WO2015192532A1

WO2015192532A1 - Method and apparatus for improving handover voice quality

Info

Publication number: WO2015192532A1
Application number: PCT/CN2014/087719
Authority: WO
Inventors: 罗小冬
Original assignee: 中兴通讯股份有限公司
Priority date: 2014-06-20
Filing date: 2014-09-28
Publication date: 2015-12-23
Also published as: CN105323803A

Abstract

A method and apparatus for improving handover voice quality, the method comprising: receiving a handover command sent by a base station controller BSC (S301); judging within a set period of time whether or not a non-critical frame is detected; if so, sending the handover command to a mobile station MS for handover; and if not, sending the handover command to the mobile station MS for handover after the set period of time expires; wherein the non-critical frame refers to a non-voice frame or a voice frame carrying no voice information (S302). Handover is performed at the moment of a noise signal by delaying handover time, thereby improving handover voice quality.

Description

Method and device for improving switching voice quality

Technical field

The present invention relates to the field of communications technologies, and in particular, to a method and apparatus for improving handover voice quality.

Background technique

Switching is a basic function in wireless communication networks. The user starts to stay in the A cell and switches to the B cell based on the wireless environment. The user equipment (User Equipment, UE for short) switches from the cell A (BTSA, Base Transceiver Station) to the cell B (BTSB). For the wireless communication system of the Global System for Mobile Communications (GSM), the switching of the voice service of the Circuit Switched Domain (CS) is a hard handover process. When the user A cell switches to the cell B, it takes a period of time for the mobile station (Mobile Station, MS for short) to synchronize with the cell B, and the data of the synchronization process user cannot be transmitted to the GSM network through the MS. At this point, the data the user is making a call will be discarded. For the process of handover, refer to FIG. 1, which is a schematic diagram of a handover scheme in the prior art. The Base Station Controller (BSC) receives the measurement information and determines the function of switching. The BSC sends a handover command (Handover Command, referred to as HO CMD) to the source base station BTS. After receiving the handover command, the source base station sends a handover command to the MS. After receiving the handover command, the MS sends the HO Access to the target base station. ), the uplink access process. After receiving the HO Access message of the MS, the target base station sends a HO Detect (Handover Detection Message) to the BSC, and performs an uplink synchronization process with the MS to send a physical message to the MS. After the MS and the target base station complete the uplink access and synchronization functions, The message that the handover is completed is sent to the BSC; the handover process is over.

The CS voice service is a basic service of the wireless communication network and a real-time service. During the call, if the voice data is not transmitted due to the switching function, the quality of the voice will be seriously affected; and due to the real-time nature of the voice, the retransmission cannot be compensated.

Summary of the invention

The embodiment of the invention provides a method and a device for improving the quality of handover voice, so as to at least solve the problem that the voice data cannot be transmitted during the handover process of the related technology cell.

To solve the above technical problem, in one aspect, the present invention provides a method for improving handover voice quality, including:

Receiving a handover command sent by the base station controller BSC;

Determining whether a non-significant frame is detected within the set time, and if so, transmitting a handover command to the mobile station MS to perform handover; if not, transmitting a handover command to the mobile station MS after the set time ends, and performing handover; Non-important frames refer to non-speech frames or speech frames that do not carry voice information.

Optionally, determining whether a non-significant frame is detected within the set time is specifically:

It is judged whether or not a non-significant frame of consecutive M frames is detected within the set time.

Optionally, determining whether a data frame is a non-significant frame includes:

When the discontinuous transmission DTX is turned off, or the discontinuous transmission DTX is turned on, and the data frame is a voice frame, the adaptive codebook gain of the data frame is extracted, according to whether the adaptive codebook gain is a non-significant frame for the data frame. Make a judgment.

Optionally, determining, according to the adaptive codebook gain, whether the data frame is a non-significant frame, specifically:

When the adaptive codebook gain is greater than a preset threshold, determining that the data frame is an important frame; wherein the important frame refers to a voice frame carrying voice information;

When the adaptive codebook gain is less than or equal to a preset threshold, it is determined that the data frame is a non-significant frame.

When the DTX is not continuously transmitted and the data frame is a non-speech frame, it is determined that the data frame is a non-significant frame.

In another aspect, the present invention provides an apparatus for improving handover voice quality, including:

a command receiving module, configured to receive a handover command sent by the base station controller BSC;

The switching module is configured to determine whether a non-significant frame is detected within the set time, and if yes, send a handover command to the mobile station MS to perform handover; if not, send a handover command to the mobile station MS after the set time ends Switching; wherein the non-important frame refers to a non-speech frame or a speech frame that does not carry voice information.

Optionally, the switching module determines whether a non-important frame of consecutive M frames is detected within a set time.

Optionally, the switching module determines whether a data frame is a non-significant frame, specifically:

When the discontinuous transmission DTX is turned off, or the discontinuous transmission DTX is turned on, and the data frame is a voice frame, the adaptive codebook gain of the data frame is extracted, according to whether the adaptive codebook gain is a non-significant frame for the frame signal. Make a judgment.

Optionally, the switching module is configured to determine whether the data frame is a non-significant frame according to the adaptive codebook gain, specifically:

When the adaptive codebook gain is greater than a preset threshold, determining that the frame signal is an important frame; wherein the important frame refers to a voice frame carrying voice information;

When the adaptive codebook gain is less than or equal to a preset threshold, it is determined that the frame signal is a non-significant frame.

When the DTX is not continuously transmitted and the frame signal is a non-speech signal, it is determined that the frame signal is a non-significant frame.

The beneficial effects of the present invention are as follows:

By delaying the switching time, the present invention switches at the moment when the noise signal is located, thereby improving the quality of the switched voice.

DRAWINGS

1 is a schematic diagram of a handover process in the prior art;

2 is a schematic diagram of a handover process in an embodiment of the present invention;

3 is a flowchart of a method for improving handover voice quality according to an embodiment of the present invention;

4 is a flowchart of another method for improving handover voice quality according to an embodiment of the present invention;

FIG. 5 is a schematic structural diagram of an apparatus for improving handover voice quality according to an embodiment of the present invention.

detailed description

The invention will be further described in detail below with reference to the drawings and embodiments. It is understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

The signals used in voice services can usually be divided into two types of signals, one is a voice signal, and the other is a noise signal. The noise signal is also often referred to as a comfort noise signal. Voice signals are basic information in the business process. In the present invention, the title is referred to as an important frame, specifically: an important frame refers to a voice frame carrying voice information; a comfort noise signal is a signal during a silent period during a call, and may generally be a mute signal or an environmental noise signal. In the present invention, it is referred to as a non-significant frame, and specifically, a non-important frame refers to a non-voice frame or a voice frame that does not carry voice information.

As shown in FIG. 1 , for the GSM prior art, generally, when the source base station receives the handover command, it can send a handover command to the MS, and then start to send the handover procedure. 2 is a schematic diagram of an improved handover scheme. The difference from the existing handover technology is that after the source base station receives the handover command of the BSC, the source base station does not immediately perform the handover process, but adds a function, that is, checks whether the uplink user plane data is in an unimportant state, if it is detected. If the user plane data is in a non-important state, the handover is initiated; otherwise, the handover is initiated for a certain period of time, and the time point of the handover is delayed from the portion of the voice signal to the portion of the noise signal, which greatly enhances the perception of voice during the handover, thereby improving Voice quality plays a big role in the application of CS voice. The present invention is directed to this idea. This patent changes the switching time to the time of the user plane data noise signal, so that the whole switching occurs mainly in the noise part, which greatly improves the voice perception of the switching.

As shown in FIG. 3, an embodiment of the present invention relates to a method for improving handover voice quality, including:

Step S301, receiving a handover command sent by the base station controller BSC;

Step S302, determining whether a non-significant frame is detected within the set time, and if yes, transmitting a handover command to the mobile station MS to perform handover; if not, transmitting a handover command to the mobile station MS after the set time is over, Switch.

In this step, first, the timer T0 is set.

Then, the detection process of the important frame and the non-important frame is performed; the detection is mainly performed based on the data of the user plane.

Secondly, the state processing of the important state and the non-important state is performed. After detecting the important frame or the non-important frame, it is necessary to set the state and perform the state conversion processing. When in the important state, if the continuous M frame detection is a non-significant frame, then the transition to the non-important state; when in the non-important state, if the continuous N frame detection is an important frame, then switch to the important state. M and N are integers greater than 1, which are preset values. M and N may be empirical values, or may be obtained through simulation tests, or may be obtained after several trial adjustments after the simulation test is obtained. The effect of switching when in a non-essential state is that since there is a false detection in the detection, by repeatedly detecting when the multi-frame data is a non-significant frame, the accuracy can be improved, and the maximum guarantee is in the non-speech. Switch when information is available.

Finally, the switchover includes the following three cases:

1. When the timer T0 does not time out and is in a non-critical state, the handover process is initiated.

2. If the timer T0 does not time out and is in an important state, the detection continues;

3. When the timer T0 times out, the handover process is initiated.

Hereinafter, a detailed description will be given by way of specific embodiments:

As shown in FIG. 4, an implementation scheme for improving handover voice quality by using AMR (Adaptive Multi Rate) is described by receiving data of an MS (ie, an application in an uplink direction), including:

Step S401, the source base station receives a handover command of the BSC. At this time, the BSC decides that the MS needs to switch to other cells. After receiving the message of the handover command, the source base station sets a timer T0 for the state detection of the handover. The setting of T0 can be considered within 1 s, and if it is too long, it also has some influence on the switching delay. The significance of the timer T0 is whether the source base station detects the non-important state during the handover in the T0 time, and if it is detected, performs handover; if it is not detected, it performs handover after the T0 expires.

In step S402, the source base station determines whether uplink DTX (discontinuous transmission) is turned on to determine a different detection and discrimination process. If the uplink DTX is off, the uplink MS always sends the SP frame (that is, the voice frame indicated by the bit on the network side. Since the voice frame may be a voice frame including a voice message, or may be a voice frame that does not include voice information, If the voice frame may be an important frame or a non-significant frame, go to step S405; if the uplink DTX is turned on, the uplink MS outputs the code according to the SP frame (speech frame) and the silence frame (SID, Silence Descriptor). If it is further determined that the SP frame or the silence frame is transmitted, the process proceeds to step S403.

In step S403, it is determined whether the frame is an SP frame; if yes, the process goes to step S405, and if no, the process goes to step S404. In this step, if the uplink DTX is enabled, the user plane on the base station side may receive the SP frame, the SID frame, and the Nodata frame (empty frame). When determining the non-important frame, other non-SP frame types need to be considered. For the voice frame, go to S404.

Step S404, the received SID frame and the Nodata frame are discriminated as non-important frames. The SID frame type includes SID_First (SID start frame) and SID_Update frame (SID update frame), which are classified as non-important frames; if the uplink does not receive valid data, the decoding result is Nodata frame, which is also classified as non- Important frame.

Step S405, when it is determined that the uplink DTX is off, or the uplink DTX is turned on, and the SP frame is received, the processing of this step is performed to extract the adaptive codebook gain of the data frame. In this embodiment, the AMR speech algorithm is used, and eight different rates need to be considered. The extraction method of each rate is slightly different. For the rates of AMR122k and 7.95k, the adaptive codebook gain can be directly in the parameter field of the encoding. Extraction; other rate, adaptive codebook The gain uses a predetermined parameter in conjunction with the fixed codebook gain, and the adaptive codebook gain needs to be further extracted by the predetermined parameter.

Step S406, determining whether the data frame is a non-significant frame or an important frame according to the adaptive codebook gain. For the AMR speech algorithm, the adaptive codebook gain is the gain of the correlation of the speech signal. The speech signal has a large correlation with the speech signal relative to the comfort noise signal, and the correlation of the noise signal is small. Therefore, the discrimination of non-significant frames can be performed by the adaptive codebook gain. When the adaptive codebook gain is greater than a preset threshold, the data frame is determined to be an important frame; when the adaptive codebook gain is less than or equal to a preset threshold, the data frame is determined to be a non-significant frame. Among them, the preset threshold value is obtained by simulation, and can also be obtained according to experience.

In step S407, the processing of the non-critical state and the important state is performed. In the process of detecting non-significant frames, these misdetections need to be filtered out due to some false detections. To this end, the concept of state is introduced. The non-essential state is that when a non-important frame of consecutive N frames is detected, it is in an unimportant state; likewise, if a continuous N frame important frame is detected, it is an important state. Set to important when starting a call.

In step S408, a handover delay determination process is performed. If an unimportant state is detected within T0, a handover command is sent to the MS to switch. The purpose of this is to switch the uplink switching point as much as possible during non-speech. If no non-significant state is detected and T0 does not time out, the previous detection process continues. If no non-critical state is detected within the time T0 and has timed out, then a handover command is sent to the MS for detection.

Step S409, the source base station sends a handover command to the MS. The MS performs a handover access procedure to the target base station, and the subsequent handover process is exactly the same as that of FIG. 1, and will not be described in detail.

Step S410, ending.

The next embodiment is an implementation scheme of switching voice quality by EFR (Enhanced Full Rate Speed Encoding), which is described by receiving data in the BSC direction, that is, the downlink direction.

Step S501, the source base station receives a handover command of the BSC. At this time, the BSC decides that the MS needs to switch to other cells. After receiving the message of the handover command, the source base station sets a timer T0 for the state detection of the handover. The setting of T0 can be considered within 1 s, and if it is too long, it also has some influence on the switching delay. The significance of the timer T0 is whether the source base station detects the non-essential state during the handover in the T0 time. If it is detected, it will be switched; if it is not detected, it will be switched after T0 expires.

Step S502, the source base station determines whether the downlink DTX (Discontinuous Transmission) is turned on to determine a different detection and discrimination process. If the downlink DTX is off, the BSC always sends the SP frame, then the step is changed. S505; If the downlink DTX is turned on, go to step S503. The downlink BSC performs corresponding VAD (Voice Activity Detection) discrimination according to the voice algorithm, and outputs an SP frame and a silence frame (SID frame). The output SP frame may also be an important frame or a non-important frame, and the SID frame is a non-significant frame.

Step S503, further determining whether the data frame is an SP frame; if yes, proceeding to step S505, and if no, proceeding to step S504. In this step, if the downlink DTX is turned on, the user plane on the base station side may receive two types of data, an SP frame and a SID frame.

Step S504, determining the received SID frame as a non-significant frame. The non-important frame of the EFR algorithm does not include a data type such as SID_FIRST, which is mainly a SID frame.

Step S505, when it is determined that the downlink DTX is off, or the downlink DTX is turned on, and the SP frame is received, the processing of this step is performed, and the adaptive codebook gain of the data frame is extracted. The EFR algorithm is a single rate processing that requires extraction of the adaptive codebook gain in 244 bits of the EFR.

Step S506, determining whether the data frame is a non-significant frame according to the adaptive codebook gain. For the EFR speech algorithm, the adaptive codebook gain is the gain of the correlation of the speech signal. The speech signal has a large correlation with respect to the comfort noise signal, and the correlation of the noise signal is small. Therefore, the discrimination of non-significant frames can be performed by the adaptive codebook gain. By detecting the size of the adaptive codebook gain, setting a threshold to determine whether the frame is an important frame or a non-significant frame, specifically: when the adaptive codebook gain is greater than a preset threshold, determining that the data frame is an important frame; When the adaptive codebook gain is less than or equal to a preset threshold, it is determined that the data frame is a non-significant frame. Among them, the preset threshold value is obtained by simulation, and can also be obtained according to experience.

In step S507, the processing of the non-critical state and the important state is performed. In the process of detecting non-significant frames, these misdetections need to be filtered out due to some false detections. To this end, the concept of state is introduced. The non-important state, that is, the non-important frame in which consecutive N frames are detected is in a non-important state, and the simultaneous detection of consecutive N-frame important frames is an important state. Set to important when starting a call.

In step S508, a handover delay determination process is performed. If an unimportant state is detected within T0, a handover command is sent to the MS to switch. The purpose of this is to switch the uplink switching point as much as possible during non-speech. If no non-significant state is detected and T0 is not exceeded, the previous detection process is continued. If no non-critical state is detected within T0 and has timed out, then a handover command is sent to the MS to switch.

Step S509, the source base station sends a handover command to the MS. The MS performs a handover access procedure to the target base station, and the subsequent handover process is exactly the same as that in FIG. 1, and this step is not described in detail.

Step S510, ending.

As shown in FIG. 5, the present invention also relates to an apparatus for improving handover voice quality by implementing the foregoing method, including:

The switching module determines whether a continuous M frame non-significant frame is detected within a set time.

The switching module determines whether a certain frame signal is a noise signal, specifically:

When the discontinuous transmission DTX is turned off, or the discontinuous transmission DTX is turned on, and the data frame is a voice frame, the adaptive codebook gain of the voice signal is extracted, and the frame code is judged to be an unimportant frame according to the adaptive codebook gain. .

The switching module determines whether the data frame is a non-important frame according to the adaptive codebook gain, specifically:

The switching module determines whether a data frame is a non-significant frame, specifically:

While the preferred embodiments of the present invention have been disclosed for purposes of illustration, those skilled in the art will recognize that various modifications, additions and substitutions are possible, and the scope of the invention should not be limited to the embodiments described above.

Industrial applicability

As described above, the method and apparatus for improving the quality of the switched voice provided by the embodiments of the present invention have the following beneficial effects: by delaying the switching time, switching is performed at the moment when the noise signal is located, thereby improving the voice quality of the handover.

Claims

A method for improving the quality of switched voices, including:

Receiving a handover command sent by the base station controller BSC;

Determining whether a non-significant frame is detected within the set time, and if so, transmitting a handover command to the mobile station MS to perform handover; if not, transmitting a handover command to the mobile station MS after the set time ends, and performing handover; Non-important frames refer to non-speech frames or speech frames that do not carry voice information.
The method for improving switching voice quality according to claim 1, wherein determining whether the non-significant frame is detected within the set time comprises:

It is judged whether or not a non-significant frame of consecutive M frames is detected within the set time.
The method for improving switching voice quality according to claim 1 or 2, wherein determining whether a certain data frame is a non-significant frame comprises:

When the discontinuous transmission DTX is turned off, or the discontinuous transmission DTX is turned on, and the data frame is a voice frame, the adaptive codebook gain of the data frame is extracted, according to whether the adaptive codebook gain is a non-significant frame for the data frame. Make a judgment.
The method for improving switching voice quality according to claim 3, wherein determining whether the data frame is a non-significant frame according to the adaptive codebook gain comprises:

When the adaptive codebook gain is greater than a preset threshold, determining that the data frame is an important frame; wherein the important frame refers to a voice frame carrying voice information;

When the adaptive codebook gain is less than or equal to a preset threshold, it is determined that the data frame is a non-significant frame.
The method for improving switching voice quality according to claim 1 or 2, wherein determining whether a certain data frame is a non-significant frame comprises:

When the DTX is not continuously transmitted and the data frame is a non-speech frame, it is determined that the data frame is a non-significant frame.
A device for improving switching voice quality, comprising:

a command receiving module, configured to receive a handover command sent by the base station controller BSC;

The switching module is configured to determine whether a non-significant frame is detected within the set time, and if yes, send a handover command to the mobile station MS to perform handover; if not, send a handover command to the mobile station MS after the set time ends Switching; wherein the non-important frame refers to a non-speech frame or a speech frame that does not carry voice information.
The apparatus for improving switching voice quality according to claim 6, wherein the switching module determines whether a non-important frame of consecutive M frames is detected within a set time.
The apparatus for improving the quality of a switched voice according to claim 6 or 7, wherein the determining, by the switching module, whether a certain data frame is a non-significant frame comprises:

When the discontinuous transmission DTX is turned off, or the discontinuous transmission DTX is turned on, and the data frame is a voice frame, the adaptive codebook gain of the data frame is extracted, according to whether the adaptive codebook gain is a non-significant frame for the frame signal. Make a judgment.
The apparatus for improving handover voice quality according to claim 8, wherein the switching module includes, according to the adaptive codebook gain, whether the data frame is a non-significant frame:

When the adaptive codebook gain is greater than a preset threshold, determining that the frame signal is an important frame; wherein the important frame refers to a voice frame carrying voice information;

When the adaptive codebook gain is less than or equal to a preset threshold, it is determined that the frame signal is a non-significant frame.
The apparatus for improving the quality of a switched voice according to claim 6 or 7, wherein the determining, by the switching module, whether a certain data frame is a non-significant frame comprises:

When the DTX is not continuously transmitted and the frame signal is a non-speech signal, it is determined that the frame signal is a non-significant frame.