WO2005101658A1

WO2005101658A1 - Analogue non-linear dynamic filter

Info

Publication number: WO2005101658A1
Application number: PCT/IB2005/051146
Authority: WO
Inventors: Hendrikus M. W. Goossens; Guido E. De Vrieze; Marcel F. Heertjes
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2004-04-15
Filing date: 2005-04-07
Publication date: 2005-10-27
Also published as: TW200611257A

Abstract

An analogue non-linear dynamic filter having a substantially semisoidal input-output transfer characteristic with a trough at a first constant level and a peak at a second constant level, and a continuously variable transfer function between the first and second levels. The first level, which is substantially zero, is defined by a small area of the input signal in the vicinity of the zero crossings thereof, and is obtained by subtracting the output of a first clipper circuit (12) from the input signal. The second constant level is obtained by applying the signal to a second clipper circuit (14) and the signal is then applied to a rectifier (16) before being multiplied by the input signal to attain a variable gain function.

Description

Analogue non-linear dynamic filter

This invention relates to an analogue, non- linear dynamic filter and, more particularly but not necessarily exclusively, to an analogue, non-linear dynamic filter for use with with a servo control circuit which controls, for example, the radial, focus and tilt tracking functions of an optical scanning unit in an optical data storage system in which data is stored on an optical storage medium. Examples of such storage media are, for instance, CD-ROM, CD-R, CD-RW, DVD, Blu Ray etc. In these examples, the optical storage medium has the shape of a disc. Optical data storage systems provide a means for storing great quantities of data on a disc. As is well known in the art, an optical disc comprises at least one track which is capable of containing data written therein. The disc may be embodied so as to be a readonly disc: the disc is manufactured with data recorded in the track, and this data can only be read from the disc. However, writeable optical discs, allowing a user to record data on a disc, are also known; in this case, a disc will normally be manufactured as a blank disc, i.e. a disc having a track structure but without data recorded in the track. Similarly, disc drives may be designed as read-only devices, i.e. devices only capable of reading information from a recorded disc. However, disc drives may also be designed for writing information into the track of a recordable disc. Physically, the information bearing portion of an optical disc is a series of pits, or bumps, arranged to form a spiral track. Data is encoded in the length of the individual pits and of the space between pits. A laser beam reflected, from the optical disc is modulated by the pits and spaces, and received by a detector whic produces a similarly modulated electrical signal, or track data signal. Generally, in an optical data storage system, as an optical disc is rotated, a signal surface on which the information signal is subject to a certain amount of vertical movement, dependent on machining accuracy, rotation accuracy, etc. In order to precisely read the information signal, the system is typically provided with a focussing function arranged and configured to maintain the distance between the objective lens (which focuses the light beam from a light source on a target) and the information storage medium constant. In other words, a focus servo system is used which drives the objective lens to compensate for the vertical movement of the signal surface so that the objective lens always accurately focuses a laser beam on the signal surface. In addition, a known type of optical storage system is provided with a function for causing the focused light spot to target track on the information storage medium. This function is referred to as the tracking function. The tracking function is accomplished whereby a detection signal (representing the distance between a target track on the information storage medium and the light spot) is generated corresponding to the light beam reflected from the information storage medium. The detection signal is converted into a digital signal by an A/D converter. Corresponding to the digital sigpnal, the objective lens is driven by a digital servo system. Thus, in summary, recording and playback devices for optical information carriers have servo control circuits in order to compensate for deviations or errors such as tracking, radial and/or focus errors, the error signals being derived from measured values of an optical scanning unit and processed in a servo unit to form control signals and these control signals being fed to the actuators of the optical scanning unit. Error signals have various different causes such as, -for example, slight curvature of the optical disc, slight inclination of the disc with respect to the drive axis, etc., as well as low-frequency shock and vibration (typically, below 50O Hz) resulting from, for example, road excitation, engine vibration or any other shock (i.e. mechanical event) causing undesired movement of the recording or playback device, and high frequency disc defects (typically beyond 5 kHz) caused by, for example, scratches, dirt spots or fingerprints on the disc itself. In the past, optical storage drives have typically been controlled by linear feedback based on the concept of bandwidth. Thus, in order to compensate for the above- mentioned low-frequency shock and vibration, the bandwidth is often given a large value so as to attain sufficient low-frequency disturbance rejection. As a consequence, a significant sensitivity to measurement noise is achieved. However, in view of the above-mentioned high frequency disc defects, which may severely corrupt the radial error signal, this is often undesirable as it causes a deterioration of system performance. In other words, in the case of known applications using linear control design, the applied bandwidth effectively represents a forced trade-off between disturbance rejection to compensate for harmonic shocks, on the one hand, and so-called playability in view of disc defects. In order to overcome this trade-off, several different proposals have been made, mostly related to the concept of variable gain, whereby on the basis of a radial error signal, the controller gain is increased beyond a certain threshold value so as to improve low- frequency disturbance rejection caused by low-frequency vibration axtd shock. For example, US Patent No. 4,722,079 describes a three-beam type optical disc player, wherein beam output disturbances due to an external disturbance such as vibration and beam output disturbances due to a local defect on the disc are distinguished from each other, and the gain of the servo unit is adjusted according "to whether the beam output disturbance is due to the external disturbance or due to the local defect, so that the main beam is controlled to correctly follow the track at all times. The system is arranged to generate a main beam for reading the data recorded on the disc, a front beam directed in front of the main beam for generating a signal utilised to form a radial error signal used to control the main beam to follow the track, and a rear beam directed to the rear of the main beam and used for generating a signal utlised to form a radial error signal. In use, a data signal recorded on the track of a disc is read with the main beam. The data signal thus read is amplified by a main beam pre-amplifier and output at a high frequency signal (RF signal) which is converted into an audio signal by a demodulating circuit, error correcting circuit and digital- to-analogue converter. Further, a front beam amplifier and a rear beam amplifier extract the low frequency components from the signals read by the front beam and the rear beam and output them as radial error detection signals. A differential amplifier obtains the difference between these radial error detection signals to form a radial error signal which is applied to a radial servo circuit. In the case where no external disturbance exists and trie disc has no local defects, only residual errors such as eccentric errors occur, and a servo gain control circuit sets the gain of the radial servo circuit to a first servo gain value. W en an external disturbance such as vibration is applied to the optical disc player, this is identified by the servo gain control circuit and, as a result, the servo gain of the radial servo circuit is set to a second servo gain (greater than the first servo gain), thereby increasing the disturbance eliminating capability of the radial servo circuit significantly, and ensuring that the main beam is quickly returned to the central line of the track. In the case where the disc has a local defect, this too can be identified by the servo gain control circuit, in response to which, it sets the servo gain of the radial servo circuit to a third servo gain (smaller than the first servo gain) so that displacement of the main beam from the current position is prevented. European Patent Application No. 0363195 describes a-n arrangement having a tracking servo loop and a focus servo loop, to control the tracking and focussing functions of the optical disc player. Disc defect detection signals are generated in response to a change in a low frequency component and a high frequency component of a read signal oxttput from a pickup, and the transfer function of at least one of the tracking servo loop and the focus servo loop is adjusted in response to the disc defect detection signal. However, in solutions based on the concept of variable gain, in principle, the amount of controller gain increase is limited by the stability of the non-linear servo control loop. In order to relax this stability limitation and hence enable a significant amount of controller gain increase, the use of a dynamic, rather than a static, non- linear input-output filter has been proposed, in which if the error signal is less than some predetermined threshold value, it is not additionally amplified, i.e. it is not subjected to any additional loop gain at all beyond that resulting from the underlying linear feedback loop, in an attempt to compensate for both the low-frequency shock and vibration and the high frequency disc defects. We have now devised an improved arrangement, and it is an obj ect of the present invention to provide an analogue non-linear dynamic filter for use in a servo control function of, for example, an optical scanning unit, which provides improved compensation for the various types of radial, focus and tracking errors which may occur. Thus, in accordance with the present invention, there is provided, an analogue filter having an input/output transfer characteristic in the time domain defined by having a constant portion at a first level and a constant portion at a second level, and a substantially quadratic transfer function therebetween. The present invention also extends to an optical scanning device comprising an amplifier for providing a gain path in respect of an input signal, and a filter as defined above, wherein an error signal is applied to the filter, and an output of said filter is applied to the amplifier. The amplifier preferably has a substantially constant gain. The present invention extends still further to an optical data storage system including an optical scanning device as defined above. The first level is preferably substantially zero, and is defined by a portion of an analogue input signal to the filter which is within a predetermined vicinity of trie zero crossings of said input signal. This substantially zero first level is beneficially obtained by providing a first clipping circuit for clipping the input signal at a predetermined level outside of a predetermined vicinity of the zero crossings of the input signal, and then subtracting the output signal from the first clipper circuit from the input signal, to provide an input signal having a dead-zone in the vicinity of the zero crossings. The size of the predetermined vicinity is preferably adjustable. The second constant level may be obtained by providing a second clipper circuit for receiving this input signal having the dead-zone and clipping the input signal at a non-zero level, which is preferably adjustable. The output signal from the second clipper circuit may be applied to a rectifier. In an optical scanning device according to an exemplary embodiment of t ie present invention, the analogue input signal applied to the analogue non-linear dynamic filter is a position error signal (obtained in any known manner), and the output of the filter is multiplied with the position error signal to realise a variable gain transfer function.

These and other aspects of the present invention will be apparent from, and elucidated with reference to, the embodiment described herein. An embodiment of the present invention will now be described by way of example only and with reference to the accompanying drawings, in which: Figure 1 is a schematic block diagram illustrating the principal components of a filter according to an exemplary embodiment of the present invention; Figures 2a - 2e are graphical representations of the transfer characteristics of the signal at various points in the filter of Figure 1; Figure 3 is a circuit diagram illustrating an input/output connection for a CD- audio in which a filter according to an exemplary embodiment of the present invention may be employed; Figure 4 is a circuit diagram of a filter for use in the circuit of Figure 3; and Figure 5 is a diagram illustrating an error signal and the non-linear gain wbich may be applied thereto using a filter according to an exemplary embodiment of the present invention.

Thus, as explained above, with reference to the prior art, within known linear control designs, the amount of disturbance rejection in optical storage drives subjected to low-frequency shock and vibration significantly improves by increasing the loop gain.

However, if the beam output disturbance is due to a disc defect, an increased loop gain results in a deterioration of playability. As a result, in known variable gain control designs the fhced trade-off between disturbance rejection and playability is avoided by specifically accounting for the actual cause of the disturbance in beam output: namely, low-frequency shock and vibration, high-frequency disc surface defects, or a combination of the two. For example, US Patent No. 5,406,536 describes an optical disc system in which a focus error signal representing the distance between an optical disc and an obj ective lens is generated using a detection signal detected by an optical detector of an optical l ead. This focus error signal (and/or a tracking error signal) is sent to a non-linear amplifying circuit which is arranged and configured to amplify a signal with a larger gain in the vicinity of a control target than at its peak. Thus, in the case of a focus error signal, this is amplified in the vicinity of a servo point, but suppressed at levels oustside this vicinity, so as to create a drive control signal for driving an objective lens in focus direction and tracking direction. In other words, the non- linear amplifying circuit is arranged to operate in one of two modes, depending on the level of the input signal. If the level of the input signal is within a predetermined vicinity of a central servo point, then the amplifying circuit applies a first , (relatively high) gain, whereas a (relatively lower) gain is applied otherwise. In the arrangement described in US Patent No. 6,046,967, a servo control circuit is provided which controls the tracking and the focus of an optical scanning unit with reference to an error signal. A linear amplifier and a non-linear amplifier are provided upstream of the servo control circuit, together with a defect detector and a controllable switch. In the event that a disturbance in the beam output is caused by external shock or vibration, the error signal is non-linearly amplified such that the tracking and focus are retained or adjusted. If a disc defect exists, such as a scratch or the like, the defect detector and the linear amplifier are additionally connected in parallel with the non- linear amplifier by means of operation of the controllable switch. The linearly amplified signal or the non- linearly amplified signal is fed to the servo control circuit as a function of the output signal of the defect detector. In other words, two different amplifiers, having two different respective loop gains are provided, which amplifiers are selectively used to amplify the error signal, depending on the cause of that error signal. Japanese Patent Application No. 09091902 describes a head position- controlling device in which the occurrence of "chattering" or minute vibrational effects is at least minimised, and the complexity of the servo control signal generating circuitry is reduced, by taking into account an estimated relative speed of a positional error signal. This positional error signal and its estimated relative speed are added together and then applied to a non-linear filter which has an input-output transfer characteristic including a dead-zone. The resultant signal is then applied to an amplifier having a predetermined gain to provide a control signal. Because of the dead-zone included in the transfer characteristic of the nonlinear filter, the signal applied to the amplifier is only non-zero if the sum of the positional error signal and its estimated speed is outside a predetermined range in the vicinity of zero. As a result, this arrangement effectively applies a zero gain to the summed signal if it is within this predetermined range, and otherwise applies a high gain thereto, irrespective of its value. We have now devised an improved arrangement in which an analogue , nonlinear, dynamic filter is provided which has a non-linear signal transfer function and which can, for example, be used in an optical data storage system, particularly for the purpose of reducing the effect of mechanical shocks on radial, focus and tilt tracking. Referring to Figure 1 of the drawings, a schematic block diagram of an arrangement including an analogue, non-linear, dynamic filter according to an exemplary embodiment of the present invention is provided for creating a signal S_out, such that (for example) the loop gain of a signal loop (for example, a radial loop) is increased if the shock (i.e. the input signal Sj_n) is above a certain level. Figure 2 illustrates graphically the signal characteristics at various points in the arrangement of Figure 1. In more detail, the filter 10 comprises a first clipper circuit 12 to which the signal S_m- is applied (see Figure 2a), which has the effect of clipping the signal outside of the range Sj_n-_Sman, which portion of the signal is then subtracted from Sj_n to give a transfer function at point A, as shown in Figure 2b, which is zero for S;_n in the range Si_n._smaiι- The signal A is then applied to a second clipper circuit 14 to "clip" the signal above another predetermined level X, giving a transfer function at point B as illustrated in Figure 2c. The signal at B is then applied to a rectifier 16 to give a signal at point C with a transfer function as shown in Figure 2d. Thus, the transfer function at point C of the circuit is continuously variable between X and a, outside of which area the transfer function is constant. The signal C is then added to S;_n by multiplier 18 to create signal S_out (see Figure 2e). It will be appreciated that the non-linear transfer function provided by the filter described herein can, for example, be used as a parallel gain path across an already available gain path in the signal loop. In other words, an extra signal is added to the error signal Sj_n (to create S_out) such that if the shock is below a certain level (i.e. in the region indicated by Si_n-_Smaiι), then the loop gain is not changed from that of the already available gain path because the additional signal applied to Sj_n to create S_out for this region is zero. However, outside of the region Si_n-_Sman, the positional error signal is significantly increased by an increasing amount, but only between the levels a and X, beyond which the positional error signal is increased by a constant amount. In other words, the arrangement described above has the effect of providing an overall gain which is a continuously variable function between two constants. As will be apparent to a person skilled in the art, an arrangement according to the following exemplary embodiment of the present invention comprises an analogue realisation of a variable gain controller with dynamic filtering, based on the following steps: 1) obtain diode signals, 2) reconstruct the radial error from these signals, 3) apply dynamic filtering, 4) apply a non-linear input-output transformation, 5) retrieve the individual diode contributions, and 6) add the separate contributions to the original diode signals to serve as input for the original control design. It will be appreciated that steps 4) and 5) can be reversed without affecting the stability properties of the system. Referring to Figure 3 of the drawings, there is provided a circuit diagram of an input/output connection for a CD-AUDIO optical data storage system by means of which steps 1), 2), 5) and 6) may be performed. The non-linear dynamic filter, i.e. the portion of the circuit which performs steps 3) and 4) mentioned above, forms an interconnection between nodes Notchln and NLOut in Figure 3, and is illustrated in Figure 4 of the drawings by means of the static input/output non-linearity, which is combined with a standard notch filter. The additional non-linear gain provided by the present invention is realised using the following five-step approach: 1) the error signal is saturated at an adjustable level (first clipper circuit 12, Figure 1), 2) this is then subtracted from the original error signal, and possibly multiplied by a gain, which provides an adjustable dead-zone characteristic; 3) this dead-zone characteristic is again saturated at an adjustable level (second clipper circuit 14, Figure 1); a sign operation is performed to obtain the variable gain caracteristic (rectifier 16, Figure 1); and 5) this variable gain characteristic is multiplied by the original error signal (multiplier 18, Figure 1) so as to create a non- linear control force characteristic. Basically, the above-described realisation of an exemplary embodiment of the present invention contains two force characteristics: i) a piecewise linear characteristic obtained at step 2) which is continuous, but non-smooth, at the switching incidents, and ii) a characteristic at step 5) which is both continuous and smooth at the switching incident, but which is no longer piecewise linear due to the quadratic behaviour in between the two switching incidents. The effect of a filter according to an exemplary embodiment of the present invention to a CDM-M3 with CD 10 chipset module is illustrated in Figure 5 of the drawings. The upper signal represents the reconstructed radial error signal after subjecting the entire CD-module to a sinusoidal disturbance at 100 Hz. In the lower signal, the additional nonlinear gain is shown. If the radial error signal is small (i.e. Si_n-_Smaiι near the zero crossings), the additional non-linear gain approximately equals zero. However, beyond a certain radial error level, the non-linear contribution increases, resulting in additional non-linear gain. It will be appreciated that the present invention is relevant to, for example, optical storage drives subjected to steady-state excitation, for example, automotive or portable applications. Optical storage drives and hard disc drives in general are alos considered to be within the field of application of the present invention, for example, for use in shock handling for electronic notebooks or telecommunications applications. An embodiment of the present invention has been described above by way of example only, and it will be apparent to a person skilled in the art that modifications and variations can be made to the described embodiment without departing from the scope of the invention as defined by the appended claims. Further, in the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. The term "comprising" does not exclude the presence of elements or steps other than those listed in a claim. The terms "a" or " an" does not exclude a plurality. In a device claim enumerating several means, several of these means can be embodied by one and the same item of hardware. The mere fact that measures are recited in mutually different independent claims does not indicate that a combination of these measures cannot be used to advantage.

Claims

CLAIMS:

1. An analogue filter having an input/output transfer characteristic in the time domain defined by having a constant portion at a first level and a constant portion at a second level, and a substantially quadratic transfer function therebetween.

2. A filter according to claim 1, wherein the first level is substantially zero.

3. A filter according to claim 2, wherein said first level is defined by a portion of an analogue input signal to the filter which is within a predetermined vicinity of the zero crossings of said input signal.

4. A filter according to claim 3, wherein said substantially zero first level is obtained by providing a first clipping circuit (12) for clipping the input signal at a predetermined level outside of a predetermined vicinity of the zero crossings of the input: signal, and then subtracting the output signal from the first clipper circuit (12) from the i-nput signal, to provide an input signal having a dead-zone in the vicinity of the zero crossings- .

5. A filter according to claim 4, wherein the size of said predetermined vicinity is adjustable.

6. A filter according to claim 5, wherein said second constant level is obtained by providing a second clipper circuit (14) for receiving said input signal having the dead-zone and clipping the input signal at a non-zero level.

7. A filter according to claim 6, wherein said non-zero level is adjustable.

8. A filter according to claim 6 or claim 7, wherein the output signal from the second clipper circuit (14) is applied to a rectifier (16).

9. An optical scanning device comprising an amplifer for providing a gain path in respect of an input signal, and a filter according to any one of claims 1 to S, wherein an error signal is applied to the filter, and an output of said filter is applied to the amplifier.

10. An optical scanning device according to claim 9, wherein the amplifer has a substantially constant gain.

11. An optical scanning device according to claim 9 or claim 10, wherein the analogue input signal applied to the analogue non-linear dynamic filter is a osition error signal, and the output of the filter is multiplied with the position error signal to realise a variable gain transfer function.

12. An optical data storage system including an optical scanning device according to claim 9 or claim 10.