US 5,430,894 AGrant
Radio Receiver Noise Suppression System
Issue Date:1995-07-04
•13 Claims
•18 Drawing Sheets
Abstract
A noise suppression apparatus includes a tuner 1 which selectively receives a radio wave signal and transforms it into an electric signal H(k) , a field information detector 2 which detects electric field information of the radio wave signal received by the tuner 1, a noise data generator 6 which generates a noise pattern W(k) on the basis of the electric field information detected by the field information detector 2, a noise canceler 4 which removes an noise component from the signal outputted from the tuner 1, and an amplifier 7 which outputs a noiseless signal outputted from the noise canceler 4. The noise canceler 4 obtains a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) fed from the tuner 1. Thereafter, the noise canceler 4 obtains a noise cancel signal S(k) on the basis of the values of the signal H(k), the noise pattern W(k), and the power spectrum .vertline.X(k).vertline..sup.2. The calculation for obtaining the noise cancel signal S(k) is carried out in accordance with the following equation: wherein, (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6.
Metadata
Assignee
- Matsushita Electric Industrial Co., Ltd.
Inventors
- Akira Nohara
- Joji Kane
Application Information
Application Number:US 0727474
Filing Date:1993-06-07
Priority Date:1992-03-11
Art Unit:261
Classifications
IPC:
H04B 110
Field of Search:
455379364328295;296;297;299;303;306;308;309;311;312;324;218;222;22599514162;163;165
Patent Drawings (18 sheets)
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a noise suppression apparatus used in radio wave communication and the like, and more particularly to a noise suppression apparatus used for a receiver.
2. Description of the Prior Art
Recently, the noise suppression using digital signal procession technology has been variously developed.
FIG. 20 is a circuit block diagram showing a conventional noise suppression apparatus. In FIG. 20, a reference numeral 20 represents a filter control section, and a reference numeral 52 represents a band pass filter group (abbreviated as BPF group). And a reference numeral 53 represents an adder. The filter control section 51 controls filter factors of the band pass filter group 52 in accordance with the noise level of sound input signal. The band pass filter group 52 includes a group of band pass filters, which separates the sound input signal into a plurality of bands. The characteristics of each band pass filter in this band pass filter group 52 is determined in accordance with the control signal of the filter control section 51. The adder 53 adds output signals fed from these band pass filters of the band pass filter group 52 and produces a noiseless sound signal.
An operation of this conventional noise suppression apparatus is explained below.
An input signal, for example a radio broadcasting sound signal containing-noise, is inputted, into the filter control section 51. The filter control section 51 judges what kind of noise components this input signal contains, and supplies the band pass filter group 52 the filter factors which cut these noise components. The band pass filter group 52 separates the input signal into a plurality of adequate band levels. Each band pass filter in the band pass filter group 52 responds to the filter factor supplied from the filter control section 51 to pass only the sound component of the input signal, and feeds it to the adder 53. The adder 53 adds the output signals of these band pass filters, so as to produce a noiseless sound signal.
In accordance with such a conventional noise suppression apparatus no significant improvement is expected in the articulation, though some increase of the S/N ratio is found. Furthermore, it is recognized that a particular noise sound remains even after the noise suppression is finished by this apparatus.
Moreover, the input signal may contain multipath noises. In order to suppress these multipath noises, there is known a method using a plurality of antennas. In this method, an antenna direction to receive radio waves is different from each other, so that the receiver can select the optimum antenna which is receiving the least multipath noise.
However, in the case where the multipath noise has once received by the receiver, this multipath noise was not removed by the signal processing. Because, the waveform of the radio wave is already distorted when received in the receiver.
SUMMARY OF THE INVENTION
Accordingly, the present invention has a purpose, in view of above-described problems or disadvantages, to provide a noise suppression apparatus capable of remarkably increasing S/N ratio as the static characteristics, suppressing multipath noises, distorting no sound signal, and being manufactured at low cost.
In order to accomplish above purposes, a first aspect of the present invention provides a noise suppression apparatus comprising: a receiving means for selectively receiving a radio wave signal to be received and transforming it through a signal transformer into an electric signal H(k); a field information detecting means for detecting electric field information of said radio wave signal received by said receiving means; a noise data generating means for generating a noise pattern W(k) on the basis of said electric field information detected by said field information detecting means; a noise cancel means for canceling a noise component of said signal received by said receiving means on the basis of said noise pattern W(k) generated by said noise data generating means, said noise cancel means including a cancel factor setting means for setting a cancel factor to the signal H(k) outputted from said receiving means, and said noise cancel means including a clamping factor setting means for generating a clamping factor, used for controlling said cancel factor in a noise canceling operation, so as to suppress adverse affection of noise removal, said noise cancel means obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the output signal H(k) of said signal transformer, and executing the noise canceling operation in accordance with the following equation:
wherein, (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(K).vertline.).gto req..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6.
Furthermore, a second aspect of the present invention provides a noise suppression apparatus comprising: a receiving means for receiving a radio wave signal an A/D converter for converting said radio wave signal from an analog signal into a digital signal; a signal transformer for transforming said digital signal into an appropriate signal form H(k); a field information detecting means for detecting electric field information of the radio wave signal; a noise data generating means for generating a noise pattern W(k) on the basis of the electric field information; a noise cancel means for receiving the signal H(k) outputted from said signal transformer and the noise pattern W(k) generated by said noise data generating means to cancel a noise component of said received radio wave signal on the basis of said noise pattern W(k); said noise cancel means obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) outputted from said signal transformer, and executing a noise cancel processing in accordance with the following equation:
wherein, (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6;
an inverse signal transformer for receiving an output of said noise cancel means and executing an inverse processing of said signal transformer; and a D/A converter for converting an output of said inverse signal transformer into a noiseless analogue signal.
Still further, a third aspect of the present invention provides a noise suppression apparatus comprising: a tuner which selectively receives a radio wave signal to be received and transforms it into an electric signal H(k); a field information detector which detects electric field information of the radio wave signal received by the tuner; a noise data generator which generates a noise pattern W(k) on the basis of the electric field information detected by the field information detector; a noise canceler which removes a noise component from said received radio wave signal, said noise canceler obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) fed from said tuner and obtaining a noise canceling signal in accordance with the following equation:
wherein, (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6; and an amplifier which outputs a noiseless signal outputted from the noise canceler.
Moreover, it is preferable that said coefficients .alpha. and .beta. are set within optimum ranges of 0.5.ltoreq..alpha..ltoreq.0.9, and 0.2.ltoreq..beta..ltoreq.0.8, respectively, in a case where said noise data generator further generates a multipath noise pattern on the basis of a multipath signal so as to cancel multipath noise. The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description which is to be read in conjunction with the accompanying drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a schematic block diagram showing the present invention;
FIG. 2 is a circuit block diagram showing a first embodiment of the present invention;
FIG. 3 is a detailed circuit diagram showing the first embodiment of the present invention;
FIG. 4 is a circuit diagram for a digital signal processing in the first embodiment of the present invention;
FIGS. 5(a) to 5(c) are graphs showing noise patterns used in the first embodiment of the present invention;
FIG. 6 is a view showing how a noise pattern is obtained in the first embodiment of the present invention;
FIG. 7 is a graph showing a relationship between the input electric field strength and the output voltage of the tuner in accordance with the first embodiment of the present invention;
FIG. 8 is a signal flowchart showing an operation of the first embodiment of the present invention;
FIG. 9 is a flowchart showing an operation of the first embodiment of the present invention;
FIG. 10 is a graph showing an improvement of S/N ratio in accordance with the present invention;
FIG. 11 is a circuit diagram showing a second embodiment of the present invention;
FIG. 12 is a circuit diagram showing a third embodiment of the present invention;
FIGS. 13(a) and 13(b) are views showing multipath signals;
FIGS. 14(a), 14(b), and 14(c) are graphs showing examples of multipath signals;
FIGS. 15(a) and 15(b) show a system specification of the embodiment of the present invention;
FIG. 16 is a circuit block diagram showing a hardware of the embodiment of the present invention;
FIG. 17 is a memory layout of the embodiment of the present invention;
FIG. 18 is a data memory map of the present invention;
FIG. 19 is a circuit block diagram showing digital components of the embodiment of the present invention; and
FIG. 20 is a circuit diagram showing a conventional noise suppression apparatus.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Hereinafter, with reference to accompanying drawings, preferred embodiments of the present invention will be explained in detail.
First Embodiment
FIG. 1 is a schematic diagram showing the concept of the present invention. In FIG. 1, a reference numeral 1 represents a tuner, which serves as a receiving means in the present invention. This tuner 1 detects a radio wave to be received and transforms it into an electric signal. AM and FM tuners are known as this type of receiver.
The tuner 1 is normally connected with an antenna 1a to improve its sensitivity. A reference numeral 2 represents a field information detecting means, which detects electric field condition at the place where the radio wave is received and at the frequency thereof.
This field information detecting means 2 chooses an intermediate-frequency signal (e.g. 10.7 MHz in case of FM broadcasting, and 450 kHz in case of AM broadcasting) from the tuner 1. By analyzing the components of this intermediate-frequency signal--such as a stereo pilot signal, the field information detecting means 2 detects the strength of the electric field and the influence of the multipath to the received signal. The detecting result of the field information detecting means 2 is transmitted to a noise suppression control means 3. The strength of the electric field can be measured by detecting a direct-current component (signal level) of the intermediate-frequency signal. On the other hand, the influence of the multipath can be measured by detecting an amplitude (AM level) of the intermediate-frequency signal. The noise suppression control means 3 receives the signal from the tuner 1 and removes the noise component of this input signal. In more detail, the noise suppression control means 3 includes a noise cancel means 4 and a noise data generating means 6. A reference numeral 5 represents an amplifier serving as an output means in the present invention, which feeds the output of the noise suppression control means 3 to a speaker or the like.
FIG. 2 is a detailed circuit diagram showing the present embodiment. As shown in FIG. 2, the noise data generating means 6 includes a noise pattern memory 6A. This noise pattern memory 6A memorizes a plurality of noise patterns (e.g. weak electric field noises A1, A2, A3, - - - , Ak, - - - , An) which are determined so as to correspond to various antenna input levels (X1, X2, X3, - - - , Xk, - - - , Xn), respectively. Here, the noise patterns need not be always memorized in the noise pattern memory 6A.
The noise data generating means 6 can analyze a generating source of noise on the basis of various components of the output from the tuner 1 and produce or synthesize a noise corresponding to, for example, the strength of the electric field.
Though various factors are considered as the source of noise generation, the noise data generating means 8 in this embodiment synthesizes the noise data on the basis of the noise change with respect to the change of the electric field strength. Namely, one of sources of noise generation is the IC used in the receiver 1. Which circuit causes the noise depends on the strength of the electric field in the noise generation by this IC. Therefore, it is possible to make a plurality of noise patterns in accordance with various strengths of the electric field.
In this embodiment, the noise patterns are calculated in advance with respect to various strengths of the electric field. These noise patterns are memorized in the noise pattern memory 6A, and are read out from this noise pattern memory 6A in the noise suppression operation. By the way, it is possible to produce these noise patterns from the calculation using the parameter of the electric field strength or the multipath. The noise suppression utilizing the multipath will be described later.
As shown in FIG. 2, the antenna input level (Xk) is directly inputted from the tuner i to the noise data generating means 6. This antenna input level (Xk) functions as the field information detecting means 2 in the present invention. In other words, no special circuit is required for providing this field information detecting means 2 in the present invention.
The noise data generating means 6 selects an optimum noise pattern Ak in accordance with the antenna input level Xk, and outputs thus selected noise pattern Ak as a noise cancel data (Ak) to the noise cancel means 4.
Next, with reference to FIG. 3, the noise cancel means 4 used in this embodiment is explained in detail. As shown in FIG. 3, the noise cancel means 4 includes a cancel factor setting means 4a, a clamping factor setting means 4b, a masking control means 4c, and a subtraction means 4d. The cancel factor setting means 4a sets a cancel factor to each frequency band of the input signal. The clamping factor setting means 4b generates a clamping factor, which is used to control the cancel factor in the canceling operation of noise so as to suppress adverse affection of the noise removal. The masking control means 4c functions to judge whether or not the clamping factor should be set by the clamping factor setting means 4b. In the subtraction operation of the noise, the masking control means 4c detects noise levels of upper and lower frequencies adjacent to the frequency of the noise to be subtracted. If the noise component of the adjacent frequency is extraordinarily large (larger than a predetermined large value), the masking control means 4c prevents the clamping factor from being set.
The cancel factor setting means 4a, the clamping factor setting means 4b, and the masking control means 4c constitute a noise suppression parameter setting means 40. The subtraction means 4d receives signals from the noise suppression parameter Setting means 40 and subtracts the noise component from the input signal fed from the tuner 1.
FIG. 4 is a circuit diagram used for a digital signal processing (abbreviated by DSP) in the first embodiment of the present invention. In FIG. 4, a reference numeral 41 represents an A/D converter, which transforms an analogue signal into a digital signal. A reference numeral 42 represents a fast hartley transformer (abbreviated by FHT), which transforms the digital signal (e.g. audio signal) inputted through one A/D 41 into an appropriate signal whose noise pattern is easy to process. An output of the fast hartley transformer 42 is calculated by the following equation (1). ##EQU1##
A reference numeral 44 represents an inverse fast hartley transformer (abbreviated by IFHT), which carries out an inverse processing of the fast hartley transformer 42. A reference numeral 45 represents a D/A converter, which performs the opposite function of the A/D converter 41.
A reference numeral 43 represents an average circuit, which obtains an average of signals representing electric field strengths inputted through the other A/D 41. A noise calculation circuit 48 receives the averaged electric field signal from the average circuit 43 and synthesizes a noise signal on the basis of thus obtained average electric field signal and the noise data memorized in the noise pattern memory 6A. The synthesized noise is fed into the noise cancel means 4, in which the output of the fast hartley transformer 42 is processed by use of this synthesized noise so as to remove the noise components thereof. Then, the (audio) signal is fed to the IFHT 44 and the D/A 45 and outputted as a noiseless audio signal.
That is, the digital signal processing (DSP) section is constituted by the FHT 42, the noise cancel means 4, the IFHT 44, the average circuit 43, the noise calculation circuit 46, and the noise pattern memory 6A.
Next, an operation of the first embodiment described in the foregoing description will be discussed below.
First of all, the tuner 1 receives a radio wave signal to be received and transforms it into an electric signal. Subsequently, the field information detecting means 2 detects an electric field strength of the signal to be received on the basis of an intermediate-frequency component of the signal received by the tuner 1. Furthermore, the field information detecting means 2 detects the multipath condition at the frequency of the signal to be received. Next, these signals are transmitted to the noise suppression control means 3, in which the noise suppression operation is carried out in the following manner.
The noise pattern is produced on the basis of the information supplied from the field information detecting means 2. In this first embodiment, several noise patterns, some of which are shown in FIG. 5, are calculated in advance and memorized in the noise pattern memory 6A. In the production of noise patterns, some of these pre-memorized noise patterns are read out from the noise pattern memory 6A. FIG. 5(a) shows a noise pattern in the case where the input level of the radio wave signal is 10 dB, and FIG. 5(b) shows a noise pattern in the case where the input level of the radio wave signal is 20 dB. Furthermore, FIG. 5(c) shows a noise pattern in the case where the input level of the radio wave signal is 60 dB.
The following explanation supposes to use only three noise patterns shown in FIGS. 5(a) to 5(c). How this embodiment synthesizes the noise patten corresponding to each of electric field strengths by using only three noise patterns will be explained below, with reference to FIG. 6.
In FIG. 6, X-axis, Y-axis, and Z-axis represent an electric field strength, a frequency, and a noise level, respectively. The noise level W.sub.dn decreases as the electric field strength dn increases. Here, let the electric field strengths di and dj have noise patterns W.sub.di(k) and W.sub.dj(k), which are functions of frequency k as shown in FIG. 6. Now, if the electric field strength dn has a value between di and dj, its noise pattern W.sub.dn(k) is obtained by linearly obtaining approximation on the basis of two noise patterns W.sub.di(k) and W.sub.dj(k). As the input electric field strength has a good linear relationship with the signal level inputted from the tuner 1, this linear approximation can be easily realized. FIG. 7 is a graph showing a relationship between the input electric field strength and the output voltage of the tuner 1.
Therefore, the noise pattern memory 6A memorizes predetermined number of noise patterns (A1, A2, A3, - - - , Ak, - - - , An) corresponding to several electric field strengths, so as to obtain a good approximation with these pre-memorized noise patterns (A1, A2, A3, - - - , Ak, - - - , An). It is possible to make additional noise patterns defined by the parameter of multipath signal. Such a modification will be described in detail in the third embodiment.
Next, the noise pattern (i.e. the noise cancel data) produced in the noise pattern generating means 6 is supplied to the noise cancel means 4 so as to subtract the noise component from the output of the tuner 1. However, it should be noted that there is a particular noise sound remaining in the case where the noise pattern is directly subtracted from the output of the tuner 1. In order to prevent this kind of particular noise from generating, the clamping factor setting means 4b sets an upper limit to the subtraction of noise in the case where the original signal is not so large even if the noise pattern is fairly large. (Namely, a value of .alpha. is controlled in the equation 4)
The subtraction of noise pattern is carried out in each frequency band of a plurality of predetermined frequency bands in this embodiment. Here, the inventors of the present invention found the remarkable fact. Under the condition that a signal having a certain frequency component is fairly large while signals having upper and lower frequency components adjacent to this certain frequency are fairly small, these small signals can be completely removed as noises without giving any adverse affection to the sound quality of the music.
Accordingly, the masking control means 4c of this embodiment controls the cancel factor used in the subtraction operation of the noise pattern, on the basis of the comparison with noise levels of upper and lower frequencies adjacent to the frequency of the noise to be subtracted. Finally, the signal, whose noise is subtracted in this manner, is outputted from the amplifier 5.
An operation of this embodiment is expressed in the signal flowchart of FIG. 8. An analogue input signal Xt is transformed into a digital signal Xi through the A/D converter 41. The FHT 42 transforms this digital signal Xi into a signal H(k) through the fast hartley transformation. On the other hand, the noise generating means 6 generates a noise pattern W(k) in accordance with an electric field d. The noise cancel means 4 receives these signals H(k) and W(k) and generates a noiseless signal S(k). This signal S(k) is inversely transformed into a digital signal Yi in the IFHT 44 and, then, is transformed into an perceivable analogue signal Yt through the D/A converter 45.
The noise cancel processing in the noise cancel means 4 is explained in more detail with reference to the flowchart of FIG. 9. First of all, in a step S1, the noise cancel means 4 obtains a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the output signal H(k) of the FHT 42. In this embodiment, the calculation of the power spectrum .vertline.X(k).vertline..sup.2 is carried out in accordance with the following equation (2).
Secondly, in a step S2, a square root of thus obtained power spectrum .vertline.X(k).vertline..sup.2 is obtained in accordance with the following equation (3).
Thirdly, the noise cancel processing is executed in a step S3. The calculation of the noise cancel processing is carried out in accordance with the following equation (4).
Wherein, (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta.
.alpha.: 0.5 (in case of a weak electric field) 0.9 (in case of a strong electric field)
.beta.: 0.4.about.0.6
With these equations, the speedy processing can be realized so as to be practically used.
In the above equation (4), two coefficients .alpha. and .beta. are independent from each other. The coefficient .alpha. relates to an improvement of S/N ratio. The other coefficient .beta. relates to a deterioration of sound quality. The smaller the coefficient .beta. becomes, the more the sound quality is deteriorated.
Although above equation (4) specifically determines values for these coefficients .alpha. and .beta., these numerical values merely show the best or optimum values applied exclusively to FM radios. If this invention is embodied in other apparatus such as a portable phone, above set values for these coefficients .alpha. and .beta. will be newly obtained. In other words, the best or optimum values for these coefficients .alpha. and .beta. should be determined by taking account of the fact as to what kind of apparatus the present invention is applied to.
If the values of these coefficients .alpha. and .beta. excurse out of above range, processing noise sound may arise or a satisfactory S/N ratio will not be obtained.
Furthermore, in a case where the signal processing is stereo type wherein L- and R-signals are independently processed, above coefficient .beta. can be set to smaller values.
Moreover, if the noise pattern is modified, above set values for the coefficients .alpha. and .beta. will be correspondingly varied.
Yet further, if the loud-speaker has poor reproduction capability and, therefore, does not strictly require sound quality, the value of the coefficient .beta. will be set to a wider range, e.g. 0.2.about.0.8.
FIG. 10 shows the result of the noise suppression. FIG. 10 shows the improvement of S/N ratio with respect to the input electric field level. A dotted line represents the result of the conventional noise suppression apparatus, while a solid line represents the result of the present invention. As apparent from FIG. 10, the noise suppression apparatus in accordance with the present invention can bring 5.about.6 dB improvement of the S/N ratio in the whole range of the electric field.
Second Embodiment
A second embodiment of the present invention will be explained below. In this second embodiment, a signal judging means 30 is provided in addition to the components of the first embodiment, as shown in FIG. 11. The same components as the first embodiment are suffixed by the same reference numerals.
In FIG. 11, the signal judging means 30 receives an output signal from the tuner 1 and judges a kind of signal to be received on the basis of this output signal from the tuner 1. For example, it is possible to classify the sound signal into several categories of a human speech sound, a piano sound, and a drum sound by analyzing the spectrum of the signal and the like. In this embodiment, the signal judging means 30 judges as to whether the received signal is a human speech sound (i.e. a conversation mode) or a music sound (i.e. a music mode).
If the signal judging means 30 judges the received signal is the music mode, the subtraction means 4d of the noise cancel means 4 is commanded not to subtract more than 60% of the signal component. That is, the clamping factor setting means 4b uniformly generates a clamping factor of 60% to every frequency component. However, it is needless to say that the value of this clamping factor can be differentiated finely in accordance with the frequency component.
In case of the conversation mode, the clamping factor setting means 4d generates no clamping factor.
In accordance with this embodiment, the signal judging means 30 judges as to whether the content of the broadcasting is news or music. Accordingly, the clamping factor can be accurately set to cancel noises on the basis of the kind of the received signal. Thus, the noise suppression can be performed without deteriorating sound quality.
Third Embodiment
Next, a third embodiment of the present invention will be explained with reference to the drawings. FIG. 12 shows a circuit block diagram of the third embodiment. The same components as those of the previous embodiments are suffixed by the same reference numerals. This embodiment has a purpose to realize the removal of the multipath noise, which was conventionally impossible, by using the DSP technology. The basic constitution of this third embodiment is similar to that of the first embodiment.
As shown in FIG. 12, the noise data generating means 6 include a multipath noise generating means 8B. This multipath noise generating means 6B basically receives a multipath signal M of the field information detecting means 2 (i.e. an output Yk from the FM tuner 1) and generates a multipath noise data K.sub.kk so as to cancel the multipath noise in the noise cancel means 4.
How the field information detecting means 2 detects the presence of the multipath noise will be explained below. Compared with AM signals, the characteristic feature of FM signals is that the frequency of the FM signal varies in response to the sound signal while its amplitude is maintained at a constant value.
After the radio wave signal is received and tuned, an FM signal is outputted from the intermediate-frequency amplifier. The amplitude of this FM signal does not vary under the condition of no presence of multipath, even in the range where the limiter of the intermediate-frequency amplifier does not work yet. However, if the multipath occurs, a distortion of the modulated sound signal is added to the amplitude of this FM signal. Therefore, the amplitude of the FM signal outputted from the intermediate-frequency amplifier varies under the presence of multipath. (Refer to FIG. 13(a))
Accordingly, the multipath signal M can be obtained by detecting and amplifying the distortion amount caused by the multipath.
Furthermore, in addition to the amplitude change, the multipath distortion is found in the frequency component when the multipath is generated. That is, the frequency component is also influenced by the multipath component. (Refer to FIG. 13(b)).
Therefore, this multipath component can be also used as the multipath signal M.
In this embodiment, the field information detecting means 2 supplies this multipath signal M (i.e. the output signal Yk of the FM tuner 1) to the multipath noise data generating means 6B. The multipath noise data generating means 6B constitutes the matrix of the multipath signal (Y1, Y2, - - - , Yk, - - - Yn) and the antenna input signal (X1, X2, X3, - - - , Xk, - - -, Xn). That is, the multipath noise pattern K.sub.kk is selected from this matrix on the basis of the multipath signal M (i.e. the output signal Yk of the FM tuner 1) and the antenna input signal Xk. It is needless to say that the multipath noise data generating means 6B can be constituted by using only the multipath signal (Y1, Y2, - - - , Yk, - - - , Yn).
FIG. 14(a) to 14(c) show several noise patterns of the multipath noise. FIG. 14(a) shows a large multipath noise signal, and FIG. 14(b) shows an intermediate multipath noise signal. And, FIG. 14(c) shows a small multipath noise signal.
The multipath noise pattern K.sub.kk is fed from the multipath noise data generating means 6B to a noise data mixer 6C, in which the multipath noise pattern K.sub.kk is added with the noise pattern A.sub.k. Thus added noise pattern (i.e. noise cancel data) A.sub.k +K.sub.kk is supplied to the noise cancel means 4, in which this noise pattern A.sub.k +K.sub.kk is subtracted from the output signal of the FM tuner 1. In this embodiment, the stereo processing is carried out for canceling right and left noises.
Returning to the specific values of the coefficients in the equation (4), the best or optimum value for the coefficient .beta., having been set in the first embodiment, can be more widely selected in this embodiment. For example, a setting of 0.2.ltoreq..beta..ltoreq.0.8 will be preferable.
In accordance with this third embodiment, the field information detecting means 2 outputs the multipath signal and the noise data generating means 6 includes the multipath noise data generating means 6B. With these means, an adequate noise suppression can be performed against the multipath noise without deteriorating the sound quality.
Though the field information detecting means 2 detects only the electric field strength and the multipath condition, it is also possible to detect a signal level of a signal adjacent to the signal to be received, influence by other channels or peripheral devices. By transmitting these signals to the noise data generating means, the noise pattern can be controlled.
FIG. 15(a) shows a system specification used in this embodiment of the present invention, and FIG. 15(b) shows a relationship between the frame length and the frame period. FIG. 18 shows a block diagram of the hardware, and FIG. 17 shows a memory layout. FIG. 18 shows a data memory map, and FIG. 19 shows a circuit block diagram of the digital components.
As this invention may be embodied in several forms without departing from the spirit of essential characteristics thereof, the present embodiments are therefore illustrative and not restrictive, since the scope of the invention is defined by the appending claims rather than by the description preceding them, and all changes that fall within meets and bounds of the claims, or equivalence of such meets and bounds are therefore intended to embraced by the claims.
Claims
What is claimed is:
1. A noise suppression apparatus comprising: a receiving means for selectively receiving a radio wave signal to be received and transforming it through a signal transformer into an electric signal H(k); a field information detecting means for detecting electric field information of said radio wave signal received by said receiving means; a noise data generating means for generating a noise pattern W(k) on the basis of said electric field information detected by said field information detecting means; a noise cancel means for canceling a noise component of said signal received by said receiving means on the basis of said noise pattern W(k) generated by said noise data generating means, said noise cancel means including a cancel factor setting means for setting a cancel factor to the signal H(k) outputted from said receiving means; and said noise cancel means including a clamping factor setting means for generating a clamping factor, used for controlling said cancel factor in a noise canceling operation, so as to suppress adverse affection of noise removal, said noise cancel means obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the output signal H(k) of said signal transformer, and executing the noise canceling operation in accordance with the following equation: ine.) wherein, (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).gtoreq..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6.
2. A noise suppression apparatus in accordance with claim 1, wherein said noise cancel means includes a masking control means for judging as to whether or not said clamping factor should be set by said clamping factor setting means.
3. A noise suppression apparatus in accordance with claim 2, wherein said masking control means detects noise levels of upper and lower frequencies adjacent to a frequency of a noise to be canceled.
4. A noise suppression apparatus in accordance with claim 3, wherein said masking control means prevents said clamping factor from being set when said noise level of said adjacent frequency is larger than a predetermined value.
5. A noise suppression apparatus in accordance with claim 2, wherein said noise cancel means includes a subtraction means for receiving signals from said cancel factor setting means, clamping factor setting means, and masking control means, in order to subtract a noise component from said signal outputted from said receiving means.
6. A noise suppression apparatus comprising: a receiving means for selectively receiving a radio wave signal to be received and transforming it through a signal transformer into an electric signal H(k); a field information detecting means for detecting electric field information of said radio wave signal received by said receiving means; a noise data generating means for generating a noise pattern W(k) on the basis of said electric field information detected by said field information detecting means; noise cancel means for canceling a noise component of said signal received by said receiving means on the basis of said noise pattern W(k) generated by said noise data generating means, said noise cancel means including a cancel factor setting means for setting a cancel factor to the signal H(k) outputted from said receiving means, and said noise cancel means including a clamping factor setting means for generating a clamping factor, used for controlling said cancel factor in a noise canceling operation, so as to suppress adverse affection of noise removal, said noise cancel means obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the output signal H(k) of said signal transformer, and executing the noise canceling operation in accordance with the following equation; ine.) where (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline..gtor eq..beta.; wherein said field information detecting means further detects multipath noise and outputs a multipath signal and said noise data generating means further generates a multipath noise pattern on the basis of the multipath signal so as to cancel the multipath noise, and said coefficients .alpha. and .beta. are set within optimum ranges of 0.5.ltoreq..alpha..ltoreq.0.9, and 0.2.ltoreq..beta..ltoreq.0.8, respectively.
7. A noise suppression apparatus comprising: a receiving means for receiving a radio wave signal; an A/D converter for converting an analogue radio wave signal from an analog signal into a digital signal; a signal transformer for transforming said digital signal into an appropriate signal form H(k); a field information detecting means for detecting electric field information of said radio wave signal; a noise data generating means for generating a noise pattern W(k) on the basis of said electric field information; a noise cancel means for receiving the signal H(k) outputted from said signal transformer and the noise pattern W(k) generated by said noise data generating means to cancel a noise component of said received radio wave signal on the basis of said noise pattern W(k); said noise cancel means obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) outputted from said signal transformer, and executing a noise cancel processing in accordance with the following equation: ine.) wherein (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6; an inverse signal transformer for receiving an output of said noise cancel means and executing an inverse processing of said signal transformer; and a D/A converter for converting an output of said inverse signal transformer into a noiseless analogue signal.
8. A noise suppression apparatus in accordance with claim 7, wherein said signal transformer is a fast hartley transformer.
9. A noise suppression apparatus in accordance with claim 8, wherein an output of said fast hartley transformer is calculated by the following equation: ##EQU2## wherein, cas(.theta.)=cos (.theta.)+sin (.theta.).
10. A noise suppression apparatus in accordance with claim 7, wherein said noise data generating means includes an average circuit for obtaining an average of signals representing electric field strengths, a noise pattern memory for memorizing a plurality of noise patterns, and a noise calculation circuit for receiving an averaged electric field signal from said average circuit and synthesizing a noise signal on the basis of thus obtained average electric field signal and a noise data memorized in said noise pattern memory.
11. A noise suppression apparatus comprising: a receiving means for receiving a radio wave signal; an A/D converter for converting an analogue radio wave signal from an analog signal into a digital signal; a signal transformer for transforming said distal signal into an appropriate signal form H(k); a field information detecting means for detecting electric field information of said radio wave signal; a noise data generating means for generating a noise pattern W(k) on the basis of said electric field information: a noise cancel means for receiving the signal H(k) outputted from said signal transformer and the noise pattern W(k) generated by said noise data generating means to cancel a noise component of said received radio wave signal on the basis of said noise pattern W(k); said noise cancel means obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) outputted from said signal transformer, and executing a noise cancel processing in accordance with the following equation: where (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline..gtor eq..beta.; an inverse signal transformer for receiving an output of said noise cancel means and executing an inverse processing of said signal transformer and a D/A converter for converting an output of said inverse signal transformer into a noiseless analogue signal: wherein said noise data generating means further generates a multipath noise pattern on the basis of a multipath signal so as to cancel multipath noise, and said coefficients .alpha. and .beta. are set within optimum ranges of 0.5.ltoreq..alpha..ltoreq.0.9, and 0.2.ltoreq..beta..ltoreq.0.8, respectively.
12. A noise suppression apparatus comprising: a tuner which selectively receives a radio wave signal to be received and transforms it into an electric signal H(k); a field information detector which detects electric field information of the radio wave signal received by the tuner; a noise data generator which generates a noise pattern W(k) on the basis of the electric field information detected by the field information detector; a noise canceler which removes a noise component from said received radio wave signal, said noise canceler obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) fed from said tuner and obtaining a noise canceling signal in accordance with the following equation: ine.) wherein (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta., 0.5.ltoreq..alpha..ltoreq.0.9, and 0.4.ltoreq..beta..ltoreq.0.6; and an amplifier which outputs a noiseless signal outputted from the noise canceler.
13. A noise suppression apparatus comprising: a tuner which selectively receives a radio wave signal to be received and transforms it into an electric signal H(k); a field information detector which detects electric field information of the radio wave signal received by the tuner; a noise data generator which generates a noise pattern W(k) on the basis of the electric field information detected by the field information detector; a noise canceler which removes a noise component from said received radio wave signal, said noise canceler obtaining a power spectrum .vertline.X(k).vertline..sup.2 on the basis of the signal H(k) fed from said tuner and obtaining a noise canceling signal in accordance with the following equation: ine.) where (1-.alpha..multidot..vertline.W(k).vertline./.vertline.X(k).vertline.).gto req..beta.; and an amplifier which outputs a noiseless signal outputted from the noise canceler; wherein said field information detector further detects multipath noise and outputs a multipath signal and said noise data generator further generates a multipath noise pattern on the basis of the multipath signal so as to cancel the multipath noise, and said coefficients .alpha. and .beta. are set within optimum hinges of 0.5.ltoreq..alpha..ltoreq.0.9, and 0.2.ltoreq..beta..ltoreq.0.8, respectively.
Patent Citations (6)
Non-Patent Literature (1)
- IEE Proceedings-G vol. 136, No. 5, Oct. 1989, Stevenage, Herts, GB, pp. 269-277, XP66014, Boussakta et al, `Prime-factor Hartley and Hartley-like Transform Calculation Using Transversal Filter-type Structures`.