JP3131437B2 - Error compensation method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial application field) The present invention relates to an error compensation method used for suppressing sound quality deterioration due to a line error in a mobile communication system such as a car telephone.

(Prior Art) In a mobile communication system such as a mobile phone, there is a problem that a bit error occurs on a propagation path and voice quality is remarkably deteriorated. Therefore, in order to suppress the deterioration of the voice quality due to bit errors, the method of limiting the amplitude of the decoded voice when the error rate of the transmission path is high is "encoder" (HL-209827).
(Reference 1). Hereinafter, this content will be described.

FIG. 8 is a block diagram showing a configuration of a decoding device including the conventional error compensation method described in Reference 1.

In FIG. 8, reference numeral 501 denotes an input terminal,
02, connected to a frame synchronization detection unit 504 and an error rate measurement unit 506. The error correction unit 502 is connected to the decoder 503, and the decoder 503 is connected to the variable amplitude limiter 507, and the variable amplitude limiter 507 is connected to the output terminal 510. The frame synchronization detecting section 504 is connected to the error correcting section 502, the decoder 503, and the error rate measuring section 506, and the error rate measuring section 506 is connected to the variable amplitude limiter 507.

Next, the operation of the above embodiment will be described. In the above embodiment, from the voice data input from the input terminal 501,
The frame synchronization detection unit 504 acquires the frame timing. The error correction unit 502 corrects an error in the audio data input from the input terminal 501 for each frame, and the decoder 503 decodes the error-corrected audio data and outputs an audio signal. The error rate measurement unit 506 measures the error rate of the transmission path from the frame synchronization bits of the audio data input from the input terminal 501, and controls the variable amplitude limiter 507. Variable amplitude limiter 507
When the error rate of the transmission path is lower than a certain value, the decoded voice output by the decoder 503 is output to the output terminal 510 as it is, and when the error rate becomes higher than the certain value, the decoded voice output by the decoder 503 is amplified. The output is limited and output to the output device 510.

As described above, even in the above-described conventional error compensation method, when the error rate of the transmission path increases, the amplitude of the output voice is limited, so that the noise level due to the transmission path error can be reduced.

(Problems to be Solved by the Invention) However, in the above-described conventional method, when the error rate is high, the decoding output is simply narrowed down. There is a problem that the sound becomes harsh and unpleasant and unpleasant.

The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide an error compensation system capable of reproducing a natural voice signal with less noise even when there are many transmission path errors. I do.

[Means for Solving the Problems] In order to achieve the above object, the present invention provides, on the encoding side, means for selecting a coded sequence into a plurality of sequences, and error correction for each of the plurality of sequences. Means for performing encoding; means for decomposing a transmitted coded sequence into a plurality of sequences on the decoding side; means for performing error correction and error detection for each of the plurality of sequences; Means for decoding a sequence and a decoded audio signal, and means for accumulating while updating information representing a drive signal of a synthesis filter of the decoded coded sequence when there is no error equal to or greater than a predetermined value as a result of error detection, Means for accumulating and updating the prediction parameters of the decoded coded sequence when there is no error equal to or greater than a certain value as a result of the error detection, and a gain of the drive signal of the immediately preceding synthesis filter and a drive signal of the current synthesis filter. gain Check whether the difference is below a certain value, and if it is above a certain value, means to limit the current gain so as to be below a certain value, and if there is more than a certain error in the immediately preceding coded signal sequence Means for restricting the pitch gain to a constant value when present and when there is no error equal to or more than a predetermined value in the current coded signal sequence, and inputting the stored driving signal and the stored prediction parameters to convert the speech signal A synthesis filter for synthesizing, and outputting a decoded speech signal decoded by the decoding means when no error exceeding a certain level exists as a result of the error detection, and synthesizing by a synthesis filter when an error exceeding a certain level exists as a result of the error detection. Means for outputting a converted audio signal.

(Operation) On the encoding side, the encoded signal sequence is classified and selected into a plurality of sequences, and an error correction code is added to each sequence.

On the decoding side, the transmitted coded sequence is decomposed into a plurality of sequences, and the coding side performs error correction on each of the sequences in a process reverse to error correction coding, and performs error detection. . The error-corrected coded sequence is decoded by the decoding means, and the parameters of the synthesis filter (the same as the prediction parameters), the driving signal of the synthesis filter, and the decoded audio signal are obtained.

When a certain error or more is not detected, the driving signal of the synthesis filter and the prediction parameter are updated and accumulated, and when the difference between the gain of the driving signal of the previous frame and the gain of the driving signal of the current frame is equal to or more than a certain value. The drive signal gain of the current frame is limited so that the difference is within a certain value, and there is a certain error or more in the coded signal sequence of the previous frame and a certain or more error in the coded signal sequence of the current frame. Limits the pitch gain to a constant value when does not exist.

Then, after synthesizing the speech from the accumulated driving signal and the prediction parameter, if the error detection does not detect a certain or more error, the decoded sound signal is output, and a certain or more error is detected. In this case, a synthesized audio signal is output.

(Example) Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

1 and 2 are block diagrams of an encoding device and a decoding device according to one embodiment of the present invention, respectively.

In FIG. 1, reference numeral 10 denotes an input terminal to which a digitized audio signal sequence is input. The encoding unit 20 encodes the input audio signal sequence and outputs the encoded signal sequence to the bit selection circuit 21. The bit selection circuit 21 divides the coded bit sequence into a prediction parameter (same as a synthesis filter parameter) and a bit string of a drive signal parameter of the synthesis filter, and separates them into error correction coding circuits.
Enter 30,31. In the error correction coding circuit, an error correction code is added to the input sequence and connected to the multiplexer circuit. As the error correction code, various known codes can be used, but a burst error occurs in mobile communication. Therefore, a Reed-Solomon code having a high correction capability for the burst error is used in this embodiment. The Reed-Solomon codes used in the error correction encoding circuits 30 and 31 have different correction capabilities depending on the sensitivity of the input sequences 30 and 31 to errors. By doing so, efficient error correction becomes possible. The buffer 40 accumulates encoded signal sequences for two frames and connects them to the interleave unit. In the interleave section, for example, as shown in FIG. 3, two frames of the frame B 'are formed from a part of the coded signal sequence of the frame A and a part of the coded signal sequence of the frame B. A frame for transmission is configured by replacing the coded signal sequence. By doing so, the burst error in one frame is randomized to two frames, so that it is possible to reduce the deterioration of the quality of synthesized speech due to a line error. 60 is an output terminal of the encoding device. The configuration and operation of the encoding unit 20 will be described later.

 Next, the decoding device will be described.

In FIG. 2, a buffer 110 accumulates an encoded signal sequence input from an input terminal 100 for two frames. Next, the ding leave section 120 performs a process reverse to that of the tent leave 50, and forms a coded signal sequence of the frame A from the frames A 'and B' in FIG. In 121, the coded signal sequence is decomposed into a bit sequence of a prediction parameter and a drive signal parameter, and each is output to error correction units 130 and 131. In steps 130 and 131, Reed-Solomon error correction decoding is performed to correct the error of the input signal sequence. At the same time, it calculates the syndrome and outputs it to the error detection units 140 and 141. The error correction code is described in detail in "Code Logic" by Takuma et al., Corona Corporation, and will not be described here.

The error detection units 140 and 141 calculate the number of erroneous symbols from the syndromes calculated by the error correction units 130 and 131, and when the number is greater than the error correction capability, E ₁ = 1, E ₂ = 1, and E ₁ = 0,
E ₂ = 0 is output to the error state determination circuit 142. Note that E ₁ and E
₂ indicates the outputs of the error detectors 140 and 141, respectively.

The error state determination circuit 142 determines an error state as follows based on the outputs of the error detection circuits 140 and 141 and the result of one frame before.

When E ₁ = 0 and E ₂ = 0, state S ₀ = no error E ₁ = 0 and E ₂ = 1 state S ₁ E ₁ = 1 and E ₂ = 0 state S ₂ E ₁ = 1 and E ₂ = When the state is 1, the state S ₃ When the state one frame before is other than S ₀ and the state of the current frame is S ₀ , the state = S ₄ The output of the error state determination circuit is the parameter memory 160, the drive signal memory 170, the cutout circuit 180, switch circuit 210
Connected to

The decoding unit 150 decodes the error-corrected coded signal, outputs the parameters of the synthesis filter (LPC parameter and pitch parameter) to the parameter memory 160, switches the driving signal of the synthesis filter to the driving signal memory 170, and switches the decoded audio signal. Output to the circuit 210. The configuration and operation of the decoding unit will be described later.

Parameter memory 160, the output of the error state determining circuit 142 when S ₀ or S _1, accumulates while updating the parameters of the synthesis filter which is the output of the decoding unit 150. Similarly driving signal memory 170 output of the determination circuit 142 when S ₀ or S _2, accumulates while updating the drive signal output from the decoding unit. Extraction circuit 180, the output of the error state determining circuit 142 is S ₁ or
Cut the last one pitch period samples of the signal from the sample point (Error = 1 and becomes immediately preceding frame) of the drive signal stored in the drive signal memory 170 when the S _3, outputs it to the driving signal generating circuit 190 I do. Drive signal generation circuit 1
In 90, the drive signal for one pitch is repeatedly connected while multiplying the drive signal for one pitch cycle by a smaller coefficient to generate a drive signal for one frame and output to the synthesis filter 200. The synthesis filter 200 receives the drive signal generated by the drive signal generation circuit 190 and synthesizes the audio signal using the parameters stacked in the parameter memory 160, that is, the parameters of the frame on which the error correction has been performed. Circuit 21
Output to 0. The switch circuit 210 has a range in which occurrence of an error can be corrected, that is, the output of the error state determination circuit is S _0.
Outputs a decoded speech signal output from the decoding unit 150 to the output terminal 220 when the occurrence of an error the error correction capability than, i.e., the output of the error state determining circuit at the output of the synthesis filter 200 when a non S ₀ A certain synthesized voice signal is output to 200.

Next, the configuration and operation of the encoding unit and the decoding unit in FIGS. 1 and 2 will be described.

FIG. 4 and FIG. 5 are block diagrams each showing an example of the configuration of the encoding unit and the decoding unit.

In FIG. 4, the frame buffer 311 is connected to the input terminal 310.
Is a circuit for accumulating one frame of the audio signal inputted to each block. Each block shown in FIG. 4 performs the following processing for each frame or each subframe using the frame buffer 311.

The prediction parameter calculation circuit 312 calculates a prediction parameter using a known method. When the prediction filter 314 is configured by continuously connecting the long-term prediction filter 351 and the short-term prediction filter 352 as shown in FIG. 6, the prediction parameter calculation circuit 312 calculates the pitch period, the pitch prediction coefficient and the linear prediction coefficient (α Parameters or K parameters, all
Called LPC parameters. ) Is calculated by a known method such as an autocorrelation method or a covariance method. The calculation method is described, for example, in the aforementioned reference 2 ("Digital Speech Processing" by Sadateru Furui, published by Tokai University Press, 1985). The calculated prediction parameters are input to the prediction parameter coding circuit 313. The prediction parameter encoding circuit 313 encodes the prediction parameter based on a predetermined number of quantization bits, and outputs a decoded value to the decoding circuit 327. Stabilization circuit 327
Checks whether the coefficient of the input prediction filter is stable and stabilizes it. The prediction parameter is LSP (Line Sp
ectrum Pair), the stability can be checked by checking whether the parameter values are arranged in order from the lowest order. When the parameters are not arranged in order, or when the interval between parameter values is narrow, Since becomes unstable or easily oscillates, it instructs the encoding circuit 313 to perform encoding again so that the parameter values are arranged in order and the interval is longer than a certain value. The encoding circuit 313 encodes the LSP parameter again and outputs the result to the stabilization circuit 327. The stabilizing circuit 327 repeats the instruction to the encoding circuit 313 until all the parameters become stable.
The stabilization circuit 327 outputs the sign of the stabilized prediction parameter to the multiplexer 325, and outputs the decoded value to the prediction filter 314, the synthesis filter 315, and the perceptual weight filter 320.
Output to The prediction filter 314 calculates a prediction residual signal using the input speech signal and prediction parameters as inputs, and outputs the signal to the density pattern selection circuit 315.

In this embodiment, the density pattern selection circuit 315 first divides a prediction residual signal of one frame into a plurality of subframes, and calculates a sum of squares of the prediction residual signal of each subframe. Next, a density (pulse interval) pattern of the drive pulse train signal in each subframe is obtained based on the sum of squares of the prediction residual signal. One example of a specific method is
Two types of density patterns having the shortest pulse interval, the number of subframes having a long pulse interval and the number of subframes having a short pulse interval are set in advance as the density pattern, and the two types of the prediction residual signal are set.
This is a method of selecting a density pattern in which the pulse interval becomes shorter in the order of the subframe having a larger sum of squares.

The gain calculation circuit 327 receives the information of the selected density pattern as input, and sets the gain of the drive signal to, for example, the standard deviation of the prediction residual signal of all subframes with short pulse intervals and the prediction residual signal of all subframes with long pulse intervals. Two types are obtained using the standard deviation of The obtained density patterns and gains are encoded by encoding circuits 316 and 328, respectively, and input to multiplexer 325, and their decoded values are input to drive signal generation circuit 317. The drive signal generation circuit 317 converts the density pattern and gain input from the encoding circuits 316 and 328, the normalized amplitude of the drive pulse input from the codebook 324, and the phase of the drive pulse input from the phase search circuit 322. Based on this, a drive signal having a variable density is generated for each subframe.

The gain of the drive pulse in the m-th subframe is
G ^(m) , the normalized amplitude of the drive pulse is g ₁ ^(m) , the number of pulses is Q _m , the pulse interval is D _m , the phase of the pulse is K _m , and the length of the subframe is L, The drive signal e _X ^(m) (n) can be described by the following equation.

Note that the phase _Km is the leading position of the pulse in the subframe. Σ (n) is a Kronecker delta function.

The drive signal generated by the drive signal generation circuit 317 is input to the synthesis filter 318, and the synthesized signal is output. The synthesis filter 318 has a relationship of an inverse filter with the prediction filter 314. The error between the input speech signal and the synthesized signal, which is the output of the subtraction circuit 319, is input to the square error calculation circuit 321 after its spectrum is transformed by the perceptual weighting filter 320. The perceptual weight filter 320 has a transfer function This filter is used to utilize the audible marking effect as in the case of the weighting filter in the conventional example.

The square error calculation circuit 321 calculates the sum of squares of the perceptually weighted error signal for each code vector stored in the codebook 324 and for each phase of the drive pulse output from the phase search circuit 322, and calculates Phase search circuit 322
Is output to the amplitude search circuit 323. The amplitude search circuit 323 searches the codebook 324 for a codeword index that minimizes the sum of squares of the error signal for each phase of the drive pulse output from the phase search circuit 322, and searches for the minimum of the sum of squares. The value is output to the phase search circuit 322, and the index of the codeword that minimizes the sum of squares is held. The phase search circuit 322 receives the information of the selected density pattern as input, changes the phase K _m of the drive pulse train in the range of 1 ≦ K _m ≦ D _m , gives the value to the drive signal generation circuit 317, and provides D _m The minimum value of the sum of squares of the error signal determined for each of the phases is received from the amplitude search circuit 323, and the phase corresponding to the smallest sum of the squares of the D _m minimum values is output to the multiplexer 325. At the same time, the amplitude search circuit 323 is notified of the phase. The amplitude search circuit 323 outputs a codeword index corresponding to the phase to the multiplexer 325.

The multiplexer 325 multiplexes the prediction parameter, the density pattern, the gain, the sign of the phase and the amplitude of the drive pulse, and outputs the multiplexed signal to the transmission line via the output terminal 326. In addition,
The output of the subtraction circuit 319 may be directly input to the square error calculation circuit 321 without passing through the perceptual weight filter 320.

Next, the decoding device shown in FIG. 5 will be described. In FIG. 5, a demultiplexer 331 separates a code input from an input terminal 330 into a prediction parameter, a density pattern, a gain, and a code of a phase and an amplitude of a driving pulse. The encoding circuits 332 and 337 decode the density pattern of the driving pulse and the sign of the gain of the driving pulse, respectively, and output them to the driving signal generating circuit 333. However, the gain of the drive signal is not output directly to the drive signal generation circuit but is output first to the limiter circuit 341. The limiter circuit holds the gain of one frame before, checks whether the gain of the current frame is within a certain value compared to the gain of the previous frame, and if it is more than the certain value, limits the gain to a certain value. By doing so, abnormal sounds caused by line errors can be suppressed. The code book 355 is the same as the code book 324 in the encoding device of FIG. 4, and outputs a code word corresponding to the index of the amplitude of the transmitted drive pulse to the drive signal generation circuit 333. The prediction parameter decoding circuit 336 is the prediction parameter coding circuit 313 shown in FIG.
And decodes the code of the prediction parameter coded by (1) and outputs it to the stabilization circuit 342. The stabilization circuit 342 performs a stabilization process of the prediction parameter based on the input prediction parameter and an error state input from the input terminal 343. FIG. 7 shows a flowchart of the stabilization processing. In the stabilization processing, first, the stabilization processing of the LSP parameters in the input prediction parameters is performed in the same manner as the stabilization circuit 327 in FIG. 4 to determine whether the magnitudes of the LSP parameter values are arranged in order from the LSP having the lower order. Check, if not arranged in order or if the interval between the LSP parameters that are the props is narrow, while adding a certain value, the parameter values are arranged in order, and
The interval between the parameters in the propeller should be more than a certain value. Next, when the error state is S ₄ , that is, when the state of the coded signal sequence of the previous frame changes from a state where an error of a certain level or more occurs to a state where no error of a certain level or more occurs in the current frame, the pitch gain is set to a constant value of 1 or less To stabilize the pitch filter which is a part of the synthesis filter 334. By the above stabilization processing, oscillation and instability of the synthesis filter caused by the line error can be prevented, and deterioration of the synthesized voice due to the line error can be suppressed. The stabilized prediction parameter is connected to a synthesis filter, and a pitch period, which is a part of the prediction parameter, is output to an output terminal 346. Drive signal generation circuit 333
Generates a drive signal having a variable density in subframe units based on the gain, amplitude, and phase of the input drive pulse in the same manner as the drive signal generation circuit 317 in the encoding device, and outputs the generated drive signal to the synthesis filter 334. Output to terminal 345. Synthesis filter 33
4 is the same as the synthesis filter 318 in the encoding device,
Receives the drive signal and the prediction parameters and buffers the synthesized signal.
Output to 338. The buffer 338 combines the input signals for each frame and outputs the combined signal to the output terminal 339.

〔The invention's effect〕

As described above, according to the present invention, when an error that cannot be corrected by the error correction capability or more occurs in a coded signal sequence, the error correction is performed not by simply limiting the output level. Since the audio signal is reproduced using the parameters of the synthesis filter of the performed frame and the drive signal, there is little noise even in a bad environment where the error rate of the transmission path is high, such as in mobile communications, and the sound is not interrupted. Natural sound can be obtained. Further, in the present invention, since interpolation processing is performed on a frame in which error correction cannot be performed at the level of the drive signal, even if a discontinuous point occurs in the waveform of the drive signal due to the interpolation processing, the signal passes through the synthesis filter. There is an effect that the discontinuous point is smoothed and the discontinuous sound cannot be recognized in the reproduced voice.

Furthermore, in the present invention, the level of the synthesized sound is attenuated at a speed based on the pitch period in the frame in which the error cannot be corrected, and the synthesis filter is stabilized, so that the quality deterioration due to the line error is greatly improved. be able to.

[Brief description of the drawings]

FIG. 1 is a block diagram of an encoding device according to one embodiment of the present invention, FIG. 2 is a block diagram of a decoding device according to one embodiment of the present invention, and FIG. 3 is an operation of an interleave unit in FIG. FIG. 4 is a block diagram showing an example of a configuration of an encoding unit in FIG. 1, FIG. 5 is a block diagram showing an example of a configuration of an encoding unit in FIG. 2, and FIG.
FIG. 4 is a block diagram showing the configuration of the prediction filter in FIG. 4, FIG. 7 is a flowchart of a stabilization process, and FIG. 8 is a block diagram of a coding apparatus including a conventional error compensation method. 20: coding section, 30: error correction coding section, 30, 110: buffer, 50: interleave section, 120: deinterleave section, 130: error correction section, 140: error detection section, 150: ... Decoding unit, 160 ... Parameter memory, 170 ... Drive signal memory, 180 ... Cutout circuit, 190 ... Drive signal generation circuit, 200 ... Synthesis filter, 210 ... Switch circuit, 312 ... Prediction parameter calculation circuit , 313: coding circuit, 314: prediction filter, 315: density pattern selection circuit, 316: coding circuit, 317: drive signal generation circuit, 318: gain calculation circuit, 319: subtraction circuit, 320: perceptual weighting filter, 321: square error calculation circuit, 322: phase search circuit, 323: amplitude search circuit, 324: code book, 325: multiplexer, 328: coding circuit, 331: … Demultiplexer, 332,337,336 …… Decoding circuit, 33 3 ... drive signal generation circuit, 334 ... synthesis filter, 338 ... buffer, 351 ... long time prediction filter, 352 ... short time prediction filter, 502 ... error correction unit, 503 ... decoder, 504 ... … Frame synchronization detector, 506… Error rate measurement unit, 507… Variable amplitude limiter.

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-1-177225 (JP, A) JP-A-62-117422 (JP, A) JP-A-50-67013 (JP, A) JP-A-3- 245370 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) H04B 14/00-14/04 H03M 13/00

Claims (1)

(57) [Claims] An encoding side includes means for selecting an encoded sequence into a plurality of sequences, and means for performing error correction encoding for each of the plurality of sequences. Means for decomposing into a plurality of streams; means for performing error correction and error detection for each of the plurality of streams; means for decoding an error-corrected coded sequence to decode a decoded speech signal; Means for updating and storing information representing the drive signal of the synthesis filter of the decoded coded sequence when there is no error, and the decoded coded sequence when there is no error exceeding a certain value as a result of error detection. Means for updating and accumulating the prediction parameters of, and checking whether the difference between the gain of the drive signal of the immediately preceding synthesis filter and the gain of the drive signal of the current synthesis filter is below a certain value, and If Means for limiting the current gain so as to be equal to or less than a certain value, and a constant pitch gain when there is an error of not less than a certain value in the immediately preceding encoded signal sequence and no error of not less than a certain value in the current encoded signal sequence. Means for limiting to a value, a synthesis filter for inputting the stored drive signal and the stored prediction parameter to synthesize a sound signal,
Outputs a decoded speech signal decoded by the decoding means when no error exceeding a certain value exists as a result of error detection, and outputs a sound signal synthesized by a synthesis filter when a certain error or more exists as a result of error detection. Means for performing error compensation.