Abstracts

Back to homepage

 If you wish to receive a copy of a paper, please ask me.

Very Low Frequency Self-Mixing Laser Diode Vibrometer, Martini, G.; Randone E.; Donati, S.; IEEE SENSORS 2012 Proceedings, pp.254-257, October 28-31, 2012, Taipei, Taiwan, ISBN: 9781457717659, IEEE Catalog Number: CFP12SEN-USB, IEEE, Piscataway, NJ USA (2012)
Abstract We present an optical vibrometer especially designed to measure small amplitude (hundreds pm to several μm) vibrations at very low frequency, down to 0.01 Hz. The instrument is based on a Self-Mixing Interferometer of the fringe-lock type, to achieve high sensitivity and a wide dynamic range. As compared to previous design based on the same principle, our vibrometer is free from fringe-lock loss due to thermal drift of the laser wavelength; this result is obtained by exploiting a separate DC feedback loop acting on the Laser Diode bias. Experiments show a very good long-term stability, glitch-free operation, with rms error of 100-pm/sqrt(Hz) over a bandwidth from less than 20 Hz to 20 kHz.

Application of Delayed Optical Feedback to the Simultaneous Measurement of Index of Refraction and Thickness of Optical Slabs, Donati, S.; Martini, G.; Fathi, M.T.; International Symposium on Physcis and Applications of Laser Dynamics 2011 (IS-PALD 2011), 7-8 Dec, 2011, Tainan, Taiwan.

Self-Mix Interferometer to Measure Transparent Plates Thickness and Index of Refraction, Donati, S.; Martini, G.; Fathi, M.T.; Proceedings of IEEE SENSORS 2011 CONFERENCE, pp.1382-1385, 28-31 Oct, 2011, Limerick, Ireland, ISBN: 9781424492886, IEEE, Piscataway, NJ USA (2011)
Abstract We present a new method to measure thickness and index of refraction of glass windows and slab, typically from a few micrometers to 1000 micrometers and over. We use a very simple setup yielding two optical phase shift readouts: one from a self-mixing interferometer based on the beam crossing the specimen, going to an external mirror and back to the laser retracing the path, and one based on a conventional, forwardgoing shear interferometer based on the double reflection at the slab walls. We show that, interestingly and unique to our method, the difference of the two phase shift readouts provides the thickness measurement, independent from refractive index. We then go back to one of the two readouts to solve for the index of refraction also. We present the experimental arrangement along with the theoretical derivation of the phase shift involved, and discuss the sources of errors and their impact on the measurement accuracy and precision. Experimental data show a repeatability of +/-1% in thickness and +/-0.1 in index of refraction, with a 825-nm laser diode emitting 5 mW. The method looks interesting as it requires very little components, is compact, and can be used with sources at different wavelength as well as state of polarization.

Self-Mixing Interferometry: a Universal Yardstick to Measure Almost Everything, Donati, S.; Martini, G.; Proceedings of LFNM 2011, 11-th International Conference on Laser & Fiber-Optical Networks Modeling, 5-8 Sept, 2011, Kharkov, Ukraine, ISBN: 9781612848136

Self-mixing interferometry: A universal yardstick for optical measurements, Donati, S.; Martini, G.; Information Optics (WIO), 2011 10th Euro-American Workshop on , vol., no., pp.1-3, 19-24 June 2011, doi: 10.1109/WIO.2011.5981440
Abstract Self-mixing interferometry (SMI) is a new scheme of interferometry having several advantages compared to standard configurations: it can work with no optical part external to the laser, is self-aligned, is immune to external stray light, and can measure also on target with diffusing surfaces. In this paper, an overview of the several measurements already demonstrated with self-mixing is presented, with application tthe measurements of kinematic quantities (displacement, vibration. etc.) as well physical (linewidth, alfa factor, etc.)

Design and Characterization of Current-Assisted Photonic Demodulators in 0.18- \mu m CMOS Technology, Betta, G.D.; Donati, S.; Hossain, Q.D.; Martini, G.; Pancheri, L.; Saguatti, D.; Stoppa, D.; Verzellesi, G.; Electron Devices, IEEE Transactions on , vol.58, no.6, pp.1702-1709, June 2011, doi: 10.1109/TED.2011.2126578
Abstract We report on the design of a current-assisted photonic demodulator (CAPD) using standard 0.18-μm complementary metal–oxide–semiconductor technology and its electrooptical characterization. The device can perform both light detection and demodulation in the charge domain, owing to a drift field generated in the silicon substrate by a majority carrier flow. Minimum-sized 10 × 10 μm2 CAPDs exhibit a direct-current charge–transfer efficiency larger than 80% (corresponding to demodulation contrast larger than 40% under sine-wave modulation) at the modest power consumption of 10 μW and a 3-dB bandwidth of>45 MHz. An excellent linearity value with an error lower than 0.11% is obtained in phase measurements. CAPDs with optimized modulation electrode geometries are finally designed, aiming at an improved contrast-versus-power tradeoff.

TOF-Range Image Sensor in 0.18 \mu m CMOS technology based on Current Assisted Photonic Demodulators, Gian-Franco Dalla Betta, Silvano Donati, Quazi Hossain, Giuseppe Martini, Lucio Pancheri, David Stoppa, Giovanni Verzellesi, Lasers and Electro-Optics (CLEO), 2011 Conference on , Proceedings, pp.1-2, 1-6 May 2011
Abstract We report on a 0.18μ m CMOS range image sensor with 120×160 array of 10×10μ m2 photonic demodulation pixels allowing for real-time 3D imaging with a worst-case accuracy of 3.3% in the distance interval [1.2-3.7] m.

Noise bias correction in accumulated modulus NMR signal recovery, G. Martini, G. Ferrante, i64) Noise and Fluctuations (ICNF), 2011 21st International Conference on, Proceedings, June 12-16, 2011, Toronto, Canada, pp.425-428,
doi: 10.1109/ICNF.2011.5994361.
Abstract We discuss Nuclear Magnetic Resonance (NMR) signal detection in unstable magnetic field B and low Signal-to-Noise Ratio (SNR) condition. To improve SNR many acquisitions are accumulated and, because of B instability, in-phase and quadrature components (I\&Q) cannot be accumulated since carrier frequency changes from one acquisition to another. Carrier frequency is removed by modulus S calculation, allowing S accumulation. Resulting accumulated S has improved SNR by a factor $\sqrt(k)$, but suffers from a noise error, sometimes called ``noise bias'', arising from Rice statistics of S. We propose a technique to compensate such an error from knowledge of the original SNR of each acquisition of I\\&Q components. Usually SNR is estimated from acquisition with zero NMR signal, by switching off B the RF generator or, in NMR Imaging (MRI), from background pixels. Our technique is new, since we estimate original SNR without switching off signal, but accumulating both S and $S^2$. We describe the compensation technique, showing both simulated results and real world results confirming goodness of our approach.

Improving Photodetector Performance by Means of Micro-Optics Concentrators, Donati, S.; Martini, G.; Randone, E.; Lightwave Technology, Journal of , IEEE, vol.29, no.5, pp.661-665, March 1, 2011, doi: 10.1109/JLT.2010.2103302
Abstract We discuss the benefits of using micro-optics concentration arrays in connection with image (or pixellated) photodetectors, in terms of: (i) recovery of area fill-factor; (ii) reduction of equivalent dark-current; (iii) mitigate dead-time issues; and (iv) improved dynamic range. As an example of application, we describe the fillfactor recovery in connection to an array of 32 x 32 6-$mu$m diameter, 50-$mu$m pitch, SPADs (Single Photon Avalanche Detector). We use a 32 x 32 array of microlenses, fabricated by polymer casting in a photoresist replica mold. We demonstrate, for the first time to the best of our knowledge, an increase by a factor x25 of the effective spectral sensitivity of the final device. The lens array itself allows a x35 recovery, and projected improvements in excess of x50 appear feasible.
http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5678611&isnumber=4357488

Uniformity of Concentration Factor and Back Focal Length in Molded Polymer Microlens Arrays, S. Donati, G. Martini, E.M. Randone, M. Fathi, J. Lee, E. Charbon, in Conference on Lasers and Electro-Optics, OSA Technical Digest (CD) (Optical Society of America, 2010), paper JThE36.

Abstract An array of 32x32 polymer microlenses, 50-μm pitch, is used in connection to an array of 6-μm SPAD, recovering a factor C=35 in sensitivity. Repeatability of concentration factor and back focal length are reported.
© 2010 The Optical Society
http://www.opticsinfobase.org/abstract.cfm?URI=CLEO-2010-JThE36
OCIS Codes: (040.0040) Detectors : Detectors, (220.0220) Optical design and fabrication : Optical design and fabrication, (220.1770) Optical design and fabrication : Concentrators, (040.1345) Detectors : Avalanche photodiodes (APDs)

Dispositivo per la Ripresa di Immagini 3D Basato su Tecnologia CMOS 180nm e Telemetria a Modulazione Sinusoidale, G.-F.Dalla Betta, Q.D. Hossain, S. Donati, G. Martini, M. Fathi, E. Randone, G. Verzellesi, D. Saguatti, D. Stoppa, L. Pancheri, N. Massari, in Atti del Convegno FOTONICA 2010 su CD-ROM (AEIT, 2010), presentazione B3.6
Abstract We present the design and a few preliminary results for a 3-D camera based on a new CMOS compatible photo-detector providing an internal demodulation mechanism. The device has on-board pixel processing of the phase-to-distance signal and works in connection with a laser illuminator emitting 0.2 to 4 W of optical power modulated at a frequency of 20 to 50 MHz, to cover a range of distance 5 to 50 m with resolution of 1-5 cm.

Recupero del Fill-Factor in Rivelatori a Immagine SPAD mediante Matrici di Microlenti , E. M. Randone, M. T. Fathi, G. Martini, S. Donati, in Atti del Convegno FOTONICA 2010 su CD-ROM (AEIT, 2010), presentazione C3.3
Abstract Using an array of 32x32 plano-convex microlenses, fabricated by co-polymer casting in a photoresist replica mold, we have been able to increase the detection sensitivity of a SPAD (Single Photon Avalanche Detector) array of 32x32 pixels, 50-μm pitch, up to a factor of 35.

Banco Ottico per la Caratterizzazione di Matrici di Microlenti , E. M. Randone, M. T. Fathi, G. Martini, S. Donati, in Atti del Convegno FOTONICA 2010 su CD-ROM (AEIT, 2010), presentazione P1.15
Abstract We describe in the following an Optical Test Bench suitable to measure both Concentration and Back Focal Length of optical microlens arrays. Individual lens within an array can be characterized without array repositioning. The Optical Test Bench is easy to use and PC-controlled. The reproducibility achieved is better than 0.15% for Concentration and less than 300nm for Back Focal Length.

Uniformity of Concentration Factor and BFL in Microlens Array for Image Detectors Applications, G. Martini, E. Randone, M. Fathi, S. Donati, in Frontiers in Optics 2009/Laser Science XXV/Adaptive Optics: Methods, Analysis and Applications/Advances in Optical Materials/Computational Optical Sensing and Imaging/Femtosecond Laser Microfabrication/Signal Recovery and Synthesis on CD-ROM (Optical Society of America, Washington, DC, 2009), presentation FWG5, ISBN: 978-1-55752-878-0
Abstract We use an array of polymer microlenses spatially matched to an array of SPADs (Single Photon Avalanche Detector) to mitigate the loss of sensitivity due to the area fill-factor. The lens array is fabricated by polymer casting in a photoresist replica mold. We report results about reproducibility of concentration factor and back focal length within an individual 32x32 array. At a C factor of 35 and a focal length F=40 μm, the spread of concentration is < 6%, and the spread of BFL is <0.5 μm. ©2009 Optical Society of America OCIS codes: (040.00040) Detectors; (040.1240) Arrays, (220.1770) Concentrators

SPAD-Array Photoresponse is Increased by a Factor 35 by use of a Microlens Array Concentrator, E. Randone, G. Martini, M. Fathi, S. Donati, in 2009 IEEE LEOS Annual Meeting Conference Proceedings, (IEEE, Piscataway, NJ, 2009) presentation TuX3, pp.324-325, IEEE Catalog Number: CFP09LEO-CDR, ISBN: 978-1-4244-3681-1, ISSN: 1092-8081, ©2009 IEEE
Abstract Using an array of 32x32 plano-convex microlenses, fabricated by co-polymer casting in a photoresist replica mold, we have been able to increase the detection sensitivity of a SPAD (Single Photon Avalanche Detector) array of 32x32 pixels, 50-μm pitch, up to a factor of 35. ©2009 IEEE OCIS codes: (040.00040) Detectors; (040.1240) Arrays, (220.1770) Concentrators

Analysis and Design of Configurable LNAs in Feedback Common-Gate Topologies, Liscidini, A.; Martini, G.; Mastantuono, D.; Castello, R., Circuits and Systems II: Express Briefs, IEEE Transactions on , vol.55, no.8, pp.733-737, Aug. 2008
Abstract A unified description of multiple feedback common-gate low-noise amplifiers (LNAs) is presented, providing analytical expressions for gain, noise figure, linearity, and stability conditions. Moreover, from the theory, a new methodology for LNA optimization is developed. This new approach, called adaptive optimization, uses the ability to reconfigure the feedback network to match the amplifier characteristics to the changing working conditions. Results of simulation of LNAs with different feedback types are shown, and they confirm the theory presented.

On the Optical Concentration Achievable by a Non-imaging Microprism Array Combined to an Image Photodetector, G. Martini, S. Donati, E. Randone, in Proceedings of OSAV 2008, St. Petersburg Russia, 12-15 May 2008, pp.232-241.
Abstract An array of non-imaging micro-prisms can offer a substantial recovery of the sensitivity loss due to area fill-factor in photodetectors. This is especially true for device in which complex circuit functions are required, that consume a large fraction of pixel area, like e.g., CMOS sensors, image rengefinders, SPAD arrays, fluorescence analysers, etc. In this paper, after a general introduction, we evaluate the achievable concentration at asymptotically low NA (input/output area ratio) and high NA (inverse square of the input NA). Then, we present the results of a ray-tracing method for the fast and efficient calculation of the optical parameters of several types of prism concentrators, i.e., truncated cone, parabolic and compound parabolic, both reflective and refractive. Parameters considered are: power efficiency, effective concentration, channel crosstalk and power penalty for output gap.

Microconcentrators to recover fill-factor in image photodetectors with pixel on-board processing circuits, Silvano Donati, Giuseppe Martini, and Michele Norgia, in Optics Express, Vol. 15, Issue 26, December 18, 2007, pp. 18066-18075
Abstract We propose an array of non-imaging micro-concentrators as a mean to recover the loss of sensitivity due to area fill-factor. This is particularly important for those image photo detectors in which complex circuit functions are required and a substantial fraction of the pixel area is consumed, like e.g., 3D camera, SPAD arrays, fluorescence analyzers, etc., but also in CMOS sensors. So far, the low fill-factor was an unacceptable loss of sensitivity precluding from the development of such devices, whereas by using a concentrator array a recovery is possible, up to the inverse square of numerical aperture of the objective lens. By ray tracing, we calculate the concentration factors of several geometries of non-imaging concentrator, i.e., truncated cone, parabolic and compound parabolic, both reflective and refractive. The feasibility of a sizeable recovery of fill-factor (up to 50) is demonstrated.
http://www.opticsinfobase.org/abstract.cfm?URI=oe-15-26-18066

Microlens array for enhancement of irradiance and fill-factor recovery in image detectors, S. Donati, G. Martini, M. Norgia, F. Ingarozza, in Proceedings of WFOPC 2007, 5th Workshop on Optical Fibres and Passive Components, Taipei (R.o.C.), 4-7 Dec. 2007. Paper Th4I
Abstract We present the result of calculations of the concentration factor C obtained by plano-convex microlenses, either in direct or reverse format, mounted in front of a detector array. C factors in the range 20 to 40 are demonstrated feasible with lens spacing of 70 μm and detector size 10 μm.

Cancellation of Amplifier Offset and 1/f Noise: An Improved Chopper Stabilized Technique, A. Agnes, A. Cabrini, F. Maloberti, G. Martini, in IEEE Trans. Circuits Syst. II, Express Briefs, Vol.54, N.6, June 2007, pp. 469-473
Abstract In this brief, a theoretical analysis on the limits of conventional chopper stabilization technique is presented and a solution conceived to improve the amplifier performance is proposed. The expected replicas of the 1/f noise at the chopping frequency and its multiples are attenuated by a modified chopping control. Simulations done by using records of real 1/f noise outputs showed that, by following the proposed approach, the spectrum of the signal is not affected while the 1/f replicas are reduced by more than 40 dB with respect to conventional techniques. The required circuit for the generation of the chopping signal is also described. The resulting overhead with respect to conventional techniques is negligible and fully acceptable.

Improved Chopper Stabilized Amplifier for Offset and 1/f Noise Cancellation, A. Agnes, F. Maloberti, G. Martini, in Proceedings of ICECS 2006, Nice, France, Dec. 10-13, 2006, pp.529-532
Abstract The limits of the conventional chopper stabilization technique for cancelling the 1/f noise are studied theoretically and with simulations. The expected replicas of the 1/f noise at the chopping frequency and its multiples are attenuated by a modified chopping control. Simulation done using records of real 1/f noise outputs show that the spectrum of the signal does not change but 1/f replicas are reduced by more than 40 dB. The required circuit for the generation of the chopping signal is also described. The resulting overhead with respect to conventional solutions is negligible and fully acceptable.

Configurable LNAs in Feedback Common Gate Topology, Antonio Liscidini, Giuseppe Martini and Rinaldo Castello, in Proceedings of WiRTeP, Wireless Reconfigurable Terminals and Platforms, Rome 10-12 April 2006, pp.113-117.
Abstract A unified description of all feedback common gate LNA is presented, providing analytical expressions for gain, noise and linearity for bipolar and CMOS technology. The flexibility of these structures suggests a novel design methodology oriented to configurable low noise amplifiers both in frequency and in performance. Finally two different prototypes tailored for cellular and WLAN receivers are reported.

1/f NOISE IN LARGE SIGNAL OPERATION OF PASSIVE COMPONENTS, GIUSEPPE MARTINI, in Fluctuation and Noise Letters, Vol. 4, No. 3 (2004) L475-L489
Abstract A model for the 1/f noise in large signal operation of linear passive one-port (e.g. carbon or polysilicon resistors) is given. Starting from the Hooge's formula, that holds when the component is operated under DC bias, it is shown that the noise current is simply proportional to the product between the conductivity fluctuation and the time-dependent large signal applied to the component. Otherwise stated, the 1/f noise exhibited by passive components in large signal operation arises from the intermodulation between the stochastic process "conductivity fluctuation" and the signal applied to the component. Detailed calculations of both autocorrelation function and power spectrum of the resulting noise current are given; different time dependence of the signal applied to the component are considered. Of particular relevance in practical applications is the case of sinusoidal signal plus an eventual DC bias; in this case the resulting noise current is a cyclostationary stochastic process, and its behaviour can be conveniently described by the cyclic autocorrelation functions or by the cyclic power spectra. The measured power spectra of the noise current of carbon resistors with DC bias and large periodic signal applied to them, are in good agreement with those calculated from the proposed model. The agreement of the measured power spectra with those predicted by the model allows to conclude that the physical origin of 1/f noise in passive components does not depend on the applied bias, DC or large signal alternating, since, in any case, the noise current is conveniently modeled by intermodulation (product) between the "conductivity fluctuation" stochastic process and the applied signal.

Noise Figure Reduction Techniques in LNA's for Wide Band Multistandard RF Receivers , Giuseppe Martini, Antonio Liscidini and Rinaldo Castello, in Noise in Devices and Circuits II, edited by F. Danneville, F. Bonani, M. Jamal Deen, M. Levinshtein, Proceedings of SPIE Vol. 5470 (SPIE, Bellingham, WA, 2004), pp. 440-447.
Abstract LNA's (Low Noise Amplifier) are widely used in wireless portable personal communication systems; the LNA noise directly affects the overall system performances. Here it is shown how the Noise Figure of a RF (Radio Frequency) Receiver can be reduced while satisfying the common constraints of impedance matching at the input, low power consumption and good linearity. The proposed Noise Figure reduction technique is based on the cold resistance approach and negative or positive feedback; a single BJT LNA, derived from the common base configuration, is considered. The Noise Figure reduction is obtained over a wide frequency band of operation, and is thus suitable for Multistandard applications. Different LNA feedback topologies are compared. It is shown that a Noise Figure lower than the limit of the common base configuration can be achieved, along with a current consuption of a few mA, over a wide frequency band of operation. Noise Figure calculations and circuit simulation results are presented and compared.

1/f Noise in Passive Components Under Time-Varying Bias, Giuseppe Martini, in Noise in Devices and Circuits, M. Jamal Deen, Z. Celik-Butler, M. Levinshtein, Editors, Proceedings of SPIE Vol. 5113 (2003), pp. 406-414.
Abstract A model for the 1/f noise in passive components undergoing time-varying bias is proposed. One-port devices exhibiting 1/f noise are considered, e.g. carbon or polysilicon resistors. Starting from the Hooge’s formula, that holds when the component is operated under DC bias, it is shown that the noise current is simply proportional to the product between the conductivity fluctuation and the time-dependant voltage applied to the component. In other words, the 1/f noise in passive components under time-varying bias arises from the intermodulation between the stochastic process “conductivity fluctuation” and the voltage applied to the component. Detailed calculations of both autocorrelation function and power spectrum of the resulting noise current are given; different time dependences of the voltage applied to the component are considered. Of particular relevance is the case of sinusoidal voltage; in this case the resulting noise current is a cyclostationary stochastic process and its behaviour can be conveniently described by the cyclic autocorrelation functions or by the cyclic power spectra. A circuit is built to measure the actual power spectra of the noise current of carbon resistors under constant or periodic voltage bias, in order to verify the results obtained from the proposed model. Very good agreement between theory and experiment is observed, and this fact supports the proposed model for 1/f noise in passive components under time-varying bias. The conclusion is that the physical origin of 1/f noise in passive components does not depend on external bias, nor in DC nor when a time-varying signal is applied to the component.

Correlation Between Intensity and Phase in Speckle Pattern Interferometry , Giuseppe Martini and Marc Sorel, in Proceedings of the Int.l Conf. Interferometry in Speckle Light - Theory and Applications, edited by P.Jacquot and J.M. Fournier, (Springer-Verlag, Berlin, 2000), pp. 27-32.
Abstract We simulate the field distribution of a fully developed speckle pattern (SP) due to diffusion of laser light from a remote rough surface. We show that, under some circumstances, the phase fluctuation and the logarithm of intensity are linked by Hilbert transform, excluding deep intensity minima. To overcome the problems arising from deep intensity minima., we propose the integration of the SP field over the finite detector area. We compute the autocorrelation function of the SP along the longitudinal axis z as a function of the detector area; we find a rapid increase of the correlation length with the detector area, up to a nearly complete correlation for integration just over a few transversal speckles. We conclude that the Hilbert transform method could open the way to compensation of SP phase errors in interferometric measurements.  

Automatic Phase Stepping in Fiber-optic ESPI by Closed-Loop Gain Switching , Giuseppe Martini, Mauro Facchini and Dario Parisi, in IEEE Trans. Instrum. Meas.- Vol.49 N.4 August 2000 pp. 823-828
Abstract In this paper we describe a new technique for automatic phase stepping in fiber-optic ESPI (Electronic Speckle Pattern Interferometry), useful in both single-point and imaging configurations. The proposed technique is based on closed-loop switching of the gain sign in the control loop used to stabilize the reference wave front phase, with the aid of a P.I.  (Proportional-Integral) controller. Simulation and experimental results are compared, showing that by a synchronization  signal or by a proper choice of the feedback loop parameters, a sequence of pi/2 steps of the reference wavefront phase, all of the same sign, is obtained. The residual fluctuation of each phase step amounts to +-2.5 mr over a bandwidth of 200Hz. This level of phase stability is well suited for ESPI aplications. This work was partially supported by the Measurement and Testing section of the Fourth framework program of the European Union.

Intermodulation noise in translinear filters, Giuseppe Martini and Vito Svelto, Alta Frequenza-Rivista di Elettronica - Vol. 11 N.2 Aprile-Giugno 1999 pp.91-94
Abstract We examine the noise properties of translinear filters operated in class A and class AB, discussing how the output noise power depends on the circuit topology. We use the simplest first order low pass filter built with four BJTs. We show that for in band useful signal the output noise does not depend on the relative position of the transistors in the filter. For class A operation we show that, with a proper choice of the circuit topology, an out of band interfering signal does not add any noise in the average to the output. For class B operation the output noise is due to intermodulation between internal noise sources and signal; the noise power grows with the signal power. For class B operation and out of band interfering signal we calculate that, depending on the circuit topology, the output noise power is reduced by 1.5 or 3.9 dB as compared with the output noise due to an in band signal of the same power.  

Correlation Between Intensity and Phase in Speckle Pattern Interferometry , Marc Sorel, Giuseppe Martini, Silvano Donati, ODIMAP II Proceedings , IEEE-LEOS, May 1999, PAVIA pp.132-137
Abstract We simulate the field distribution of a fully developed speckle pattern (SP) due to diffusion of laser light from a remote surface. We show that the phase fluctuation and the logarithm of intensity fluctuation are linked by Hilbert transform, excluding points where the field vanishes. To overcome the problems arising from zero field points, we propose the integration of the SP field over the finite detector area. We compute the autocorrelation function of the SP along the longitudinal axis z, as a function of the diffuser and the detector areas; we find a rapid increase of the correlation length with the detector area, up to a nearly complete correlation for integration just over a few transversal speckles. The Hilbert-transform method could open the way to compensation of SP phase errors in interferometric measurements.  

Noise in Class AB Translinear Filters , Giuseppe Martini and Vito Svelto, Alta Frequenza-Rivista di Elettronica - Vol.10 N.4 Luglio-Agosto 1998 pp.53-56
Abstract A specific statistical approach to describe the noise properties of non linear circuits is used. The noise properties of translinear filters operated in class AB are considered. This kind of filter has a dynamic range larger then the maximum signal to noise ratio, and exhibit signal to noise ratio saturation at high signal level. We show how the noise properties depend on the circuit design parameters.