (2018). [CrossRef]
18. R. A. Shafik, M. S. Rahman, A. R. Islam, and N. S. Ashraf, “On the error vector magnitude as a performance metric and comparative analysis,” in IEEE ICET, Peshawar, Pakistan, 2006, pp. 27–31
