KR0164200B1 - End-to-end call quality automatic measurement system - Google Patents

Abstract

The present invention relates to a mobile phone tangible call quality measurement system capable of measuring sound quality by a voice recognition function, which is connected to a general landline telephone to perform automatic call generation and incoming calls, call sound quality measurement, and remote call control using a modem. MQM which detects Land Quality Module (1) and measurement vehicle and performs automatic call generation and reception by mobile phone, measurement of call sound quality, measurement path display by GPS (Global positioning system) and remote control using wireless modem. (Mobile Quality Module) (2), and the data measured by the LQM (1) and MQM (2) by cross-combining analysis to measure call statistics, voice quality evaluation for each voice channel belonging to a specific channel, details of the cause of incomplete calls Voice recognition capable of speech recognition in a mobile phone tangible call quality measurement system with a call analysis module (CAM) 3 that performs analysis and sound quality measurement location record display function By embedding the chip in the LQM (1) and MQM (2), by solving the problems of the prior art that is different from the sound quality that the real person feels by a simple method such as SINAD or tone level measurement accuracy By performing sound quality measurement, there is an effect that enables more effective haptic sound quality measurement.

Description

Mobile phone tangible call quality measurement system capable of measuring sound quality by voice recognition function

1 is an overall configuration diagram of a mobile phone tangible call quality measurement system (ECAMS) to which the present invention is applied.

2 is a block diagram of an embodiment of an LQM according to the present invention.

3 is a block diagram of an embodiment of an MQM according to the present invention.

4 is a flowchart of a sound quality measurement by the speech recognition function according to the present invention.

* Explanation of symbols for main parts of the drawings

11: LQM main control part 12,32: multi-protocol interface part

13,33: automatic dialing unit 14,34: line converting unit

15,35: notch filtering unit 16,36: tone detection unit

17,37: SINAD measuring part 18,38: A / D conversion part

19,39 amplification section 20,40 codec section

21,41: speech recognition processing unit 22,42: selection unit

31: MQM main control unit 43: GPS receiver

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a mobile phone tangible call quality measurement system capable of measuring sound quality by a speech recognition function. In particular, the present invention relates to automatic call generation, call state tracking, and continuous measurement of electric field strength required for performing field tests of a cellular mobile phone system. It relates to an integrated end-to-end call quality automatic measurement system (hereinafter referred to as 'ECMAS') that performs sound quality measurement using voice recognition as well as functionality.

ECMAS and similar equipment include TQNM of SAFCO, EXP-2001 of LCC, and Buzzard of ROTADATA, UK. These devices all measure the end-to-end immersive call quality of analog mobile phones (AMPS), called end-to-end automatic call, Incoming and sound quality measurement capabilities.

SAFCO's TQNM measures the sound quality at 1004 Hz on downlink (route from mobile phone to mobile phone) on the voice channel and 590 Hz on uplink (from mobile phone to landline). Simultaneously transmit the signal to measure the signal to noise and distortion ratio (SINAD).

In addition, the sound quality measurement method of LCC EXP-2001 is also performed by SINAD. Sound quality measurement method of UK ROTADATA company Buzzard is divided into frequency test and noise test.

The frequency test transmits 622Hz, 1179Hz, and 1950Hz to the uplink and downlink alternately to measure and score the degree to which the reception level deviates from the reference level, and the noise test By simultaneously transmitting 590 Hz to the uplink and 1004 Hz to the downlink, the degree to which the reception level deviates from the reference level is measured and scored. As a result, it is judged that it is a favorable number if the low trace number is 80 points or more.

However, the conventional call quality measurement system has a problem that can not perform accurate sound quality measurement according to the human voice by measuring the sound quality measurement by SINAD.

Therefore, the present invention devised to solve the problems of the prior art adds the call sound quality measurement function by the speech recognition function to the conventional analog mobile phone end-to-end immersive call sound quality measurement method To provide better analog mobile phone call quality measurement system (ECAMS) that can improve analog mobile phone call quality by performing better tangible call quality measurements and by analyzing the detailed causes of incomplete calls. There is this.

The present invention for achieving the above object is connected to a normal landline telephone, automatic call generation and incoming calls, call sound quality measurement, LQM (Land Quality Module) to perform a far-end control function using a modem and a measurement vehicle mounted on a mobile phone Mobile Quality Module (MQM) that performs automatic call generation and reception, call quality measurement, measurement path display by GPS (Global positioning system), and remote control using a wireless modem, and measurement by the LQM and MQM Mobile phone immersive call with CAM (Call Analysis Module), which performs combined call analysis of measured data to measure call statistics, evaluate the sound quality of each voice channel belonging to a specific channel, detailed analysis of the cause of call incompleteness, and record call measurement location A quality measurement system, comprising: LQM main control means for performing overall control of the LQM; Multi-protocol interface means connected to said LQM main control means for performing a matching function; Automatic dialing means for dialing under the control of the LQM main control means via the multiprotocol interface means; Line converting means for converting and outputting the input signal connected to the telephone line; Notch filtering means for filtering and outputting a signal received through the line conversion means; Tone detection means for receiving an output of the notch filtering means and detecting and outputting a tone signal; The signal to noise and distortion ratio (SINAD) indicating the distortion level of the signal-to-noise is received by the output of the notch filtering means, or is output by the control of the LQM main control means to output a tone signal for SINAD measurement by MQM. SINAD measuring means; An analog / digital (A / D) converting means for receiving the output of the tone detecting means or the output of the SINAD measuring means and converting it into a digital signal and outputting the digital signal to the multiprotocol interface means; Amplifying means for amplifying and outputting an audio signal input from said line converting means; Codec means for receiving an amplified speech signal of the amplifying means and receiving the compressed signal and outputting the compressed signal; Speech recognition processing means for recognizing the speech signal inputted from the codec means, outputting a speech recognition ready signal and data to the multi-protocol interface means, and outputting the speech signal stored under the control of the LQM main control means to the codec means; And an LQM having selection means for receiving a decompressed-decoded transmission speech signal of the codec means and a tone signal of the SINAD measurement means, selected by the control of the LQM main control means, and outputting to the line conversion means. MQM main control means for performing overall control; Multi-protocol interface means connected to said MQM main control means for performing a matching function; Automatic dialing means for dialing under the control of the MQM main control means via the multi-protocol interface means; Line converting means for converting the dialing signal inputted from the automatic dialing means into a line through the test mobile telephone or outputting the incoming dialing signal received through the test mobile telephone to the automatic dialing means; Notch filtering means for filtering and outputting a signal received through the test mobile telephone; Tone detection means for detecting the tone signal by receiving the output of the notch filtering means; SINDA measuring means for receiving the output of the notch filtering means for measuring and outputting the SINAD indicating the degree of distortion of the signal-to-noise, or output the tone signal for SINAD measurement to the LQM under the control of the MQM main control means; An analog / digital (A / D) conversion means for receiving the output of the tone detection means or the output of the SINAD measurement means and converting it into a digital signal and outputting the digital signal to the multiprotocol interface means; Amplifying means for amplifying and outputting an audio signal inputted from the test mobile phone; a codec means for receiving an amplified voice signal of the amplifying division and outputting it by compression encoding or receiving a compressed coded audio signal and performing reverse compression decoding and outputting it; ; Speech recognition processing means for recognizing a speech signal inputted from said codec means, outputting a speech recognition ready signal and data to said multi-protocol interface means, and outputting a speech signal stored under control of said MQM main control means to said codec means; Selecting means for receiving a decompressed decoded transmission voice signal of the codec means and a tone signal from the SINDA measuring means, and selecting the same by a control signal of the MQM main control means and outputting the tone signal to the test mobile telephone; And an MQM having a GPS receiving means for receiving current position information through an antenna and outputting the current position information to the multiprotocol interfering means.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention;

1 shows an overall configuration diagram of an ECAMS to which the present invention is applied.

ECAMS consists of a Land Quality Module (LQM) (1) that is connected to a regular landline telephone line in an office, a Mobile Quality Module (MQM) 2 that is mounted in a vehicle, and a call analysis module that analyzes measured data. (3).

When the LQM (1) attempts to make a call through a landline telephone, the call is wirelessly transmitted to the mobile telephone through a public switching telephone network (PSTN) and a mobile telephone switching office (MTSO). The call is received by the MQM 2 and the call is made.

After the call path is formed, call sound quality is measured by voice recognition or signal to noise and distortion ratio (SINAD) measurement unit.

The LQM (1) is connected to a regular wired telephone line and its main functions are automatic call generation and incoming call for automatic call reception, voice recognition or uplink call quality measurement using SINAD measurement unit 17, Directs the MQM remote control function using a modem.

The LQM (1) is largely divided into hardware and software. As a hardware component, a main controller (486PC) which automatically calls / receives calls at a predetermined time using a modem and a voice recognition and SINAD measurement unit for measurement It is composed of a remote control device. The software consists of control software that controls the entire LQM 1.

The MQM (2) is mounted on a measurement vehicle to automatically call generation for incoming and outgoing calls, automatic call reception for incoming calls, downlink call sound quality measurement using voice recognition or SINAD measurement, and LQM remote control using a wireless modem. Also, it measures received signal strength of control channel and voice channel, measures sound quality for each voice channel assigned to specific control channel, collects system parameter statistics, and records location by GPS (Global Positioning System).

The MQM (2) is largely divided into hardware and software. The hardware consists of a main xontroller (486PC), a GPS receiver, and voice recognition and SINAD measurement for sound quality. It consists of a participant control system.

CAM (3) is a software (S / W) tool that analyzes and analyzes the result data measured and generated by MQM (2) and LQM (1) into uplink and downlink. to be.

The main functions include call completion rate and call quality statistics analysis, cause of call incompleteness analysis, voice quality evaluation for each voice channel belonging to a specific channel, and call status indicator for each measurement path by GPS.

Call analysis software, an analysis tool of the CAM 3, can be performed on a general personal computer (PC). The advantage of ECAMS is that it allows detailed analysis of the causes of incomplete calls.

Call completion cause analysis includes no service, no page, control channel access failure, radio frequency (RF) channel shortage, channel assignment check failure, and dropped during a call due to SAT (Supervisory Audio Tone) failure. It is also possible to analyze the cause of disconnection due to handoff failure.

These features can be used to improve analog mobile phone call quality. For example, the upper part of the cause of poor reception of an analog mobile phone system is lost in the process of calling the user's phone in the analog mobile phone system.

Thus, if ECAMS is used in combination with the tool (too1) within an analog mobile phone system, then it is possible to identify which path in the analog mobile phone system has been lost.

2 is a temporary block diagram of a Land Quality Module (LQM) according to the present invention, in which FIG. 11 is an LQM main control unit, 12 is a multi-protocol interface unit, 13 is an automatic dialing unit, 14 is a line conversion unit, and 15 Is notch filtering part, 16 is tone detection part, 17 is SINAD measurement part, 18 is analog / digital conversion part, 19 is amplification part, 20 is codec part, 21 is voice recognition processing part, 22 is Each selection part is shown.

Looking at the operation of the LQM (1), first, the automatic call generation is generated in the LQM main control section 11 by the operator's test request, and made in the automatic dialing section 13 through the multi-protocol interface section 12, and vice versa. The automatic call incoming is received by the automatic dialing unit 13 and made by the LQM main control unit 11 through the multi-protocol interface unit 12.

After the call is connected, the sound quality is measured by the SINAD or speech recognition function, and the selection of the two functions is performed by the SINAD measurement unit 17 and the speech recognition processing unit through the multi-protocol interface unit 12 at the LQM main controller 11 at the request of the operator. By controlling 21.

In the case of the SINAD measurement, the 590 Hz tone transmitted from the MQM 2 is input to the line converter 14 through a regular landline. The line converter 14 performs a function of converting a two-wire telephone line into four lines, and is for distinguishing transmission and reception.

The signal converted by the line converter 14 is filtered by the notch filtering unit 15 and input to the tone detector 16 and the SINAD measurement unit 17.

The tone detector 16 receives the filtered signal from the notch filtering unit 15, detects the tone signal, and outputs the tone detection signal to the A / D converter 18.

The SINAD measurement unit 17 receives the filtered signal from the notch filtering unit 15, measures the distortion degree of the signal-to-noise, and outputs the measured SINAD analog signal to the A / D converter 18.

The A / D conversion unit 18 converts the tone detection signal input from the tone detection unit 16 or the SINAD analog signal input from the SINAD measurement unit 17 into a digital signal to convert the LQM through the multi-protocol interface unit 12. Output to main control section 11 to analyze the results of the test.

On the other hand, the 1004Hz tone signal for downlink SINAD measurement is generated in the SINAD measurement unit 17 under the control of the LQM main control unit 11 and output to the selection unit 22.

The selecting unit 22 performs a function of selecting and outputting one of a speech recognition measurement and a SINAD measurement by the LQM main control unit 11, and selecting a tone output signal input from the SINAD measurement unit 17 to convert the line to a line conversion unit. Output through 14.

In the sound quality measurement by the speech recognition function, the voice signal transmitted from the MQM 2 is amplified by the amplifier 19 through the amplifier 19 and input to the codec unit 20.

The codec unit 20 compresses and encodes the voice signal received through the amplifier 19 in a form suitable for the voice recognition processor 21 and outputs the compressed signal to the voice recognition processor 21.

The speech recognition processor 21 is configured by using a speech recognition digital signal processor (DSP) chip. The speech recognition processor 21 recognizes a compressed coded speech signal input from the codec unit 20 and prepares a speech recognition ready signal and a speech. Recognition data is output to the LQM main control unit 11 via the multi-protocol interface unit 12.

The LQM main control unit 11 analyzes the sound quality measurement result using the data received from the speech recognition processing unit 21.

On the other hand, in the case of LQM voice transmission, the voice stored in the voice recognition processing unit 21 is output to the codec unit 20 under the control of the LQM main control unit 11.

The codec unit 20 decompresses and decodes the voice signal input from the voice recognition processor 21 to output to the selector 22. The selector 22 converts the voice signal received from the codec unit 20 into line conversion. Output through the unit (14).

3 is a block diagram of an embodiment of a mobile quality module (MQM) according to the present invention, in which 31 is an MQM main control part, 32 is a multi-protocol interface part, 33 is an automatic dialing part, 34 is a line conversion part, and 35 Is notch filtering unit, 36 is tone detection unit, 37 is SINAD measurement unit, 38 is analog / digital (A / D) conversion unit, 39 is amplification unit, 40 is codec unit, 41 is speech recognition processing unit, 42 is A selection unit 43 denotes a GPS (Global Positioning System) receiving unit, and 44 denotes a test mobile phone, respectively.

Looking at the operation of the MQM (2), first, the automatic call generation is generated in the MQM control unit 31 at the request of the operator and made in the automatic dialing unit 33 through the multi-protocol interface unit 32, the line conversion unit 34 2 lines are converted into 4 lines, and then transmitted from the test mobile phone 44 to the antenna. In the automatic call reception, the signal from the antenna is received by the test mobile phone 44 and received by the automatic dialing unit 33 through the line converting unit 34 and the MQM main control unit 31 through the multiprotocol interface unit 32. Is entered.

After the call is connected, the sound quality is measured by the SINAD or speech recognition function, and the selection of the two functions is made by the SINAD measurement unit 37 and the speech recognition processing unit through the multi-protocol interface unit 32 in the MQM main control unit 31 at the request of the operator. By controlling 41.

In the case of the SINAD measurement, when a 1004 Hz tone transmitted from the LQM 1 is received by the test mobile 44, the test mobile phone 44 outputs the received tone signal to the notch filtering unit 35. The filter is filtered by the notch filtering unit 35 and input to the tone detector 36 and the SINAD measuring unit 37.

The tone detector 36 receives the filtered signal from the notch filtering unit 35, detects a tone signal, and outputs a tone detection signal to the A / D converter 38.

The SINAD measurement unit 37 receives the filtered signal from the notch filtering unit 35, measures the distortion degree of the signal-to-noise, and outputs the measured SINAD analog signal to the A / D converter 38.

The A / D conversion unit 38 converts the tone detection signal input from the tone detection unit 36 or the SINAD analog signal input from the SINAD measurement unit 37 into a digital signal to convert the MQM through the multi-protocol interface unit 32. Output to the main control section 31 so that the results of the test can be analyzed.

On the other hand, the 590Hz tone signal for uplink SINAD measurement is generated in the SINAD measurement unit 37 under the control of the MQM main controller 31 and output to the selection unit 42.

The selecting unit 42 performs a function of selecting and outputting one of the SINAD measurements under the control of the MQM main controller 31, and selecting the tone output signal input from the SINAD measuring unit 37 to test mobile telephone 44 )

In the sound quality measurement by the speech recognition function, the voice signal transmitted from the LQM 1 is amplified by the amplification unit 39 and input to the codec unit 40.

The codec unit 40 compresses and encodes the voice signal received through the amplifier 39 in a form suitable for the voice recognition processor 41 and outputs the compressed signal to the voice recognition processor 41.

The speech recognition processor 41 is configured by using a speech recognition digital signal processor (DSP) chip. The speech recognition processor 41 recognizes a compressed coded speech signal input from the codec unit 40 and prepares a speech recognition ready signal and a speech. Recognition data is output to the MQM main control unit 31 via the multi-protocol interface unit 32.

The MQM main control unit 31 analyzes the sound quality measurement result using the data received from the speech recognition processing unit 41.

On the other hand, in the case of MQM voice transmission, the voice stored in the speech recognition processing section 41 is output to the codec section 40 under the control of the MQM main control section 31.

The codec unit 40 decompresses and decodes the voice signal input from the voice recognition processor 41 to output to the selector 42. The selector 42 transfers the voice signal received from the codec unit 40 for a test movement. Output through the phone 44.

4 is a flowchart illustrating a sound quality measurement by the speech recognizer according to the present invention.

Sound quality measurement by ECAMS voice recognition function is as follows. In the analog mobile telephone system, after the voice channel by the control channel is allocated, the voice is transmitted first in the incoming call. In the outgoing call, the voice recognition chip has been built into the voice recognition chip. Recognizes voices sent by If the call state is good, the voice recognition score is excellent. If the call state is poor, the low score is recorded.

The transmitting call transmits a voice that is different from or the same as the voice sent from the incoming call immediately after voice recognition, and the incoming call contains a voice recognition chip that has been trained to recognize the voice to be sent from the calling call. Perform recognition. Likewise, if the call state is good, the voice recognition score is excellently recorded, and if the call state is poor, the low score is recorded.

In this way downlink and uplink call quality is measured.

Referring to Figure 4 as follows.

The word "communication" is trained in the LQM speech recognition chip (100), and the word "move" is stored in the LQM speech recognition chip (101). The term "move" is trained in the MQM speech recognition chip (102) and the term "communication" is stored in the MQM speech recognition chip (103).

The call is attempted (104) to determine if the call is successfully connected (105), if the call connection fails, retry, and if successful, transmitting a voice (e.g. mobile) stored in the LQM (1) (106) and the MQM (2). Speech recognition (eg, moving) is scored (107). A voice (e.g., communication) stored in the MQM 2 is transmitted (108) after being voice recognized at the MQM (2), and voice recognition (e.g., communication) at the LQM (1) is scored (109).

In this way, sound quality is measured based on the voice recognition score. If the call is not terminated by determining whether the call is terminated (110), repeating the sound quality measurement, and after the call is terminated after passing the call idle time (for example, 40 seconds) (111) and determining the end of the call test (112). When the call test ends, the call test ends. In this way, the voice recognition is performed by the calling party and the called party alternately performing voice recognition during a call.

The present invention configured and operated as described above uses a voice recognition chip to solve the problems of the prior art, which is different from the sound quality felt by a real person by a simple method of SINAD or tone level measurement. By performing the measurement, it is possible to measure the haptic sound quality more effectively.

Claims (2)

LQM (Land Quality Module) (1), which is connected to a regular landline telephone and performs automatic call generation and reception, call quality measurement, remote call control using modem, and automatic call generation and incoming by mobile phone Mobile Quality Module (MQM) 2 which performs call sound quality measurement, measurement path indication by GPS (Global positioning system), and remote control using a wireless modem, and by the LQM (1) and MQM (2) The CAM (Call Analysis Module) (3) performs the combined call analysis of measured data to perform measurement call statistics, voice quality evaluation for each voice channel belonging to a specific channel, detailed analysis of the cause of call incompleteness, and call recording location display. A mobile phone tangible call quality measurement system, comprising: LQM main control means (11) for performing overall control of the LQM (1); Multi-protocol interface means (12) connected to the LQM main control means (11) to perform a matching function; Automatic dialing means (13) for dialing under the control of the LQM main control means (11) via the multi-protocol interface means (12); A line converting means (14) connected to a telephone line for converting an input signal and outputting the line; notch filtering means (15) for filtering and outputting a signal received through the line changing means (14); Tone detection means 16 for receiving the output of the filtering means 15 detects and outputs a tone signal; Signal to Noise And Distortion (SINAD) indicating the degree of distortion of the signal-to-noise received from the output of the notch filtering means 15; SINAD measuring means (17) for measuring and outputting a ratio or outputting a tone signal for SINAD measurement to the MQM (2) under the control of the LQM main control means (11); An analog / digital (A / D) converting means (18) for receiving the output of the SINAD measuring means (17), converting it into a digital signal, and outputting the digital signal to the multiprotocol interface means (12); Input from An amplifying means 19 for amplifying and outputting an audio signal, and a codec means for receiving the amplified speech signal of the amplifying means 19 and compressing and outputting the audio signal, or receiving and compressing the decoded speech signal and performing reverse compression decoding. 20) a voice signal input from the codec means 20, and outputs a voice recognition ready signal and data to the multi-protocol interface means 12, and the voice signal stored under the control of the LQM main control means 11; A speech recognition processing means 21 for outputting the signal to the codec means 20, and a decompressed decoded transmission speech signal of the codec means 20 and a tone signal of the SINAD measurement means 17. MQM main control means 31 which performs overall control of the LQM 1 and the MQM 2 with the selection means 22 selected by the control signal of the means 11 and output to the line changing means 14. ; MQM group connected to the main control means (31) multi-protocol interface device 32 to perform the matching function; Automatic dialing means (33) for dialing under the control of the MQM main control means (31) via the multi-protocol interface means (32); Converts the dialing signal input from the automatic dialing means 33 and outputs it through the test mobile phone 44, or converts the incoming dialing signal received through the test mobile phone 44 into the automatic dialing means ( Line converting means 34 for outputting to 33; Notch filtering means (35) for filtering and outputting a signal received through the test mobile telephone (44); Tone detection means (36) for receiving the output of the notch filtering means (35) and detecting and outputting a tone signal; Receives the output of the notch filtering means 35 and measures and outputs a SINAD indicating the distortion of signal-to-noise, or outputs a tone signal for SINAD measurement to the LQM 1 under the control of the MQM main control means 31. SINAD measuring means 37; An analog / digital (A / D) converting means for receiving the output of the tone detecting means 36 or the output of the SINAD measuring means 37 and converting it into a digital signal and outputting the digital signal to the multiprotocol interface means 32; 38); Amplifying means (39) for amplifying and outputting a voice signal input from the test mobile telephone (44); Codec means (40) for receiving an amplified speech signal of the amplifying means (39) and compressing and outputting the amplified speech signal; Recognizes the voice signal input from the codec means 40, outputs a voice recognition ready signal and data to the multi-protocol interface means 32, and stores the voice signal stored under the control of the MQM main control means 31. Speech recognition processing means 41 for outputting to the means 40; The decompressed transmission voice signal of the codec means 40 and the tone signal of the SINAD measurement means 37 are received and selected by the control signal of the MQM main control means 31 to the test mobile phone 44. Selecting means 42 for outputting; And an MQM (2) having a GPS receiving means (43) for receiving current position information through an antenna and outputting the current position information to the multiprotocol interface means (32). The system of claim 1, wherein the speech recognition processing means (21, 41) comprises a speech recognition digital signal processor (DSP) for recognizing speech through training and storing the speech. .