Classifications

• • G—PHYSICS
  • G10—MUSICAL INSTRUMENTS; ACOUSTICS
  • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
  • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
  • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
  • G10L21/038—Speech enhancement, e.g. noise reduction or echo cancellation using band spreading techniques
• • G—PHYSICS
  • G10—MUSICAL INSTRUMENTS; ACOUSTICS
  • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
  • G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
  • G10L19/04—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using predictive techniques
  • G10L19/16—Vocoder architecture
  • G10L19/18—Vocoders using multiple modes
  • G10L19/24—Variable rate codecs, e.g. for generating different qualities using a scalable representation such as hierarchical encoding or layered encoding

Definitions

• Sampling rate conversion device encoding device, decoding device, and methods thereof
• the present invention relates to a sat transform apparatus, an encoding apparatus, a decoding apparatus, and methods thereof.
• ADSL Asymmetric Digital Subscriber Line
• W-CDMA Wideband-Code Division Multiple Access
• ITU dnnationalization is a representative scheme for encoding narrowband signals.
• G. 726, G. 729 equal power S standardized in the Telecommunication Union). Also, ITU-T (International) is a typical method for encoding wideband signals.
• the scalable function is a function that can decode the audio signal even with some power of the coding code.
• it is necessary to decode high-quality voice signals using all coding codes in good channel conditions, and transmit only part of the coding code in bad channel conditions. Can reduce the frequency of occurrence of packet loss.
• effects such as the efficiency of network resources at the time of communication between multiple points can be obtained.
• a signal of sampling rate power S8 kHz is encoded using the method of G. 726, G. 729, etc. standardized by ITU-T, and the error signal is further added at a sampling rate of 16 kHz.
• coding it is possible to realize quality improvement and scalability by extending the signal band.
• FIG. 1 is a block diagram showing a typical configuration of a coding device that performs scalable coding.
• the number of layers N 3
• the audio signal (audio signal, audio signal, etc.) input to the downsampling unit 12 through the input terminal 11 is downsampled from 32 kHz to 16 kHz, and the first layer code input unit 13 Given to The first layer coding unit 13 determines the first coding code so as to minimize the perceptible distortion between the input acoustic signal and the decoded signal generated after the coding.
• This first encoded code is sent to the multiplexing unit 26 and also sent to the first layer decoding unit 14.
• First layer decoding section 14 generates a first layer decoded signal using the first encoded code.
• the up-sampling unit 15 up-samples the sampling frequency of the first layer decoded signal from 16 kHz to 24 kHz, and supplies this signal to the subtractor 18 and the adder 21.
• the acoustic signal input to the downsampling unit 16 via the input terminal 11 is downsampled from 32 kHz to 24 kHz and supplied to the delay unit 17.
• the delay unit 17 delays the down-sampled signal by a predetermined time length.
• the subtractor 18 obtains the difference between the output signal of the delay unit 17 and the output signal of the up-sampling unit 15, generates a second layer residual signal, and is given to the second layer code input unit 19.
• Second layer mark The coding unit 19 encodes the second layer residual signal so as to aurally improve the quality, determines a second coding code, and multiplexes the second coding code. 26 and second layer decoding unit 20 are provided.
• the second layer decoding unit 20 performs a decoding process using the second code and the code to generate a second layer decoded residual signal.
• the adder 21 takes the sum of the first layer decoded signal and the second layer decoded residual signal described above to generate a second layer decoded signal.
• the up sampling unit 22 upsamples the sampling frequency of the second layer decoded signal from 24 kHz to 32 kHz and supplies this signal to the subtractor 24.
• the acoustic signal input to the delay unit 23 through the input terminal 11 is delayed by a predetermined time length, and is provided to the subtractor 24.
• the subtractor 24 takes the difference between the output signal of the delay unit 23 and the output signal of the upsampling unit 22 to generate a third layer residual signal.
• This third layer residual signal is applied to third layer code block 25.
• the third layer coding unit 25 performs code coding so that the third layer residual signal can be aurally improved in quality, determines the third coding code, and outputs the result to the multiplexing unit 26. Give that code a code.
• the multiplexing unit 26 multiplexes the encoded code obtained from the first layer coding unit 13, the second layer coding unit 19, and the third layer coding unit 25, and outputs the multiplexed code through the output terminal 27. Do.
• Patent Document 1 Japanese Patent Application Laid-Open No. 2000-68948
• An object of the present invention is to provide a sampling rate conversion device and an encoding device capable of reducing the circuit scale and reducing the amount of processing operation of the code, and decoding the signal encoded by the encoding device. Apparatus, and methods of providing these. Means to solve the problem
• the effective frequency band of the spectrum is expanded in the frequency domain to obtain a signal in the time domain.
• ⁇ ⁇ get an equivalent signal to the case of up-sampling.
• the sampling rate conversion device of the present invention converts the input time domain signal into the frequency domain to obtain the first spectrum, and extends the frequency band of the first spectrum obtained; And an insertion means for inserting the second spectrum into the expanded frequency band of the first space after expansion.
• the signal equivalent to the up-sampled signal in the time domain is converted by converting the input time domain signal into the frequency domain signal and extending the frequency band of the obtained spectrum. You can get In addition, the circuit scale of the encoding device can be reduced, and the amount of processing operation of the code can also be reduced.
• the code device of the present invention is a conversion means for obtaining a first spectrum of Na point by frequency analysis of an input sampling frequency signal of Fx with analysis length 2 ⁇ Na, and a first spectrum obtained
• An extension means for extending the frequency band of Petal to Nb points and a second spectrum to be inserted in the expanded frequency band of the first spectrum after extension are specified, and a code representing the second spectrum is specified.
• the encoding apparatus of the present invention has a configuration as described above, wherein the second spectrum is generated based on the first vector.
• the second spectrum is obtained by performing frequency analysis on an input signal with a sampling frequency of Fy at a point 2'Nb, Na ⁇ k ⁇ It adopts a configuration that is determined to be similar to the spectrum included in the Nb frequency band.
• the expanded spectrum can be determined based on the spectrum of the original signal. Since it is possible to obtain an extended spectrum with higher accuracy.
• the coding means divides the frequency band of Na ⁇ k ⁇ Nb into two or more sub-bands, and
• a configuration is adopted in which a coded code representing two spectra is output.
• the encoding apparatus of the present invention has a configuration as described above, wherein the signal with the sampling frequency Fx is a signal decoded by a lower layer in a layer code.
• the present invention can be applied to hierarchical coding configured by coding units of multiple layers, and hierarchical coding can be realized with only minimum sampling transformation.
• the decoding apparatus of the present invention is an acquisition unit that analyzes a signal of sampling frequency Fx with analysis length 2'Na to acquire a first spectrum of a frequency band of 0 ⁇ k ⁇ Na, and a code Decoding means for receiving a second code and decoding a second spectrum in the Na ⁇ k ⁇ Nb frequency band, and combining the first and second spectra to obtain a 0 ⁇ ⁇ k ⁇ Nb frequency band And a conversion means for converting a spectrum included in a frequency band of 0 ⁇ k ⁇ Nb into a time domain signal.
• the decoding apparatus of the present invention has a configuration as described above, wherein the second spectrum is generated based on a spectrum of a frequency band of 0 ⁇ k ⁇ Na.
• the width of the frequency band of the combined spectrum obtained by the generation means is equal to a predetermined width after the combination.
• the apparatus further comprises: a force for inserting a specified value in the high band part of the spectrum or means for discarding the high band part of the combined spectrum.
• the decoding apparatus of the present invention has a configuration as described above, wherein the signal with the sampling frequency of Fx is a signal decoded by a lower layer in a layer code.
• the circuit scale of the encoding device can be reduced, and the amount of processing operation for encoding can also be reduced.
• FIG. 1 is a block diagram showing a typical configuration of a coder performing scalable coding.
• FIG. 2 is a block diagram showing the main configuration of a spectrum coder according to a first embodiment.
• FIG. 3A is a diagram showing the first spectrum
• FIG. 3B is a diagram showing the spectrum after the effective frequency band is extended.
• FIG. 4A Diagram for explaining in principle the effect of processing to extend the effective frequency band of extra space
• FIG. 5 shows the main configuration of the wireless transmission device according to the first embodiment-[FIG. 6] shows the internal configuration of the code device according to the first embodiment-[FIG. 7] according to the first embodiment
• Fig. 8 shows an internal configuration of a spectral code portion-[Fig. 8]
• FIG. 9 shows the main configuration of the radio receiving apparatus according to the first embodiment-[FIG. 10] shows the internal configuration of the decoding apparatus according to the first embodiment-[FIG. 11] according to the first embodiment 12 shows the internal configuration of the spectrum decoding unit-[FIG. 12]
• FIGS. 12A and 12B show the process performed by the band expanding unit according to the first embodiment.
• FIG. 13 A diagram showing how a spectrum is generated through the processing in the coupling unit and time domain conversion unit according to the first embodiment.
• FIG. 14A A block diagram showing the main configuration of the transmitting side when the coding apparatus according to Embodiment 1 is applied to a wired communication system.
• FIG. 14B A block diagram showing the main configuration of the receiving side when the decoding apparatus according to Embodiment 1 is applied to a wired communication system.
• FIG. 15 A main configuration of the decoding apparatus according to Embodiment 2 is shown-[FIG. 16] An internal configuration of a spectrum decoding unit according to Embodiment 2 is shown-[FIG. 17] Embodiment 2 Figure for explaining in more detail the processing of the correction unit according to
• FIG. 18 A diagram for illustrating the process of the correction unit according to the second embodiment in more detail.
• FIG. 19 A diagram for further explaining the operation of the spectrum decoding unit according to Embodiment 2.
• FIG. 20A A diagram for further explaining the operation of the spectrum decoding unit according to Embodiment 2.
• 20B A diagram for further explaining the operation of the spectrum decoding unit according to Embodiment 2.
• FIG. 21 A diagram showing the main configuration of a communication system according to Embodiment 3.
• FIG. 22 A diagram showing the main configuration of a communication system according to Embodiment 4.
• FIG. 2 is a block diagram showing a main configuration of spectrum coder apparatus 100 according to Embodiment 1 of the present invention.
• Spectrum coding apparatus 100 has sampling rate conversion section 101, input terminal 102, spectrum information identification section 106, and output terminal 107.
• the sampling rate converter 101 further includes a frequency domain converter 103, a band expander 104, and an extended spectrum adder 105.
• a signal sampled at the sampling rate Fx is input to the spectrum coding apparatus 100 via the input terminal 102.
• Frequency domain conversion section 103 frequency-analyzes this signal by analysis length 2′Na.
• a signal in the intermediate domain is converted to a signal in the frequency domain (frequency domain conversion) to calculate a first extra space Sl (k) (0 ⁇ k ⁇ Na).
• the obtained first spectrum Sl (k) is given to the band expanding unit 104.
• frequency analysis uses a modified discrete cosine transform (MDCT).
• MD CT analysis is performed by superposing adjacent frames and analysis frames in half by half, and using the orthogonal basis where the first half of the analysis frame is an odd function and the second half is an even function, distortion between frames is canceled. It has the characteristic of being As a frequency analysis method, it is also possible to use discrete Fourier transform (DFT), discrete cosine transform (DCT) or the like.
• DFT discrete Fourier transform
• DCT discrete cosine transform
• Extended spectrum assigning section 105 assigns an extended spectrum S1, (k) (Na ⁇ k ⁇ Nb) to which external power is also input to the frequency band expanded by band expanding section 104, and spectrum information identifying section 106. Output to
• Spectrum information identification section 106 sets information required for identifying expanded spectrum S 1, (k) among the spectrums given from extended spectrum application section 105 as a code and output terminal 107. Output through.
• This code ⁇ code is information representing the sub-band energy of the extended spectrum S1 '(k), information representing the effective frequency band, and the like. Details of this will also be described later.
• FIG. 3A shows a first spectrum Sl (k) provided by frequency domain conversion section 103
• FIG. 3B shows spectrum Sl (k) after the effective frequency band is expanded in band expansion section 104.
• the band expanding unit 104 secures an area in which new spectrum information can be stored in a frequency band in which the frequency k of the first spectrum Sl (k) is expressed in the range of Na ⁇ k ⁇ Nb.
• the size of this novel region is represented by Nb-Na.
• Nb is a sampling rate Fx of an externally applied signal through input terminal 102.
• the relationship between the analysis length 2'Na of the frequency domain transform unit 103 and the sampling rate Fy of the signal decoded by the decoding unit (not shown) is determined. Specifically, Nb is set by the following equation.
• the sampling rate F y of the signal decoded by the decoding unit is determined by the following equation.
• FIGS. 4A and 4B are diagrams for explaining in principle the effect of the process of extending the effective frequency band of the spectrum performed in band expanding section 104.
• FIG. FIG. 4A shows a spectrum Sa (k) obtained when the signal of the sampling rate Fx is frequency-analyzed by the analysis length 2′Na.
• the horizontal axis represents frequency, and the vertical axis represents spectral intensity.
• the Nyquist theorem power of the effective frequency band of the signal is also 0 ⁇ FxZ2.
• the range of the frequency index k is 0 ⁇ k ⁇ Na
• the frequency resolution of the spectrum Sa (k) is FxZ (2′Na).
• the spectrum Sb (k) obtained by frequency analysis with analysis length 2 ⁇ Nb is shown in Figure 4B
• the effective frequency band of the signal is extended to 0-FyZ 2
• the range of frequency index k is 0 ⁇ k ⁇ Nb.
• the frequency resolution FyZ (2′Nb) of the scan vector Sb (k) is equal to FxZ (2′Na). That is The spectrum Sa (k) in the band 0 ⁇ k ⁇ Na and the spectrum Sb (k) are equal.
• the spectrum Sb (k) when the band of the spectrum Sa (k) (0 ⁇ k ⁇ Na) is expanded to Nb is analyzed after the signal of sampling Fx is upsampled to the sampling Fy. It means that it agrees with the spectrum obtained by frequency analysis with a length 2 ⁇ ⁇ b.
• sampling rate conversion section 101 converts an input time domain signal into a frequency domain signal, and extends the effective frequency band of the obtained spectrum to obtain a signal in the time domain.
• a spectrum equivalent to the spectrum obtained by frequency conversion of the upsampled signal can be obtained.
• sampling rate conversion unit 101 Since the signal output from sampling rate conversion unit 101 is a signal in the frequency domain, when a signal in the time domain is required, a time domain conversion unit is provided to perform reconversion to the time domain. Do. In the above example, since sampling rate conversion section 101 is installed in spectral code device 100, the signal in the frequency domain is input to spectrum information identification section 106 without being converted back to the time domain signal. , The encoding code is generated
• the spectral information identification unit 106 outputs The code rate of the coded code is different. That is, part of the processing in the sampling rate conversion unit 101 has a great influence on coding. This means that the spectral code device 100 simultaneously realizes the conversion of the sampling rate of the input signal and the code ⁇ .
• the process performed by force spectrum information identification unit 106 described with the case where extended spectrum is applied to the original spectrum in extended spectrum application unit 105 as an example is as follows: Since it is necessary to output the information necessary for specifying the extended spectrum as a code, it is sufficient if the extended spectrum to be given is specified, and the extended spectrum is not necessarily given. It doesn't have to be. Furthermore, although upsampling has been described as an example of sampling rate conversion, the above principle can also be applied to downsampling.
• FIG. 5 is a block diagram showing the main configuration of radio transmitting apparatus 130 when encoding apparatus 120 according to the present embodiment is mounted on the transmitting side of the radio communication system.
• the wireless transmission device 130 includes an encoding device 120, an input device 131, an AZD conversion device 132, an RF modulation device 133, and an antenna 134.
• the input device 131 converts the sound wave W11 heard by the human ear into an analog signal which is an electrical signal, and outputs the analog signal to the AZD conversion device 132.
• the AZD converter 132 converts this analog signal into a digital signal and outputs it to the coder 120 (signal Sl).
• the encoder 120 encodes the input digital signal S1 to generate an encoded signal, and outputs the encoded signal to the RF modulator 133 (signal S2).
• the RF modulator 133 modulates the coded signal S 2 to generate a modulated coded signal, and outputs the modulated coded signal to the antenna 134.
• the antenna 134 transmits the modulation coded signal as the radio wave W12.
• FIG. 6 is a block diagram showing an internal configuration of the above-mentioned coder device 120. As shown in FIG. Here, the case of hierarchical coding (scalable coding) will be described as an example.
• Code decoding apparatus 120 includes input terminal 121, downsampling section 122, first layer code decoding section 123, first layer decoding section 124, delaying section 126, spectrum coding section 100a, multiplexing It has a part 127 and an output terminal 128.
• the acoustic signal S1 of the sampling rate Fy is input to the input terminal 121.
• the downsampling unit 122 downsamples the signal S1 input through the input terminal 121 to generate and output a signal of the sampling rate Fx.
• the first layer code unit 123 codes this down-sampled signal, and outputs the obtained code unit code to the multiplexing unit (multiplexer) 127, and also performs first layer decoding. Also output to the part 124.
• the first layer decoding unit 124 generates a decoded signal of the first layer based on the coded code.
• the delay unit 126 gives a delay of a predetermined length to the signal S 1 input via the input terminal 121.
• the magnitude of this delay is equal to the time delay that occurs when the signal passes through the downsampling unit 122, the first layer coding unit 123, and the first layer decoding unit 124.
• Spectrum coding section 100 a is a sample signal output from first layer decoding section 124. Spectrum coding is performed using the signal S3 of the coding rate Fx and the signal S4 of the sampling rate Fy output from the delay unit 126, and the generated code code S5 is output to the multiplexing unit 127.
• the multiplexing unit 127 multiplexes the code obtained with the first layer code 123 and the code S5 obtained with the spectral code 10a, and outputs the output terminal 128 as an output code S2. Output through.
• This output code S2 is given to the RF modulator 133
• FIG. 7 is a block diagram showing an internal configuration of the above-described spectrum coding unit 100a.
• the spectrum coding unit 100a has the same basic configuration as the spectrum coding unit 100 shown in FIG. 2, and the same components are denoted by the same reference numerals. Omit.
• a feature of spectrum encoding section 100a is that extended spectrum 31 '&) ⁇ & ⁇ 1 ⁇ ⁇ ? ⁇ ) Is applied using the input signal S3 at sampling rate Fy. That's right. According to this, since the target signal for determining the extended spectrum Sl '(k) is given, the accuracy of the extended spectrum Sl' (k) is improved, and as a result, the effect of leading to the quality improvement is obtained.
• the frequency domain conversion unit 112 performs frequency analysis on the signal S4 of the sampling rate Fy input through the input terminal 111 with an analysis length of 2 ⁇ Nb, and the second spectrum S 2 (k) (0 ⁇ k ⁇ Nb Ask for).
• the relationship represented by (Expression 1) holds for the sampling frequencies Fx and Fy and the analysis lengths Na and Nb.
• Spectrum information identifying section 106 determines a coding code representing extended spectrum SI ′ (k).
• the expanded spectrum S1 ′ (k) is determined using the second vector S2 (k) obtained by the frequency domain conversion unit 112.
• the spectrum information identification unit 106 determines the code ⁇ code through two steps of determining the shape of the expanded spectrum S 1, (k) and determining the gain of the expanded spectrum S 1, (k).
• the expanded spectrum Sl '(k) is determined using the band 0 ⁇ k ⁇ Na of the first spectrum Sl (k).
• the first spectrum Sl (k) separated by a fixed value C on the frequency axis is copied to the expanded spectrum S1, (k).
• C is a predetermined fixed value, and it is necessary to satisfy the condition of C ⁇ Na.
• information for representing the shape of the extended spectrum Sl '(k) is not output as a code.
• the fixed value C has a predetermined range T.
• the value T of the variable T when similar may be output as part of the sign code.
• the expanded spectrum S 1, (k) is expressed by the following equation.
• the gain of the extended spectrum Sl, (k) is determined to coincide with the band of the band Na ⁇ k ⁇ Nb of the second spectrum S2 (k). Specifically, the deviation V of the beam is calculated according to the following equation, and an index obtained by quantizing this value is output as a code ⁇ code from the output terminal 107.
• the extended spectrum SI ′ (k) may be divided into a plurality of sub-bands, and the code and code may be determined independently for each sub-band.
• the expanded spectrum SI ′ (k) may be divided into a plurality of sub-bands, and the code and code may be determined independently for each sub-band.
• T In the step of determining the shape of Sl, (k), T, represented by (Eq. 4) may be determined for each subband and output as a code ⁇ code, or a common T 'may be output. Only one decision may be made and output as a code / code. Then, in the step of determining the gain of the expanded spectrum Sl '(k), the deviation V (j) of the beam is calculated for each subband, and this value is calculated. The index obtained by denitrification is output as a code code via the output terminal 107. The amount of variation of power per subband is expressed by the following equation.
• j represents a subband number
• BL (j) is a frequency index corresponding to the minimum frequency of the j-th subband
• BH (j) is a frequency corresponding to the maximum frequency of the j-th subband Represents an index.
• a scalable function can be realized by outputting a code / code for each sub-band as described above.
• the mode for LPC analysis of the signal of the sampling rate Fy as shown in FIG. But it may be 10 Ob). That is, the signal of the sampling rate Fy can be subjected to LPC analysis to obtain LPC coefficients, and the LPC coefficients can be used to determine the extended spectrum Sl '(k). In this configuration, LPC coefficients can be DFT converted into spectral information, and this spectrum can be used to determine the extended spectrum S1 '(k).
• the circuit scale of the coding apparatus can be reduced, and the amount of processing calculation of code ⁇ can also be reduced.
• LPF low pass filter
• sampling rate conversion alone causes a delay of 5 ms or more.
• the occurrence of such a delay is a problem when it is applied to two-way voice communication, the reaction of the other party seems to be delayed.
• the cutoff characteristic can be made steep even if the order is relatively small, and the delay does not increase as much as the FIR filter.
• IIR filters it is impossible to design filters such as FIR filters in which the amount of delay occurring at all frequencies is constant. This means that, in the scalable code system, when the input signal power is also subjected to the sampling rate conversion, the input signal must be given a certain amount of delay according to the time delay of the signal after the sampling rate conversion.
• the IIR LPF is used, the amount of delay with respect to the frequency is not constant, which causes the problem that the subtraction processing can not be performed accurately.
• the coding apparatus according to the present embodiment can solve these problems that occur in scalable coding systems.
• FIG. 9 is a block diagram showing a main configuration of radio reception apparatus 180 that receives a signal transmitted from radio transmission apparatus 130. Referring to FIG. 9
• the wireless receiver 180 includes an antenna 181, an RF demodulator 182, a decoder 170, a DZA converter 183, and an output device 184.
• the antenna 181 receives a digital encoded acoustic signal as the radio wave W12, generates a digital reception code ⁇ acoustic signal of the electric signal, and provides the same to the RF demodulator 182.
• the RF demodulator 182 demodulates the received coded acoustic signal from the antenna 181 to generate a demodulated coded acoustic signal S 11 and supplies it to the decoder 170.
• Decoding device 170 receives the digital demodulation coded acoustic signal S11 from RF demodulator 182, performs decoding processing to generate digital decoded acoustic signal S12, and outputs it to DZA converter 183. give.
• the DZA converter 183 converts the digital decoded audio signal S12 from the decoder 170 into an analog decoded audio signal and supplies it to the output device 184.
• the output device 184 converts an analog decoded audio signal, which is an electrical signal, into air vibration and outputs it as a sound wave W13 so as to be heard by the human ear.
• FIG. 10 is a block diagram showing an internal configuration of the decoding device 170 described above. Again, the floor The case of decoding a layer-coded signal will be described as an example.
• the decoding device 170 includes an input terminal 171, a separating unit 172, a first layer decoding unit 173, a vector decoding unit 150, and an output terminal 176.
• the code S 11 hierarchically coded from the RF demodulator 182 is input to the input terminal 171.
• the separation unit 172 separates the demodulated coded acoustic signal S11 input through the input terminal 171, and uses the code code for the first layer decoding unit 173 and the code code for the spectrum decoding unit 150.
• Generate First layer decoding section 173 decodes the decoded signal of sampling rate Fx using the encoded code obtained by demultiplexing section 172, and outputs decoded signal S 13 to spectrum decoding section 150.
• Spectrum decoding unit 150 performs spectrum decoding, which will be described later, on code S 14 separated by separation unit 172 and signal S 13 of sampling rate Fx generated by first layer decoding unit 173. To generate a decoded signal S12 of the sampling rate Fy and output it via the output terminal 176.
• FIG. 11 is a block diagram showing an internal configuration of spectrum decoding section 150 described above.
• Spectrum decoding section 150 includes input terminals 152 and 153, frequency domain conversion section 154, band extension section 155, decoding section 156, coupling section 157, time domain conversion section 158, and output terminal 1 59. Have.
• the signal S 13 sampled at the sampling rate Fx is input to the input terminal 152. Further, a code ⁇ code S14 related to the extended spectrum Sl '(k) is input to the input terminal 153.
• the frequency domain conversion unit 154 performs frequency analysis on the time domain signal S13 input from the input terminal 152 with an analysis length of 2′Na to calculate a first spectrum Sl (k).
• the frequency analysis method uses a modified discrete cosine transform (MDCT). MDCT performs analysis by overlapping the previous and next adjacent frames with the analysis frame by half, and using the orthogonal basis where the first half of the analysis frame is an odd function and the second half is an even function. It is characterized by being The first spectrum Sl (k) thus obtained is given to the band extending unit 155.
• MDCT discrete Fourier transform
• DCT discrete cosine transform
• the region where the spectrum can be given is secured so that the band of the first spectrum Sl (k) is 0 ⁇ k ⁇ Nb.
• the first spectrum Sl (k) whose band has been expanded is output to the combining unit 157.
• decoding section 156 decodes code code S14 related to expanded spectrum Sl '(k) input through input terminal 153 to obtain expanded spectrum Sl' (k), and combines them. Output to section 157.
• the combining unit 157 combines the first spectrum Sl (k) given by the band extending unit 155 with the extended spectrum Sl ′ (k). This coupling is realized by inserting the extended spectrum Sl '(k) in the band Na ⁇ k ⁇ Nb of the first spectrum Sl (k). The first spectrum S l (k) obtained by this processing is output to the time domain conversion unit 158.
• the time domain conversion unit 158 performs time domain conversion processing corresponding to the inverse conversion of the frequency domain conversion performed by the spectral encoding unit 100a, and performs multiplication and superposition addition of appropriate window functions to obtain time. Generate signal S12 of the region. The signal S12 in the time domain generated in this manner is output as a decoded signal through the output terminal 159.
• FIG. 12A shows the first spectrum Sl (k) provided by the frequency domain transform unit 154.
• FIG. 12B shows a spectrum obtained as a result of the processing of the band expanding unit 155, and an area where new spectrum information can be stored is secured in a band in which the frequency k is represented by Na ⁇ k ⁇ Nb.
• the size of this novel region is expressed as Nb-Na.
• Nb represents the relationship between the sampling rate Fx of the signal supplied from the input terminal 152, the analysis length 2 ⁇ Na of the frequency domain conversion unit 154, and the sampling rate Fy of the signal decoded by the spectrum decoding unit 150. Depending on the following equation, Nb can be set.
• the sampling rate Fy of the signal decoded by the spectral decoding unit 150 is determined by the following equation. [Number 8]
• FIG. 13 is a diagram showing how a spectrum is processed by combining section 157 and time domain conversion section 158 and a decoded signal is generated.
• Coupling section 157 inserts extension spectrum 31 and &) ⁇ & ⁇ 1 ⁇ ⁇ ? ⁇ ) Into the band of Na ⁇ k ⁇ Nb of the first spectrum Sl (k) whose band is extended. And sends the first spectrum 31 &) (0 ⁇ 1 ⁇ ⁇ ? ⁇ ) Obtained by this process to the time domain conversion unit 158.
• the decoding apparatus of the present embodiment it is possible to decode a signal encoded by the coding apparatus according to the present embodiment.
• the encoding apparatus or the decoding apparatus according to the present embodiment is applied to a wireless communication system
• the coding apparatus or the decoding apparatus according to the present embodiment is described.
• the standardization device can also be applied to a wired communication system as described below.
• FIG. 14A is a block diagram showing the main configuration on the transmitting side when the coding apparatus according to the present embodiment is applied to a wired communication system.
• the same components as those shown in FIG. 5 are denoted by the same reference numerals, and the description thereof will be omitted.
• Wired transmission apparatus 140 includes coding apparatus 120, input apparatus 131, and AZD conversion apparatus 13. 2 and the output is connected to the network N1.
• the input terminal of the AZD conversion device 132 is connected to the output terminal of the input device 131.
• the input terminal of the code device 120 is connected to the output terminal of the AZD converter 132.
• the output terminal of the code device 120 is connected to the network N1.
• the input device 131 converts the sound wave W11 that can be heard by the human ear into an analog signal that is an electrical signal and supplies the analog signal to the AZD conversion device 132.
• the AZD converter 132 converts an analog signal into a digital signal and supplies the digital signal to the coder 120.
• the coder unit 120 codes the input digital signal to generate a code, and outputs the code to the network N1.
• FIG. 14B is a block diagram showing the main configuration on the receiving side when the decoding apparatus according to the present embodiment is applied to a wired communication system.
• the same components as shown in FIG. 9 are assigned the same reference numerals and descriptions thereof will be omitted.
• Wired receiver 190 has receiver 191 connected to network N 1, decoder 170, DZA converter 183, and output device 184.
• An input terminal of the receiving device 191 is connected to the network N1.
• the input terminal of the decoder 170 is connected to the output terminal of the receiver 191.
• the input terminal of the DZA converter 183 is connected to the output terminal of the decoder 170.
• the input terminal of the output device 184 is connected to the output terminal of the DZA converter 183.
• Receiving device 191 receives the digital coded audio signal from network N 1, generates a digital received audio signal, and provides it to decoding device 170.
• Decoding apparatus 170 receives the received acoustic signal from receiving apparatus 191, performs decoding processing on the received acoustic signal, generates a digital decoded audio signal, and provides DZA converting apparatus 183.
• the DZA conversion unit 183 converts the digital decoded speech signal from the decoding unit 170 to generate an analog decoded speech signal and supplies it to the output unit 184.
• the output device 184 converts an analog decoded acoustic signal, which is an electrical signal, into air vibration and outputs it as a sound wave W13 so as to be heard by the human ear.
• FIG. 15 is a block diagram showing the main configuration of decoding apparatus 270 according to Embodiment 2 of the present invention.
• Decoding apparatus 270 has the same basic configuration as decoding apparatus 170 shown in FIG. 10, and the same components are denoted by the same reference numerals and the description thereof will be omitted. .
• the feature of the present embodiment is that the maximum frequency index Nb of the first spectrum Sl (k) (0 ⁇ k ⁇ Nb) after combination is corrected to a desired value Nc, so that the desired sampling rate can be obtained. To generate a decoded signal.
• Spectrum decoding section 250 includes coding code S 14 separated by separation section 172, signal S 13 of sampling rate F x generated by first layer decoding section 173, and input terminal 27 1. Spectral decoding is performed using the coefficient Nc (signal S21) input via the interface. Then, the obtained decoded signal of the sampling rate Fy is output through the output terminal 176.
• FIG. 16 is a block diagram showing an internal configuration of spectrum decoding section 250 described above.
• Coefficient Nc input through input terminal 271 is applied to correction unit 251 and time domain conversion unit 158 a.
• the correction unit 251 determines an effective band of the first spectrum Sl (k) (0 ⁇ k ⁇ Nb) given by the combining unit 157 based on the coefficient Nc given by the input terminal 271 (signal S21). Correct to 0 ⁇ k ⁇ N c. Then, the first spectrum Sl (k) (0 ⁇ k ⁇ Nc) after band correction is given to the time domain conversion unit 158a.
• the time domain conversion unit 158 a performs the first spectrum S l (k) (0 ⁇ k ⁇ Nc) given from the correction unit 251 under the analysis length 2 ⁇ Nc according to the coefficient Nc given via the input terminal 271. ) Is subjected to conversion processing, multiplication and superposition addition of appropriate window functions are performed, and a signal of time domain is generated and output through an output terminal 159.
• FIGS. 17 and 18 are diagrams for describing the process of the correction unit 251 in more detail.
• FIG. 17 shows the process of the correction unit 251 in the case of Nc Nb.
• the band of the first spectrum Sl (k) (signal S21) given by the coupling unit 157 is 0 ⁇ k ⁇ Nb. That
• the correction unit 251 deletes the spectrum in the range of Nc ⁇ k ⁇ Nb so that the band of the first spectrum Sl (k) is 0 ⁇ k ⁇ Nc.
• the first spectrum Sl (k) (0 ⁇ k ⁇ Nc) (signal S22) obtained as a result is supplied to the time domain conversion unit 158a, and the time domain decoded signal S23 is generated.
• FIG. 18 similarly shows the processing in the case of force Nc> Nb, which is the processing of the correction unit 251.
• the band of the first spectrum Sl (k) (signal S25) given from the coupling unit 251 is 0 ⁇ k ⁇ Nb as in FIG.
• the correction unit 251 extends the Nb ⁇ k ⁇ Nc band so that the band of the first spectrum Sl (k) is 0 ⁇ k ⁇ Nc, and assigns a specific value (for example, a zero value) to the region.
• the first spectrum Sl (k) (0 ⁇ k ⁇ Nc) (signal S26) obtained as a result is applied to the time domain conversion unit 158a, and a time domain decoded signal S27 is generated.
• spectrum decoding section 250 will be further described using FIG. 19, FIG. 20A and FIG. 20B.
• the code / code input via the input terminal 153 fluctuates for each frame. That is, in the band of the first spectrum Sl (k) output from the combining unit 157, 0 ⁇ k ⁇ Na (band R1), 0 ⁇ k ⁇ Nb1 (band R2), 0 ⁇ as shown in FIG. It is assumed that there are three bands of k ⁇ Nb2 (band R3) (however, NaNb ⁇ Nb2), and one of these bands is selected for each frame.
• FIG. 20A is a diagram for explaining the operation of spectrum decoding section 250 when coefficient Nc is equal to Nb2
• FIG. 20B is a diagram for explaining the operation of spectrum decoding section 250 when coefficient Nc is equal to Nbl. is there.
• the band of the spectrum obtained in the i-th frame is any one of Rl, R2, and R3.
• processing 1 inserts a zero value into the band of Nbl ⁇ k ⁇ Nb2
• processing 2 inserts a zero value into the band of Na ⁇ k ⁇ Nb2
• processing 3 removes the band of Nbl ⁇ k ⁇ Nb2
• processing 4 represents the processing of inserting a zero value into the band of Na ⁇ k ⁇ Nbl.
• the correcting unit 251 performs the first spectrum Sl (k) (0 ⁇ k ⁇ Nb2) is output to the time domain conversion unit 158a.
• the spectral band is R2, that is, the band of the first spectrum S l (k) is 0 ⁇ k ⁇ Nb 1. Therefore, the correction unit 251
• the first spectrum Sl (k) (0 ⁇ k ⁇ Nb2) is converted to a time domain conversion unit after the band of the one-vector Sl (k) is expanded to Nb2 and a zero value is inserted in the band Output to 158a.
• the correction unit 251 selects the first spectrum Sl. After expanding the band of (k) to Nb2 and inserting a zero value in the range of Na ⁇ k ⁇ Nb2, the first spectrum Sl (k) (0 ⁇ k ⁇ Nb2) is converted to the time domain conversion unit 158a. Output to
• the band of the spectrum is R2 in the second frame to the fourth frame and the ninth frame, that is, the band of the first spectrum Sl (k) is 0 ⁇ k ⁇ Nbl
• the first spectrum S l (k) (0 ⁇ k ⁇ Nb 1) is output to the time domain conversion unit 158 a without any processing.
• the spectral band is R3, that is, the band of the first spectrum Sl (k) is 0 ⁇ k ⁇ Nb2, and therefore the correction unit 251 After removing the band of Nbl) k ⁇ Nb2, the first spectrum Sl (k) (0 ⁇ k ⁇ Nbl) is output to the time domain conversion unit 158a.
• the correction unit 251 determines the first spectrum Sl After expanding the band of (k) to Nbl and inserting a zero value into the band of Na ⁇ k ⁇ Nbl, the first spectrum Sl (k) (0 ⁇ k ⁇ Nbl) is converted to the time domain conversion unit 158a. Output to
• FIG. 21 is a diagram showing the main configuration of a communication system according to Embodiment 3 of the present invention.
• the feature of the present embodiment is to cope with the case where the effective frequency band of the first extra frame Sl (k) received on the receiving side fluctuates with time depending on the condition of the communication network (communication environment). It is to be.
• Layer code section 301 performs the layer coding process described in Embodiment 1 on the input signal of sampling rate Fy, and generates a scalable coded code.
• the generated code ⁇ code power band 0 ⁇ ⁇ ⁇ ⁇ k ⁇ information about Ne (R 31), band Ne ⁇ k ⁇ Nf (R 32), and band N f ⁇ k ⁇ Ng (R 33) It shall be composed.
• the hierarchical code unit 301 supplies the code unit code to the network control unit 302.
• the network control unit 302 transfers, to the hierarchy decoding unit 303, the coding sequence provided by the hierarchy coding unit 301.
• the network control unit 302 discards part of the coding code transferred to the hierarchical decoding unit 303 according to the state of the network. Therefore, the coded code input to the hierarchical decoding unit 303 is the coded code composed of the information R31 to R33 and the coded code of the information R33 when there is no code to be discarded.
• Hierarchical decoding section 303 applies the hierarchical decoding method shown in Embodiment 1 or 2 to a given coded code to generate a decoded signal.
• the sampling rate of the decoded signal can be set by a desired coefficient Nc, and the sampling rate Fz of the decoded signal is FyNc /. It becomes Ng.
• Embodiment 4 even if the effective frequency band of the first extraneous signal Sl (k) received on the receiving side varies with time depending on the conditions of the communication network, The side can stably obtain a decoded signal of a desired sampling rate.
• FIG. 22 is a diagram showing the main configuration of a communication system according to Embodiment 4 of the present invention.
• the feature of the present embodiment is that one code ⁇ code generated by one layer code unit can be decoded for a plurality of layer decoding units having different sampling rates (different in decoding ability) that can respectively decode. Even if they are transmitted simultaneously, the receiving side is to respond to this and obtain decoded signals of different sampling rates.
• Layer code section 401 applies the coding process shown in the first embodiment to the input signal of sampling rate Fy to generate a scalable coded code.
• the generated code ⁇ code is information on band 0 ⁇ ⁇ k ⁇ Nh (R41), information on band Nh ⁇ k ⁇ Ni (R42), information on band Ni ⁇ k ⁇ Nj (R43) ⁇ It shall consist of this.
• Layer code section 401 applies this encoded code to first layer decoding section 402-1, second layer decoding section 402-2 and third layer decoding section 402-3. .
• the first layer decoding unit 402-1, the second layer decoding unit 402-2, and the third layer decoding unit 402-2 are implemented for the given encoded code.
• a decoded signal is generated by applying the hierarchical decoding scheme shown in the first embodiment or the second embodiment.
• the sampling rate F2 of this decoded signal is Fy'NiZNj.
• the sampling rate F3 of this decoded signal is Fy'NhZNj.
• the transmitting side can transmit the code without considering the decoding capability of the receiving side, it is possible to reduce the load on the communication network. Can.
• decoded signals of these multiple types of sampling rates have a simple configuration and few
• V can be generated with the amount of operation.
• the code device or the decoding device according to the present invention can also be installed in a communication terminal device and a base station device in a mobile communication system, and thereby has the same operation effect as described above.
• a communication terminal apparatus and a base station apparatus can be provided.
• the encoding apparatus and the decoding apparatus according to the present invention have an effect of realizing scalable encoding with a simple configuration and a small amount of calculation, and can be applied to the use of a communication system such as an IP network.

Landscapes

• Engineering & Computer Science (AREA)
• Computational Linguistics (AREA)
• Quality & Reliability (AREA)
• Signal Processing (AREA)
• Health & Medical Sciences (AREA)
• Audiology, Speech & Language Pathology (AREA)
• Human Computer Interaction (AREA)
• Physics & Mathematics (AREA)
• Acoustics & Sound (AREA)
• Multimedia (AREA)
• Compression, Expansion, Code Conversion, And Decoders (AREA)
• Transmission Systems Not Characterized By The Medium Used For Transmission (AREA)

Priority Applications (5)

Applications Claiming Priority (2)

Related Child Applications (2)

Publications (1)

Family

ID=34386230

Family Applications (1)

Country Status (5)

Families Citing this family (28)

Citations (2)

Family Cites Families (25)

• 2003
  • 2003-09-30 JP JP2003341717A patent/JP4679049B2/ja not_active Expired - Lifetime
• 2004
  • 2004-09-29 US US10/573,812 patent/US7756711B2/en active Active
  • 2004-09-29 CN CN201310064311.XA patent/CN103177730B/zh not_active Expired - Lifetime
  • 2004-09-29 EP EP04788282A patent/EP1669981A4/en not_active Withdrawn
  • 2004-09-29 CN CN2004800256756A patent/CN1849647B/zh not_active Expired - Lifetime
  • 2004-09-29 WO PCT/JP2004/014215 patent/WO2005031705A1/ja not_active Ceased
  • 2004-09-29 EP EP10151098A patent/EP2172931A1/en not_active Withdrawn
• 2010
  • 2010-02-18 US US12/708,290 patent/US8195471B2/en active Active
• 2012
  • 2012-05-03 US US13/463,653 patent/US8374884B2/en not_active Expired - Fee Related

Patent Citations (2)

Non-Patent Citations (5)

Also Published As

Similar Documents

Legal Events