GB2479359A - Virtual feedback circuit arrangement for ambient noise-cancelling (ANC) earphones - Google Patents

Abstract

An electronic circuit for an ambient noise-cancelling (ANC) earphone 1 incorporates the analogous electrical circuit of a particular, real acoustical system so as to mimic the properties of the acoustical system in an electrical feedback network, thereby deriving electrical signals from the mimic circuit which can also be used to drive, in parallel, the real acoustical system in order to provide a given transfer characteristic to any chosen part of the acoustical system. The electronic circuit comprises: an amplifier 50 for coupling noise-cancelling signals to an electro-acoustic transducer, such as a microspeaker 24, inside the earphone 1; and an electrical impedance connected in negative feedback path over the amplifier 50, where the value of the impedance is influenced by an electrical counterpart of the analogous electrical circuit. Consequently, the resultant noise-cancelling signal at the listener's eardrum (5) is snatched precisely to an incoming ambient noise signal at the same position.

Description

Virtual feedback circuit arrangement The present invention relates to an electronic circuit arrangement for application to ambient noise-cancelling earphones and, in particular, to such an arrangement for use with ear-bud type earphones. The arrangement is especially suitable for use with ambient noise-cancelling ("ANC") earphones utilising the feed-forward principle, in which the intention is to detect ambient noise that is on its way to the ear-drum of an earphone user, to generate electrical signals representing the detected noise, to electronically invert the electrical signals, thereby generating an electronic cancellation signal which is then converted into a noise-cancelling acoustic waveform intended to arrive at the eardrum at the same time as the detected ambient noise and (because of the electronic inversion) to destructively interfere with it. A feed-forward ANC arrangement is disclosed, for example, in US-A-5, 138,664.

It will be appreciated that ANC is a term of art, and is not used herein to imply complete cancellation of ambient noise; merely that the ambient noise is significantly reduced from the level at which it would otherwise be perceived by the listener.

In order to efficiently achieve the desired destructive interference, it is necessary to ensure that the noise-cancelling acoustic waveform arrives at the eardrum position without significant amplitude or phase changes, at least over the frequency range 50 Hz to 2 kHz or thereabouts. Indeed, in order to achieve a reduction in perceived noise of 20 dB or greater, the amplitudes of the two signals must be matched to within about ±0.9 dB, and, simultaneously, the relative phase difference between the two signals must be less than ±5°.

Significant difficulties arise in achieving this, however, primarily because of the nature of the complex electro-acoustic system, comprising an earphone structure coupled to a human ear-canal, to be traversed.

The present invention aims to address the above-mentioned difficulties and, according to one embodiment of the invention, there is provided an electronic circuit arrangement for an ambient noise-cancelling earphone containing an electro-acoustic transducer for generating ambient noise-cancelling acoustic signals to be directed in use into the ear canal of a listener wearing the earphone, the arrangement comprising means for developing ambient noise-cancelling electrical signals, and further comprising amplifier means for coupling said noise-cancelling electrical signals to said electro-acoustic transducer, and an electrical impedance connected in a negative feedback path over said amplifier means; wherein the value of said impedance is influenced by an electrical counterpart of at least one acoustical component of said transducer.

Preferably the earphone comprises an ear-bud having at least one acoustical leakage path through which ambient noise can reach the ear of said listener and wherein said negative feedback path is also coupled to ground by way of a further impedance; the value of said further impedance being influenced by an electrical counterpart of said at least one acoustical leakage path.

It is also preferred that the value of said further impedance is influenced by an electrical counterpart to the acoustical parameters of the listener's ear canal.

In preferred embodiments of the invention, the earphone comprises part of a feed-forward ambient noise cancellation system.

In one preferred embodiment of the invention, said transducer comprises a microspeaker, and the circuit arrangement is configured such that it generates, from an electrical input signal indicative of inverted ambient noise, an electrical output signal for driving the microspeaker; the arrangement being such that the resultant acoustic signal at the listener's eardrum is substantially identical to the electrical input signal, both in terms of amplitude and phase, over a predetermined range of frequencies.

In one such preferred embodiment of the present invention, an analogous electrical circuit comprises, in part at least, a model of the acoustic pathway from the microspeaker to the listener's eardrnm.

In another preferred embodiment of the present invention, analogous models of a plurality of acoustic pathways are used in an electronic circuit to implement a complete ambient noise-cancelling signal- processing circuit, thereby enabling accurate and effective noise-cancellation to be achieved with a relatively small number of electronic components.

It will be appreciated from the foregoing that the invention employs the principle that an acoustical circuit can be mathematically modelled, similarly to the way in which electrical circuits and mechanical circuits can be modelled, where sound pressure is the driving variable (analogous to voltage in an electrical circuit, and force in a mechanical circuit), and using circuit elements of acoustic inertance, acoustic compliance and acoustic resistance, analogous to the counterpart electrical properties of inductance, capacitance and resistance, respectively.

In order that the invention may be clearly understood and readily carried into effect, one embodiment thereof will now be described, by way of example only, with reference to the accompanying drawings, of which: Figure 1 shows schematically a typical structure of a feed-forward type ANC ear-bud coupled to an ear canal of a listener; Figure 2 shows graphically a measured ear-canal response of a typical ear-bud type earphone; Figure 3 shows an analogous electrical circuit representative of the acoustic structure of the earphone of Figure 1 coupled the an ear canal; Figure 4 shows, graphically, simulated amplitude and phase responses of an ear-bud type earphone derived from an analogous electrical circuit; Figures 5a and 5b show simplified analogous electrical circuits of the ear-bud coupled to an ear canal; Figure 6a shows an electronic circuit with feedback, using a non-inverting amplifier with gain; Figure 6b shows impedance components whose values are derivable by analysis of an analogous electrical circuit connected into an arrangement in accordance with one example of the invention; Figure 7 shows, graphically, phase and amplitude responses at a listener's eardrum attributable to the arrangement of Figure 6b; Figures 8a and 8b show, in full and simplified forms respectively, analogous electrical circuits representative of the electro -acoustic components and acoustical pathways of a typical ear-bud type earphone; Figure 9 illustrates in block diagrammatic form a complete ANC system based on simplified analogous circuit configurations; and Figure 10 shows the analogous circuit configurations utilised in constructing the system of Figure 9.

By translating an acoustical circuit into an equivalent or "analogous" electrical circuit by converting the acoustical components into their respective electrical counterparts via their fundamental, physical units (mks), which are common to all three systems (acoustical, electrical and mechanical), the inventor has created an electronic circuit arrangement capable of being used (inter alia) with feed-forward ambient noise-cancelling (ANC) earphones. The circuit arrangement is especially useful when applied to ANC earphones utilising thin rubber flanges that seal the outlet conduit of the earphone into the entrance of the listener's ear-canal, and the following embodiment will be described in that context. Such earphones are sometimes referred to as "in-ear" earphones, or "ear-bud type" earphones, and they are now widely used, especially for portable communications and entertainment applications whilst the listener is travelling, including listening to music and, in conjunction with cellular telephone handsets, for hands-free calls and conversations.

Although the thin rubber ear-bud flanges might appear to effectively "seal" the earphone assembly into the listener's ear-canal, an earphone thus positioned and located does not provide an effective acoustic seal between the listener's ear canal and the ambient environment, because low-frequency sound vibrations can still pass through the rubber flanges themselves. In addition, acoustic coupling impedance pathways having known acoustic transmission characteristics are frequently built in to the earphone structures in order to adjust the acoustic performance for a desired frequency response at the listener's ear, as disclosed in US 4,852,177, and these inadvertently provide leakage pathways which allow external sound energy to be transmitted directly through the actual strncture of the earphone and into the ear-canal. Such leakage pathways are often implemented as very small apertures (diameter <1 mm) bearing acoustically resistive nyion mesh material, or similar, situated between the outer ambient and the internal space situated in front of the internal microspeaker or the space behind it, or situated between these two internal spaces themselves (or some combination thereof), and these contribute to the complexity of the acoustic structure of the earphone.

In general, environments for travellers are seldom quiet, and very high levels of ambient noise can be present in air travel, subway trains and motor transport. Consequently, it is a great advantage to incorporate an ANC system into ear-phones and headsets such that the music and communications are intelligible, and such that the listener is not required to increase the listening volume to an excessively high level in order to overcome the background noise; a practice which is most undesirable, for health reasons among others.

In a feed-forward ANC system, as has been mentioned above, incoming ambient-noise acoustic signals are detected and transduced into electrical signals which are processed to create phase-inverted noise cancellation electrical signals which, in turn, are transduced into noise cancellation acoustic signals at the ear of the listener, such that when an ambient-noise acoustic signal and its phase-inverted noise cancellation acoustic counterpart arrive, substantially simultaneously, at the tympanic membrane, destructive interference occurs because the phase-inverted (cancellation) signal is ideally of equal magnitude and opposite polarity to the ambient noise signal, and therefore the resultant, summed signal is ideally zero. In principle, this is an elegant way to create an active noise-cancelling system, but in practice very substantial practical difficulties are involved.

These difficulties are significantly contributed to by the different acoustic and electronic pathways followed by, or imposed upon, the incoming noise and generated (phase-inverted) cancellation signals. This will now be described with reference to Figure 1, which shows in partial cross- section the structure of a typical ANC ear-bud configured for feed-forward ANC.

The ear-bud 1 comprises a body moulding 10, made of plastics material, which defines a first housing compartment 12, a second housing compartment 14 and an outlet port 16. The first housing compartment is divided by a septum 18 into a frontal volume 20 and a rear volume 22; the septum supporting a microspeaker 24 whose front or output surface faces the outlet port 16. A small acoustic coupling port 26 is formed through the septum 1 8 and covered with an acoustic resistance medium 28. The rear volume 22 is coupled to the external ambient air via one or more rear vent apertures 34. The second housing compartment 14 houses a microphone 30, which detects, by way of an inlet aperture 32 formed in the body moulding 10, ambient noise on its way into the ear of a user of the ear-bud, and is formed with one or more rear vents such as 34. The microphone 30 is connected to an electrical cable 36 which conveys electrical signals from the microphone 30 to a processing circuit (not shown) in which the aforementioned inversion and other signal processing to be described takes place. A second electrical cable 38 connects the processed signals to the microspeaker 24 which then converts these processed signals into a noise-cancelling acoustic signal which is projected through the outlet port 16 into the user's ear canal. A rnbber flange 40, comprising the ear-bud seal, is supported on the outer wall of the outlet port 16.

The ear-bud is placed into the entrance of the listener's ear canal 3, the flange 40 providing a comfortable fitment and partial seal, enabling sound from the ear-bud's microspeaker 24 to be projected along the canal 3 to the ear-drum 5.

Typically, the microphone 30 comprises a small electret microphone, typically 4 mm or 6 mm in diameter, orientated outwards, as shown. The electrical flex connections 38 and 36 conveniently link the microspeaker 24 and the microphone 30 electrically to a small "pod" unit (not shown) that houses a battery supply and the electronic processing circuitry.

Generally, the wiring to the microspeaker is hermetically sealed in place with glue, as indicated at 42, so as to acoustically isolate the microspeaker rear volume 22 from tl1e microphone housing compartment 14.

As mentioned above, it is common to introduce one or more defined acoustic leakages, such as 26/2 8, in order to rnodifi the frequency response to provide high-quality sound reproduction. Such leakages typically are formed as acoustic resistors, made by sealing a thin, acoustically resistant nylon mesh (or similar) 28 over a small diameter (< 1 mm), short length (< 1 mm) aperture 26 in the housing. It is beneficial to deploy such a resistance either between the front volume 20 and the ambient, or between the front and rear volumes 20, 22. This is also useful for preventing a total hermetic seal of the earphone in the ear of the user, which otherwise can cause an unpleasant "blocked ear" feeling. In practice, only one of these acoustic couples is required. For all of the examples used herein, a front-volume to rear-volume resistive acoustic couple has been chosen, as shown. However, the invention is equally applicable to any configuration of one or more acoustic couples.

These resistive impedances are very critical components, however, and even small changes in value have a significant influence on the frequency response and overall transfer function of the earphone.

In any event, in this constrnction, ambient noise signals travel from the ambient surroundings through the earphone leakages (such as those associated with the flange 40, the rear vents such as 34, the acoustic coupling port 26, and the diaphragm of the microspeaker 24) to the ear-canal and thence to the ear-drum. This complex route followed by the ambient noise from ambient to ear is, of course exclusively an acoustic pathway and it contrasts significantly with the pathway encountered by the counterpart noise-cancelling signal, which involves acoustical-to-electrical transduction by microphone 30; electronic signal processing; electrical-to-acoustic transduction by microspeaker 24 and acoustic transmission into the frontal volume 20 of the ear-bud and thence via the ear-canal to the eardrum; this latter part of the pathway, from the microspeaker 24 to the ear-drum 5, being sometimes called the driver-to-ear or transducer-to-ear path.

As will be appreciated, these differing acoustic and acoustic/electronic pathways for the ambient noise and its noise-cancelling counterpart have very different transfer characteristics, which significantly modify the amplitude and phase cl1aracteristics of the respective signals, disturbing the match between the two. This disturbance requires precise compensation if effective cancellation of the ambient noise signal is to be achieved.

Figure 2 shows a typical frequency-dependent amplitude response of an ear-bud type earphone, measured on an ear-canal simulator comprising a tube having the same average dimensions as a human ear canal (approximately 7.5 mm diameter and 22 mm in length), with an entrance aperture suitable for accepting an ear-bud fitment, and with the end remote from the earphone terminated with a microphone located approximately at the ear-drnm position. This arrangement generates signals that are representative of the sound pressure level at the ear-drum.

As can be seen, the response is by no means uniform and flat, and the associated phase response also varies widely. It will be appreciated that, in order to achieve a flat transfer function from the microspeaker to the ear-drum position, a considerable amount of complex filtering would be required. This is additionally complicated by the simultaneous requirement to provide a flat, zero-phase response.

Accordingly the inventor has realised that, if a supplementary active electronic circuit can be used to drive the electrical mimic of an acoustic network so as to create a desired electrical response at the ear-canal node (for example) in the mimic circuit, then that same electronic circuit output signal can be applied to the real acoustic network itself to create exactly the same desired acoustic response in the ear-canal.

The acoustic structure of the earphone can be analyzed as a lumped-element model. This is valid up to frequencies where its physical dimensions approach about 1/6 X which, for the earphone (say 10 mm) is about 5.7 kHz. The microspeaker itself, as is known in the prior art, can be represented in simplified form by a serial L-C-R network, where the three network elements correspond respectively to the acoustic mass (primarily of the voice-coil and diaphragm), the acoustic compliance of the diaphragm, and the acoustic resistance of the diaphragm suspension.

The values of these parameters can be obtained by well-known Thiele-Small analysis methods, and converted into electrical component equivalents, such that acoustic mass, or inertance (MA), is represented by an electrical inductance (L); acoustic compliance (CA) is represented by a capacitance (C); and acoustic resistance (RA) is represented by an electrical resistance (R), as described by Beranek in Acoustics (1993 edition) American Institute of Physics, New York (1996), ISBN 0-883 18- 494-X. In addition, the structural elements of the ear-phone body (front-and rear-volumes 20, 22; front-rear resistive couple 26, 28; outlet-port 16 and so on) can also be considered as individual acoustical components, enabling their properties to be calculated from first principles, as can the ear-canal, too, in a simplified form.

For example, the acoustic compliance, CA, of a volume of air, V, is given by the formula: C4 = (1) P0 where y is the adiabatic ratio of the specific heat capacities of air (-4.4 for a diatomic gas at STP), and Po is the static pressure (i0 N.m2).

Consequently, the ear-canal, having a volume of about 795 mm3, has an -12 5 -1 acoustic compliance of 5.68 (10) m.N Similarly, the acoustic mass, MA, of the air in the outlet port 16, for example, can be calculated using the formula: M = = kg.rn4 (2) A s7 s where MM is the (mechanical) mass of the air in the port, S and 1 represent the cross-sectional area and length of the port, respectively, and p is the density of air at STP. The port is essentially a cylinder and so, if the internal diameter is 3.2 rrim, and the length is 6 mm, the acoustical mass is 8.80 (102) kg.m4.

Figure 3 shows an analogous electrical circuit of the significant elements of the acoustic structure of a typical earphone, from which the measured response of Figure 2 was derived, coupled to an ear-canal. The serial L-C-R network that represents the microspeaker 24, as noted above, is shown as Li, Ci, Ri and R2, where Ri corresponds to the intrinsic resistance of the diaphragm suspension and R2 corresponds to the extrinsic, or damping, resistance attached to the frame (usually as a resistive mesh patch overlying the apertures behind the rear of the diaphragm).

With further reference to Figure 3, the microspeaker 24 is represented by the pressure source, Vi, and the serial L-C-R network, as described above, in parallel with a front-rear couple (L2, R3), of which the inductive component L2 is associated with the small aperture 26 of the resistor structure. The front of the microspeaker' s diaphragm is necessarily coupled to the front volume 20 of the earphone (compliance C3), and thence via its outlet port i 6, having inductance L4, into the ear-canal compliance C4. The rear of the microspeaker 24 is coupled to its rear volume 22, having compliance C2, which is vented directly to the outside ambient via several relatively large apertures such as 34, typically of i mm diameter, through the thickness of the casing (about 0.5 mm).

The rear vents 34 have only a very small inductance, L3 (i mH), which in practice is negligible, and couple to the external ambient, shown as R4.

Strictly, the latter should be a frequency-dependent impedance to represent radiation impedance in free space, but it is negligible and can be assumed to be zero for the current purposes.

Table 1 lists the analogous circuit components representing the acoustic elements of the earphone and ear-canal, and their references with respect to Figure 3.

Component Physical Parameter Electrical Reference Value inductor Li microspeaker: diaphragm & 157 mH voice coil mass capacitor Cl microspeaker: diaphragm 0.69 tF compliance resistor Ri microspeaker: suspension 28) resistance resistor R2 microspeaker: damping 2000 «= resistance inductor L2 front-rear couple: acoustic mass 52 inH resistor R3 front-rear couple: acoustic i 000 «= resistance capacitor C2 earphone rear volume: 0.2i pF compliance inductor L3 earphone rear vent: acoustic i mH mass resistor R4 radiation impedance i «= (negligible) capacitor C3 earphone front volume: 0. ii pF compliance inductor L4 front outlet port: acoustic mass 20 inH capacitor C4 ear-canal volume compliance 0.52 pF voltage source pressure source i V vi When the microspeaker parameters have been analysed and evaluated, and other parameters calculated from first principles (acoustic masses and acoustic compliances), the remaining parameters can be estimated by successive fitting. Figure 4 shows simulated responses of both the frequency-dependent amplitude and phase characteristics of the analogous earphone/ear-canal circuit of Figure 3. The simulated amplitude response matches well the real earphone frequency response measurement of Figure 2, thus confirming the validity of the model, and confirming that the analogous circuit of Figure 3 is a direct electrical mimic of the acoustic and electro-acoustic network of a particular earphone coupled to a human ear-canal.

For ambient noise-cancellation applications, operating only within a predetermined and relatively restricted frequency range (typically 20 Hz to 3 kHz), the analogous circuit of Figure 3 can be simplified considerably, as follows, and as shown in Figure 5a. In these circumstances, the impedances of the inductors are negligible, and hence they can be removed (i.e. replaced in each case by a zero impedance link). Moreover, because the rear vents 34 connect directly to free acoustic space (equivalent to a dissipative electrical "ground"), and have negligible impedance, the rear terminal of the microspeaker 24 can be considered to be connected to ground. Consequently, the capacitive element C2 associated with the rear volume element 22 is short-circuited, and therefore can be removed. In addition, the front-rear couple (now just R3, associated with acoustic resistance 28), being coupled to the rear volume, also becomes effectively connected to ground. The serial, resistive elements of the microspeaker, Ri and R2 can be lumped together (R5). Finally, now that the frontal outlet-port inductance has been replaced by a zero impedance link, this connects the capacitance C3 associated with the front volume 20 to the capacitance C4 associated with the ear-canal volume, in parallel, and so the two can be lumped together and represented by one single capacitance (acoustical compliance), C5.

For clarity of explanation, Figure 5a can be rearranged and re-drawn as shown in Figure 5b (wherein the suffix "B" is used to identify components that are identical to those described above in relation to Figure 5a). This shows that the initial analogous circuit of Figure 3 has been simplified to a ground-referenced voltage source, V1B, driving through a serial R-C network (Cl B, R5B) and then through a parallel R-C combination (R3B, C5B) down to ground, where a connecting node NiB represents the voltage at the ear-canal/front-volume capacitor, C5B, in the analogous network. This voltage corresponds to the sound pressure in the ear-canal in the corresponding real acoustic network.

It can be seen that the pressure (voltage) source, Vi, is now ground-referenced, and is no longer a "floating", differential arrangement. This allows Vi to be configured as the output node of an active electronic component such as an operational amplifier, such that a supplementary active electronic circuit arrangement can be built around the simplified analogous circuit of the microspeaker.

The inventor has further realised that, if a supplementary active electronic circuit can be used to drive the electrical mimic of an acoustic network so as to create a desired electrical response at the ear-canal node (for example) in the mimic circuit, that same electronic circuit output signal can be applied to the real acoustic network itself to create exactly the same desired acoustic response in the ear-canal itself It will be recalled that ambient noise-cancellation requires a signal having very precisely defined amplitude and phase characteristics to be transferred into an earphone structure coupled to a human ear-canal, such that the cancellation signal arrives at the eardrnm position without significant amplitude or phase changes over the frequency range 50 Hz to 2 kHz, or thereabouts. However, Figures 2 and 4 clearly show that, when the earphone itself is coupled to a constant amplitude signal source, the frequency-dependent amplitude and phase signals at the ear-drum differ significantly from the source signal.

This example of the present invention achieves the desired signal transfer to the ear-canal and ear-drum by incorporating the analogous network as part of a feedback loop, using an operational amplifier. The principle is illustrated using a simple resistor network in Figure 6a, which shows an operational amplifier 50 in a "non-inverting amplifier with gain" configuration, where the signal input Vm is driven into the amplifier's non-inverting input, and a feedback resistor, RB, connects the amplifier's output to its inverting input, which is also connected to ground via resistor RA. When a positive voltage is applied to the non-inverting input, the output voltage increases until the resultant voltage on the inverting input is equal to that on the non-inverting input.

This property (that the amplifier drives its inverting input to substantially the sanie voltage as the input signal present on its non-inverting input) is an irriportant factor in this implementation of the present invention.

The voltage at the inverting input node, VN, is equal to the input signal, Viii, and is related to the output voltage, V0, by the potentiometric division afforded by RA and RB. The gain of the system is given by the following known relationship, which also applies for complex impedances as well as purely resistive components.

Gciin = = 1+-(3)

R

This circuit type can be combined with the simplified analogous circuit of Figure 5b, as shown in block form in Figure 6b. The inverting input of the operational amplifier 50 is connected to the node joining the ear-canal components (represented by impedance ZA, connected to ground) to the microspeaker and some earphone components (i.e. node NiB in Figure 5b), and its output is connected to drive the microspeaker components (ZB). Consequently, the amplifier feedback circuit ensures that the voltage on the inicrospeaker-to-ear-canal node closely tracks the input voltage V. In addition to the analogous circuit, however, the amplifier output signal can also be used to drive a real earphone in an ear-canal in parallel and, because the analogous circuit is a mimic of the real electroacoustic network, the acoustic signal at the ear-canal also tracks the electrical input signal exactly, both in terms of amplitude and phase.

Figure 6b shows such a circuit arrangement driving an earphone in an ear-canal simulator in order to measure the transfer function from the electrical input signal, V1, to the ear-canal microphone signal, V0, and test the effectiveness of the invention. The result is shown in Figure 7, which shows flat amplitude and phase responses up to several hundred Hz, in contrast to the highly non-linear responses of Figures 2 and 4.

Because of the somewhat extensive circuit simplification process in the example described above, especially the entire removal of the inductive components, the analogous circuit example described above is not valid in the higher frequency ranges where the inductive reactances become significant. However this is acceptable for application in ambient noise-cancellation, which is not required to operate above several kHz in earphone applications. By matching the analogous circuit more closely to the real electroacoustic network that it mimics, however, the frequency range can be extended considerably, even beyond 10 kHz. Similarly, by minimising the acoustic masses of the acoustic components, a better match is achieved to a simplified, electrical, inductor-free mimic circuit.

This circuit arrangement can be usefully extended to incorporate additional analogous circuitry. In particular, an "ambient-to-ear" transfer function can be added to the mimic circuit to provide a complete ambient noise-cancelling processor using only a single operational amplifier. The ambient-to-ear pathway can be followed from left to right through the individual components in the analogous circuit of Figure 3, with the incoming ambient signal passing in through the rear vents, L3, to the rear volume C2, and then passing through the microspeaker and front-rear couple, in parallel, to the front volume C3, and via the outlet port L4 to the ear-canal C4. This arrangement is shown in Figure 8a, with the driving pressure source V1 now coupling in via the rear vents, L3, rather than being an integral part of the microspeaker. This circuit can be simplified in exactly the same way that the "driver-to-ear" circuit of Figure 3 was reduced to the circuit of Figure 5a, by elimination of the inductors and so on. Figure 8b shows a simplified version of Figure 8a: an analogous circuit of the acoustic ambient-to-ear path, using only four components.

Feed-forward ambient noise-cancellation requires that, at the ear-drum, the acoustic cancellation signal is an inverted, exact replica of the acoustic ambient-to-ear signal at the same position. Accordingly, noting that the circuit of Figure 6 can transfer a signal to the ear-drum without amplitude or phase error, it will be appreciated that, if this circuit can be driven with a signal that represents the signal, at the ear-canal, of the ambient-to-ear path, then the resultant circuit arrangement will fulfil the entire ANC processing requirement.

The circuit of Figure 8b generates exactly this required signal at node NiB, and so by incorporating this simplified analogous network of the ambient-to-ear function at the non-inverting input of the amplifier, the resultant circuit implements both driver-to-ear compensation and also ambient-to-ear transfer function. The non-inverting input of the amplifier is a high-impedance port, and so it does not load the passive components.

A complete ambient noise-cancellation system based on this implementation of the invention is shown in block form in Figure 9, in which the output from an external microphone, on the shell of the earphone, is fed via an adjustable-gain, inverting, buffer amplifier, via the ambient-to-ear network of Figure 8b to the non-inverting input of the amplifier, configured as above, for copying the signal to the ear-canal without amplitude or phase distortion. These electronic components (doubled to form a stereo pair if required), together with additionally required elements, such as stabilised power-supply devices, can be implemented as a single integrated-circuit to provide a low-cost ANC earphone processor.

Figure 10 shows a schematic diagram version of the complete ANC circuit of Figure 9. (The component numbering is consistent with previous figures, in that the component affix "B" indicates a second occurrence of the same component in the circuit.) In practical terms, the basic circuit is not intrinsically DC stable because the feedback path impedance (ZB in Figure 6b) contains a serial capacitor Cl B (representing the microspeaker diaphragm compliance), and so the DC gain (equation 3) is not defined. This can be remedied by the inclusion of a high-value stabilising resistor, R(stab), in parallel with said capacitor (Ci B), as shown in Figure 10. This additional resistor, together with R3B (the front-rear couple resistance), determines the DC gain, and stabilises the circuit, although the inclusion of R(stab) does modify the transfer function of the circuit slightly at low-frequencies and so its value is preferably kept high (typically 200 k) or greater), and hence the use of an amplifier with a low input-offset voltage specification (say, <1 mV) is desirable in order to minimise any consequent output offset voltage.

Additionally, in practice it is likely that both the microphone and the microspeaker will be AC coupled to the ANC circuitry via serial capacitors, and this would introduce electronic, low-cut filtering, the effects of which can be taken account of during the design of the circuitry. Also, because the noise-cancellation is only required over a restricted range of frequencies, it is advantageous to include both high-cut and low-cut filtering into the system. Electronic filtering to incorporate these features can be included in the circuits of Figures 9 and 10, as will be appreciated by those skilled in the art, either as part of the existing circuit paths, or as additional circuitry.

Claims (12)

1. CLAIMS: 1. An electronic circuit arrangement for an ambient noise-cancelling earphone containing an electro-acoustic transducer for generating ambient noise-cancelling acoustic signals to be directed in use into the ear canal of a listener wearing the earphone, the arrangement comprising means for developing ambient noise-cancelling electrical signals, and further comprising amplifier means for coupling said noise-cancelling electrical signals to said electro-acoustic transducer, and an electrical impedance connected in a negative feedback path over said amplifier means; wherein the value of said impedance is influenced by an electrical counterpart of at least one acoustical component of said transducer.
2. 2. A circuit arrangement according to claim 1, wherein the earphone comprises an ear-bud having at least one acoustical leakage path through which ambient noise can reach the ear of said listener and wherein said negative feedback path is also coupled to ground by way of a further impedance; the value of said further impedance being influenced by an electrical counterpart of said at least one acoustical leakage path.
3. 3. A circuit arrangement according to claim 2, wherein the value of said further impedance is influenced by an electrical counterpart to the acoustical parameters of the listener's ear canal.
4. 4. A circuit arrangement according to any preceding claim, wherein the earphone comprises part of a feed-forward ambient noise cancellation system.
5. 5. A circuit arrangement according to any preceding claim, wherein said transducer comprises a microspeaker, and the circuit arrangement is configured such that it generates, from an electrical input signal indicative of inverted ambient noise, an electrical output signal for driving the microspeaker; the arrangement being such that the resultant acoustic signal at the listener's eardrum is substantially identical to the electrical input signal, both in terms of amplitude and phase, over a predetermined range of frequencies.
6. 6. A circuit arrangement according to any preceding claim, wherein said first-mentioned impedance value is derived from an analogous electrical circuit representing, in part at least, a model of the acoustic parameters of said transducer.
7. 7. A circuit arrangement according to any preceding claim, wherein impedance values derived from analogous models representing a plurality of acoustic pathways are used to implement a complete ambient noise-cancelling signal-processing circuit, thereby enabling a desired degree of noise-cancellation to be achieved with a relatively small number of electronic components.
8. 8. A circuit arrangement according to any preceding claim wherein said amplifier means comprises an operational amplifier.
9. 9. A circuit arrangement according to any preceding claim, configured at least in part as an integrated circuit.
10. 10. A circuit arrangement according to any preceding claim configured for use in a headset for a cellular phone.
11. 11. A circuit arrangement according to any preceding claim configured for use in a cellular phone handset.
12. 12. An electronic circuit arrangement substantially as herein described with reference to the accompanying drawings.