KR20000068950A - Transmission system with improved reconstruction of missing parts - Google Patents

Abstract

In the transmission system, the signal to be transmitted by the first node 2 is applied to the encoder 8 and transmitted by the transmitter 10 to the second node 6. At the second node 6, a signal from the transmission medium 4 is received by the receiver 12 and transmitted directly via the interpolator 14 to the selector 18. By delaying signal decoding in the event of a transmission error, which results in missing frames, it is possible to complete the missing frames by interpolation.

Description

Transmission system with improved reconstruction of missing parts}

A transmission system according to the introduction is known from US Pat. No. 4,907,277.

Such a transmission system is used in applications where the source signal must be transmitted through a transmission system having a limited transmission capability or stored in a storage medium having a limited storage capacity. Such applications include the transmission of voice signals over the Internet, the transmission of voice signals from mobile phones to base stations and from base stations to mobile phones, and the storage of voice signals in CD-ROM, solid-state memory or hard disk drives. And so on.

In order to reduce the bit rate of the signal to be transmitted or stored, a source encoder is used which derives the encoded signal from the input signal. Such source encoders may operate according to different operating principles. Examples of such operating principles include PCM, DPCM, RPE, CELP, transform coding (subband coding, DCT), and the like.

In the case of an unreliable transmission medium such as a mobile radio channel, encoded signal portions may not be correctly received by the second node. In order to maintain the quality of the decoded signal, the second node includes an interpolator to complete the missing portion of the recovered signal. The above-mentioned U.S. patent discloses a method of obtaining a quality, complete missing part by interpolating the part surrounding the missing part. A disadvantage of the transmission system according to the US patent is that the decoding delay is increased due to the interpolation process.

The present invention includes a first node having a source encoder for deriving an encoded signal from an input signal, the first node further comprising a transmitter for transmitting the encoded signal to a second node via a transmission medium. And the second node comprises recovery means for deriving a recovery signal from the encoded signal.

The invention also relates to a node, a recovery means and a recovery method.

1 shows a first embodiment of a transmission system according to the invention in which interpolation is used to complete a received signal.

2 shows a detail of the interpolator 14 according to FIG. 1.

3 is a detailed view of an apparatus in which interpolator 14 and selector 18 perform a function.

4 is a timing diagram for recovery of a signal received in the transmission system according to FIG.

5 is a detailed view of the error detector 16.

6 is a second embodiment according to the present invention for requesting retransmission to be used to complete a received signal.

7 is a first possible timing diagram for recovery of a signal received in the transmission system according to FIG. 6;

8 is a second possible timing diagram for recovery of a signal received in the transmission system according to FIG. 6;

9 is an embodiment of an audio processing function at a second node according to the present invention;

It is an object of the present invention to provide a transmission system according to the introduction in which the average delay value is reduced.

In order to achieve the above object, in the transmission system according to the present invention, the recovery means is configured to immediately provide the recovered signal if the quality of the received encoded signal satisfies a certain quality criterion, and the recovery means And if the quality of the encoded signal does not satisfy the quality criterion, provide the recovered signal after a recovery delay.

The present invention is based on the recognition that increasing the recovery delay under poor signal conditions increases the likelihood of recovering the source signal under poor signal conditions. Increasing the recovery delay enables interpolation to complete the missing portion of the decoded signal or requests the transmission of the missing portion of the encoded signal by the first node. It is observed that the recovery delay is only increased under poor signal conditions to make a signal delay that is not greater than necessary under good signal conditions.

In one embodiment of the invention, the quality criterion is the integrity of the received encoded signal, and wherein the recovery means comprises completion means for completing the recovered signal if the received signal is not complete. do. According to this embodiment, the source signal is directly recovered when the received signal is complete at the given moment in time. If a portion of the decoded signal is missing, the recovery delay is increased, allowing for retransmission to allow interpolation or to complete the missing portion of the encoded signal. It is observed that interpolation does not have to be performed on an encoded signal but it is also possible to perform interpolation on a decoded frame of the signal.

Another embodiment of the invention is characterized in that the recovery means is arranged to increase the recovery delay in accordance with the quality of the received encoded signal.

By reducing the quality of the received encoded signal and at the same time increasing the recovery delay, it becomes easier to recover the source signal better. If a large portion of the received encoded signal is missing, for example, additional time is needed to complete the signals with interpolation or retransmission by the first node.

Another embodiment of the present invention is characterized in that the recovery means is installed to reduce the recovery delay after detecting that there is no recovered signal.

If it is determined that the source signal is not present, by reducing the recovery delay, the fact that the recovery delay is not greater than absolutely necessary for obtaining a good recovery of the source signal is obtained.

In another embodiment of the present invention, the source signal includes a voice signal, and the second node introduces comfort noise having a level that increases as the quality of the received encoded signal decreases. It comprises a comfort noise introduction means.

By lowering the quality of the received encoded signal and introducing more comfort noise, the degradation of the transmission quality results in a state that can be similarly heard using analog transmission systems that are familiar to the user.

In another embodiment of the present invention, the transmission system includes a full dual voice transmission system, and wherein the second node introduces a side tone having a level decreasing with the recovery delay. Tone introduction means is characterized in that it comprises a.

The importance of increasing the delay of decoding is that the adverse effect of echo is more pronounced. The source of the echo is the introduction of side tones that are very common in telephone systems. Side tones are introduced to provide the user with some acoustic feedback of the transmitted audio signal. By reducing the level of the side tone signal, the adverse effect of the echo is reduced.

The present invention will now be described with reference to the drawings.

In the transmission system according to FIG. 1, a signal to be transmitted by the node 2 is applied to the input of the source encoder 8. The encoded signal at the output of the source encoder 8 is present as a frame transmitted by the transmitter 10 to the second node 6. Operation of the transmitter 10 includes channel coding and modulation.

At the second node 6, the signal from the transmission medium 4 is processed by the receiver 12. The operation of the receiver 12 includes demodulation and channel decoding. Recovery means in accordance with the inventive concept comprise an apparatus 13 and a decoder 20. At the output of the receiver 12 there is a received encoded signal. This received encoded signal is applied to quality determining means 16 which determines the quality of the received encoded signal. The quality determining means 16 comprises a decoder for a block code, for example a Reed Solomon code. In this case, the quality criterion may be whether or not the received encoded signal contains an uncorrectable error.

If the quality criteria are met (no calibration error), then the current frame of the received encoded signal is complete. Also, once all previous frames are complete, the received encoded signal is passed directly to the decoder 20 via the selector 18. The selector 18 receives the appropriate control signal from the quality determining means and passes the received decoded signal directly to the decoder 20. At the output of decoder 20, the recovered signal can be used.

If the quality criteria are not met (an uncorrectable error exists), the selector 18 is instructed to send the previous frame back to the decoder 20. The interpolator 14 is also instructed to complete the missing frame from the frame surrounding the indirectly received frame.

After interpolation is performed, the interpolated frame is passed to the decoder. Interpolation will delay the recovery of the interpolated frame, and then the supply of the recovered source signal. It is observed that reselecting the undelayed received encoded signal to be input to the encoder is not possible because this results in significant disturbance. Only in the silent period is it changed to non-delayed recovery. When such a silent period is detected by the quality detecting means 16, the selector 18 is instructed to convey the output signal of the receiver 12 back to the decoder 20.

In the interpolator 14, the output signal of the receiver 12 is applied to the cascade connection of the delay element 22 and the delay element 24. The delay value of these delay elements is equal to the duration of one frame. The interpolator 26 is provided for calculating the interpolated frame from the frames at the input of the delay element 22 and at the output of the delay element 24.

If the signal to be transmitted is a speech signal, the encoded signal may be made based on linear prediction. The frame includes an indication of a plurality of prediction coefficients and an indication of an excitation signal. The recovered signal is obtained by applying an excitation signal to the synthesis filter and has parameters derived from the prediction coefficients. If frame i-1 is incorrectly received, prediction coefficient α _K [i-1] is calculated as follows.

(A)

The interpolated prediction coefficient may be, for example, an arc sine function of a log area ratio, a line spectral pair, and a reflection coefficient. The representation of these prediction coefficients is well known to those skilled in the art.

Since the excitation signals can be very different between frames, the interpolated excitation signal can be very different from the original signal. Thus, the excitation signal is not normally interpolated but the previous frame is used.

If the frame is received incorrectly, the selector 18 is instructed to pass the output of the interpolator 26 to the decoder 20. If the frame is correctly received, for one frame period, the delayed received encoded signal available at the output of delay element 22 is delivered to decoder 20.

3 shows a device combining the functions of the interpolator 14 and the selector 18. The apparatus can also provide interpolation of the integrity of one or two missing frames.

The input signal of the device is applied to the cascade connection with the delay element 30, the delay element 32 and the delay element 34. The apparatus receives three control signals from the quality determining means 16. The first control signal ERR indicates that the frame present on the input of the device is in an error state. The second control signal DELAY indicates the current delay of the device. The third signal INTERPOALATE # represents a plurality of frames that must be complete by interpolation.

When all the frames are correctly received, the value of the signal ERR is FALSE, the value of the signal DELAY is equal to zero, and the value of the signal INTERPOALATE # is also equal to zero. In this case, the selector 40 is instructed to deliver the input signal of the delay element 30 directly to the decoder 20. This case is shown in graphs 50 and 52 of FIG. 4 for frames 1, 2 and 3.

If the frame is received in an error state, the signal ERROR will be equal to one. While the plurality of errors indicated by ERR are less than or equal to the value of DELAY, the selector will deliver the same output signal to decoder 20. In this case, since the plurality of errors 1 are greater than the value of DELAY, the logic circuit 42 selects the selector so that it is present again in the previous frame, i.e. frame 3, at the output until the interpolated frame 3 'becomes available. 40 will be instructed. This is indicated in the graph 50 showing that the frame 3 is fed back to the decoder.

It is known how many frames are dropped upon reception of the first correction frame and as a result how many frames should be interpolated. These multiple frames are signaled by a signal INTERPOALATE #. At the same instant, the value DELAY is set as the maximum function of the current DELAY value and the plurality of missing frames. Only the frame 4 in the graph 30 is missing between the frames 3, 5. As soon as frame 5 is correctly received, the signals DELAY, INTERPOALATE # are set to one. This causes the selector 36 to deliver the output signal of the delay element 32 (ie the frame 3) to the first output of the interpolator 38. At the second input of the interpolator 38 an input signal of the delay element 30 (ie frame 5) may be used.

As soon as the interpolation is complete, the output signal of the interpolator 38 is transmitted by the selector 40 to the decoder 20. This can be seen in the graph 50 in which the interpolated frame 4 ′ is delivered after the second instant of frame 3.

The value 1 of the signal DELAY causes the selector 40 to pass to the input of the output decoder 20 of the delay element 30 after the interpolated frame 4 '. Because frame 5 can be used at the output of the delay element at that moment, frame 5 is passed to decoder 20 following frame 4 'as can be seen in graph 50. From the graph 50, it can be seen that the delay increases after the occurrence of a transmission error. If the next single error occurs, there will be a discontinuity because the interpolation result is present sooner or later. Since the value of DELAY remains the same, the next frame to be delivered to decoder 20 is selected from the output of delay element 30.

If two frames are missing from the input signal as shown in the graph 52 in which the frames 8 and 9 are missing, interpolation of two or more frames must be performed. Upon detection of the incorrectly received frame 8, the signal ERR will be set to one. Since the number of errors and the value of DELAY are equal to 1, the selector 40 is instructed to keep delivering the output signal of the delay element 30 (i.e., the frame 7) to the decoder 20. When the next frame 9 is also wrong, the value of ERR is increased to two. Since the number of errors ERR is now greater than the value of DELAY, the selector 40 is again instructed to deliver the previous frame (ie frame 7) to the decoder 20.

Immediately after that, if the frame 10 is correctly received, the quality determining means 16 sets the signal INTERPOLATE # to a value of two. Thereby the selector 36 is instructed to deliver the output of the delay element 34 to the first input of the interpolator 38 instead of the output of the delay element 32. The interpolator 38 now calculates the LPC parameters of the missing frame according to the following.

(B)

(C)

As soon as the interpolated frames 8 ', 9' are determined, they are then passed to the decoder 20 as shown in the graph 50. Since the value of INTERPOLATE # is 2, the value delay is also set to 2 to instruct the selector 40 to transmit the output signal of the delay element 32 to the decoder 20. After the interpolated frame is delivered to the decoder 20, the frame 10 is present at the output of the delay element 32. Thus, frame 10 is delivered to decoder 20 directly after frame 9 '.

The frame 13 is also missing from the graph 52. As long as the number of errors indicated by the signal ERR is smaller than the delay value, the selector 40 keeps transmitting the output signal of the delay element 32 to the decoder 20. Once the interpolated frame (i.e., frame 13) is calculated after the value of INTERPOLATE # is received from the quality detecting means, the selector 40 decodes the interpolated frame instead of the output signal of the delay element 32. To pass). Since the number of errors ERR does not exceed the value of DELAY, discontinuity does not occur.

If a long period without source signal is detected, the delay value is reset to zero, so that the delay is not greater than necessary. This can be seen from the graphs 50 and 52 in which the value of DELAY is reduced to zero after three frames without source signals are received.

In the quality determining means 16 according to FIG. 5, the output signal of the channel decoder is applied to the error detector 66 and the silence detector 60. If the receiver includes a channel decoder that already has an output indicating that an uncorrectable error exists, the error detector includes a single buffer that holds the error signal until the next frame is present at the receiver's output. If not, the error detector 66 determines whether the received frame contains an uncorrectable error.

Error detector 66 sends a count signal to error counter 64 for each non-correctable frame to increase the count value. The count value is passed to the maximum value determiner 62 and to an output terminal representing the signal ERR. This maximum value determiner 62 accepts a count value as long as the count value is larger than its current output value. The output signal of the maximum value determiner 62 exhibits a delay value as described with reference to FIG. 3.

Error detector 66 displays a signal indicating receipt of the first correct frame after one or more uncorrectable frames. This signal is used to reset the error counter, which causes the buffer 16 to display its contents as an output signal INTERPOLATE # to start interpolation to complete the non-correctable frame.

The silence detector determines the presence of a predetermined number of frames that carry very small or zero signals. When this number is reached, the silence detector resets the value of DELAY to zero at the output of the maximum value determiner 62 to set the delay value in the interpolator 14 (or device 14 + 18) to zero.

In the embodiment of the invention according to FIG. 6, the source signal to be transmitted is applied to the encoder 8 to obtain an encoded signal. The output of the encoder 8 is connected to the input of the buffer 72. The output signal (frame) of the buffer 72 is provided to the transmitter 10 for transmission to the second node 4. The buffer 72 is configured to temporarily store the frames provided to the transmitter 10 and may use the frames when requested to be transmitted by the second node 4.

At the second node 4, the signal from the first node 2 is obtained by the receiver 12 from the transmission medium. The output of the receiver 12 is connected to the first input of the selector 74, the input of the buffer memory 7, and the input of the error detector 76. The error detector 76 determines whether the frames transmitted by the first node 2 are correctly received by the second node 4. If the frames are received correctly, the error detector 76 instructs the selector 74 to deliver to the decoder 20 the output signal of the receiver 12. This is also shown in FIG. 7.

Graph 79 shows a sequence of frames generated from encoder 8. Graph 80 shows the sequence of frames at the output of receiver 12 and graph 81 shows the sequence of frames delivered to decoder 20. From FIG. 7, it is evident that the frames 1, 2, 3 are correctly received by the receiver 12 and they are delivered directly to the decoder 20. When the error detector 76 detects a transmission error, the selector 74 is instructed to deliver the previous frame to the decoder 20. The error detector 76 sends a request for transmission of an incorrectly received frame via the transmitter 78. The buffer 7 exists to store the frames following the incorrectly received frame before the retransmission frame is received. This is necessary to be able to deliver the frames to the decoder 20 in the correct procedure.

In graph 80, it can be seen that frame 4 is received in an error state. Error detector 76 will request retransmission of frame 4. The transmitter takes frame 4 from its buffer 72 and retransmits it to node 2. With regard to FIG. 7, it is assumed that the first node retransmits the frame 4 before transmitting the frame 5. This can be done using an acknowledgment procedure in which receipt of each frame is confirmed by the second node. In the same way the wrong reception of frames 8, 9 and 12 is handled.

8 shows a situation when a negative response is used in which reception of incorrectly received frames is delivered to a first node. From graph 86, it can be seen that frame 5 is received before retransmitting frame 4. In order to maintain the correct order of the frames, the frame 5 is temporarily stored in the buffer memory 7 until the retransmitted frame 4 is received. After the retransmitted frame 4 is delivered to the decoder 20, the selector 74 delivers the delayed frame 5 to the decoder 20. The selector 74 keeps delivering delayed frames to the decoder 20 to prevent discontinuities until the next error occurs. If two frames are missing as indicated for frames 7 and 8 of graph 86, retransmission may already occur before subsequent frames are sent to second node 4. Subsequently, no buffering is required at the second node, and the received frames are delivered directly to the decoder 50. While an error occurs, the same frame 7 is repeatedly delivered to the decoder 50. However, as can now be seen from graphs 85 and 86, a delay is introduced into the first node.

After the silent period occurs, the delay values at the first node and the second node are set to zero as can be seen from the graphs 85, 86, 87.

In the second node according to Fig. 9, audio processing according to the present invention is shown. The signal from the first node is received by the receiver 90 and communicated to the control device 92 which performs the functions of the error detector and interpolator described in connection with the previous figures. The first output of the control device conveying the received frame is delivered to a decoder 94. The output signal of the decoder 94 is connected to the first input of the adder 97. The output of adder 97 is coupled to loudspeaker 102 via amplifier 98.

The second node further includes a microphone 104 coupled to the encoder 110 via an amplifier 106. The output of the encoder 110 is connected to the input of the transmitter 112 which transmits the encoded signal to the first node.

According to one type of the invention, the level of comfort noise is introduced depending on the amount of delay. In addition, the delay value DELAY is transmitted to the comfort noise generator 96 by the control device 92. The comfort noise generator 96 introduces the comfort noise signal into the loudspeaker 102 by supplying the adder 97 a noise signal having a level which increases with the delay value.

According to another type of the present invention, an amount of side tones is introduced which increases and decreases while the delay value increases. Side tones are introduced so that the user has the possibility to hear his / her own voice in the loudspeaker indicating that the system is operating. However, side tones can cause echoes that can be annoying in case of large transmission delays. By increasing the delay and reducing the side tone level, annoying echoes are prevented from occurring due to the increased decoding delay. The introduction of the variable level side tone is made by the amplifier 100 controlled by the delay value from the control device 92.

Claims (11)

A first node having a source encoder for deriving an encoded signal from an input signal, the first node further comprising a transmitter for transmitting the encoded signal to a second node via a transmission medium; A transmission system comprising recovery means for deriving a recovery signal from the encoded signal, The recovery means is configured to provide the recovered signal immediately if the quality of the received encoded signal satisfies a certain quality criterion, and wherein the recovery means is provided if the quality of the received encoded signal does not meet the quality criterion And provide the recovered signal after a recovery delay. The method of claim 1, The quality criterion is the integrity of the received encoded signal, and wherein the recovery means comprises completion means for completing the recovered signal if the received signal is not complete. The method of claim 1, And the recovery means is configured to increase the recovery delay in accordance with the quality of the received encoded signal. The method according to any one of claims 1 to 3, And the recovery means is installed to reduce the recovery delay after detecting that there is no recovered signal. The method according to any one of claims 1 to 4, The source signal includes a voice signal, and the second node includes comfort noise introduction means for introducing comfort noise having a level that increases as the quality of the received encoded signal decreases. A transmission system, characterized in that. The method according to any one of claims 1 to 5, The transmission system includes a full dual voice transmission system, and the second node includes side tone introduction means for introducing a side tone having a level that decreases with the recovery delay. Transmission system. The method of claim 2, And the completion means is configured to request by the first node to retransmit the missing portion of the encoded signal. The method of claim 2, And said completion means is adapted to derive the missing portion of said encoded or recovered signal by interpolating the portion surrounding said encoded or recovered signal. 10. A node for use in a transmission system, comprising: recovery means for deriving a signal recovered from the encoded signal The recovery means is configured to provide the recovered signal immediately if the quality of the received encoded signal satisfies a certain quality criterion, and wherein the recovery means is provided if the quality of the received encoded signal does not meet the quality criterion And provide the recovered signal after a recovery delay. A recovery means comprising a decoder for deriving a signal recovered from an encoded signal, The recovery means is configured to provide the recovered signal immediately if the quality of the received encoded signal satisfies a certain quality criterion, and wherein the recovery means is provided if the quality of the received encoded signal does not meet the quality criterion Means for providing said recovered signal after a recovery delay. A method of decoding an encoded signal, comprising deriving a signal recovered from the encoded signal, the method comprising: The method may decode the encoded signal immediately after reception if the quality of the received encoded signal satisfies a certain quality criterion, and after the recovery delay if the quality of the received encoded signal does not meet the quality criterion. And providing the recovered signal.