IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 19, NO. 4, JULY/AUGUST 2013
1500807
Linewidth Sharpening via Polarization-Rotated Feedback in Optically Injected Semiconductor Laser Oscillators Thomas B. Simpson, Member, IEEE, Jia-Ming Liu, Fellow, IEEE, Mohammad AlMulla, Nicholas G. Usechak, Senior Member, IEEE, and Vassilios Kovanis
Abstract—Combining optical injection and polarization-rotated optical feedback in a semiconductor laser can induce selfreferenced periodic output that is widely tunable by simply varying the dc-bias points of the system’s master and slave lasers. We observed a feedback-induced reduction of the fundamental periodone oscillation linewidth by more than two orders of magnitude relative to the injection-only case. Performance was found to be negatively affected by the interference between the external injection signal and the residual feedback in the same polarization. The nonlinear dynamics of the optically injected semiconductor laser can be used to minimize sensitivity to fluctuations in the operating points. However, the use of the nonlinear dynamics at high oscillation frequencies is limited by the decreasing strength of the interaction between the circulating intracavity optical field and the carrier density. Index Terms—Injection locking, RF oscillators, tunable oscillators.
I. INTRODUCTION HE nonlinear dynamics induced via the optical injection of a semiconductor laser offer a new path to improve the performance of low-noise photonic oscillators. Over a wide range of operating conditions, the injected optical signal perturbs the output of a slave laser so that it exhibits periodic dynamics instead of a steady-state output. Simply by controlling the operating points of the master and slave lasers, through the bias currents and operating temperatures that dictate the free-running offset frequencies and the relative injection power, the oscillation/pulsation frequency can be widely tuned over the microwave and millimeter-wave bands [1]. When the optical
T
Manuscript received October 30, 2012; revised December 21, 2012; accepted December 24, 2012. Date of publication January 18, 2013; date of current version May 13, 2013. The work of T. B. Simpson and J.-M. Liu was supported in part by the Air Force Research Laboratory through a contract with Optimetrics, Inc. The work of N. G. Usechak and V. Kovanis was supported by the Air Force Office of Scientific Research under Grant 12RY09COR. The views and opinions expressed in this paper (88ABW-2012-5555) are those of the authors and do not reflect the official policy or position of the United States Air Force, Department of Defense, or the U.S. Government. T. B. Simpson is with the L-3 Applied Technologies, Inc., San Diego, CA 92121 USA (e-mail:
[email protected]). J.-M. Liu and M. AlMulla are with the Department of Electrical Engineering, University of California, Los Angeles, CA 90095 USA (e-mail:
[email protected];
[email protected]). N. G. Usechak and V. Kovanis are with the Air Force Research Laboratory, Wright–Patterson AFB, OH 45433 USA (e-mail:
[email protected];
[email protected]). Color versions of one or more of the figures in this paper are available online at http://ieeexplore.ieee.org. Digital Object Identifier 10.1109/JSTQE.2012.2237016
Fig. 1. Calculated P1 oscillation frequency of a semiconductor laser under optical injection as a function of the amplitude and detuning of the injected signal. The red lines show the ranges for the calculations summarized in Figs. 13 and 14.
output is detected by a conventional high-speed photodiode, the generated photocurrent reproduces the high-frequency oscillation/pulsation. Fig. 1 shows a calculation of the oscillation frequency as a function of the injection strength and the frequency detuning of the master laser signal [2]. The range of period-one (P1) dynamics, oscillation at the fundamental resonance frequency of the dynamic system, is separated from stable locking by a Hopf bifurcation, and it surrounds regions of more complex periodic and aperiodic/chaotic dynamics. The figure shows curves of constant P1 frequency. Note that near the Hopf bifurcation there are ranges where these curves are parallel to the detuning axis, while for large positive offset frequencies they are essentially parallel to the injection strength axis. These conditions represent very different sensitivities to changes in the injection parameters. Previous work demonstrated that optical injection can be used in combination with optoelectronic feedback for a novel photonic microwave oscillator [3]. When the optical injection induces the P1 oscillation, the feedback does not have to provide additional loop gain. It acts as the self-referencing input to narrow and stabilize the oscillation characteristics. Here, we combine optical injection with polarization-rotated optical feedback
1077-260X/$31.00 © 2013 IEEE
1500807
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 19, NO. 4, JULY/AUGUST 2013
Master Laser
-10
Slave Laser
Circulator
-20 x
x
Microwave Spectrum Analyzer
Optional LO Laser
PD
PC
Bias Tee
x Function Generator
Fig. 2. Schematic of the experimental apparatus. VOA, variable optical attenuator; PC, polarization controller; PD, photodiode. As depicted in the figure, the master and LO lasers were packaged with optical isolators while the slave laser had no isolator. The figure also identifies the propagation direction of the light within the setup using dashed arrows. The double-sided arrows and dots surrounded by circles represent the horizontal and vertical polarizations, respectively, and have been included to succinctly communicate the purpose of each component in the overall system.
to demonstrate an all-photonic variation. The optical injection induces a P1 periodic oscillation in the output of the slave laser, and simultaneously, the nonlinear gain characteristic of this laser also acts as a narrow-band microwave filter and feedback loop gain element. In a conventional semiconductor distributed feedback (DFB) laser, the orthogonally polarized modes have very different profiles with very different resonance frequencies and gain and loss characteristics. Rotating the polarization of part of the optical output and feeding it back into the slave laser along with the external injection effectively produce an optoelectronic feedback current which accompanies the optical injection, rather than a second optical injection into the oscillating mode. The feedback is now nonresonant with the optical cavity modes of the laser and primarily modifies the carrier density in the gain medium. This technique bypasses the losses, complexity, and amplifier 1/f-noise of the microwave circuit elements in a conventional optoelectronic oscillator (OEO) [4] while simultaneously being less sensitive to feedback path-length fluctuations relative to conventional optical feedback techniques. II. EXPERIMENTAL APPARATUS AND MEASUREMENTS Fig. 2 is a schematic of the experimental layout that uses an existing apparatus [5]. All optical components are connected by single-mode, nonpolarization-preserving fiber. The lasers are single-mode DFB lasers oscillating at approximately 1557 nm. The free-running characteristics of the slave laser, as well as its nonlinear dynamics under optical injection, have been described previously [5]. The master laser is packaged with an optical isolator, and an optical circulator is used to further isolate the master laser from unwanted feedback. Both lasers are temperature stabilized, and modulation currents can be added to the dc-bias currents of either the master or the slave laser. The laser outputs are polarized, and fiber polarization rotators are used to adjust the polarization of the master laser to match that of the slave laser and to rotate the polarization of the feedback signal to be orthogonal to the oscillating intracavity field of the slave laser. The latter is accomplished by monitoring the output
Signal (dBm)
PC
Bias Tee
VOA
-30 -40 -50 -60 -70 5.9
6
6.1
6.2 6.3 6.4 Frequency (GHz)
6.5
6.6
Fig. 3. Power spectrum of the monitor photodiode signal under optical injection only (dotted black traces) and with polarization-rotated feedback (solid red traces).
power spectrum of the slave laser without optical injection, and then adjusting the polarization of the feedback so that there is no evidence of external cavity modes in the microwave spectrum of a fast photodetector monitoring the output. An optional third DFB laser (labeled “LO laser” in Fig. 2) is used as a tunable local oscillator. Sweeping the optical frequency of this laser by varying its operating temperature and mixing its output field with the slave laser output generates low-resolution (∼100 MHz) optical spectra with the microwave spectrum analyzer [5]. Fig. 31 shows the typical spectrum of the photodiode signal around the P1 pulsation frequency with and without polarization-rotated optical feedback. Without feedback, the spectrum consists of a single, fairly broad feature with a fullwidth half-maximum (FWHM) of approximately 2 MHz. The feedback causes multiple peaks to appear, with a frequency separation determined by the round-trip feedback delay time. The delay time can be adjusted simply by changing the length of the fiber in the delay path. To date, the best spectrum had a central peak about 15–20 dB stronger than the side peaks, with details shown in Fig. 4. The ∼15-kHz (FWHM) peak was more than two orders of magnitude narrower than that obtained by external injection only. This linewidth was dominated by jitter on time scales >10 ms. On longer time scales, the peak with the largest amplitude slowly hopped/shifted between 2–3 adjacent peaks during observation periods of up to several minutes. This instability is simply a reflection of the fact that the experimental apparatus was not systematically isolated from environmental changes in the room. In particular, effects due to optical path length changes in the feedback loop were observed, as discussed below. Since the long term stability of the apparatus was not optimized, it was not rigorously characterized.
1 Note Figs. 3–12 use the following convention: pure injection locking (IL) data are represented by dotted black traces, IL with polarization-rotated feedback (PRF) is shown in solid red traces, IL with optoelectronic feedback (OEF) is shown in dashed green traces, IL with both PRF and OEF is represented in dash– dotted blue traces, and comparisons between different states/configurations of the same type of feedback are represented with the appropriate color and dotted gray traces.
SIMPSON et al.: LINEWIDTH SHARPENING VIA POLARIZATION-ROTATED FEEDBACK
1500807
0
-30 Normalized Signal (dB)
-5
Signal (dB)
-40
-50
-60
-70
-10 -15 -20 -25 -30 -35 -40
-0.4
-0.2
0 0.2 Offset Frequency (MHz)
-45 -0.4
0.4
Fig. 4. High-resolution details of the strongest peak under simultaneous optical injection and polarization-rotated optical feedback. A ∼15-kHz FWHM was found using a 10-kHz resolution bandwidth (RBW).
-0.2
0 0.2 Offset Frequency (MHz)
0.4
Fig. 6. Higher resolution (3-kHz RBW) details of the strongest peak under simultaneous optical injection and polarization-rotated feedback (IL+PRF) (solid red) and optoelectronic feedback (IL+EOF) (dashed green) reveals an FWHM as narrow as 7 kHz.
-20
-30 -30 Signal (dBm)
Signal (dBm)
-20
-40
-50
-40
-50
-60 -60
-20
-10
0
10
20
Offset Frequency (MHz) Fig. 5. Power spectrum of the monitor photodiode signal for the laser under simultaneous optical injection and optoelectronic feedback using a 30-kHz RBW.
We observed that the amplitude of the total (external injection plus feedback) injected optical signal fluctuated slowly in time and determined that this was due to the interference between the injected master laser signal and the residual feedback signal in the same polarization. The slow phase fluctuations between master injection signal and feedback signal in the same polarization and at the same wavelength caused time-varying fluctuations in the output of the 2 × 2 fiber coupler used to combine these signals. Therefore, in our apparatus, there was residual feedback that remained in the original polarization at the point where the two optical signals were combined. The residual signal was enough to cause feedback amplitude fluctuations of 5–10%, though this depended on the specific injection amplitude of the master laser and that of the feedback signal. To determine the relative importance of the residual feedback coherently interfering with the injected signal from the master laser, we substituted optoelectronic feedback for the polarization-rotated feedback. A second fast photodiode (not shown in Fig. 2) was used to convert the optical output from the slave laser into a modulation current that was fed back to the slave laser along with the dc bias through the bias tee depicted
-80
-60
-40
-20 0 20 40 Offset Frequency (MHz)
60
80
Fig. 7. Power spectrum of the photodetector signal for a laser operated at the point of low sensitivity to fluctuations and subject to OEF (dashed green traces). This spectrum, centered at 5.94 GHz and taken with a 30-kHz RBW, is similar to that obtained in Fig. 5. Here, it is compared to the spectrum obtained under IL only (dotted black traces).
in Fig. 2. Fig. 5 shows the resulting spectrum of the laser under simultaneous optical injection and optoelectronic feedback. In Fig. 6, for a direct comparison, the central peak of the optically injected laser with optoelectronic feedback and the same optically injected laser with polarization-rotated optical feedback are shown. The former peak in Fig. 6 was distinctly narrower, though jitter continued to dominate the spectral width, and the hopping between peaks was similar. The jitter could be due to relative fluctuations in the operating points of the two lasers. By making use of the varying nonlinear dynamics, we were able to isolate the key source of jitter as amplitude fluctuations of the injected signal from the master laser into the oscillating mode of the slave laser. Recall that in Fig. 1, there are operating points where the pulsation frequency is insensitive to changes in the detuning of the master laser. By observing changes in the amplitude of a modulation current added to the slave laser, we found that these points were relatively insensitive to slave laser current fluctuations [6]. We next investigated the use of feedback at a low-sensitivity
1500807
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 19, NO. 4, JULY/AUGUST 2013
-15
-30
-20
-35
-25
-40 Signal (dBm)
-35 -40 -45
-45 -50 -55
-50
-60
-55
-65
-60
-70 0 5 Offset Frequency (MHz)
-75 -60
10
Fig. 8. The power spectrum of the photodetector signal for an IL + OEF laser operated near the point where the P1 frequency is locally minimized (dashed green traces). This result is contrasted with an IL + OEF laser where the system has been detuned from the low-sensitivity point (dotted gray traces). Note that the point of reduced sensitivity to current fluctuations remains, even with the feedback.
operating point to see if the special features described previously remained when the configuration was modified by the addition of optoelectronic or polarization-rotated feedback. Optoelectronic feedback provided the best performance for frequencies less than the 8-GHz bandwidth of the photodetector; therefore, we biased the slave laser at approximately 12 mA to produce a P1 oscillation/pulsation frequency at about 6 GHz. Fig. 7 shows the change to the overall spectral feature with the addition of the feedback. The spectrum looks similar to that seen at an operating point away from the low-sensitivity operating point. However, a more detailed look at the individual peaks of the spectrum with optoelectronic feedback showed that the reduced sensitivity to current fluctuations in the slave laser remained. Fig. 8 shows the reduction of the noise pick-up sidebands when the laser was operated at the reduced-sensitivity operating point verifying that the sideband reduction occurs for much lower frequencies than could be measured without feedback. Similarly, with polarization-rotated optical feedback, the spectra seen at and away from the low-sensitivity operating point showed the same corresponding behavior, as can be seen in Fig. 9. Note that the noise pick-up sidebands about each peak are reduced and the sideband features due to the round-trip delay are also weaker. However, the overall jitter due to the fluctuations of the injected signal at the slave laser polarization remained. Finally, we combined polarization-rotated optical feedback with optoelectronic feedback. The polarization-rotated feedback was on the shorter feedback path and was the dominant feedback strength because we operated the laser at a higher P1 frequency point. Fig. 10 shows the power spectrum of the photodetector signal. There are features separated by the 20MHz frequency spacing determined by the polarization-rotated feedback loop (∼10 m of fiber). These features are themselves composed of features separated by the 1-MHz frequency spacing determined by the longer delay of the optoelectronic feedback loop (∼200 m of fiber). The dominant central peak has a width of less than 3 kHz, as shown in Fig. 11, but the overall
-40
-20 0 20 Offset Frequency (MHz)
40
60
Fig. 9. Details of the power spectrum of the photodetector signal for an IL + PRF laser operated near the point where the P1 frequency is locally minimized (solid red traces). This result is contrasted with an IL + PRF laser where the system has been detuned from the low-sensitivity point (dotted gray traces). The point of reduced sensitivity to current fluctuations remains, even with the feedback.
Signal (dBm)
-5
-40 -60 50
0
50
-30 Signal (dBm)
-10
-40 -50 -60 -70 -8
-6
-4
-2 0 2 4 Offset Frequency (MHz)
6
8
Fig. 10. Power spectrum of the photodetector signal for an IL+PRF+OEF laser. (Top) Lower resolution showing the ∼20-MHz-spaced features due to the short polarization-rotated feedback loop. (Bottom) ∼1-MHz-spaced features due to the longer optoelectronic feedback loop. -30 -35 FWHM = 3 kHz -40 Signal (dBm)
Signal (dBm)
-30
-45 -50 -55 -60 -65 -70 -75 -0.2
-0.1
0 0.1 Offset Frequency (MHz)
0.2
Fig. 11. Details of the power spectrum of the photodetected signal of the laser operated under simultaneous optoelectronic and polarization-rotated feedback (IL+PRF+OEF) (dash–dotted blue traces). The dominant peak is well fit by a 3-kHz Lorentzian function as shown (solid black traces).
2 0.1
1.5
0.05 0.05 0.1 0.15 0.2 Normalized Injection Strength (AU)
-75 Period-One Frequency (GHz)
30
-80
-85 19.7
19.8
19.9 Frequency (GHz)
20
20.1
25
20
15
10
Fig. 12. Comparison of the narrowing of the jitter-dominated pulsation at higher frequencies by optoelectronic, IL+OEF (dashed green traces), and polarization-rotated optical feedback, IL+PRF (solid red traces), with respect to the IL (dotted black traces) case.
stability of the configuration is still degraded by the amplitude jitter due to interference between the externally injected signal and the residual optical feedback in the original polarization direction. At higher pulsation frequencies, beyond the 8-GHz bandwidth of the photodiodes used in this experiment, we observed superior performance of the polarization-rotated optical feedback relative to the optoelectronic feedback. Fig. 12 shows the relative performance of the polarization-rotated optical feedback compared to the optoelectronic feedback when the pulsation frequency is increased to nearly 20 GHz. The shift in the polarization-rotated peak is due to the drift during the experiment. While superior to the optoelectronic feedback when circuit losses become important, the polarization-rotated optical feedback is not as effective at the higher frequencies as it is at lower frequencies. III. DISCUSSION The drop in effectiveness of polarization-rotated optical feedback in reducing the P1 oscillation/pulsation linewidth at high frequencies can be understood by examining the coupledequation model for the circulating field amplitude and carrier density. This model has been successfully used to model the nonlinear dynamics of the optically injected semiconductor laser [1], [5]. If we assume that the optical spectrum is dominated by two strong optical frequency components [7] for large P1 frequencies, as has been previously observed, then the resulting leading-order terms for the normalized optical field,(1 + a)eiφ and carrier density n ˜ can be cast in the form (1 + a)eiφ ≈ a0 eiφ 0 + a1 ei(Ωτ 1 +φ 1 ) γn γs J˜ 1 − (a20 + a21 ) n ˜≈ γs 2 + a2 ) + 2a0 a1 Ω γn + (a 1 0 1 γn × sin(Ω1 τ + φ1 − φ0 ) + O
γn Ω1
(1)
2 (2)
(c)
0
0.05 0.1 0.15 0.2 0.25 Normalized Injection Strength (AU)
(b)
0.15
1.5 0.1 1 0.05 0.5 10
0 60
20 30 40 50 Offset Frequency (GHz)
70
Normalized Carrier Density (AU)
-70
0.15
2 Normalized Intensity (AU)
IL IL + PRF IL + OEF
(a)
2.5
Period-One Frequency (GHz)
Signal (dBm)
-65
Normalized Intensity (AU)
-60
1500807
Normalized Carrier Density (AU)
SIMPSON et al.: LINEWIDTH SHARPENING VIA POLARIZATION-ROTATED FEEDBACK
(d)
60 50 40 30 20 10 10
20 30 40 50 Offset Frequency (GHz)
60
Fig. 13. Dependence of the peak-to-peak variations of the normalized intensity and normalized carrier density (a and b) and P1 frequency (c and d) as a function of the amplitude of the injection parameter (a and c) and master laser offset frequency (b and d). The laser parameters are the same as those used to generate the mapping in Fig. 1.
where the normalization is with respect to the free-running values, Ω1 is the pulsation frequency, γs is the carrier decay rate, γn is the differential gain rate, and J˜ is the normalized pump parameter, which is the difference between the slave laser bias current and the threshold current, divided by the threshold current. We are taking γn /Ω1 as the smallness parameter, with the observation that the solutions of interest have Ω1 equal to or greater than the free-running relaxation resonance frequency, which is typically much larger than the differential gain rate. The key point is that the P1 oscillation term in the carrier equation scales with the smallness parameter. As Ω1 increases, the optical power fluctuations have a weaker influence on the carrier density, and thus, there is a weaker coupling of the polarizationrotated optical feedback to the laser mode. This analysis is further supported by numerical calculations of the full nonlinear coupled-equation model of the system [2]. We looked at the modulation characteristics of the gain medium as the P1 oscillation/pulsation frequency is changed by varying the master laser offset or the injection signal strength. The calculations were made using the parameters for the P1 mapping of Fig. 1. The results are summarized in Figs. 13 and 14. In Fig. 13(a) and (c), the strength of the injected optical signal is varied, with the offset between master and slave laser held fixed at 10 GHz. The injection parameter is ξ = γηc ai , where η is the injection rate, γc is the cavity photon decay rate, and ai is the injection amplitude normalized to the amplitude of the free-running laser. Fig. 13(a) shows the changes in their respective peak-to-peak variations of the intensity and the carrier density, both normalized with respect to the free-running values, while Fig. 13(c) shows the changes to the P1 frequency. In Fig. 13(b) and (d), the offset frequency between the master and the slave laser is varied, while the injection parameter is fixed at 0.05. The changes to the optical and carrier amplitudes and P1 frequencies depend on the specific parameter change. Recall that the two-frequency solution for the carrier density (2) predicts that the carrier density should scale inversely with
1500807
IEEE JOURNAL OF SELECTED TOPICS IN QUANTUM ELECTRONICS, VOL. 19, NO. 4, JULY/AUGUST 2013
Normalized Carrier to Intensity Ratio
0.1 0.978 1.027x 1.067 1.2797x
0.09 0.08 0.07 0.06 0.05 0.04 0.03 0.02 10
20
30 40 50 Period-One Frequency (GHz)
60
Fig. 14. Calculated dependence of the ratio of the normalized peak-to-peak carrier oscillations to normalized peak-to-peak intensity oscillations as a function of the P1 frequency. Two sets of data are shown with the injection parameter varied in the one (solid red traces) and the offset frequency varied in the another (dashed blue traces). Also shown for comparison are best fit power-law trendlines that are close to an inverse dependence for the calculated curves (solid black traces).
the P1 frequency for a given intensity level. Therefore, we replot the data from Fig. 13 making the P1 frequency the independent variable and normalizing the carrier density peak changes to the intensity peak changes. The resulting calculated data are shown in Fig. 14, along with calculated power-law trendlines. All data now closely approximate the predicted inverse dependence. As the P1 frequency increases, the carriers play a less influential role in the nonlinear dynamics, though there is no strict threshold that marks a transition as is the case with the Hopf bifurcation. When the carrier dynamics are insignificant, the external optical injection leads primarily to a shift in the laser cavity resonance frequency and regenerative amplification of the injected optical signal. Both optoelectronic and polarization-rotated feedback, each of which operates through the carrier density, are less influential at high P1 frequencies due to the reduced response of the carriers. IV. CONCLUSION In conclusion, polarization-rotated optical feedback provides a self-referencing signal to stabilize the tunable P1 oscillations of a semiconductor laser subject to external optical injection. Nonlinear dynamics generate operating conditions where the P1 frequency exhibits reduced sensitivity to fluctuations of master and slave laser operating points. Residual optical feedback in the original polarization induces amplitude jitter due to interference with master laser injection in our fiber-coupled system. Investigations are underway to control these fluctuations for improved performance. REFERENCES [1] T. B. Simpson, J.-M. Liu, A. Gavrielides, V. Kovanis, and P. M. Alsing, “Period-doubling cascades and chaos in a semiconductor laser with optical injection,” Phys. Rev. A, vol. 51, pp. 4181–4185, 1995. [2] S. Chan, S. Hwang, and J.-M. Liu, “Period-one oscillation for photonic microwave transmission using an optically-injected semiconductor laser,” Opt. Exp., vol. 15, pp. 14921–14935, 2007.
[3] S. Chan and J.-M. Liu, “Tunable narrow-linewidth photonic microwave generation using semiconductor laser dynamics,” IEEE J. Sel. Topics Quantum Electron., vol. 10, no. 5, pp. 1025–1032, Sep./Oct. 2004. [4] D. Eliyahu, D. Seidel, and L. Maleki, “RF amplitude and phase-noise reduction of an optical link and an opto-electronic oscillator,” IEEE Trans. Microw. Theory Tech., vol. 56, no. 2, pp. 449–456, Feb. 2008. [5] T. B. Simpson, “Mapping the nonlinear dynamics of a distributed feedback semiconductor laser subject to external optical injection,” Opt. Commun., vol. 215, pp. 135–151, 2003. [6] T. B. Simpson, J.-M. Liu, M. AlMulla, N. G. Usechak, and V. Kovanis, “Tunable photonic microwave oscillator self-locked by polarizationrotated optical feedback,” in Proc. IEEE Int. Freq. Control Symp., May 2012, pp. 1–5. [7] S.-Z. Chan, “Analysis of an optically injected semiconductor laser for microwave generation,” IEEE J. Quantum Electron., vol. 46, no. 3, pp. 421– 428, Mar. 2010.
Thomas B. Simpson (M’95) received the Ph.D. degree in applied physics from Harvard University, Cambridge, MA, USA, in 1983, investigating infrared multiphoton excitation of polyatomic molecules. He then served for four years in the U.S. Army, assigned to the Harry Diamond Research Laboratories, where he investigated optical and nonlinear optical properties of doping superlattices and other novel materials. In 1987, he joined Jaycor (subsequently purchased by L-3 Communications, and now part of L-3 Applied Technogies, Inc.) as a Senior Scientist. He has investigated a variety of topics in photonics, with the nonlinear dynamics of semiconductor lasers as a principal research area. He has published more than 40 technical papers and has one patent. Dr. Simpson is a member of the American Physical Society, the Sigma Xi, and the Tau Beta Pi. Jia-Ming Liu (M’83–SM’85–F’08) received the B.S. degree in electrophysics from National Chiao Tung University, Hsinchu, Taiwan, in 1975, and the S.M. and Ph.D. degrees in applied physics from Harvard University, Cambridge, MA, USA, in 1979 and 1982, respectively. He was an Assistant Professor in the Department of Electrical and Computer Engineering, State University of New York at Buffalo, from 1982 to 1983 and was a Senior Technical Staff Member of the GTE Laboratories, Inc., from 1983 to 1986. He is currently a Professor of Electrical Engineering at the University of California, Los Angeles, USA. He is also a Chair Professor of Photonics at National Chiao Tung University, and has been a Distinguished Visiting Professor in many universities. He is a pioneering researcher in ultrafast lasers, ultrafast laser–material interactions, and laser dynamics. He has more than 200 scientific papers, 14 book chapters, 12 U.S. patents, and two books. He has delivered more than 160 conference presentations and technical seminars, including 70 invited talks. His research interests include nonlinear optics, ultrafast optics, solid-state lasers, semiconductor lasers, photonic devices, optical wave propagation, nonlinear laser dynamics, chaotic communications, chaotic radar, nanophotonic imaging, and graphene photonics. Dr. Liu became a Licensed Professional Electrical Engineer in 1977. He is a Fellow of the Optical Society of America, the American Physical Society, and the Guggenheim foundation.
Mohammad AlMulla was born in Kuwait, in 1985. He received the B.S. degree in electrical engineering from Kuwait University, Kuwait, in 2008, and the M.S degree in electrical engineering from the University of California at Los Angeles, USA, in 2012, where he is currently working toward the Ph.D. degree in electrical engineering. His current research interests include optical communication systems, semiconductor lasers, and nonlinear dynamics.
SIMPSON et al.: LINEWIDTH SHARPENING VIA POLARIZATION-ROTATED FEEDBACK
Nicholas G. Usechak (M’07–SM’12) was born in Long Branch, NJ, in 1976. He received the B.S. degrees (with high Hons.) in both electrical engineering and engineering physics from Lehigh University, Bethlehem, PA, USA, in 2000. He received the M.S. and Ph.D. degrees both in optical engineering from the Institute of Optics, University of Rochester, Rochester NY, USA, in 2003 and 2006, respectively, where his dissertation focused on experiments, simulations, and the theory of FM mode-locked fiber lasers. He was at Trumpf Photonics, Cranbury, NJ, USA, for a year following graduation characterizing high-power semiconductor laser arrays, automating experiments, and modeling the thermal effects of solder interfaces using transient temperature-extraction experiments to ground those models. After Trumpf, he joined the Air Force Research Laboratory, Wright–Patterson Air Force Base, OH, USA, where he is currently a Senior Electronics Engineer. At Lehigh, he was a presidential scholar during the academic year 1999–2000. At the University of Rochester, he conducted his experimental work in the Laboratory for Laser Energetics where he was a Frank J. Horton Fellow. His research interests include nonlinear fiber optics, fiber lasers, ultrafast optics, high-speed test and measurement, high-power semiconductor lasers, mode-locked lasers, parametric processes, optical clock generation, novel gain media, and diverse waveform generation. Dr. Usechak is a member of the Optical Society of America, the Tau Beta Pi, and the Sigma Xi.
1500807
Vassilios Kovanis studied physics at the University of Athens, Athens, Greece, followed by graduate study at Temple University, Pennsylvania, PA, USA, and completed the Ph.D. dissertation in condensed matter theory at the University of New Mexico, Albuquerque, NM, USA. In September 1989, he joined the Nonlinear Optics Center at the Air Force Weapons Laboratory, Kirtland Air Force Base. He remained with that organization for the next 11 years, working on multiple projects of optical and electronic technologies. During that period, he also held research faculty positions in the Departments of Applied Mathematics and Electrical Engineering, University of New Mexico, and was a National Research Council Fellow between 1992 and 1994. Subsequently, he did a stint in corporate Research and Development Laboratories with Corning, Inc., Corning, NY, USA, as a Senior Research Scientist managing technical interactions with telecommunications system houses and with BinOptics Corporation, Ithaca, NY, USA, as a Program Manager for next-generation photonic product development. Between 2003 and 2005, he was a Faculty Member in the Department of Applied Mathematics, Rochester Institute of Technology. He returned to Air Force Research Laboratory in June 2005, where he is currently with the Photonics Technologies Branch at the Sensors Directorate of the Air Force Research Laboratory, Wright–Patterson Air Force Base, OH, USA. He has published extensively on issues of optical injection locking, coherence collapse, optical coupling of semiconductor lasers as well in controlling, synchronizing and communicating with chaotic waveforms.
