A photonic radar warning receiver was conceived and practically demonstrated. The system was very flexible in terms of frequency measurement range and resolution. This was achieved using a cascaded grating which provided different time delays for different wavelengths. The system was thus able to become reconfigured to operate at both broadband, low resolution and narrow band, high resolution modes.
© 2013 Optical Society of America
Broadband nature, immunity to electromagnetic interference, and system reconfigurability, have made photonic technology a very attractive candidate for wide band signal processing applications [1–6]. One application is threat frequency identification in electronic warfare environment. This is usually done by radar warning receivers. They must be able to provide a very accurate frequency measurement over a wide band frequency range. Receivers implemented using traditional microwave techniques usually operate at only one mode; wide band mode, or high resolution mode. Two types of receiver are thus required. One has to cover a broad frequency range and the other must exhibits high resolution characteristics. The wide band receiver is called instantaneous frequency measurement (IFM) receiver and the high resolution one, is called scanning receiver.
Rigorous spectrum analysis is usually achieved via a scanning receiver. Here an extremely narrow oscillator tone is tuned across the measurement band and is then mixed with the incoming signal. This scanning process is precise and has exceptional noise rejection characteristics, but is essentially serial in nature. It is possible that a smart threat could operate only at frequencies where the scanning receiver is not tuned. In this case the threat would be missed. To ensure a 100% probability of intercept, the full spectrum must be monitored continuously and potential threat signals identified instantaneously. Since identifying the exact frequency spectrum of a threat could require a significant scan time, it is important to narrow the frequency range to reduce the scan time. IFM receivers are used for this purpose. They provide an early indication of threat classification and also suggest frequency ranges in which to focus more sophisticated spectrum analysis resources to find the exact RADAR threat spectrum. Having identified the exact spectrum of the threat, it is possible to pattern match this spectrum against a database of known threats to obtain a recommended countermeasure and to take further appropriate action to mislead the threat.
Attempts have been done to implement both scanning and IFM receivers using photonic technology. In an attempt, a photonic scanning receiver was implemented based on an electrically tuned fiber Bragg grating (FBG) . The system exhibited a 50 MHz resolution over a 2–18 GHz bandwidth. Another scanning receiver was demonstrated in . The system employed a Fabry-Perot interferometer to analyze the microwave sidebands on an optical carrier. The system was demonstrated over a frequency operation range of 40 GHz with 90 MHz resolution. In another work, multichannel chirped FBG was employed to enable amplitude comparison of power fading functions generated by double sideband modulated optical carriers propagating through a dispersive medium . In another attempt, a polarization domain interferometer was employed to measure amplitude and frequency of an RF signal . The system was based on constructive and destructive modes. A frequency measurement error less than 200 MHz was achieved over a 1–18 GHz frequency range. A dual parallel Mach-Zehnder modulator was employed to achieve a reconfigurable IFM system in . The system exhibited a frequency measurement range of 1–12 GHz with 100 MHz resolution. Another microwave frequency measurement system was demonstrated in . The system operated based on a null biased Mach-Zehnder interferometer. A fixed relationship between the optical power at the output of the system and the RF frequency was established. This enabled frequency measurement of the input microwave signal. System operation was demonstrated over a 6–18 GHz bandwidth with a ±300 MHz resolution. A chipped Mach-Zehnder modulator was employed to implement frequency to power mapping in . A high resolution of 50 MHz was achieved over a limited bandwidth. In another work, stimulated Brillouin scattering was employed to achieve adjustable measurement range and resolution . The system was demonstrated at both 12 GHz bandwidth with ±250 MHz resolution, and 2 GHz bandwidth with 50 MHz resolution. Frequency to power mapping was also employed in  where the same RF modulated optical carriers traversed through different types of fiber (polarization maintaining, and dispersion compensation fiber) to achieve frequency dependent power difference. A 70 MHz resolution was achieved over a 10.5 GHz bandwidth.
In this paper we present a photonic frequency measurement system capable of being reconfigured to operate at both wide band, low resolution and narrow band, high resolution modes. This means a radar warning receiver consisting of both traditional IFM systems, and scanning receivers, could be potentially replaced by such system.
Figure 1(a) shows block diagram of the proposed photonic frequency measurement system. An RF tone (cos Ωt) was divided into two equal portions. The first portion, was mixed by an optical carrier (cosω◦t) within mixer1. The resulted signal was then delayed by amount of τ and mixed by the second portion of the RF tone within mixer2. The resulted signal was input to an envelope detector, and low-pass filtered.
The output of the envelope detector consists of a tone at frequency Ω and a low frequency component ( ). Therefore, using a low-pass filter (LPF), the low frequency component can be extracted and measured using a low frequency voltmeter. The output voltage VDC exhibits an oscillating behavior. Thus, provided τ is chosen such that the amplitude traces half a single period over the desirable frequency range, it is possible to measure the RF frequency by measuring the output voltage. It can also be seen that for a smaller amount of τ the frequency measurement range will be increased. In this case; however, since the slope of the voltage curve is decreased, less accuracy will be achieved. In the electronic warfare environment, it would be desirable that a frequency measurement system operates at a board band mode to be able to detect any threat signal over a wide band frequency range and to obtain a coarse measurement, while being able to become reconfigured to provide a more accurate measurement if required. To achieve such reconfigurability, the amount of τ must be adjustable.
Having developed a concept to measure the frequency of RF signals using photonic technology, it is now possible to design a reconfigurable photonic frequency measurement system.
3. System setup
In this section, we conceive an experimental set up based on the concept of Section 2. Figure 1(b) presents the experimental setup of the proposed reconfigurable photonic frequency measurement system. A high frequency signal generator (SG) produced an RF tone with angular frequency Ω = 2πf. This was divided into two equal portions using a 3 dB power divider. These porions fed two arms of the system. These arms were labeled optical path and RF path. In the optical path, the first portion modulated an optical carrier using a Mach-Zehnder modulator (MZM1) biased at quadrature point (Vq). The output signal, was input to port1 of an optical circulator, and sent out from port2, to traverse through a cascaded grating. The cascaded grating provided different delays for different wavelengths (λ1 − λ3). To ensure correct polarization at the MZM2 input, both circulator and cascaded grating were chosen to be polarization maintaining. The delayed signal was output from port3 and was modulated again by the second portion of the RF tone. This portion was delayed using a piece of coaxial cable in the RF path. The modulation was done within MZM2 biased at Vq. The resulted signal was amplified using an Erbium doped fiber amplifier (EDFA). The amplified signal was then detected by a photo-detector (PD), low-pass filtered and finally measured by a digital voltmeter. The output voltage of the system can be described as :Fig. 1(b). The cascaded grating provided different delays for different wavelengths, thus by changing wavelength of the laser source, it would be possible to achieve different delays in the optical path and consequently change the amount of τ. From Eq. (1) it is evident that the total delay of τ + τ′i could be controlled by selecting different wavelengths on the laser source. This should make it possible to reconfigure this system to operate at different frequency measurement ranges.
Having conceived a reconfigurable photonic frequency measurement system, it is now time to demonstrate the system performance.
4. System demonstration
The system was configured as depicted in Fig. 1(b). The factor G was calculated to be 1.1. Vπ was 5 V for both MZMs. PRF, and P° were set to 0, and 10 dBm, respectively. The input impedance of the low-pass filter was 50 Ω. The photo-detector was connected to the low-pass filter input thus ZPD=50 Ω. Input impedance of the MZM was also Zin=50 Ω. The delay τ was chosen to be 9 ps. Note that to cover a 40 GHz bandwidth in a half period, a 12.5 ps delay is required. However, since the slope of the voltage curve is nearly flat at the end of the band (35–40 GHz), selecting a smaller delay would increase the slope of the voltage curve at high frequencies. This would result in achieving a better accuracy. Values of τ′1, and τ′2 were 49, 169 ps, respectively. Note that while the amount of τ can be arbitrarily determined by changing the coaxial cable length or the length of fiber patch cords between MZM1, and MZM2, the amount of τ′1, and τ′2 were imposed by the cascaded grating characteristics. The laser wavelength was first set to λ1 = 1548.1 nm. Then the signal generator swept a frequency band of .04–40 GHz with 50 MHz steps. The output voltage was recorded at each step. This procedure was then repeated for λ2 = 1548.6, and λ3 = 1553.3 nm. Table 1 shows the overall differential delay caused by each wavelength.
The measurement results along with prediction made by Eq. (1) are shown in Fig. 2(a). Excellent agreement between measurement and prediction is evident. For frequencies above 32 GHz some fading can be seen. This could be attributed to high frequency loss of the coaxial cable. Due to oscillatory behavior of the system, for each delay, there is a number of distinct bands within which the frequency measurement is unambiguous. Each band starts from a peak and ends at the next trough. For delays 89, and 289 ps, there are 7, and 23 bands, respectively. To better understand the concept of the bands, two adjacent bands for 289 ps delay curve are placed within colored bars in Fig. 2(a). For a delay of 9 ps, there is only one band 55.5 GHz wide. Therefore, the whole band can not be seen in the graph. These measurements were then used to calculate the input frequency using Eq. (1). Figure 2(b) shows measured frequency against the input frequency. To better illustrate the system behavior, four different vertical axes were chosen. For larger amount of delay (narrower bandwidth) a higher accuracy has been achieved as predicted. Table 2 shows frequency measurement error for different delays. Both maximum error and mean error were provided. At wide band operating mode (9 ps delay), a maximum error of 1.95 GHz, and a mean error of 450 MHz was recorded over a .04–40 GHz frequency range. For 89, and 289 ps delays, maximum errors of 0.51, and 0.1 GHz, and mean errors of 25, and 3 MHz were observed, respectively. Note that the frequency measurement with high resolution (0.1 GHz maximum error) was achieved within any of the unambiguous measurement bands. However, this band was identified in one step before in a lower resolution mode (89 ps), thus the measured frequency with a 0.1 GHz error is unambiguous over the whole band. Here the importance of the system reconfigurability can be inferred. A bandwidth of 0.04–40 GHz can be simultaneously monitored with a delay of 9 ps. In case of any threat identification, a rough measurement of the threat frequency will be conducted. Then a better accuracy can be achieved by switching to 89, and 289 ps delays, respectively. After obtaining an accurate measurement, the system can be reconfigured to wide band operating mode, to monitor the whole band again.
Note that this system was assembled using discrete components. There were tens of centimeters of optical fiber connecting MZM1, and MZM2 to the optical circulator, and the optical circulator to the cascaded grating. Remember τ is the differential delay caused by the physical length difference between the optical path and the RF path. To achieve a wide unambiguous frequency range, the amount of τ has to be adjusted to a few picoseconds. This means the physical length difference has to be reduced to a few millimeters. To achieve this length difference, extra length in the RF path would be required. This was achieved by inserting a length of coaxial cable in the RF path as shown in Fig. 1(b). However, the coaxial cable itself puts limitation on the system bandwidth. Wider bandwidth could thus be achieved via integration. This would remove the need for the coaxial cable.
It would be imaginable that two modules operate in parallel. One always operates in wide band mode, while the other can be reconfigured to operate at both 89, and 289 ps delay modes. This way no threat signal will be missed when switching to high resolution modes. Further improvement in measurement accuracy could be achieved using larger amount of delay by introducing more channels to the cascaded grating. To implement such system, there will be no need to any extra component.
The dynamic range of the system is mainly determined by the system sensitivity. The upper limit of the dynamic range in a radar warning receiver is usually determined by the limiters placed before the receiver. For the stand alone system; however, the upper limit of the system dynamic range is determined by the maximum RF power allowed at each MZM input. The lower limit of the dynamic range is determined by the sensitivity of the system. The system sensitivity could be measured and improved if required using lock-in amplification method as discussed in . This would improve the dynamic range accordingly.
A reconfigurable photonic frequency measurement system was practically demonstrated. The system was able to become reconfigured to operate at both wide band, and high resolution modes. This was achieved using a cascaded grating which exhibited different delays for different wavelengthes. The system is very flexible in terms of being extended to achieve wider frequency measurement range or higher frequency measurement resolution. The system operation was demonstrated over a frequency range of 0.04–40 GHz with 100 MHz error.
References and links
1. J. Capmany and D. Novak, “Microwave photonics combines two worlds,” Nat. Photonics (London) 1(6), 319–330 (2007) [CrossRef] .
2. A. J. Seeds and K. J. Williams, “Microwave photonics,” J. Lightwave Technol. 24(12), 4628–4641 (2006) [CrossRef] .
3. J. Yao, “Microwave photonics,” J. Lightwave Technol. 27(3), 314–335 (2009) [CrossRef] .
4. R. A. Minasian, “Photonic signal processing of microwave signals,” IEEE Trans. Microwave Theory Tech. 54(2), 832–846 (2006) [CrossRef] .
6. H. Emami, N. Sarkhosh, E. Lopez, and A. Mitchell, “Photonic feed for sinuous antenna,” J. Lightwave Technol. 30(16), 2725–2743 (2012) [CrossRef] .
7. P. Rugeland, Z. Yu, C. Sterner, O. Tarasenko, G. Tengstrand, and W. Margulis, “Photonic scanning receiver using an electrically tuned fiber Bragg grating,” Opt. Lett. 34(24), 3794–3796 (2009) [CrossRef] [PubMed] .
8. S. T. Winnal and A. A. Lindsay, “A Fabry-Perot scanning receiver for microwave signal processing,” IEEE Trans. Microwave Theory Tech. 47(7), 1385–1390 (1999) [CrossRef] .
9. L. V. T. Nguyen and D. B. Hunter, “A photonic technique for microwave frequency measurement,” IEEE Photon. Technol. Lett. 18(10), 1188–1190 (2006) [CrossRef]
10. M. V. Drummond, P. Monteiro, and R. N. Nogueira, “Photonic RF instantaneous frequency measurement system by means of a polarization domain interferometer,” Opt. Express 17(7), 5433–5438 (2009) [CrossRef] [PubMed] .
11. W. Li, N. H. Zhu, and L. X. Wang, “Reconfigurable instantaneous frequency measurement system based on dual-parallel Mach-Zehnder modulator,” IEEE Photon. Journal 4(2), 427–436 (2012) [CrossRef]
12. J. Dai, K. Xu, X. Sun, J. Niu, Q. Lv, J. Wu, X. Hong, W. Li, and J. Lin, “A simple photonic-assisted microwave frequency measurement system based on MZI with tunable measurement range and high resolution,” IEEE Photon. Technol. Lett. 22(15), 1162–1164 (2010) [CrossRef] .
13. J. Li, S. Fu, K. Xu, J. Q. Zhou, P. Shum, J. Wu, and J. Lin, “Photonic-assisted microwave frequency measurement with higher resolution and tunable range,” Opt. Lett. 34(6), 743–745 (2009) [CrossRef] [PubMed] .
15. J. Zhou, S. Fu, P. P. Shum, S. Adita, L. Xia, J. Li, X. Sun, and K. Xu, “Photonic measurement of microwave frequency based on phase modulation,” Opt. Express 17(9), 7217–7221 (2009) [CrossRef] [PubMed] .
16. N. Sarkhosh, H. Emami, L. A. Bui, and A. Mitchell, “Reduced cost photonic instantaneous frequency measurement system,” IEEE Photon. Technol. Lett. 20(18), 1521–1523 (2008) [CrossRef] .
17. N. Sarkhosh, H. Emami, L. A. Bui, and A. Mitchell, “Microwave photonic instantaneous frequency measurment with improved sensitivity,” in Proceedings of 2009 IEEE MTT-S International Microwave Symposium Digest (MTT) (Institute of Electrical and Electronics Engineers, Boston, 2009), 165–168 [CrossRef] .