GB2069280A - Process of testing for a sound control system - Google Patents

Abstract

In active sound control systems, sounds are generated to cancel noise, and correct sound generation is achieved by generating signals representative of the noise, passing the signals through a filter and generating sounds from the filter output. Difficulty occurs in determining the characteristic for the filter but the process of the invention provides an effective method. A computer 21 is used to provide signals for a loudspeaker N positioned in the path of noise to be cancelled. Signals are recorded by the computer 21 which are representative of noise received at microphones d and p. The noise received is then regenerated by a speaker S and signals representing the noise received at the microphones d and p are again recorded. The various signals recorded provide the values of parameters in an expression identifying the required filter, and if the filter is a digital filter its coefficients can be derived from the recorded signals. <IMAGE>

Description

SPECIFICATION Process of testing for a sound control system The present invention relates to a process of testing applicable to the manufacture of a sound control system of the type specified below.

The sound control system of the type specified aims to reduce noise in a certain region by providing a sound source which generates pressure variations tending to cancel pressure variations in the region due to noise and therefore to quieten the region.

In this specification a sound control system of the type specified comprises sound receiving means for receiving sounds from a first point or points at a second point or points and deriving signals representative of the sounds received, a means for generating sounds, at one or more third points, which are dependent on the said signals, the sound generated at the third point or points being such that it tends to cancel, at a fourth point or points, the sound received from the first point or points.

An embodiment of a known sound control system of the type specified is illustrated in Figure 1 where a noise source N depicted for convenience by a loudspeaker transmits noise along a duct 10. The sound control system comprising a microphone d connected by way of a filter 11 and a power amplifier 12 to a loudspeaker s generates sounds dependent on noise from the source N with the aim of cancelling noise at a point p.

A major problem with a noise control system of the type specified is the difficulty of providing a suitable transfer function between the second and third point or points, for example between the microphone d and the loudspeakers.

According to a first aspect of the present invention there is provided a process of testing applicable to the manufacture of a sound control system of the type specified comprising transmitting test sounds over a frequency range of interest from the first point or points, simultaneously receiving first resultant sounds at the second point or points and generating first resultant signals representative thereof, transmitting the same test sounds from the first point or points, simultaneously receiving second resultant sounds at the fourth point or points and generating second resultant signals representative thereof, transmitting from the third point or points sounds representative of the first resultant signals and simultaneously receiving third resultant sounds at the fourth point or points, transmitting from the third point or points sounds representative of the second resultant signals and simultaneously receiving fourth resultant sounds at the second point or points, and subtracting signals representative of the fourth resultant sounds from signals representative of the third resultant sounds.

Advantageously, transmitting test sounds from the first point or points and receiving the test sounds at the second point or points is carried out simultaneously with receiving these sounds at the fourth point or points.

The above process of testing is carried out with the object of determining an optimum transfer function Td which is the transfer function between the second and third point or points. This transfer function is derived as follows: If Ta is any transfer function between the second and third point or points, then S=TaD D where S is sound transmitted from s and D is sound received at d, and assuming, referring to Figure 1 for example, an overall flat characteristic of unity gain for the microphone d, the amplifier 12 and the speakers over the frequency range of interest.

When noise N from the source N is present, the sound at d is D = Ads S + Adn N (where Ads is the transfer function from s to d and Adn is the transfer function from N to d) and the sound at p is P=ApsS+ApnN N (where Aphis the transfer function from s to p and Apn is the transfer function from N to p).

From equations 1, 2 and 3 p, ApsTaAdnN +ApnN.

1 - Toads Without cancelling P = Apn N and thus Noise reduction = P (with cancelling) = 1 + TaApsAdn/Apn P (without cancelling) 1 - TaAds = 1 - Ta (AdsApnApsAdn)fApn 1 - TaAds and if Td = Apn ---------------------------------------- Equation 5 Ad5Apn - ApaAdn 1 - Ta/Td Noise reduction = 1 - Ta/Td 1 - TaAds That is if T5 = Td, the point p will be in silence and hence the required value for Td is as given in equation 5.

In addition to its basis in the transfer function Td the present invention springs from the reaiisation that the functions which make up Td can be measured by the process of testing of the present invention.

It is therefore an advantage of the present invention that the difficulty of providing filters such as the filter 11 shown in Figure 1 is largely overcome since the present invention allows a filter to be set up using parameters derived from quantities measured in the process of testing.

For example by recording the signal received at the fourth point or points when the test sounds are transmitted from the first point or points, and recording the difference obtained by the said subtraction of signals, two functions are obtained which are representative of the output and input, respectively, of the required filter. Known techniques can then be used to design the filter 11 in analogue form. However this technique is complicated since for satisfactory results a high order filter of say fifteen to twenty stages is required. Thus it is better to use other known techniques to set up a digital filter. Such other techniques includes the steps of cross correlation and auto correlation to derive filter coefficients. These techniques are discussed later.

Thus according to the second aspect of the present invention there is provided a method of making a sound control system of the type specified wherein the means for receiving sounds and deriving signals representative of the sounds received comprises a first transducer for generating electrical signals representative of sounds received, and a digital filter, the means for generating sounds comprises a second transducer for generating sounds derived from the signals at the filter output, the method including carrying out the process of testing according to the first aspect of the invention, deriving coefficient values for the digital filter from the second resultant signals, and the said difference obtained by the said subtraction of signals, and setting the filter coefficients in accordance with the values derived.

In the process of testing according to the first aspect of the present invention white noise is the best form for test sounds transmitted from the first point or points but this cannot be achieved simply by connecting a white noise generator to a loudspeaker in view of the fall in frequency response of the loudspeaker at low frequencies. For this reason a pre-emphasis filter is preferably used to compensate for the loudspeaker characteristic.

In operation and testing of the arrangement shown in Figure 1 sounds reach the point p by means of plain waves only since those frequencies which would propagate by reflection from the duct walls are not transmitted from the first point. Thus in this case there is only one component of noise from the source n reaching the point p and it is said to be a single degree of freedom system. However in other situations a point at which sound is to be cancelled may receive sounds by several paths, for example directly from the noise source or by way of a reflecting surface. In this case noise received can be said to have several components. Additionally, if the sound travels in three dimensions, the noise source may not be a simple source but may be itself composed of many components.One way of distinguishing these components is by representing them as the multipole moments of the source i.e. as monopole, dipole, quadrupole components etc. Multipole moments of a source are described in A.J. Kempton's PhD thesis (1977 University of Cambridge)"The Identification of Efficient Sources of Aerodynamic Sound". In general an infinite number of components exist, but often only the first few are significant. For example if only the monopole and dipole are considered significant then the source is said to have two components. Assuming that the sound arrives at the point where noise is to be reduced from the source by direct path, the noise received can be said to consist of two components. Clearly if a reflecting surface is also included in the system a further two components also occur. Thus it can be seen that sound systems having many significant components usually occur in practice. The present invention can be applied to such systems by considering each component separately and setting up a filter for a sound control system specific to that component. For example where sound has two components two systems including two filters are required.

In general, when the sound system is composed of two or more components it is not always possible to measure each component with a single microphone, instead the weighted mean of a few microphone signals can be used. Similarly the cancelling sources may have to be composed of more than one loudspeaker. Each microphone is then located at a respective one of the second points and each loudspeaker at a respective one of the third points.

Certain embodiments of the invention will now be described by way of example with reference to the accompanying drawings, in which: Figure 1 is a block diagram of a prior art sound control system, Figure 2 is a block diagram of a sound-control system according to the second aspect of the present invention, Figure 3 is a block diagram of a test arrangement used in carrying out a process according to the first aspect of the invention, Figures 4a and 4b show a flour diagram of a process according to the first aspect of the invention, Figures 5a and 5b illustrate the transfer function of a pre-emphasis filter used in the arrangement of Figure 3, Figures 6a and 6b are block diagrams of a digital filter used in Figures 2 and 3, Figures 7a and 7b illustrate the transfer function of a filter used in the arrangement of Figure 2, and Figure 8 is a block diagram showing in more detail circuits used in a typical implementation of the filter of Figures 6a and 6b.

Where the sound control system required is to be of the general type shown in Figure 1 but is to employ a digital filter then the arrangement of Figure 2 may be used. The microphone d is connected by way of a pre-amplifier 13to an anti-aliasing filter 14. Aliasing is a phenomenon which occurs in systems employing sampling techniques. A signal which is applied to a sampling device at a frequency which is more than half the sampling frequency gives sampled outputs which suggest that a low frequency, or alias signal, is present. The anti-aliasing filter 14 is a low pass filter which has a cut-off at about 500 Hz so that signals having frequencies above this frequency are severely attenuated.The output of the anti-aliasing filter is applied to an anlogue to digital converter 15 which samples incoming signals at a rate of 1,000 samples per second and it will thus be seen that the anti-aliasing filter 14 has a cut-off at half the sampling frequency.

Samples from the analogue-to-digital converter 15 are applied to a digital filter 16 which may be either in hardware form or may be in the form of a computer programmed to act as a digital filter.

Signals from the digital filter 16 pass to a digital-to-analogue converter 17 and then to an anti-aliasing filter 18 which has the same characteristic as the filter 14. The filter 18 "smooths" the samples from the converter 17 so that high frequency signals present in the stepped output of the converter 17 are removed. It is necessary to remove these signals since the digital filter 16 and the remainder of the system are not designed to cope with signals above 500 Hz and such signals could cause unpleasant effects at the point p.

Signals from the anti-aliasing filter 18 are amplified in a power amplifier 19 and passed to the loudspeaker s.

The process of the present invention is used to set up the digital filter 16 by providing coefficients for this filter. The process will be described using the test arrangement shown in Figure 3. In this test arrangement signals representing noise are generated by a computer 21 and passed by way of a transmitting arrangement 22 which comprises component circuits which are the same as those within a dashed line 22' of Figure 2, that is a digital-to-analogue converter, an anti-aliasing filter and a power amplifier. Signals received by the microphone d are passed through a receiving system 23 comprising component circuits which are the same as those shown within a dashed line 23' of Figure 2, that is a pre-amplifier, an anti-aliasing filter and an analogue-to-digital converter.In addition the computer 21 provides signals for a further transmitting system 22' which is identical with the system 22 and a microphone is provided at the point p and connected by way of receiving system 23' which is identical to the receiving system 23.

In using the test arrangement of Figure 3 the computer 21 is programmed according to a flowchart shown in Figures 4A and 4b. A series of random numbers is generated in an operation 25, these numbers specifying white noise when passed to the analogue-to-digital converter in the transmitting system 22. However in order to pre-emphasize the noise generated, the random numbers are processed in an operation 26 which shapes the spectrum of the noise produced so that it compensates for the response of the loudspeaker N.

The process 26 of shaping the spectrum is carried out using a digital filter, that is using the computer 21 to act as a digital filter. Digital filters are described in the book "Digital Filters: Analysis and Design" by Andreas Antoniou, published by McGraw Hill, 1979. Asuitablefilter is described below in connection with Figures 6a and 6b.

It is well known that computers can be employed as digital filters and suitable programming is a routine matter and is therefore not described in the specification. A typical filter characteristic for use in shaping the spectrum is shown in Figures 5a (gain) and 5b (phase) but it will be realised that the filter used in practice will be constructed after measuring the characteristic of the loudspeaker N, the filter being designed to provide equal powers at almost all frequencies in the frequency range considered.

The digital output from the computer 21 is passed in an operation 27 to the transmitting system 22 and as a result the loudspeaker N generates a sound which is received by the microphones d and p. Signals from the microphone d are processed by the receiving system 23 and as a result digital signals are input to the computer 21 in an operation 28. These signals are then stored in an operation 29.

Simultaneously signals from the microphone p are converted into digital input signals in an operation 31 and stored in an operation 32.

Using the convention specified above the store 29 now stores a series of numbers representing a product AdnN and the store 32 stores a series of numbers representing the product ApnN.

The next steps in the testing procedure are first to replay the signals stored in operation 29 and then replay those stored in operation 32. Thus in operations 33 and 35 signals representing AdnN are output through the transmitting system 22' to the speaker 2 and received by way of the receiving system 23' and then stored in an operation 36. When this step has been carried out signals representing ApnN are output to the speakers in operation 37 and received by way of the microphone d and the receiving system 23 in an operation 38.

The information stored in operation 26 is a series of numbers representing ApsAdnN and the output from operation 38 is a series of numbers representing AdsApnN and thus an operation 39 provides output signals representative of the denominator of the required transfer function Td. The numerator of this function is available from the signals stored in operation 32. After carrying out the operations illustrated in the flow diagram of Figure 4a information for calculating the transfer function Td is available.

The method described above referring to Figures 3 and 4a assumes that the transmitting and receiving systems 22, 22', 23 and 23' are of unity gain over the frequency band of interest. The characteristic of the transmitting system 22 is not critical since the exact form of sound produced by the speaker N is not important; what is needed is a test sound to drive the overall system. The characteristic of the receiving system 23' is also not critical because, in operation, no significant noise should be received at p.

While a practical working system can be achieved by calculating the transfer function Td and connecting a circuit with this function between the microphone d and the speakers, the results obtained depend on the approach of the receiving system 23 and the transmitting system 22' to unity gain. An improvement in performance is achieved by regarding the points N, d, s and p as being the appropriate terminals of the computer 21. Hence the circuit with the transfer function Td should be connected between the output terminals of the receiving system 23 and the input terminals of the transmitting system 22' (as shown in Figure 2), unless allowance is made for the characteristics of these systems.

Instead of working out the transfer function it is usually advantageous to carry out the operations given in the flow chart of Figure 4b by employing the computer 21 to generate coefficients for the digital filter 16 of Figure 2.

Signals from the subtract operation 39 and signals corresponding to those stored in operation 32 are each auto-correlated and then cross correlated in an operation 41 to determine the matrix of correlations and a vector. The matrix is inverted in operation 42 and the vector is multiplied by the inverse in operation 43 to provide a set of filter coefficients.

The process of obtaining the filter coefficients is known as "system identification" and a number of suitable methods is given in the paper by ström, K.J. and Eykhoff, P. in "System Identification - A Survey", Automatica, Volume 7, pages 123 to 162,1971.

Following from this paper the determination of the filter coefficients is now briefly described.

If ApnN = y(k) and (ApsAdn - AdsApn)N = u (k), where 1 S k S M autocorrelation of u(k) and y(k) is given by

respectively; and cross-correlation of u(k) and y(k) is given by

where i varies from 0 to n.

If these correlations are written as a symmetric matrix

and also as a vector in the form of a column matrix

then by the "least squares" system identification method the coefficients required p where p = [a1, a2, a3 an, bo, bl, bn]T are given by p = M-1 C, where M- is the inverse of the matrix M and [ .. ]T is the transpose of [ .. ].

These coefficients can be used to set a hardware digital filter or to program the computer to act as a digital filter. In either case the digital filter can be of the type shown in Figures 6a and 6b (these figures illustrating one of the digital filters mentioned in the above mentioned book entitled "Digital Filters: Analysis and Design").

Regarding the filter as hardware, an input terminal 44 is connected to the input of a network 45 having its output connected to an adder 46. An output terminal 47 connected at the output of the adder 46 is also connected to a network 48 which passes signals back to another input of the adder 46.

Each of the networks 45 and 48 is as shown in Figure 6b where an input terminal 50 is connected to n delay circuits D1 to Dn connected in series. The input terminal 50 is also connected to the first of a series of multipliers Mo to Mn, the other multipliers in the series being connected to the outputs of the delay circuits D1 to Dn, respectively. The output of the multiplier Mn is connected by way of adder circuits SO to Sun 1 connected in series between the multiplier and an output terminal 52. The adder circuits So to Sun 1 receive a further input from the multipliers Mo to Mn-1, respectively.Each of the multipliers in Figure 6b is shown with a further input designated aO to an, respectively which represent means for setting the factor or coefficient used in multiplication. Since the circuit shown in Figure 6b is a digital circuit the coefficient aO to an may be held in respective registers and, in effect, counted down to zero in each multiplication process. Typically n is equal to fifteen to twenty and each of the delays D1 to Dn is approximately 1,000th of a second. Table I below gives a typical set of coefficients bo to b17 for the network 45 and a further typical set of coefficients aO to a17 for the network 48.

TABLE I aO = O bo = .3698036E+01 a1 = -.1350325E+01 b1 = -.6868942E+01 a2 = .7919101E-01 b2 = .3075350E+01 a3 = .1815468E+00 b3 = .6152971E+00 a4 = -.5384790E-01 b4 = -.5155725E+00 a5 = .2295512E+00 b5 = -.1731848E+00 a6 = -.1397968E+00 b6 = .9032223E+OO a7 = -.2212853E+00 b7 = -.3302465E+00 a8 = -.1756457E-01 b8 = -.1278819E+01 = = .6651361E+00 bg = .1806996E+01 a10 = -.2898495E+00 b10 = -.1636288E+01 all = -.4032059E-01 b" = .1498719E+01 a12 = .9706490E-01 b12 = -.1124145E+01 a13 = -.2686708E+00 b13 = .6521518E+OO a14 = .8005762E-01 b14 = .4333774E+OO a15= .5010519E-01 b15= .5338715E+00 a16 = .5203353E-01 b16 = -.4896732E+00 a17 = -.5112985E-01 b17 = .1648639E+00 Figures 7a and 7b show the transfer characteristic of atypical filter of the type shown in Figure 6a obtained using the coefficients given in Table I.

The operation 26 of shaping the spectrum may, as has been mentioned, be carried out with a computer acting as a digital filter. Such a filter may be as shown in Figures 6a and 6b and Table II below gives a typical set of coefficients for such a filter but in this particular filter, the network 48 of Figure 6a is omitted and thus only coefficients bo to b30 are given. In both Table I and Table II the coefficients are given in scientific notation that is first and second number separated by an E and when the first number is multiplied by 10 raised to the power indicated by the second number an ordinary decimal number results. The coefficients given in Table II provide the transfer characteristic shown in Figures 5a and 5b.

TABLE II bo = .1600000E+00 b15 = .1076707E+OO b1 = .3729979E-01 b16 = .8436136E-01 b2 = .1685366E+00 b17 = .9159869E-01 b3 = .7374379E-01 b18 = .9779616E-01 b4 = .1489818E+00 b19 = .9580757E-01 b5 = .8166648E-01 b20 = .8338795E-01 b6 = .1422934E+00 b21 = .8094792E-01 b7 = .7625830E-01 b22= .6817611E-01 b8 = .1275564E+00 b23 = .5824334E-01 b9 = .7465453E-01 b24= .5214405E-01 b10 = .1080737E+OO b25 = .3892642E-01 b11 = .6656469E-01 b26 = .3437206E-01 b12 = .7866539E-01 b27 = .2563632E-01 b13 = .6692573E-01 b28 = .1564867E-01 b14 = .6087035E-01 b29 = .1428388E-01 b30 = .4781688E-02 An example of the circuit diagram of a typical digital filter 16 based on Figures 6a and 6b and constructed from integrated circuits is shown in Figure 8.

A RAM having two areas 53a and 53b for input and output push-down stacks, respectively, is connected to a common data bus 55 which is also coupled to a multiplier 56, an output latch 57 and to receive signals from an analogue-to-digital converter 15, for example that A/D converter in the receiver 23. The latch 57 is connected to a digitai-to-analogue converter 17 such as that in the transmitter 22'. Separate buses connect a filter weight ROM 58 to the multiplier 56 and the multiplier output to an adder/accumulator 59. The operation of the filter is controlled by a clock/sequencer 60. The input and output stacks are in this example contained in one RAM with the most significant bit of the address specifying input or output. The multiplier output is calculated continuously and thus changes a short time after every input change.The clock/sequencer 60 may be formed from an oscillator, a counter and a ROM arranged in a similar way to a microcode sequencer in a computer. A filter weight counter and a stack counter are also provided (but not shown) to address the ROM 58, and the RAM areas 53a and 53b, respectively. The control bits in the ROM correspond to:: Bit 0 ADC - start conversion Bit 1 ADC - bus buffer enable (bbe) Bit 2 Stack - select input (0) or output (1) Bit 3 Stack - increment address counter Bit 4 Stack - write Bit 5 Stack - bus buffer enable Bit 6 Filter weight counter increment Bit 7 Filter weight counter - reset Bit 8 Add/Accumulate - start Bit 9 Add/Accumulate - buss buffer enable Bit 10Add/Accumulate-clearaccumulator Bit 11 Output latch - latch data from bus To control a digital filter with sixteen input and sixteen output weights the ROM, in this example, contains the code shown in the Table Ill.

TABLE III Bits Counters Comments

0 1 2 3 4 5 6 7 8 9 10 11 E 0 o X z U) iL B 1 0 0 1 0 0 0 1 0 0 0 0 1 0 0 ADC start, increase stack and clear filter weight counter 0 1 0 0 0 0 0 0 0 0 0 0 2 0 0 ADC bbe 0 0 0 0 1 0 0 0 0 0 0 0 3 0 0 Write to input stack 0 0 0 0 0 0 0 0 1 0 0 0 4 0 0 Add/Acc start 0 0 0 0 0 0 0 0 0 1 0 0 5 0 0 Add/Accbbe 0 0 1 0 1 0 0 0 0 0 0 1 6 0 0 Write to output stack, output to latch 0 0 0 1 0 0 0 0 0 0 1 0 7 1 0 Clear Acc. increase stack 0 0 0 1 0 0 1 0 0 0 0 0 8 2 1 Increase stack and weights 0 0 0 0 0 1 0 0 0 0 0 0 9 2 1 Stack babe 0 0 0 0 0 0 0 0 1 0 0 0 10 2 1 Add/Ace start 0 0 0 1 0 0 1 0 0 0 0 0 11 3 2 Increase stack and weights 0 0 0 0 0 1 0 0 0 0 0 0 12 3 2 Stack babe 0 0 0 0 0 0 0 0 1 0 0 0 13 3 2 AddlAce start 0 0 0 1 0 0 1 0 0 0 0 0 50 0 15 Increase stack and weights 0 0 0 0 0 1 0 0 0 0 0 0 51 0 15 Stack babe 0 0 0 0 0 0 0 0 1 0 0 0 52 0 15 Add/Ace start 0 0 0 1 0 0 1 0 0 0 0 0 53 1 16 Increase stack and weights 0 0 1 0 0 1 0 0 0 0 0 0 54 1 16 Stack babe 0 0 0 0 0 0 0 0 1 0 0 0 55 1 16 Add/Ace start . . . . . . . . . . . . .. . ..

0 0 0 1 0 0 1 0 0 0 0 0 98 0 31 Increase stack and weights 0 0 1 0 0 1 0 0 0 0 0 0 99 0 31 Stack babe 0 0 0 0 0 0 0 0 1 0 0 0 100 0 31 AddlAce start 1 0 0 1 0 0 0 1 0 0 0 0 1 1 0 Start cycle again In operation the counterforthe ROM in the clock/sequencer 60 cycles through its states providing the bits in columns 0 to 11 of Table III and these bits cause the operations shown in the list above to occur. The operations corresponding to ROM counter counts 1 to 7, read in each input signal from the A/C, write to the input stack, calculate a new output in dependence on the previous cycle and write the output to the output stack and the output latch. The adder/accumulator is then cleared ready for the next cycle of operations.

Operations 8 to 49 make calculations corresponding to the network 45 of Figure 6a, each coefficient being used in a respective sub-cycle of three operations, for example operations 8,9 and 10 or 11, 12 and 13. Since such sub-cycles are repetitive operations 14to 49 are not shown in Table III.

Operations 50 to 100 not all of which are shown carry out similar sub-cycles corresponding to the network 48 in Figure 6a and the cycle then repeats. When the stack counter is full in operation 49 and is incremented in operation 50 it reverts to zero and similarly the filter weight counter reverts to zero at the beginning of each new cycle.

RAMS, ROMS, Multipliers, counters, oscillators and adder/accumulators can be obtained as integrated circuits for the construction of the digital filter shown in Figure 8.

A process according to the present invention of measuring a transfer characteristic may, in some environments, suffer from the disadvantage that background noise is picked up by the microphones d and p'. This problem can be overcome by storing the signals output in operation 27 and using them in the cross and autocorrelation process 41 using the "instrumental variable method" described in a paper entitled "Refined Instrumental Variable Methods of Recursive Time-Series Analysis by Jakeman, A.J. and Young, P.C. published in Report No. AS!R 13 (1978) from the Centre for Resource and Environmental Studies (CRES), Australian National University and also in the International Journal of Control 1979. The correlation matrices formed in this procedure will be slightly different and use is made of the fact that the background noise is uncorrelated with the signal output in operation 27.

Since the characteristics of the arrangement shown in Figure 3 may introduce frequency ranges of poor transmission in the paths between the speakers and the microphones d and p' it is sometimes advisable to pre-emphasize signals output in operations 33 and 37. This pre-emphasis will be performed by two identical filtering operations carried out between operations 28 and 29, and between operations 31 and 32. Again this pre-emphasis can be carried out using the computer 21 to act as a digital filter. Although the signals transmitted in the operations 33 and 37 are then modified forms of AdnN and Ann, the correct value of the transfer function Td is obtained because the pre-emphasis occurs in both the numerator and denominator on the right had side of equation 1 and therefore cancels.

Since, referring to Figure 2, a loop is formed from the loudspeakers to the microphone d and back through the circuits 13 to 19 to the speakers, it is possible that if the filter characteristic Ta is not exactly equal to Td then oscillation could occur round this loop. In order to overcome this problem further digital filters, having the characteristic 1 1-loop gain the loop being that mentioned above, may be used to filter the outputs from Figure 4a to emphasize critical parts of the transfer function. The computer 21 can also be programmed for this purpose or separate filters can be used.

While certain embodiments of the invention have been specifically described it will be realised that the invention can be put into practice in many other ways. In particular the process of the invention and the settings for digital filters corresponding to the filter 16 can be derived without the use of a computer. Other methods of system identification may be used to determine the coefficients for the filter 16, for example the "maximum likelihood estimator" or the "instrumental variable method" both mentioned in the paper by Astrnm and Eykhoff may be used. Furthermore filters corresponding to the digital filter 16 need not be in the form of a programmed computer. As explained earlier in this specification, although Figures 1,2 and 3 show a duct 10 the invention is not restricted to operations within a duct.

Claims (17)

1. A process of testing applicable to the manufacture of a sound control system of the type specified comprising transmitting test sounds over a frequency range of interest from the first point or points, simultaneously receiving first resultant sounds at the second point or points and generating first resultant signals representative thereof, transmitting the same test sounds from the first point or points, simultaneously receiving second resultant sounds at the fourth point or points and generating second resultant signals representative thereof, transmitting from the third point or points sounds representative of the first resultant signals and simultaneously receiving third resultant sounds at the fourth point or points, transmitting from the third point or points sounds representative of the second resultant signals and simultaneously receiving fourth resultant sounds at the second point or points, and subtracting signals representative of the fourth resultant sounds from signals representative of the third resultant sounds.

2. A process according to Claim 1 wherein receiving first resultant sounds and generating first resultant signals are carried out simultaneously with receiving the second resultant sounds and generating the second resultant signals.

3. A process according to Claim 1 or 2 wherein the first and second resultant signals are stored when generated, and the stored signals are used to generate the sounds transmitted from the third point or points.

4 A method of making a sound control of the type specified wherein the means for receiving sounds and deriving signals representative of the sounds received comprises a first transducer for generating electrical signals representative of sounds received, and a digital filter, the means for generating sounds comprises a second transducer for generating sounds derived from the signals at the filter output, the method including carrying out the process of testing according to Claim 1,2 or3, deriving coefficient values for the digital filter from the second resultant signals, and the said difference obtained by the said subtraction of signals, and setting the filter coefficients in accordance with the values derived.

5. A method according to Claim 4 wherein the coefficient values are derived by considering the second resultant signals as input signals for the digital filter, and by considering the said difference, obtained by the subtraction of signals, as output signals of the filter.

6. A method according to Claim 4 wherein the coefficient values are derived by autocorrelating and cross-correlating the second resultant signals and the said difference obtained by the subtraction of signals, considering these correlations as a symmetric matrix and a vector in the form of a column matrix, and then employing the "least squares" system identification method to give the coefficient values.

7. A method according to Claim 4, 5 or 6 wherein the digital filter is formed by a programmed digital computer.

8. A process or method according to any preceding claim wherein in the sound control system of the type specified and the said process the first, second, third and fourth point or points are constituted by a group of points in a first, a second, a third and a fourth location, respectively.

9. A process or a method according to Claim 8 wherein in the sound control system the first location is the cross-section of a duct, the sound receiving means comprises a microphone, and the means for generating sounds comprises a loudspeaker.

10. A process according to Claim 8 or 9 wherein the sounds transmitted from the first and third locations are transmitted using a loudspeaker at each of these locations, the sounds received at the second and fourth locations are received using a microphone at each of these locations.

11. A process or method according to any of Claims 1 to 7 wherein in the sound control system of the type specified and the said process at least one of the first, second, third and fourth point or points are constituted by groups of points in first locations, second locations, third locations and fourth locations, respectively.

12. A process or a method according to Claim 11 wherein in the sound control system of the type specified the sound receiving means comprises a plurality of microphones, each at one of the said second locations, and/or the means for generating sounds comprises a plurality of loudspeakers, each at one of the third locations.

13. A process according to Claim 11 or 12 wherein the sounds transmitted from the first and third locations are transmitted using a plurality of loudspeakers at each of these locations and/or the sounds received at the second and fourth locations are received using a plurality of microphones at each of these locations.

14. A process of testing applicable to the manufacture of a sound control system of the type specified, the process being substantially as herein before described.

15. A method of making a sound control system of the type specified, the method being substantially as hereinbefore described.

16. A sound control system of the type specified made using the method of any of Claims 4 to 9, 11, 12 and 15.

17. A sound control system of the type specified made using a method or process of any preceding claim.