Abstract

A magneto-optical data storage system utilizing single-mode fiber is capable of providing high signal-to-noise ratio (SNR) recording if laser noise sources are properly managed. In particular, mode partition noise (MPN) associated with use of a Fabry–Perot laser diode can be a significant problem in a fiber-based system. The various mechanisms leading to MPN as well as to laser phase noise are discussed in the context of a system constructed with polarization-maintaining fiber. The primary noise mechanisms include spurious fiber-endface reflections and errors in the quarter-wave plate on the recording head. An understanding of these effects is essential for fabrication of a fiber-based recording system with suitable SNR performance.

© 2002 Optical Society of America

Full Article  |  PDF Article

References

  • View by:
  • |
  • |
  • |

  1. J. P. Wilde, J. F. Heanue, A. A. Tselikov, J. E. Hurst, “Magneto-optical disk drive technology using multiple fiber-coupled flying optical heads. Part I. System design and performance,” Appl. Opt. 40, 691–706 (2001).
    [CrossRef]
  2. K. Petermann, Laser Diode Modulation and Noise (Kluwer Academic, Dordrecht, The Netherlands, 1988).
    [CrossRef]
  3. E. C. Gage, S. Beckens, “Effects of high frequency injection and optical feedback on semiconductor laser performance,” in Optical Data Storage, Y. Tsunoda, M. R. de Haan, eds., Proc. SPIE1316, 199–204 (1990).
    [CrossRef]
  4. G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
    [CrossRef]
  5. J. U. de Arrudam, J. Blake, “Mode partition noise interferometry,” in Optical Fiber Sensors, Vol. 16 of 1997 OSA Technical Digest Series (Optical Society of America, Washington, D.C., 1997), pp. 590–593.
  6. J. W. Goodman, Statistical Optics (Wiley, New York, 1985), p. 162.
  7. S. Inoue, Y. Yamamoto, “Longitudinal-mode-partition noise in a semiconductor-laser-based interferometer,” Opt. Lett. 22, 328–330 (1997).
    [CrossRef] [PubMed]
  8. J. U. de Arruda, J. Blake, “Mode-partition noise interferometric conversion function,” Opt. Lett. 23, 1179–1181 (1998).
    [CrossRef]
  9. G. P. Agrawal, N. K. Dutta, Semiconductor Lasers, 2nd ed. (Van Nostrand Reinhold, New York, 1993), p. 255.
  10. The complete optical path and resulting data signal level were modeled with diffract, a product of MM Research Inc., Tucson, Ariz.

2001

1998

1997

1993

G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
[CrossRef]

Agrawal, G. P.

G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
[CrossRef]

G. P. Agrawal, N. K. Dutta, Semiconductor Lasers, 2nd ed. (Van Nostrand Reinhold, New York, 1993), p. 255.

Beckens, S.

E. C. Gage, S. Beckens, “Effects of high frequency injection and optical feedback on semiconductor laser performance,” in Optical Data Storage, Y. Tsunoda, M. R. de Haan, eds., Proc. SPIE1316, 199–204 (1990).
[CrossRef]

Blake, J.

J. U. de Arruda, J. Blake, “Mode-partition noise interferometric conversion function,” Opt. Lett. 23, 1179–1181 (1998).
[CrossRef]

J. U. de Arrudam, J. Blake, “Mode partition noise interferometry,” in Optical Fiber Sensors, Vol. 16 of 1997 OSA Technical Digest Series (Optical Society of America, Washington, D.C., 1997), pp. 590–593.

de Arruda, J. U.

de Arrudam, J. U.

J. U. de Arrudam, J. Blake, “Mode partition noise interferometry,” in Optical Fiber Sensors, Vol. 16 of 1997 OSA Technical Digest Series (Optical Society of America, Washington, D.C., 1997), pp. 590–593.

Dutta, N. K.

G. P. Agrawal, N. K. Dutta, Semiconductor Lasers, 2nd ed. (Van Nostrand Reinhold, New York, 1993), p. 255.

Gage, E. C.

G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
[CrossRef]

E. C. Gage, S. Beckens, “Effects of high frequency injection and optical feedback on semiconductor laser performance,” in Optical Data Storage, Y. Tsunoda, M. R. de Haan, eds., Proc. SPIE1316, 199–204 (1990).
[CrossRef]

Goodman, J. W.

J. W. Goodman, Statistical Optics (Wiley, New York, 1985), p. 162.

Gray, G. R.

G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
[CrossRef]

Heanue, J. F.

Hurst, J. E.

Inoue, S.

Petermann, K.

K. Petermann, Laser Diode Modulation and Noise (Kluwer Academic, Dordrecht, The Netherlands, 1988).
[CrossRef]

Ryan, A. T.

G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
[CrossRef]

Tselikov, A. A.

Wilde, J. P.

Yamamoto, Y.

Appl. Opt.

Opt. Eng.

G. R. Gray, A. T. Ryan, G. P. Agrawal, E. C. Gage, “Control of optical-feedback-induced laser intensity noise in optical data recording,” Opt. Eng. 32, 739–744 (1993).
[CrossRef]

Opt. Lett.

Other

G. P. Agrawal, N. K. Dutta, Semiconductor Lasers, 2nd ed. (Van Nostrand Reinhold, New York, 1993), p. 255.

The complete optical path and resulting data signal level were modeled with diffract, a product of MM Research Inc., Tucson, Ariz.

J. U. de Arrudam, J. Blake, “Mode partition noise interferometry,” in Optical Fiber Sensors, Vol. 16 of 1997 OSA Technical Digest Series (Optical Society of America, Washington, D.C., 1997), pp. 590–593.

J. W. Goodman, Statistical Optics (Wiley, New York, 1985), p. 162.

K. Petermann, Laser Diode Modulation and Noise (Kluwer Academic, Dordrecht, The Netherlands, 1988).
[CrossRef]

E. C. Gage, S. Beckens, “Effects of high frequency injection and optical feedback on semiconductor laser performance,” in Optical Data Storage, Y. Tsunoda, M. R. de Haan, eds., Proc. SPIE1316, 199–204 (1990).
[CrossRef]

Cited By

OSA participates in CrossRef's Cited-By Linking service. Citing articles from OSA journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (13)

Fig. 1
Fig. 1

Simplified diagram of the recording system based on PM fiber. The three primary sources of laser noise discussed in this paper are shown. They are (1) a spurious reflection from the fiber launch surface, (2) a spurious reflection from the head end of the fiber, and (3) imperfections in the quarter-wave plate (QWP1) on the head. LBS, leaky beam splitter; PBS, polarizing beam splitter.

Fig. 2
Fig. 2

Free-space Michelson interferometer setup used to measure MPN with a 660-nm FP laser. NPBS, nonpolarizing beam splitter.

Fig. 3
Fig. 3

Average output signal versus path-length difference from the interferometer setup of Fig. 2.

Fig. 4
Fig. 4

Noise power versus path-length difference from the interferometer setup of Fig. 2. The noise is measured at a frequency of 1 MHz with a resolution bandwidth of 30 kHz. The calculated shot-noise level and the measured intensity noise level are shown for the condition of a bright fringe (i.e., when all the optical power, ∼125 µW in this case, appears in the output arm).

Fig. 5
Fig. 5

Measurement of the MPN power level as a function of laser modulation frequency. The modulation depth for this experiment was maintained at 28%. Here the excess noise power corresponds to noise above the electronic plus shot-noise level.

Fig. 6
Fig. 6

Illustration of the way in which the pulse train from the front surface reflection is temporally shifted from the main beam pulse train. For deep modulation, an appropriate temporal shift prevents the two beams from interfering and hence precludes noise generation.

Fig. 7
Fig. 7

Minimum-noise curves, as described by Eq. (2), relating laser modulation frequency to fiber length.

Fig. 8
Fig. 8

Noise power measured by the differential detector when a fiber launch reflection interferes with the return main beam. PM fiber (Polarization-Maintaining and Absorption-Reducing style) designed for use at 660 nm and manufactured by Fujikura is used. The fiber launch surface is straight cleaved and AR coated (0.25% reflectivity). The laser modulation frequency controls the degree of temporal overlap between the two pulse trains entering the detection system. This result is obtained with use of a DFB laser and a benchtop optical setup having a one-way optical path difference of 1.62 m (L f = 0.97 m with n = 1.46, plus L 0 = 0.2 m of free space). The nominal optical power entering the differential detector is approximately 50 µW.

Fig. 9
Fig. 9

Excess laser noise power (above the shot plus electronic noise level) for errors in the head QWP. The model results assume P opt = 50 µW, τ c = 20 ps, S = 0.44 A/W, G = 2500 Ω, a load resistance of 50 Ω, and a resolution bandwidth of 30 kHz.

Fig. 10
Fig. 10

Measurements of the noise power versus optical power for three different settings of the head QWP angle. The calculated shot-noise-limit curve is also shown.

Fig. 11
Fig. 11

Plot of the relative noise-level constituents, along with the total relative noise level, as a function of the optical power. The relative noise levels are derived when we normalize the absolute levels by the optical power. In this way the RPN becomes independent of optical power, much like the more common RIN.

Fig. 12
Fig. 12

Data of Fig. 10 recast in terms of relative noise power. The data are fit by use of RPN as the free parameter.

Fig. 13
Fig. 13

CNR as a function of the RPN. Optimum system performance requires the optical path to be constructed such that the RPN is less than -145 dB/Hz.

Equations (31)

Equations on this page are rendered with MathJax. Learn more.

τ=2i+12frf, i = 0, 1, 2,.
frf=c2i+14(nLf+L0, i = 0, 1, 2,  low-noise condition.
Vshot=G2qSPoptB1/2,
RIN=VintVdc21B,
RINeff=RINCMRR.
RPN=VpolVdc21B,
total relative noisedB/Hz=10 logVen2+Vshot2+Vpol2+Vint2/CMRRVdc2B.
CNRdB=10 logVcVn2.
Einput=ExtEyt, Sfiber=100α, SQWP1=AB-B*A*, Sdisk=100-1.
A=cos2 ρexpiδ/2+sin2 ρexp-iδ/2,
B=2i sin ρ cos ρ sinδ/2.
S=ABCDforward pathS˜=A-C-BDreturn path.
Eout=R45° · SQWP20° · S˜fiber · S˜QWP1 · Sdisk·SQWP1 · Sfiber · Einput.
Eout,xt=12Q+2αRExt+i2αR-α2Q*Eyt,
Eout,yt=12-Q+2αRExt-i2αR-α2Q*Eyt,
QA2+B*2,
RAB-A*B*/2i=ImAB.
Eout,xt=12QExt+2RExt-τ+i2REyt-τ-iQ*Eyt-2τ,
Eout,yt=12-QExt+2RExt-τ-i2REyt-τ-iQ*Eyt-2τ.
ΔIt=|Eout,x|2-|Eout,y|2 =iQ2ExtEy*t-2τ+2QRExtEx*t-τ-2QREyt-τEy*t-2τ -i4R2Ext-τEy*t-τ+c.c.,
ΔIt=iQ2EtE*t-2τ+2QREtE*t-τ-2QREt-τE*t-2τ+c.c..
ΔĨt=iQ2EtE*t-2τ)+c.c.
ΔĨt=iQ2EtE*t-2τ+c.c.=iQ2ΓE2τ+c.c.,
Et1Et2E*t3E*t4=ΓEt1-t3ΓEt2-t4+ΓEt1-t4ΓEt2-t3.
ΓΔĨT=-Q4EtE*t-2τEt-TE*t-2τ-T+|Q|4EtE*t-2τE*t-TEt-2τ-T+c.c.=-Q4ΓE2τΓE2τ+ΓE2τ+TΓE2τ-T+|Q|4ΓE2τΓE-2τ+ΓETΓE-T+c.c.=-Q4ΓE22τ+ΓE2τ+TΓE2τ-T+|Q|4|ΓE2τ|2+|ΓET|2+c.c.
ΓΔĨT=ΔĨtΔĨt-T =|Q|4|ΓET|2-Q4ΓE2τ+TΓE2τ-T+c.c.
PΔĨω=ΓΔĨT =|Q|4- |ΓET|2 expiωTdT-Q4- ΓE2τ+TΓE2τ-TexpiωTdT+c.c.
PΔĨω=|Q|4- |ΓET|2 expiωTdT+c.c.=2|Q|4- |ΓET|2 expiωTdT.
=- |ΓET|2 expiωTdT- |ΓET|2dT=τc|ΓE0|2=τcPopt2,
PΔĨω=2τcPopt2|Q|4.
PΔV=2τcPoptSG2|Q|4.

Metrics