KR960001114B1 - Noise controlling method of wireless telephone - Google Patents

Abstract

detecting an error generation by imputing ADPCM information from the receiving part; setting error state value corresponding to the error generation frequency after comparing multiple reference value at each step by computing error generation number; detecting the error generation due to the existence of error generation number; lowering the output level of ADPCM information due to the error state value.

Description

Noise suppression method of wireless telephone

1 is a circuit block diagram of a conventional radiotelephone system.

2 is a circuit block diagram of a wireless telephone system to which a noise suppression method according to a preferred embodiment of the present invention is applied.

3 is a flowchart illustrating an algorithm of a noise suppression method according to a preferred embodiment of the present invention.

4 is a circuit block diagram of a wireless telephone system to which a noise suppression method according to another embodiment of the present invention is applied.

5 is a flowchart illustrating an operation of the error detector of FIG.

6 is a flowchart illustrating an operation process of the volume adjusting unit of FIG.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a noise level detection method of a digital radio telephone. Specifically, a code word corresponding to a noise level is detected at an output of an adaptive differential pulse code modulation (ADPCM) circuit in a digital radio telephone. The present invention relates to a method for lowering the noise level according to the frequency of word generation.

FIG. 1 is a circuit diagram of a conventional digital radiotelephone, which is a functional block diagram of a digital radiotelephone according to the 1988 CCITT G.721 Recommendation 32 Kbps ADPCM voice coding scheme. In FIG. 1, a radiotelephone is largely composed of a transmitter 100, 100 'and a receiver 200, 200', which is connected to a transmission / reception connection unit 6 to thereby establish a channel with another radiotelephone. Information is transmitted through each other.

In addition, since the transmitting radio telephone and the receiving radio telephone have the same circuit configuration, the description of the transmitting radio telephone is replaced by the whole. The circuit operation of the transmitter 100 will be described. The analog signal of the audio signal provided from the microphone 1 or the telephone input line is converted into the digital signal of the PCM code by the analog / digital conversion unit 2, and the digital signal corresponding to the voice signal is converted into the ADPCM encoding unit 3 Is converted into a 4-bit ADPCM code word, and the data provided by the data generator 4 and the ADPCM code word are coded in a form suitable for channel transmission by the channel coding unit 5, and coded. The signal is transmitted through the channel 12 by the transmission / reception connection unit 6. Data provided by the data generator 4 is generated to maintain a channel or to transmit information.

Next, the receiver 200 will be described. The signal transmitted through the channel 13 is converted into a base band signal by the transmission and reception connection unit 6 and input to the channel decoding unit 7.

At this time, the channel decoding unit 7 separates the baseband signal into data and ADPCM information, and provides data to the data receiving unit 8 and ADPCM information to the ADPCM decoding unit 9. The ADPCM decoding unit 9 converts the input 4-bit ADPCM information into a PCM code or other digital code, and the converted digital code is converted into an analog signal by the digital / analog converting unit 10 and then a speaker Provided by (11).

The ADPCM decoding unit 9 may convert an 8-bit or 13-bit PCM code or other digital code according to the type of the digital / analog converter 10.

As in the conventional wireless telephone system having such a configuration, when the voice information is transmitted and received using the ADPCM code, the state of the communication channel 12 may be unstable (bad) and may cause an error.

In this way, when an error occurs, the voice level suddenly increases due to the characteristics of the ADPCM method, and the receiver receives an unpleasant noise. In the conventional method for removing such noise, an error detection code is added to the ADPCM code or the data information provided by the data generator 4, which is voice information, and upon reception, the error detection code is analyzed to analyze the voice information. Detecting whether there is an error is proposed.

However, in this method, since an error detection code must be added to the ADPCM code, which is voice information, and then analyzed at the time of reception, the transmitter requires a processor for generating an error detection code, and a receiver for analyzing the error detection code is provided at the receiver. Is required.

Therefore, the circuit configuration of the radiotelephone system is complicated, and there is a problem in that the production cost increases because two expensive processors must be added.

In addition, since the above-described method is an indirect detection method that reads an error detection code and estimates whether an error occurs in voice information, there is a problem that an error may be detected if an error only occurs in the error detection code itself.

In particular, when the error occurrence frequency is high according to the characteristics of the ADPCM method, there is a problem in that noise occurs as the volume of the receiving side suddenly increases. In order to solve this problem, the present invention is to provide a noise suppression method of a wireless telephone that detects an error in the continuous ADPCM code provided by the channel decoding unit of the receiver and then lowers the noise level according to the frequency of error occurrence. There is this.

Another object of the present invention is to provide a noise level detection method of a radiotelephone in which a processor for error detection is implemented only in a receiving unit to realize circuit simplification and reduction of production cost.

According to one aspect of the present invention for achieving the above object, a noise level detection method of a wireless telephone is a noise level detection method of a wireless telephone comprising a transmitter and a receiver for transmitting and receiving a channel signal in a 4-bit ADPCM code method, the receiver Detecting whether an error occurs by comparing ADPCM information by inputting ADPCM information, and comparing the plurality of reference values set in stages to determine the frequency of error by counting the number of error occurrences, and simultaneously Setting an error state value corresponding to each error occurrence frequency; if there is no error, detecting the error again according to the existence of the error occurrence number; and outputting ADPCM information according to the error state value Reducing the level. The step of detecting the occurrence of an error again includes comparing the error occurrence variable to '0', clearing an error state value if the error occurrence variable is '0', and setting the error occurrence variable to '0'. Or comparing the third reference value, clearing the error occurrence variable and the error state value if the error occurrence variable is greater than the third reference value, and subtracting the error variable value if the error occurrence variable is smaller than the third reference value. It increments by 1˝ and makes an error state value become a preset error state value.

Hereinafter, a first embodiment of the present invention will be described in detail with reference to the accompanying drawings. 2 is a circuit block diagram of a radiotelephone system to which the noise suppression method of this embodiment is applied.

The configuration shown in FIG. 2 is similar to that of FIG. 1 except that the error adjusting unit 13, which is a microprocessor, is additionally connected between the channel decoding unit 7 and the ADPCM decoding unit 9 in FIG. Since the components are the same, the same reference numerals are used to refer to components that perform the same function.

The function of the error adjusting unit 13 inputs a 4-bit ADPCM code out of a signal (4-bit ADPCM code and channel maintenance and information transmission data) output from the channel decoding unit 7 in the receiving unit 100. By detecting the error occurrence of the ADPCM code due to noise and the like, and providing the output level of the voice information reduced in stages according to the frequency of the error occurrence to the ADPCM decoding section (9).

The ADPCM decoding unit 9 receives the voice information, which is a level adjusted ADPCM code, converts it into a digital signal of the transmitting unit, that is, an 8-bit or 13-bit PCM or other digital signal, and then provides it to the digital / analog converter 10. do. Therefore, the ADPCM decoding unit 9 converts the 4-bit ADPCM code corresponding to the output level adjusted by the error adjusting unit 13 into a digital signal of a predetermined bit and provides it to the digital / analog converter 10. As a result, the audio signal output from the speaker 11 corresponds to the output level.

The noise suppression method according to the present embodiment will now be described with reference to the flow chart of the circuit configuration and operation process shown in Figs. Prior to the description of the present embodiment will be described to determine whether the error of the ADPCM code as follows. The circuit configuration of the radiotelephone shown in Figs. 1 and 2 is based on the 32 Kbps ADPCM voice coding scheme of CCITT G.721.

According to the above recommendation, the ADPCM circuit is implemented to convert an analog signal to a digital signal into a codeword of 4 bits per sample.

In the ADPCM circuit, signal level tracking is performed by an adaptive quantizer before quantization. When an error occurs in a channel during transmission, the ADPCM decoding unit 9 causes a noise signal to be mixed into the output. In addition, it also increases the output level of noise.

If the output level of noise caused by the error is directly provided to the speaker 11 through the digital / analog converter 10, an extremely loud noise (noise) can be heard by the user using the wireless telephone. According to the above recommendation, all four codewords use 15 codewords except {0000} among 16 codewords (# 0000 #, # 0001 #, ~ # 1111 #).

That is, since the codeword of {0000} out of the 16 codewords is not used, it may be determined as an error when this codeword is detected. Therefore, in the present invention, if a codeword of {0000} among the codewords is generated, it is determined as an error, and if such codewords are frequently detected, it is determined that the channel state is bad (that is, an error is generated).

As described above, in the present invention, the reference signal for determining whether noise is generated is the codeword of {0000}.

As shown in FIG. 3, when 4 bits of ADPCM information is input from the signal output from the channel decoding unit 7 (step S1), it is determined whether an error (noise) occurs by the reference signal (step S2). If the inputted ADPCM information is in error, the processor of the error adjusting unit 13 proceeds to step S3, and if not otherwise, the processor proceeds to step 9.

In step S3, the error generation variable err is increased by one to count the number of error occurrences. The error occurrence variable err is compared with the first reference value n1 already set in step 4 to determine the frequency of error occurrence, and when it is small, it is determined that an error has occurred but not much has occurred. The process is proceeded to step S6 in order to clear the variable ST1 indicating the error state information and to compare it with the second reference value n2 set larger than the first reference value n1 when it is large.

In step S6, when the error occurrence variable err is smaller than the second reference value n2, the flow advances to step S7 to determine that there is a slight error condition, and sets the variable ST1 to '1'. On the contrary, if it is large, the flow advances to step S8 to determine that there are many error states and sets the variable STI to # 2.

As described above, in steps S2 to S8 in the noise level detection algorithm, an error state corresponding to each frequency is set in order to count and determine the error occurrence frequency in the ADPCM information. Steps S13 to S27 are steps for changing the ADPCM information provided to the ADPCM decoding unit 9 according to the set error state value STI. Step S13 compares whether the ST1 value set previously is "0", and if "0", proceeds to step S14 to provide ADPCM information output from the channel decoding unit 7 to the ADPCM decoding unit 9 as it is. If not 0, go to Step S15 to compare whether the STI value is # 1 or higher.

In step S15, when the STI value is # 1, the process proceeds to step S16, where it is determined that there is a slight error state, and the output level of the ADPCM information provided to the ADPCM decoding unit 9 is slightly lowered.

On the other hand, if it is not # 1 as in STI (that is, if # 2 is larger than # 1), it is determined that many error conditions have occurred in step S17, and the ADPCM provided to the ADPCM decoding unit 9 is provided. The output level of the information is lowered than the above example. As the output level is lowered according to the frequency of error occurrence, the user cannot hear the noise level.

On the other hand, when the process proceeds from step S2 to step S9, the error occurrence variable err is compared with "0" to check the number of error occurrences. If err = 0, the process proceeds to step S9a to clear the STI value, and then immediately proceeds to step S13 to transmit the above-described normal ADPCM information to the ADPCM decoding unit 9. On the contrary, if err ≠ 0, the process proceeds to step S10, where the err value is compared with the third reference value n3 set larger than the second reference value n2. In step S10, if the err value is larger than the third reference value n3, the process proceeds to step S11 to clear the err value and the STI value, and if it is small, the process proceeds to step S12, where the err value is increased by ˝1 계 and In addition, the previous STI (preset error state value) is stored in the error state information variable STI. When the steps S11 and S12 are performed, the process proceeds directly to step S13. As described above, the step S9 to step S12 is a process of restarting the error state check, and when the err value is less than the third reference value while the error does not occur, the err value is changed to the third value. It shows the process of clearing all variables (err, STI) after making it larger than the reference value.

Of course, the reference values have a relationship of n1 < n2 < n3 and can be arbitrarily set in the manufacturing process of the wireless telephone according to the type and component of the wireless telephone product.

In addition, as in steps S13 to S17, the magnitude of decreasing the output level of ADPCM information according to each error occurrence frequency can also be arbitrarily adjusted in the manufacturing process of the radiotelephone.

4 is a circuit block diagram of a radiotelephone system to which a noise suppression method according to another embodiment of the present invention is applied.

4, the error detection unit 14 is connected between the ADPCM decoding unit 9 of the channel decoding unit 7, and the ADPCM decoding unit 9 and the digital / analog conversion unit are shown in FIG. Except for connecting the volume adjusting section 15 between the (10) and the same as the configuration of FIG. 1, the components that perform the same function is denoted by the same reference numerals as FIG. As described above, the 4-bit ADPCM uses 15 codewords out of the 16 possible codewords and does not use the codeword {0000}.

Therefore, if a {0000} codeword in which 4 bits of the transmitted ADPCM information is mode 0 is detected, this is a criterion that can be determined as an error. If this codeword is frequently detected, the channel state can be determined to be bad.

If this codeword is detected frequently and the output level is reduced, the user will not hear suddenly loud noises.If the level is muted completely, the user may be mistaken for a malfunction, but if it is output at a small level, the user may notice a bad channel condition. have. In order to limit the output level of the ADPCM in such a noisy channel, it is possible to use a method of suppressing a specific codeword according to the bit error rate.

For example, codewords of +7 and -7 can be suppressed to +6 and -6, respectively, at an error rate of 1%. In very noisy channels, the {4,5,6,7} and {-4, -5, -6, -7} codewords can be forced to 3 and -3. As a method of suppressing the output level, it is possible to directly reduce the volume level according to an error rate before or after the digital / analog converter 10, 10 'after ADPCM decoding without suppressing a specific codeword.

In this method, the method of suppressing a specific codeword takes into account the complexity of the algorithm that considers the sign of the codeword in classifying the codeword, classifies the codeword according to the absolute value, changes the codeword to the corresponding codeword, and then reapplies the corresponding codeword. You can get the same effect without going through it. Another embodiment of the present invention will be described with reference to FIG. The input signal from the microphone 1 or the telephone line or the like is digitized by the PCM coder (Coder-DECoder) or the analog / digital converter (2). The digitized signal is converted by the ADPCM encoding section 3 into an ADPCM signal, which is a 4-bit codeword for each sample.

This ADPCM signal is voice information and the data generator 4 generates data for maintaining a channel or transmitting information. The ADPCM signal and the data signal are shaped by the channel coding unit 5 according to the channel shape and transmitted through the wireless or wired communication channel 12 through the transceiver 6. On the receiving side, the signal from the channel 12 is converted into a baseband signal by the transceiver 6 'and transferred to the channel decoder 7'. The channel decoding unit 7 'separates the data information and the ADPCM voice information from the channel information, transfers the data to the data receiving unit 8', and inputs an unused codeword through the error detection unit 14 '. After checking that the error rate is determined according to the frequency, the ADPCM information is transmitted to the ADPCM decoding unit 9 '. The ADPCM decoding unit 9 ') converts ADPCM information, which is 4 bits per sample, into 8 bits or more digital information.

This digital information adjusts the volume according to the error rate determined by the error detection unit 14 '. This volume adjustment can be performed either before or after the digital / analog converter 10'. Shifting makes it easy to adjust 6 decibel volume levels per 1-bit shift.

The digital / analog converter 10 'converts the digital information into analog and transmits it to the user through the speaker 11' or transmits it to a telephone line or the like. 5 and 7 are algorithms of the error detector 14 ′ and the volume adjuster 15 ′ according to the embodiment to determine the channel condition and adjust the output volume level. The error detection unit 14 'checks whether all four bits of information are 0 in the ADPCM information S1 inputted from the channel decoding unit 7' as shown in Fig. 5 (S2). A zero of all four bits is an error, indicating that the channel is bad.

If all are 0, the parameter that counts the number of errors, that is, the error occurrence variable err is increased by one (S3).

It is checked whether the increased error occurrence variable err is smaller than the first predetermined reference value n1 for error rate determination (S4). At this time, if the value is smaller than the first reference value n1, an error occurs but not much occurs, so the variable ST1 indicating the status information of the error is set to 0 (S5).

In step S4, if the error generation variable err is not smaller than the first reference value n1, it is checked whether the error generation variable err is smaller than the second reference value n2 (S6). At this time, if it is small, the value of the state variable STI is set to 1 with a slight error (S7). If the value of the error occurrence variable err is not smaller than the second reference value n2, the value of the variable STI is set to 2 (S8). In step S2, if all of the four-bit codewords are not zero, it is checked whether the error generation variable err is zero (S9).

At this time, if the value of the variable Err is 0, it is determined that the channel state is good, and the error state information variable STI is set to 0 (9a) and the process is passed to ADPCMELZHELDQN 10 '. In step 9, if the error generation variable err is not 0, it is determined whether it is larger than the third reference value n3 (S10).

If it is not greater than the third reference value n3, the error occurrence variable err is increased by one, so that the error occurrence variable err is the third reference value n3 when the current variable STI is 0, 1, or 2. To make it larger, induce the value of variable STI to become zero after a certain time. If the value of the error occurrence variable err is greater than the third reference value n3 (S10), the values of err and STI are set to 0 (S11), and the error status check for the next ADPCM information is restarted (S2). ).

The ADPCM information which sets the error state through the error detection unit 14 'converts the ADPCM information, which is 4 bits per sample, into 8 or more bits of digital information (usually 13 bits) in the ADPCM decoding unit 9'.

The ADPCM decoded digital information adjusts the output volume according to the error level of the ADPCM in the volume adjusting unit 15 '. The process is as shown in FIG. Referring to FIG. 6, the operation of the volume adjusting unit 15 ′ will be described. First, it is checked whether the error status information variable STI is 0 (S18).

If the variable STI is 0, it is determined that the error rate is low and outputs the analog signal through the digital / analog converter 10 'as it is in the normal volume (S19). If the variable STI is 1 (S20), it is determined that there is a slight error and the output volume level is reduced.However, shifting the ADPCM decoding output to the right by 1 bit can reduce the 6 decibel volume, so that the variable ST1 is reduced. When it is -1, it shifts by the prescribed level and outputs it to the digital-to-analog converter 10 '(S22).

If the digital / analog converter 10, 10 'is a PCM method, a part for changing to PCM is added. In this way, when a transmission error occurs, noise is applied from the ADPCM decoding units 9 and 9 'and the noise output level can be limited.

Here, n1, n2, and n3 for determining the error state information variable STI are n1 < n2 < n3, and the value can be determined according to the type and situation of the product. In addition, the method of suppressing the noise output level according to the present embodiment may also be changed according to the type and situation of the product.

As described above, according to the present invention, since an error can be detected in the ADPCM information itself, a hardware device for adding and determining an error detection code to the ADPCM information is not required, and a sudden increase in noise level can be achieved. By suppressing the user to hear a weak level of noise has the advantage of solving the inconvenience in use.

Although the present invention shows an embodiment that is applied to a radiotelephone, the present invention is not limited to this, it is obvious to those skilled in the art that it can be applied to all systems that transmit and receive with ADPCM information. something to do.

Claims (5)

In the noise suppression method of a secondary telephone line consisting of a transmitter and a receiver for transmitting and receiving a channel signal in the ADPCM code method, the step of inputting the ADPCM information at the receiver to compare whether a predetermined code word to detect whether or not an error occurs; Counting the number of occurrences and comparing the plurality of reference values set in stages to determine the frequency of error occurrences, and at the same time setting an error state value corresponding to each of the frequency of occurrences of errors; Detecting whether the error occurs again according to whether or not, and lowering an output level of ADPCM information according to the error state value. The method of claim 1, wherein the error state value setting comprises: a counting step of increasing an error generating variable by one; comparing the error generating variable with a first reference value; and in the step, the error state value when the first reference value is large. When the first reference value is small, comparing the error occurrence variable with the second reference value, and if the second reference value is large, sets the error state value to ˝1˝, and conversely, And setting the error state value to " 2 " if it is small. The method of claim 1, wherein the output level lowering step of the ADPCM information is output in three stages: a normal output level, an output level lowered to a predetermined value, and an output level lowered significantly larger than the predetermined value according to the magnitude of the error state value. Noise suppression method of a wireless low frequency generator, characterized in that for adjusting the level. 2. The method of claim 1, wherein the step of detecting the occurrence of an error again comprises the steps of comparing whether the error occurrence variable is '0', clearing an error state value if the error occurrence variable is '0', Comparing the reference value with a third reference value if it is not? 0, clearing the error occurrence variable and the error state value if the error occurrence variable is greater than a third reference value, and error if the error occurrence variable is less than a third reference value. A noise suppression method for a wireless telephone, characterized in that the variable value is increased by ˝1˝ and the error state value is set to a predetermined error state value. 5. The noise suppression method according to claim 2 or 4, wherein the first reference value, the second reference value, and the third reference value < second reference value <