Patent Application
20060192706
This application is based upon and claims the benefit of priority from prior Japanese Patent Application P2005-054909 filed on Feb. 28, 2005; the entire contents of which are incorporated by reference herein.
1. Field of the Invention
The present invention relates to a high frequency compensator for compensating high frequency component of a compressed digital audio signal such as a mini disk (MD), a moving picture experts group 1 audio layer 3 (MP3), and a digital audio broadcasting (DAB), and relates to a reproducing device includes the high frequency compensator and reproduces a digital audio signal.
2. Description of the Related Art
With respect to a compact disc (CD), an audio signal is a stereo signal having the sampling frequency of 44.1 kHz and the data bit length of 16 bits. A data rate can be expressed by the number of bits required for 1 second. The data rate of the CD is 1411.2 kbps. With respect to the MD, the data volume to be encoded is reduced by advanced transform acoustic coding (ATRAC) and ATRAC 3 which is an improved version of the ATRAC. The ATRAC and the ATRAC 3 are compression technologies for decimating a component that a human being can hardly hear and the data before and after big sound by utilizing the characteristic of human auditory sense. The ATRAC 3 is capable of the data compression rates of two times (mini disk long play mode (MDLP) 2, 132 kbps) and four times (MDLP 4, 66 kbps) those of the ATRAC. Therefore, the ATRAC 3 can compress an amount of data up to 1/20 level of the CD. However, as the compression rate becomes higher, the sound quality becomes deteriorated.
It is said that the audible frequency band of humans is from 20 Hz to 20 kHz. However, the audible frequency band varies between individuals, and it is considered that the upper limit for ordinal people is between 14 kHz and 15 kHz. Compression technologies have been developed in consideration of such audible frequency band, and pure sound (sine wave) is used for measuring the audible frequency band. However, in a real music or natural sound, sounds of a plurality of frequencies exist in mixed state. Therefore, there has been a widespreading idea that although a sound outside of the audible frequency band cannot be heard alone, it affects human audibility in terms of soundscape, or the like.
In Japanese Patent No. 3576941 and Japanese Patent No. 3576951, there is disclosed a frequency decimating device in which one of a pair of strongly correlated bands is removed, and in which the spectrum in a range of the other band is shortened to be onto a low frequency side, based on a spectrum distribution of a signal to be subjected to the decimating processing.
In Japanese Patent No. 3576942, there is disclosed a frequency compensating system in which when a signal is compensated, a portion having a higher spectrum distribution correlation of the spectrum of the signal to be compensated, is added along an envelope onto the high frequency side of the signal to be compensated, thereby expanding the band.
In Japanese Patent Laid-open No. 2004-80635, there is disclosed a signal encoding device which the spectrum in a limited band is encoded, the limited band corresponding to a predetermined frequency band of the input time series signal, and on the decoding side, in order to obtain the time series signal in the frequency band to be expanded based on a spectrum map, mapping information showing a mapping method is adaptively created, and the encoded spectrum in the limited band and the mapping information are output.
When a digital audio signal is compressed with a high compression rate and recorded, for example, in the MDLP 4, the frequency spectrum is only present up to 12 kHz level, as shown in
In view of the aforementioned situation, the present applicant proposes a high frequency compensator with which a compressed and expanded digital audio signal is subjected to ½ down-sampling, then is subjected to double up-sampling, and the digital audio signal having undergone double up-sampling is filtered with a digital low-pass filter so as to compensate the high frequency component of the original digital audio signal.
When a digital audio signal is compressed with a high compression rate such as the MDLP 4, the frequency spectrum is present only up to 12 kHz level. Therefore, the high frequency interpolation can be preferably carried out with the aforementioned high frequency compensator. However, when it is compressed with a low compression rate such as the MDLP 2, the frequency spectrum is present up to 18 kHz level. The information to be lost through the ½ down-sampling of the digital audio data compressed with a low compression rate is large, and there is a possibility that the sound quality is not improved.
The object of the present invention is to provide a high frequency compensator capable of compensating information lost through the compression of a digital audio signal compressed with a low compression rate, and is to provide a reproducing device including the high frequency compensator.
A first aspect of the present invention inheres in a high frequency compensator encompassing, a down sampler configured to perform 1/2 down-sampling for a compressed and expanded digital audio signal, an up sampler configured to perform double up-sampling for an output signal of the down-sampler, a digital low-pass filter configured to filter an output signal of the up sampler, a digital high-pass filter configured to filter an output signal of the digital low-pass filter, and an adder configured to add an output signal of the digital high-pass filter to the compressed and expanded digital audio signal, and to output a digital audio signal generated by the adding.
In the high frequency compensator according to the first aspect, the digital low-pass filter may be a digital filter for executing over-sampling processing, and a parameter of the digital low-pass filter is calculated by solving a conditional expression by utilizing Sampled-data control theory based on a predetermined condition, the condition expression being set to design an infinite impulse response filter in order that an error signal of an original analog audio signal having a limited band and an analog audio signal, which is generated by converting a digital signal generated after the original analog audio signal is converted into the digital signal, becomes small.
In the high frequency compensator according to the first aspect, the digital high-pass filter may have a plurality of cut-off frequencies, and one of the plurality of cut-off frequencies is selected.
A second aspect of the present invention inheres in a reproducing device encompassing a high frequency compensator according to the first aspect, wherein a high frequency component of one of a digital audio signal read from a recording medium and a digital audio signal received from an external device is compensated by the high frequency compensator in order that the digital audio signal is reproduced.
Various embodiments of the present invention will be described with reference to the accompanying drawings. It is to be noted that the same or similar reference numerals are applied to the same or similar parts and elements throughout the drawings, and the description of the same or similar parts and elements will be omitted or simplified.
The up sampler 41 performs double up-sampling for the supplied digital audio signal, and inserts a sampling point with sampling value “0” at the midpoint of each sampling value (zero insertion). The digital audio signal having undergone the double up-sampling generates an axisymmetric spectrum (imaging component) with respect to its axis of symmetry, for example, 11 kHz (when the original sampling frequency is 44.1 kHz).
The digital audio signal having undergone the double up-sampling by the up sampler 41 is supplied to an infinite impulse response (IIR) or a finite impulse response (FIR) digital low-pass filter 42 together with an imaging component and is subjected to over-sampling. The over-sampling processing includes the double up-sampling processing in the up sampler 41.
As shown in
A high frequency component having a higher frequency than a desired cut-off frequency is extracted by a digital high-pass filter 43 from the digital audio signal subjected to the over-sampling processing by the up sampler 41 and the digital low-pass filter 42, and is supplied to an adder 44.
The digital high-pass filter 43 has a plurality of cut-off frequencies, and one of the cut-off frequencies is selected in accordance with a compression rate of the original digital audio signal. That is, when the digital audio signal has a low compression rate, the spectrum is present up to a relatively high frequency. Therefore, the interpolation is not needed in this range, and a relatively high cut-off frequency is selected. Accordingly, it is possible to obtain a high frequency compensator capable of optimum interpolation for information having been lost through the compression of the digital audio signal compressed with a low compression rate, according to the compression rate.
The adder 44 adds the supplied high frequency component to the original compressed and expanded digital audio signal, and outputs a result of the addition as a digital audio signal with the compensated high frequency component. As described above, a digital audio signal compressed by MDLP 2 and having the spectrum of about 18 kHz level or less as shown in
It should be noted that adjustment of high frequency compensating effects is possible by providing an amplifier connected between the digital high-pass filter 43 and the adder 44 to adjust gain having less than one.
In this manner, according to the high frequency compensator according to the first embodiment, the information lost through the compression of the digital audio data compressed with the low compression rate can be compensated.
The digital signal up-sampled to M times is given to a digital low-pass filter 53 and subjected to over sampling processing. The digital signal subjected to the over-sampling processing is converted into an analog signal by a digital to analog (D/A) converter 54, and then is supplied to a subtracter 56 after having been filtered by a post-filter 55 having a characteristic P(s). The subtracter 56 calculates an error signal “e” of the analog signal from the signal model 50 (when necessary, it is delayed) and the analog signal filtered by the post-filter 55, and outputs the error signal.
When a parameter of the digital low-pass filter 42 is calculated, a conditional expression is set in order to design an IIR digital low-pass filter 53 in order that the error signal “e” becomes small in the above described signal decompression system model. Next, a formula is obtained by converting this conditional expression approximately to a finite dimensional discrete time system. The obtained formula is solved by Sampled-data control theory based on a predetermined condition, whereby the parameter of the digital low-pass filter 53 (digital low-pass filter 42) is calculated.
It should be noted that the thus obtained digital low-pass filter is an IIR digital low-pass filter. However, for digital signal processing, an FIR filter is often used due to its ease of programming and implementation. In this case, a derived impulse response of the IIR filter is calculated, accordingly, it can be accurately realized as an FIR filter.
In the method for designing a digital low-pass filter as described above, the characteristics P(s) of an analog part such as the post-filter 55 is also taken into consideration. These analog characteristics are discretized (digitized) in a form without losing the information between the sampling points by lifting of the sampling value control theory. Therefore, an optimized design becomes possible even for the response between the sampling points. For the detailed description of this designing method, refer to Japanese Patent Laid-open No. 2001-127637.
Accordingly, it is possible to obtain a high frequency compensator capable of compensating the information lost through the compression of the digital audio signal compressed with a low compression rate. In addition, a spectrum having an approximate frequency to that of an original sound before compression can be obtained, and high frequency interpolation can be carried out for further approximating to the original sound.
The mini disk 11 is driven to rotate by a spindle motor 13 so as to have a predetermined constant linear velocity. The optical pickup 15 is driven by a motor 16 and moves in a radial direction of the mini disk 11. The magnetic head 19 is driven by a head driver 20 at the time of recording and moves in a radial direction of the mini disk 11, and its position is controlled in order that the magnetic head 19 sandwiches the same track together with the optical pickup 15.
The spindle motor 13, the optical pickup 15, and the motor 16 are driven by a servo controller 17. In addition, the servo controller 17 includes a tracking controller 32 and a focus controller 33. A signal detected by the optical pickup 15 is transmitted to a radio frequency (RF) amplifier 22 to be amplified. A focus error signal and a tracking error signal among signals amplified by the RF amplifier 22 are supplied to the focus controller 33 and the tracking controller 32, respectively.
An address signal among the signals amplified by the RF amplifier 22 is transmitted to an address decoder 21 to be decoded, and supplied to an encoder/decoder 23. The address signal decoded by the encoder/decoder 23 is utilized for the position control of the magnetic head 19 by the head driver 20. In addition, the address signal decoded by the encoder/decoder 23 is transmitted to a system controller 18, and used for the driving control of the spindle motor 13, the optical pickup 15, and the motor 16 by the servo controller 17.
A data signal among the signals amplified by the RF amplifier 22 is transmitted to the encoder/decoder 23 to be decoded, and is transmitted to a dynamic random access memory (DRAM) 25 through an anti-shock memory controller 24. The data (digital audio data) transmitted to the DRAM 25 is temporarily recorded therein, and then is transmitted to an audio encoder/decoder 26 through the anti-shock memory controller 24. The audio encoder/decoder 26 expands the transmitted data, and gives it to a high frequency compensator 34 as the high frequency compensator described in the first embodiment. The high frequency compensator 34 carries out the high frequency interpolation of the given data (digital audio signal) as described in the first embodiment. The digital signal with the compensated high frequency is output through a D/A converter 28.
The digital audio signal entered through an A/D converter 27 is compressed and encoded by the audio encoder/decoder 26, and is transmitted to the DRAM 25 through the anti-shock memory controller 24. The data transmitted to the DRAM 25 is temporarily recorded therein, and then is transmitted to the encoder/decoder 23 through the anti-shock memory controller 24 and is decoded therein. Subsequently, the data is recorded in the mini disk 11 by the magnetic head 19 and the optical pickup 15.
The anti-shock memory controller 24 and the DRAM 25 utilize a time difference of a time needed for recording to the DRAM 25 and a time needed for reading from the DRAM 25, thereby preventing sound skipping caused by vibration or the like. The system controller 18 is connected to a display 29, a timer 30, and an input unit 31, and controls operations of the servo controller 17, the encoder/decoder 23, and the anti-shock memory controller 24 while causing designated information displayed on the display 29 in response to the input, or the like, from the input unit 31.
According to the second embodiment, it is possible to obtain a reproducing device capable of compensating the information lost through the compression of the digital audio signal compressed with a low compression rate, and capable of acquiring a reproduced sound approximate to the original sound before the compression.
Various modifications will become possible for those skilled in the art after receiving the teachings of the present disclosure without departing from the scope thereof.
| Number | Date | Country | Kind |
|---|---|---|---|
| P2005-054909 | Feb 2005 | JP | national |
