CN1649421A - Self adaptive multiple rate encoding and transmission method for voice - Google Patents

Abstract

This invention relates to an adaptive multi-rate (AMR) code/decode and transmitting method in UMTS, which overcomes the short coming of losing AMR encode frames caused by switching in voice service, applies multimode AMR encode frame output method to configurate multi-voice transmission format combination and buffer store the AMR code frames, not discarding the delayed frames and set reduction for controlling AMR code mode at the wireless access network interface used in UMTS directly.

Description

The adaptive multi-rate coding of speech and transmission method

Technical field

The present invention relates to a kind of method of in universal mobile telecommunications system (UMTS), carrying out the coding of self adaptation and many speed and transmitting the data that described coding produces, insert and the method for control adaptive multi-rate (AMR) coded frame and the method that on middle core net (CN) and wireless access network (RAN) interface, transmits adaptive multi-rate (AMR) coded-bit in the method that is specifically related to improve the method for adaptive multi-rate (AMR) encoder encodes, the data that output to adaptive multi-rate (AMR) decoder are handled, the wireless access process.

Background technology

The technical specification (TS) of 3G (Third Generation) Moblie partner plan (3GPP) is made following regulation to adaptive multi-rate (AMR) encoder:

Stipulate among-3GPP the TS26.071, as input, the voice signal of pulse code modulation (pcm) form that perhaps voice signal of extended formatting is converted to 13 bit accuracy is as input with the voice signal of the precision pulse code modulation (pcm) form that is 13 bits for adaptive multi-rate (AMR) encoder.The sample rate of encoder is 8000 samples/sec, the piece that encoder is formed with the sample of the pulse code modulation (pcm) form of 160 13 bit accuracy is that unit encodes, and produces coding mode and be 12.2,10.2,7.95,7.40,6.70,5.90,5.15, the speech coding frame or the coded frame of mourning in silence of one of 4.75kbps.

Stipulate among the 3GPP TS26.171, as input, the voice signal of pulse code modulation (pcm) form that perhaps voice signal of extended formatting is converted to 14 bit accuracy is as input with the voice signal of the precision pulse code modulation (pcm) form that is 14 bits for adaptive multi-rate (AMR) encoder.The sample rate of encoder is 16000 samples/sec, the piece that encoder is formed with the sample of the pulse code modulation (pcm) form of 320 14 bits is that unit encodes, and produces coding mode and be 23.85,23.05,19.85,18.25,15.85,14.25,12.65,8.85, the speech coding frame or the coded frame of mourning in silence of one of 6.6kbps.

Stipulate among-3GPP TS26.071 and the TS26.171 that adaptive multi-rate (AMR) encoder can be according to the per 20 milliseconds of bit rates that change a speech coding frame of order that change coding mode.

Stipulate among-3GPP TS26.101 and the TS26.201, in the content of each frame that adaptive multi-rate (AMR) encoder is exported, should indicate the coding mode of this frame and the indication coding mode and the request coding mode of adaptive multi-rate (AMR) encoder, and the coding mode of frame, indication coding mode and request coding mode all are unique.

Among the technical specification 3GPP TS25.415 in the universal mobile telecommunications system (UMTS) in the appendix A to the adaptive multi-rate in the speech calling procedure (AMR) coded frame the transmission in network made indicative explanation, therefrom can obtain to draw a conclusion:

-core net (CN) is carried out the mutual of signaling message with the same wireless access network of RANAP (RAN), send the attribute of RAB (RAB) to wireless access network (RAN), this attribute has comprised the bit size of each class in the bit size of each pattern of 20 milliseconds of Speech frames of adaptive multi-rate (AMR) and each pattern.

-radio network controller (RNC) is arranged RAB stream combination indication (RFCI), the pattern correspondence of 20 milliseconds of Speech frames of the same adaptive multi-rate of RFCI (AMR) according to the attribute of RAB (RAB).

-radio network controller (RNC) disposes some dedicated channels (DCH) of collaborative work according to RFCI, the same RFCI of a transformat (TF) among the DCH is corresponding, therefore the corresponding one or more transformat combination of RFCI identify (TFCI), and described TFCI is also just corresponding with the pattern of 20 milliseconds of Speech frames of adaptive multi-rate (AMR).

-radio network controller (RNC) sends core net-radio access network interface initialization frame (Iu initialisationframe), and the some RFCI that its institute of notice core net can provide to code pattern converter and the son stream of related RAB (RAB) thereof are big or small.

Some DCH of-travelling carriage (UE) configuration collaborative work, in layoutprocedure, need RRC signaling message mutual between radio network controller (RNC) and the travelling carriage (UE), travelling carriage (UE) can inform that its audio coder ﹠ decoder (codec) is received and the RAB (RAB) of the speech that will send stream (sub-flow) structure according to transformat (TF), and this RAB (RAB) stream (sub-flow) structure is exactly each class bit of 20 milliseconds of Speech frames of adaptive multi-rate (AMR).

The codec that the information of-some RFCI that will obtain from core net-wireless access network (Iu) interface initialization frame (Iu initialisation frame) and the son of related RAB (RAB) stream size thereof is given code pattern converter uses for configuration.

-sending when containing the Service Data Unit (SDU) of Speech frame in core net-radio access network interface, RFCI is set to the corresponding value of pattern with 20 milliseconds of Speech frames of this adaptive multi-rate (AMR) in core net-wireless access network users face (Iu UP) frame;

In RNC, with this RFCI core net-wireless access network users face (Iu UP) frame is decomposed on some DCH of collaborative work, these DCH have different separately bit protection ranks;

The TFCI of radio frames is selected by medium Access Layer MAC;

Travelling carriage (UE) receives the SDU that contains Speech frame from some DCH of collaborative work, and they are reverted to 20 milliseconds of Speech frames of adaptive multi-rate (AMR), can indicate the structure of Speech frame with TFCI to voice codec.

By afore mentioned rules as seen, each 20 milliseconds speech input is converted into the coded frame of some specific adaptive multi-rate (AMR) pattern.

Configuration to the DCH transmission channel format of 20 milliseconds of Speech frames of above-mentioned mapping adaptive multi-rate (AMR) among the technical specification 3GPP TS25.944 in the universal mobile telecommunications system (UMTS) has provided example, and in example, be described being input to radio frame segmentation from transport channel data, among 4.1.1.3.1.2 that specifically describes at 3GPP TS25.944 and the 4.2.1.3.1.2 as seen.

In universal mobile telecommunications system (UMTS) system, transmission data in the transmission channel are sent to compound channel, paste through CRC, transmission block cascade/code block is cut apart, operating procedures such as chnnel coding, become the input bit sequence of rate-matched module, the output bit sequence that is complementary with the bit number of the radio frames of physics channel through rate-matched output, when the punching to rate-matched has maximum punching ratio to limit, those transmission data that can make the punching number surpass this ratio restriction can initiatively be abandoned, when the restriction of maximum punching ratio is not carried out in the punching of rate-matched, the excessive transmission data of those punching ratios also can't correctly be reduced at receiving terminal.For above-mentioned reasons, be equivalent to the physical channel rate of physical channel radio frames bit number, can determine the validity of transmission channel format combination; Effectively those transmission channel format corresponding to 20 milliseconds of Speech frame patterns of adaptive multi-rate (AMR) make up in the transmission channel format combination of sets, have determined the many speed of efficient adaptive (AMR) Speech frame pattern.

In universal mobile telecommunications system (UMTS) system, the data that are sent to transmission channel start from logic channel, each logic channel has a priority level with respect to other logic channel, data send in corresponding transmission channel during Transmission Time Interval in succession, each Transmission Time Interval (TTI) presents the specific duration for its pairing transfer channel, transformat combination be chosen in the beginning in the reference transmission time interval time implement, the described reference transmission time interval represents to have the Transmission Time Interval of transfer channel of the minimum duration of Transmission Time Interval.When the data on the scheduling logic channel are to transmission channel, carry out the transmission channel format combination selection, preferential transmission channel format combination of from effective transmission channel format combination of sets, selecting to transmit the high priority logic channel; When not dispatching the logic channel that transmits 20 milliseconds of Speech frames of adaptive multi-rate (AMR) and other high priority logic channel simultaneously to transmission channel the time, can make 20 milliseconds of Speech frames of adaptive multi-rate (AMR) that can not be scheduled be dropped.

When travelling carriage when carrying out the switching of universal mobile telecommunications system (UMTS), the Radio Link between travelling carriage and the wireless access network changes: this change is the temporary transient interruption of Radio Link during direct-cut operation; Soft handover is unremitting switching, but the decline of temporary transient Radio Link physical channel rate was often arranged between transfer period.This shows that the variation because of this physical channel between transfer period can cause losing of some adaptive multi-rate (AMR) 20 milliseconds of Speech frames.

The real-time of speech requires Speech frame to be transmitted and to receive with fast as far as possible speed, travelling carriage obtains an adaptive multi-rate (AMR) Speech frame from the WAP (wireless access protocol) functional unit and deciphers for per 20 milliseconds, when in the Speech frame that the WAP (wireless access protocol) functional unit can't provide or provide when wrong, adaptive multi-rate (AMR) can avoid a small amount of Speech frame to lose the negative effect of being brought with the error concealment technology.

The side that transmission contains adaptive multi-rate (AMR) Speech frame sends with the per 20 milliseconds fixed rates of dispatching frames, when when having a frame to send for certain 20 milliseconds, abandoning this frame, at next 20 milliseconds of next Speech frames of scheduling, continuous per 20 milliseconds one adaptive multi-rate (AMR) Speech frame comes from the coding of adaptive multi-rate (AMR) voice encryption device to speech.For each speech of 20 milliseconds, the aerial transmitting time of its adaptive multi-rate (AMR) Speech frame is that fix, immutable.

The notion of adaptive multi-rate (AMR) is that the connection between travelling carriage (UE) and the network provides a kind of many rate capacities, this ability is reflected on adaptive multi-rate (AMR) coding mode of the voice data that outputs to logic channel, the current techniques standard is not described the restriction that the excellent means of utilizing adaptive multi-rate (AMR) notion optimization transmission control and the Speech frame that can allow are lost, thus, the current techniques standard of universal mobile telecommunications system (UMTS) system has been reserved the space for the application that promotes adaptive multi-rate (AMR) notion.

Summary of the invention

The technical problem that solves

The technical problem to be solved in the present invention is to reduce losing of adaptive multi-rate (AMR) 20 milliseconds of Speech frames in the voice call process, although adaptive multi-rate (AMR) can avoid Speech frame to lose the negative effect of being brought with the error concealment technology, but error concealment technology this avoided acting on just losing of individual frame there is good effect, because it removes to make up lost frames with paid-in adjacent Speech frame, so between lost frames that make up and the actual loss frame error is arranged, the very little or even consecutive hours in interval between lost frames, it is very big that the error of the lost frames of back can become.

Adaptive multi-rate (AMR) Speech frame to be lost in following three places more outstanding, the one, the losing of the Speech frame that the unexpected decline of physical channel rate causes, the 2nd, the losing of the caused Speech frame of high priority logic channel of priority scheduling burst, the 3rd, the losing of the Speech frame between transfer period.

It is invalid that the unexpected decline of physical channel rate can make the many speed of original efficient adaptive (AMR) Speech frame transformat combination become, thereby cause that adaptive multi-rate (AMR) Speech frame loses, adapt to this variation, the one, change adaptive multi-rate (AMR) speech coding pattern, generally need the stand-by period more than 40 milliseconds, this is because travelling carriage obtains the time that information that physical channel rate changes is wanted 20 milliseconds of radio frames from the radio frames of receiving, changes adaptive multi-rate (AMR) speech coding pattern and will wait for that at least 20 milliseconds time just can come into force; The 2nd, use new transformat combination (TFC), when it is when needing the transformat combination (TFC) of configuration, to need a signaling layoutprocedure.

When the high priority logic channel of burst need be dispatched, when carrying out transformat combination (TFC) selection, it is no longer selected originally to transmit adaptive multi-rate (AMR) Speech frame transformat combination, thereby causes that adaptive multi-rate (AMR) Speech frame loses, and just steals frame.Non-speech service proportion increases in the UMTS system, and the complexity and the quantity of signaling message all are greatly improved, the more important thing is and in the UMTS system, use the compound channel that is mapped to physical channel to carry all professional logic channels, unlike GPRS with the physical channel of Packet Service separately with the physical channel of voice service, make, when physical channel has been seized in the scheduling of packet traffic channel, especially when the priority of the logic channel of Packet Service is higher than the speech logic channel, the number of times of the stolen frame of voice channel increases greatly, needs the mechanism that reduces the negative effect of stealing frame.

Between transfer period,, must cause that adaptive multi-rate (AMR) Speech frame loses if direct-cut operation just has the temporary transient interruption of Radio Link; If soft handover and the decline of temporary transient Radio Link physical channel rate is arranged, may cause that also adaptive multi-rate (AMR) Speech frame loses.

Technical scheme

The present invention has provided technical scheme aspect the transmission reception to adaptive multi-rate (AMR) speech in order to solve the problems of the technologies described above on the transmission reception on coding control, the transmission channel and core net-wireless access network (Iu) interface.

---generate and transmit the method for adaptive multi-rate (AMR) coded frame in the travelling carriage

The coding and the scheme of transmission in travelling carriage of the present invention is a kind of method that generates and transmit adaptive multi-rate (AMR) coded frame in travelling carriage, in the voice call process, it is 20 milliseconds Speech frame that adaptive multi-rate (AMR) encoder obtains length by the sampling to the input voice signal, each Speech frame contains the sample of 20 milliseconds of speeches, to 20 milliseconds of Speech frame codings, what described coding produced is adaptive multi-rate (AMR) coded frame that meets universal mobile telecommunications system (UMTS) standard, need the wave point functional unit to send the voice data of adaptive multi-rate (AMR) coded frame form, the wave point functional unit sends described voice data by the transformat of each transmission channel, it is characterized in that:

Send coded command to adaptive multi-rate (AMR) encoder, described coded command has been specified a plurality of adaptive multi-rate (AMR) pattern, effective Speech frame coded sequence that adaptive multi-rate (AMR) encoder is produced 20 milliseconds of Speech frames coding according to described coded command, it is the set of a plurality of adaptive multi-rate (AMR) speech coding frame, it is an adaptive multi-rate (AMR) coded frame of mourning in silence, when described Speech frame coded sequence was a plurality of adaptive multi-rate (AMR) speech coding frame, the pattern of its adaptive multi-rate (AMR) coded frame was consistent with the adaptive multi-rate in the described coded command (AMR) pattern;

To wave point functional unit output Speech frame coded sequence;

In the wave point functional unit, send adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence to the process of transmission channel, comprise, in this Speech frame coded sequence, select an adaptive multi-rate (AMR) coded frame and select a transformat combination, the transformat combination of picking out, comprised the transformat that transmits all class bits of adaptive multi-rate (AMR) coded frame of picking out, adaptive multi-rate (AMR) coded frame of picking out has been dispatched on the transmission channel with this transformat combination.After the transmission of a selected adaptive multi-rate (AMR) coded frame, abandon remaining adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence.

Wave point functional unit in the above-mentioned scheme that coding is controlled and sent in travelling carriage is the functional unit that air interface is provided for the user data of travelling carriage, its input is these user data of speech, grouping and signaling, what export is radio frames, and it will meet the communication protocol of Radio Resource control (RRC), Radio Link control (RLC), medium access control unit (MAC) and physical layer (PHY).

Use above-mentioned in travelling carriage coding control and the scheme that sends, when send on the speech logic channel be non-quiet frame the time; If compare with preceding transformat combination (TFC) when transmitting non-quiet frame when once dispatching the speech logic channel to transfer channel, when needs are scheduled for the non-voice transmission logic channel of high priority that transfer channel gets on when changing, perhaps change when causing that valid transport format combination (TFC) set changes when the maximum transmission rate of physical channel, it no longer is the suitable transformat combination (TFC) of modal selection of the non-coded frame of mourning in silence of the preceding adaptive multi-rate that once transmits on transmission channel (AMR) that transformat combination (TFC) when dispatching the speech logic channel to transfer channel is selected, and this is to have the pattern of the non-coded frame of mourning in silence of multiple adaptive multi-rate (AMR) to choose from because of current.The benefit of bringing like this is: produce the non-coded frame of mourning in silence of adaptive multi-rate (AMR) of various modes owing to once encode, in the time can't arriving transformat combination (TFC), may find the transformat combination (TFC) of the non-coded frame pattern of mourning in silence of another adaptive multi-rate (AMR) for previous modal selection; When the multiple transformat combination (TFC) of corresponding various modes is available, can select for use the transformat combination (TFC) of the highest pattern of bit rate to remove to transmit speech.

Can determine class (class) bit of its pairing adaptive multi-rate (AMR) pattern-coding frame according to the transformat (TF) of the Radio interface protocols functional unit of travelling carriage configuration, determine its pairing adaptive multi-rate (AMR) pattern according to the transformat combination (TFC) of configuration again, like this, just can obtain the optional scope of the pattern the coded command from the transport format combination set (TFCS) of configuration.

In travelling carriage, determine that according to Radio interface protocols the coding control of pattern in the coded command and the scheme that sends are: according to the above-mentioned method that in travelling carriage, generates and transmit adaptive multi-rate (AMR) coded frame, each adaptive multi-rate (AMR) pattern in the described coded command is corresponding with a speech transmissions format combination in the transport format combination set of travelling carriage, and described speech transmissions format combination is the transformat combination that comprises all class bit transformats of a kind of adaptive multi-rate (AMR) pattern-coding frame.

Doing the benefit of bringing like this is: travelling carriage can make the pattern of the frame of adaptive multi-rate (AMR) encoder output be complementary with the transformat combination (TFC) that changes, travelling carriage Radio interface protocols functional unit can inform the variation on the Radio Link adaptive multi-rate (AMR) encoder in time, when particularly Radio Link is in quick variable condition, for example during high-speed motion, the Radio interface protocols functional unit can detect this quick variable condition, just can require encoder to adapt to this quick variation in the mode of exporting multimodal adaptive multi-rate (AMR) coded frame.

---adaptive multi-rate (AMR) coding of network side and the method that sends

Similar in the scheme of the coding of the network side of universal mobile telecommunications system (UMTS) and transmission coexists travelling carriage, the present invention is given in a kind of method that generates and transmit adaptive multi-rate (AMR) coded frame of the network side of universal mobile telecommunications system (UMTS), in the voice call process, code pattern converter in core net becomes adaptive multi-rate (AMR) coded frame with the speech coding conversion of signals of input, described adaptive multi-rate (AMR) coded frame is placed on the Iu user plane frame and sends to radio network controller (RNC), radio network controller (RNC) sends voice data on transmission channel, it is characterized in that:

Send coded command to code pattern converter, described coded command has been specified a plurality of adaptive multi-rate (AMR) pattern, code pattern converter is by described coded command, speech coding signal to input carries out encoding operation, the Speech frame coded sequence of the effective 20 milliseconds of length speeches that produced, it is the set of some (more than or equal to 1) individual adaptive multi-rate (AMR) speech coding frame, it is an adaptive multi-rate (AMR) coded frame of mourning in silence, when described Speech frame coded sequence was many (greater than 1) individual adaptive multi-rate (AMR) speech coding frame, the pattern of its adaptive multi-rate (AMR) coded frame was consistent with the adaptive multi-rate in the described coded command (AMR) pattern;

With all adaptive multi-rate (AMR) coded frame in core net-radio access network interface (Iu) user plane frame voice-over frame coded sequence, send core net-radio access network interface (Iu) user plane frame of described voice-over frame coded sequence to radio network controller (RNC);

In radio network controller (RNC), dispatch adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence to the process of transmission channel, comprise, in this Speech frame coded sequence, select an adaptive multi-rate (AMR) coded frame and select a transformat combination, the transformat combination of picking out, comprised the transformat that transmits all class bits of adaptive multi-rate (AMR) coded frame of picking out, adaptive multi-rate (AMR) coded frame of picking out has been dispatched on the transmission channel with this transformat combination.After the transmission of a selected adaptive multi-rate (AMR) coded frame, abandon remaining adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence.

Determine that by radio network controller (RNC) scheme that the coding of pattern in the coded command generates and sends is at universal mobile telecommunications system (UMTS) network side, method according to generation in above-mentioned universal mobile telecommunications system (UMTS) network side and transmission adaptive multi-rate (AMR) coded frame, radio network controller (RNC) sends core net-radio access network interface (Iu) user plane rate controlled frame to core net, sequence number sign (RFCI n indicator) territory of a plurality of RAB stream combination indications in core net-radio access network interface (Iu) user plane rate controlled frame is corresponding one by one with a plurality of RAB stream combination indications (RFCI), these RAB stream combination indications (RFCI) are corresponding with adaptive multi-rate (AMR) pattern again, after core net is received described core net-radio access network interface (Iu) user plane rate controlled frame, to the coded command that code pattern converter sends, a plurality of adaptive multi-rate (AMR) pattern of appointment in the coded command is exactly described RAB stream combination indication (RFCI) pairing adaptive multi-rate (AMR) voice mode.

---the principle of delayed delivery and decoding and according to

Lose for avoiding Speech frame, the present invention proposes the method that transmission channel or physical channel is stopped to transmit adaptive multi-rate (AMR) the coded frame delayed delivery during the voice data, and corresponding mobile station adaptive multi-rate (AMR) postpones the method for decoding.Delayed delivery will be in conjunction with the method for above-mentioned generation and transmission Speech frame coded sequence, because channel when sending once more after postponing and the channel when being ready for sending originally may be distinguished bigger, the pattern of original adaptive multi-rate (AMR) speech can not be continued to use, the pattern of suitable adaptive multi-rate (AMR) must be sought.

The characteristic of speech is in real time: end-to-end time delay is little, maximum allow end-to-end time delay will according to the people to audio session feel decide, general subjective evaluation end-to-end time delay is acceptable between the 200-300 millisecond, is that the quality of service that can not cause tangible time delay aspect descends as long as method of the present invention controls to end-to-end time delay below 200 milliseconds therefore.

The method of the following explanation many speed of travelling carriage delayed adaptation (AMR) earlier decoding then, is explained to the sending method of the buffer type of travelling carriage and network side respectively.

---the method for the delay decoding of travelling carriage

The scheme that travelling carriage adaptive multi-rate (AMR) decoder that the present invention proposes postpones decoding is: the method for the adaptive multi-rate that a kind of processing of travelling carriage receives (AMR) coded frame, travelling carriage receives adaptive multi-rate (AMR) coded frame from the transmission channel that transmits speech, adaptive multi-rate (AMR) coded frame that adaptive multi-rate (AMR) decoder is received travelling carriage is characterized in that by the per 20 milliseconds of decodings outputs of finishing an adaptive multi-rate (AMR) coded frame of the order after arrive first:

Be that a voice call is set up the placement adaptive multi-rate, (AMR) lower limit of coded frame buffer area and this buffer area, receive the adaptive multi-rate of described voice call for the first time from travelling carriage, (AMR) coded frame begins, with the adaptive multi-rate of receiving, (AMR) the speech coding frame is stored in the buffer area, adaptive multi-rate in buffer area, when (AMR) number of coded frame reaches described lower limit, adaptive multi-rate, (AMR) decoder begins adaptive multi-rate that travelling carriage is received, (AMR) coded frame decoding

When travelling carriage is in voice (SPEECH) pattern, check that adaptive multi-rate (AMR) speech coding frame increases in per 20 milliseconds of buffer areas number: when described number less than 1 the time, initiatively add adaptive multi-rate (AMR) reception lost frames; When described number greater than 1 and buffering area in when having under voice (SPEECH) pattern reception lost frames that initiatively add, delete described under voice (SPEECH) pattern the reception lost frames of interpolation initiatively.

Use the benefit of the method for this buffer memory adaptive multi-rate (AMR) coded frame to be, it makes, the transmit leg of adaptive multi-rate (AMR) coded frame do not introduced in the background for another example like that, must abandon this frame in case an adaptive multi-rate (AMR) coded frame just can't be sent in two radio frames of 20 milliseconds.Use method of the present invention, even in follow-up radio frames, send adaptive multi-rate (AMR) coded frame in these 20 milliseconds of voice datas, this adaptive multi-rate (AMR) coded frame still can be decoded with correct order, described follow-up transmission relates to the frame of selecting a kind of pattern in various modes adaptive multi-rate (AMR) the speech coding frame from sequence and sends, and sending described adaptive multi-rate (AMR) speech coding frame both can be that radio network controller (RNC) also can be a travelling carriage.

Adaptive multi-rate (AMR) coded frame is used the rlc layer pattern of transparent mode, not to carrying the protocol Data Unit append serial numbers of adaptive multi-rate (AMR) coded frame, the recipient sorts according to the time sequencing of receiving adaptive multi-rate (AMR) coded frame, if not having adaptive multi-rate (AMR) coded frame in two per 20 milliseconds radio frames can use for the recipient, the recipient just inserts one and receives lost frames, and the type that receives lost frames is that the middle type of receipt (RX_TYPE) of the table (Table lc) of the 4.1.3 of 3GPP TS26.101 is the type of free of data (NO_DATA).Owing to average adaptive multi-rate (AMR) the speech coding frame transmission of per 20 millisecond of one frame in the methods of the invention in uneven mode, just need the reception lost frames that those mistakes that is caused by inhomogeneous transmission are added be deleted from buffer area, the negative effect that elimination is caused by inhomogeneous transmission Speech frame, this utilize the buffer area memory space adapt to speech transmit temporary transient method of pausing make adaptive multi-rate (AMR) speech coding frame when coding is finished mutually between time gap farthest kept.

Travelling carriage has 2 kinds of mode of operation-voice (SPEECH) patterns and comfort noise (COMFORT_NOISE) pattern, transmit leg sends quiet frame discontinuously under comfort noise (COMFORT_NOISE) pattern, and the reception lost frames that the recipient initiatively adds are effective.For the reception lost frames that initiatively add under voice (SPEECH) pattern, wherein part receives lost frames has many speed (AMR) speech coding frame that the many speed in opposite end (AMR) voice encryption device produces and that lost in transport process corresponding with it, another part does not then have, and this part that does not have caused by inhomogeneous send mode of the present invention.Those do not have the reception lost frames of corresponding many speed (AMR) speech coding frame is unnecessary, do not play the many speed of compensating missing (AMR) speech coding frame, if can not be deleted by the deletion action of inhomogeneous send mode, the time delay and the quality that can cause speech descend, and therefore need carry out certain limitation to the reception lost frames that initiatively add:

According to buffer memory adaptive multi-rate (AMR) coded frame of travelling carriage and the method for reseptance of interpolation and deletion reception lost frames, set up the number limits value of the reception lost frames that under voice (SPEECH) pattern, initiatively add of described buffer area, when the number of the reception lost frames that initiatively add under voice in the buffer area (SPEECH) pattern reaches described number limits value, under voice (SPEECH) pattern, no longer add receiving lost frames.

To the reception lost frames number of initiatively adding carry out above-mentioned restriction be because: be limited from the number that sends many speed (AMR) coded frame that is delayed that travelling carriage can hold to the path of the RNC of receiving mobile, the number of the reception lost frames that initiatively add under the therefore corresponding with it SPEECH pattern also is limited, should have described number limits value to embody this finiteness.

Simultaneously when the reception lost frames of finding initiatively to add produce influencing of increase time delay to speech, promptly, when many speed (AMR) coded frame number of the storage of finding described buffer area has surpassed the buffer area higher limit of regulation, can adopt the method for many speed (AMR) coded frame that initiatively abandons in the buffer area to eliminate the counter productive that time delay increases.

For the radio network controller (RNC) that sends adaptive multi-rate (AMR) coded frame, it can use the reception buffer area lower limit of recipient's travelling carriage to control the number of adaptive multi-rate (AMR) coded sequence that it keeps when switching in sending buffer area, therefore need a kind of mechanism to transmit the reception buffer area lower limit of recipient's travelling carriage, the present invention proposes travelling carriage sends method from the message that contains described buffer area lower limit to network side, transmit this lower limit with the method to core net and radio network controller (RNC), control time delay with this when the network side delayed mode sends, travelling carriage also can be as reference when setting with the limit value of the buffer area of discontinuous manner delivery plan.

---generate and transmit the method for adaptive multi-rate (AMR) coded frame in the travelling carriage with discontinuous manner

Travelling carriage has the technical scheme of the transmission of interruption to be: according to above-mentioned generating in travelling carriage and transmission adaptive multi-rate (AMR) coded frame method, during the radio frames of travelling carriage time-out transmission voice-over data, the Speech frame coded sequence from adaptive multi-rate (AMR) encoder is carried out buffer memory, after the described EOT end of timeout, adaptive multi-rate (AMR) coded frame in transmitting the Speech frame coded sequence is in the process of transmission channel, in the mode of per 20 a milliseconds of scheduling above adaptive multi-rate (AMR) coded frame, handle the part or all of Speech frame coded sequence that is buffered to transmission channel.

Travelling carriage suspends the situation that sends radio frames to be had multiple, all belong to the category that suspends the radio frames that sends the voice-over data, switch the sub-district is modal, other also have: the dynamic channel adjustment, promptly travelling carriage switches to another physical channel from a physical channel.Though image duration, travelling carriage did not suspend the transmission radio frames steathily, it belongs to the category that suspends the radio frames that sends the voice-over data.

When referring to that the sub-district is between transfer period during the described travelling carriage time-out transmission radio frames, promptly, between transfer period the Speech frame coded sequence from adaptive multi-rate (AMR) encoder is carried out buffer memory in the sub-district, after switching is finished, adaptive multi-rate (AMR) coded frame in transmitting the Speech frame coded sequence is in the process of transmission channel, in the mode of per 20 milliseconds of scheduling adaptive multi-rate (AMR) coded frame that (do not comprise) more than, handle the part or all of Speech frame coded sequence that is buffered to transmission channel.

Adaptive multi-rate (AMR) coded frame that the buffer memory sub-district can't send between transfer period; behind handover success; adaptive multi-rate (AMR) coded frame of these buffer memorys is sent with per 20 milliseconds of more than one speed of scheduling; just can realize not having between transfer period and suspend the phenomenon that speech is broadcasted; the benefit that this method that Here it is is brought is that it has overcome switches the uncomfortable effect that the Speech frame that causes is lost the hearer that brings.

When suspending, described travelling carriage refers to during sending radio frames during the time-out voice data transmission during to transfer channel of other logic channel that priority scheduling is higher than the speech priority of logical channels time; just mean; the technical scheme of the transmission that buffer memory arranged of travelling carriage when being interrupted the voice data transmission by the high priority logic channel is: according to above-mentioned generating in travelling carriage and transmission adaptive multi-rate (AMR) coded frame method; during other logic channel that priority scheduling is higher than speech priority of logical channels time-out voice data during to transfer channel sends; Speech frame coded sequence from adaptive multi-rate (AMR) encoder is carried out buffer memory; after voice data sends and recovers; adaptive multi-rate (AMR) coded frame in transmitting the Speech frame coded sequence is in the process of transmission channel; in the mode of per 20 milliseconds of scheduling adaptive multi-rate (AMR) coded frame that (do not comprise) more than, handle the part or all of Speech frame coded sequence that is buffered to transmission channel.

Buffer memory adaptive multi-rate (AMR) coded frame on the speech logic channel of high priority transmission that logic channel interrupts that is scheduled, improve or after the high priority logic channel stops to send in physical channel rate, adaptive multi-rate (AMR) coded frame of these buffer memorys is sent with per 20 milliseconds of more than one speed of scheduling, just can solve the problem of the stolen frame of voice data that this class brought by priority of logical channels, the benefit that this method that Here it is is brought.

Above-mentioned travelling carriage has in the technical scheme of transmission of interruption, and travelling carriage has the technical scheme of transmission of buffer memory and travelling carriage in the technical scheme of being interrupted the transmission that buffer memory arranged of voice data when sending by the high priority logic channel when switching, the mode of per 20 milliseconds of scheduling adaptive multi-rate (AMR) coded frame that (do not comprise) more than to transmission channel all arranged, the method that realizes this mode has multiple, the present invention provides a concrete method, promptly, adaptive multi-rate (AMR) coded frame that described per 20 milliseconds of scheduling (do not comprise one) more than one is to dispatch the transmission channel of a complete adaptive multi-rate (AMR) coded frame to 10 millisecond Transmission Time Interval (TTI) to the mode of transmission channel.A complete adaptive multi-rate (AMR) coded frame is dispatched to the method on the transmission channel of 10 milliseconds of Transmission Time Intervals (TTI), take 10 milliseconds transformat (TF), all classes of an adaptive multi-rate (AMR) coded frame all are dispatched on the transmission channel.

The radio network controller of network side (RNC) detects whether adaptive multi-rate (AMR) coded frame that receives is the frame that is delayed, a kind of method of detection is exactly to check per 20 milliseconds of adaptive multi-rate of receiving (AMR) coded frame, in case find to receive in 20 milliseconds in frame more than 1 or 10 milliseconds and just receive a frame, just show and received the frame that is delayed; The another kind of method that detects is exactly to send signaling message to wireless access network, and this message indicates the initial frame number and the frame number of several transmission radio frames that are delayed frame.Radio network controller (RNC) to a kind of method that core net sends adaptive multi-rate (AMR) coded frame of band delay sign is: will postpone extension field (spare extension) lining that sign can be placed on core net-radio access network interface (Iu) user plane frame of carrying this adaptive multi-rate (AMR) coded frame.

---network side transmits the method for adaptive multi-rate (AMR) coded frame in the mode of being interrupted

The technical scheme of the transmission that interruption is arranged of universal mobile telecommunications system (UMTS) network side is: according to a kind of method that generates and transmit adaptive multi-rate (AMR) coded frame of above-mentioned network side in universal mobile telecommunications system (UMTS), and determine the method that the coding of pattern in the coded command generates and sends by radio network controller (RNC) at universal mobile telecommunications system (UMTS) network side, radio network controller (RNC) is set up buffer area, the Speech frame coded sequence that the Iu user plane frame of receiving during the transmission of time-out radio frames by the core net transmission is carried is buffered in this buffer area, after the described EOT end of timeout, adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence that scheduling is buffered adopts the mode of per 20 milliseconds of scheduling adaptive multi-rate (AMR) coded frame that (do not comprise) more than to transmission channel in the process of transmission channel.

The technical scheme that universal mobile telecommunications system when switching in the sub-district (UMTS) network side sends is: according to the sending method that interruption is arranged of universal mobile telecommunications system (UMTS) network side, be by the minizone within the radio network controller (RNC) between transfer period during described time-out radio frames sends.

The technical scheme that universal mobile telecommunications system (UMTS) network side when switching between radio network controller (RNC) sends is: according to the sending method that interruption is arranged of universal mobile telecommunications system (UMTS) network side, be between radio network controller (RNC) between transfer period during described time-out radio frames sends.

Technical scheme in the transmission of being interrupted UMTS (UMTS) network side of voice data when sending by the high priority logic channel is: according to the sending method that interruption is arranged of UMTS (UMTS) network side, during described time-out radio frames sends be the time-out voice data transmission of other logic channel during to transfer channel that be higher than the speech priority of logical channels by priority scheduling caused fail to send AMR (AMR) coded frame in the Speech frame coded sequence that those core net-radio access network interfaces (Iu) user plane frame carries during.

---network side transmits the method for adaptive multi-rate (AMR) coded frame that is delayed

When travelling carriage has used the sending method that interruption is arranged, can produce adaptive multi-rate (AMR) coded frame be delayed, cause that the radio network controller (RNC) that sends travelling carriage sends voice data with per 20 milliseconds of speed that surpass a Speech frame coded sequence to core net, core net also is bound to send to the radio network controller (RNC) of receiving mobile with the per 20 milliseconds speed that surpass a Speech frame coded sequence, the transmission technology scheme of handling universal mobile telecommunications system (UMTS) network side of this class frame is: according to a kind of method that generates and transmit adaptive multi-rate (AMR) coded frame of above-mentioned network side in universal mobile telecommunications system (UMTS), and determine the method that the coding of pattern in the coded command generates and sends by radio network controller (RNC) at universal mobile telecommunications system (UMTS) network side, the Iu user plane frame that described core net sends to the voice-over frame coded sequence of radio network controller (RNC) comprises the delay sign of this Speech frame coded sequence, when postponing sign indication Speech frame coded sequence is when being delayed, and radio network controller (RNC) (does not comprise one) with per 20 milliseconds of scheduling more than one, and adaptive multi-rate (AMR) coded frame is dispatched this Speech frame coded sequence to the mode of transmission channel.

Sending adaptive multi-rate (AMR) coded frame or transmit to transmit adaptive multi-rate (AMR) coded frame that is delayed with discontinuous manner when, all to use above-mentioned network side the mode of per 20 milliseconds of scheduling adaptive multi-rate (AMR) coded frame that (do not comprise) more than to transmission channel, here provide a concrete method, promptly, adaptive multi-rate (AMR) coded frame that described per 20 milliseconds of scheduling (do not comprise one) more than one is to dispatch the transmission channel of a complete adaptive multi-rate (AMR) coded frame to 10 millisecond Transmission Time Interval (TTI) to the mode of transmission channel.A complete adaptive multi-rate (AMR) coded frame is dispatched to the method on the transmission channel of 10 milliseconds of Transmission Time Intervals (TTI), take 10 milliseconds transformat (TF), all classes of an adaptive multi-rate (AMR) coded frame all are dispatched on the transmission channel.

In the technical scheme of the transmission that interruption is arranged of above-mentioned universal mobile telecommunications system (UMTS) network side, and when switching in relevant sub-district, when switching between radio network controller (RNC), when being interrupted the voice data transmission by the high priority logic channel, with core net with the per 20 milliseconds speed that surpass a Speech frame coded sequence when radio network controller (RNC) sends voice data, need control the time-delay of speech, therefore the number that is buffered in the Speech frame coded sequence in radio network controller (RNC) buffer area is limited, concrete method is, after the buffer area lower limit that outnumbers the purpose travelling carriage of the Speech frame coded sequence of buffer memory, begin to abandon the Speech frame coded sequence of buffer memory.

In above-mentioned all technical schemes, be that travelling carriage or network side use the above-mentioned send mode that interruption is arranged all to need each Speech frame of 20 milliseconds is encoded to the frame of a plurality of adaptive multi-rate (AMR) pattern, because through after the buffer delay, the channel that it will be faced has variation.The many speed of multiple self adaption (AMR) pattern is to be interrupted to send needed technical characterictic thus.

Beneficial effect

The invention enables the choice mechanism that adaptive multi-rate (AMR) pattern was arranged in the TFC choice phase, reduce in the prior art in the TFC choice phase and lose with the TF Speech frame that causes that do not match because of single adaptive multi-rate (AMR) pattern, among the present invention, the output of adaptive multi-rate (AMR) encoder is multimodal, and this multi-mode has the transmission channel format combination to be complementary with it.Change for the effective transmission channel format combination (TFC) that descends suddenly because of physical channel rate or scheduling burst high priority logic channel causes suddenly, the output of multimodal adaptive multi-rate (AMR) encoder can be used for mating described effective transmission channel format combination (TFC) and changes suddenly in the method for the present invention, descends suddenly or the frame losing of the Speech frame that scheduling burst high priority logic channel causes thereby reduce because of physical channel rate.

Caching mechanism among the present invention has overcome per 20 milliseconds of the prior art must be finished the transmission of an adaptive multi-rate (AMR) Speech frame or abandon this restriction, can accomplish delayed delivery because use method of the present invention, make to switch like this and the coded frame of adaptive multi-rate (AMR) the encoder output of scheduling burst high priority logic channel during seizing still can obtain transmitting, and prior art abandons these in meeting and is switched and dispatches adaptive multi-rate (AMR) coded frame that burst high priority logic channel has been seized Radio Resource.

With frame is different steathily in the prior art, those Speech frames of being dispatched burst high priority logic channel and influencing can not be subjected to lingeringly to use the Speech frame different pattern adjacent with its front to send in the inventive method, perhaps postpone to send behind the certain hour, can not cause abandoning of Speech frame.

The generation of the adaptive multi-rate that the present invention provides (AMR) coding and the method for transmission have utilized the bit rate that physical channel provided to transmit voice data to the full extent, this utilization be embodied in 20 milliseconds speech input can carry out in the certain pattern scope and the regular hour scope in select and the adaptation of this selection to physical channel on.

Description of drawings

Fig. 1 is the block diagram that travelling carriage generated and transmitted the embodiment of adaptive multi-rate (AMR) coded frame.

Fig. 2 is that the adaptive multi-rate in the coded command (AMR) pattern count is the processing schematic diagram of 2 coding module among Fig. 1.

Fig. 3 is that the adaptive multi-rate in the coded command (AMR) pattern count is the processing schematic diagram of 3 coding module among Fig. 1.

Fig. 4 is travelling carriage generated and transmitted the embodiment of adaptive multi-rate (AMR) coded frame with cache way a block diagram.

Fig. 5 is the block diagram of embodiment of the delay decoding of travelling carriage.

Fig. 6 is the block diagram that network side generated and transmitted the embodiment of adaptive multi-rate (AMR) coded frame.

Fig. 7 is that network side transmits the embodiment of adaptive multi-rate (AMR) coded frame and the block diagram that network side transmits the embodiment of adaptive multi-rate (AMR) coded frame that is delayed in the mode of being interrupted

Embodiment

Describe the preferred embodiments of the present invention in detail below in conjunction with accompanying drawing.

---travelling carriage generates and transmits the embodiment of adaptive multi-rate (AMR) coded frame

Referring to Fig. 1, shown in embodiment wherein, in travelling carriage, adaptive multi-rate (AMR) encoder converts speech to the signal of telecommunication from sound, through after the filtering, realized from the analogue voice signal to the digital voice signal 1 conversion again through over-sampling, coding module is digital voice signal 1 coding by the mode of 20 millisecond of one frame, coding module produces 1 Speech frame coded sequence 2 for per 20 milliseconds, each Speech frame coded sequence comprises: pattern indication, several adaptive multi-rate (AMR) speech coding frame or 1 adaptive multi-rate (AMR) quiet frame.Whether the pattern indication provides the situation of Speech frame coded sequence and adaptive multi-rate (AMR) coded frame thereof, that is, be effective Speech frame coded sequence, the number of the adaptive multi-rate in the Speech frame coded sequence (AMR) coded frame and the coding mode of each frame.Transport format combination selection module in the wave point functional unit is input with the Speech frame coded sequence 2 of adaptive multi-rate (AMR) voice encryption device and grouping busihess data 4 and signaling data 5, a selected transformat combination determines to transmit the transformat of speech and the voice data transmission piece 6 of this form correspondence according to this transformat combination.When Speech frame coded sequence 2 is effective Speech frame coded sequences, voice data transmission piece 6 is exactly an adaptive multi-rate (AMR) coded frame, this adaptive multi-rate (AMR) coded frame is 1 in all adaptive multi-rate (AMR) speech coding frame (non-quiet frame) of being comprised of Speech frame coded sequence 2, or that 1 quiet frame of being comprised of Speech frame coded sequence 2.By the corresponding grouping busihess data 4 of transport format combination selection module output and the transmission block of signaling data 5 is respectively packet data transmission piece 8 and signaling data transmission block 7.Other module of wave point functional unit is mapped to voice data transmission piece 6, packet data transmission piece 8 and signaling data transmission block 7 on the physical channel and sends.The one Speech frame mode control module of being responsible for to adaptive multi-rate (AMR) encoder output mode selection signal 9 is arranged in the wave point functional unit, mode select signal 9 outputs to coding module, when mode select signal 9 comprises a plurality of adaptive multi-rate (AMR) coding mode, mode select signal 9 should comprise a plurality of adaptive multi-rate (AMR) speech coding frame as coded command indication coding module when exporting non-quiet frame, and has indicated the pattern of these frames.

Form 1 is the composition structure of a Speech frame coded sequence 2 and the example that bit number distributes thereof.

Form 1

Fig. 2 is the processing schematic diagram of coding module that can produce the frame sequence of 2 kinds of adaptive multi-rate (AMR) pattern, digital voice signal 1 is simultaneously to 100 and 101 outputs, the structure of the counterpart in Fig. 1 speech processing capacity of 100 same universal mobile telecommunications system (UMTS) standard TS26.071 (or TS26.171) signal (Overview of audio processing functions) is basic identical, and difference is to have drawn among Fig. 2 to output to the coding mode index signal 17 of 100 speech coding module.What the voice activation detection module was exported is that voice activation detects sign (Voice Activity Detector flag) 10,100 speech coding module output speech coding frame 12, what 101 speech coding module was exported is speech coding frame 19, speech coding frame 12 and every frame bit number of 19 depend on coding mode index signal 17 and 18 respectively, 2 tunnel outputs of coding mode index signal 17 and 18 1 pattern 1 tunnel that to be coding mode control modules resolve into 2 patterns in the mode select signal 9.Discontinuous transmission and the many speed of operational module output adaptive (AMR) frame type signal 11 are to multi-path voice code multiplexing module and speech coding module, 11 indications of adaptive multi-rate (AMR) frame type signal: whether information bit 14 is efficient adaptive multi-speed coding frames, the pattern of self-adaptive multi-speed coding frame, the pattern of described self-adaptive multi-speed coding frame is the pattern of the non-Speech frame of mourning in silence or the pattern of quiet frame, when the pattern of self-adaptive multi-speed coding frame was quiet frame, information bit 14 was that mourning in silence of comfort noise sending module output detected frame 13.101 speech coding module output adaptive many rate speech coded frame 19 and adaptive multi-rate (AMR) frame type signal 16 thereof, the pattern of adaptive multi-rate (AMR) frame type signal 16 expression adaptive multi-rate speech coding frames 19.Multi-path voice code multiplexing module will be merged into Speech frame coded sequence 2 from the information bit 14 and the adaptive multi-rate speech coding frame 19 of discontinuous transmission and operational module, the method that merges is: when adaptive multi-rate (AMR) frame type signal 11 indication information bits 14 were invalid, the pattern that Speech frame coded sequence 2 is set was designated as invalid Speech frame coded sequence; When adaptive multi-rate (AMR) frame type signal 11 indication information bits 14 were quiet frame, the pattern that Speech frame coded sequence 2 is set was designated as quiet frame, information bit 14 is put in the Speech frame coded sequence of Speech frame coded sequence 2; When adaptive multi-rate (AMR) frame type signal 11 indication information bits 14 are non-quiet frame, information bit 14 and 19 together is put in the Speech frame coded sequence of Speech frame coded sequence 2, the pattern that Speech frame coded sequence 2 is set simultaneously is designated as the indicated pattern of adaptive multi-rate (AMR) frame type signal 11 and 16.

Adaptive multi-rate (AMR) pattern count in coded command of form 2 expressions is the information bit 14 of coding module generation in 2 o'clock and the formation of the non-quiet frame 19 of adaptive multi-rate speech coding, the formation of grouping busihess data 4 and signaling data 5, and the sign of transmitting the corresponding transmission channel of these user data, and the pattern and the priority (the more little priority of numerical priority value is high more) of relevant logic channel.

User data	Information bit			14			The non-quiet frame 19 of adaptive multi-rate speech coding			??5	??4
												The AMR type of user data in each Transmission Time Interval	Category-A	Category-B	The C class	Category-A	Category-B	The C class	Arbitrarily	Arbitrarily
Transmission channel	????1	????2	????3	????1	??2	??3	??5	??4
									The pattern of the logic channel of carrying user data	????TM	????TM	????TM	????TM	??TM	??TM	??AM	??UM
The priority of the logic channel of carrying user data	????4	????4	????4	????4	??4	??4	??2	??6

Form 2

The attribute and the parameter, particularly Pei Zhi transformat (TF) of each transmission channel in the form 3 expression forms 2.

Transmission channel		????1	????2	????3	????4	????5
							Chnnel coding		????Turbo	Convolution	Convolution	????Turbo	????Turbo
Code rate		????1/3	????1/2	????1/2	????1/3	????1/3
							????TTI		????20ms	????20ms	????20ms	????20ms	????20ms
?TFI	????0	????1×0	????0	????0	????0×60	????0×60
							????1	????1×39	????1×103	????1×60	????1×60	????1×60
	????2	????1×81	????1×87		????2×60	????2×60
							????3	????1×61	????1×53
	????4	????1×42

Form 3

Adaptive multi-rate (AMR) the voice encryption device pattern of the transformat of configuration support is 12.2kbps, 7.4kbps and 4.75kbps as can be seen from form 3, Speech frame pattern control section can comprise in above-mentioned 3 kinds of patterns any 2 kinds to the mode select signal 9 that the AMR voice encryption device sends in the wave point functional unit, for example (12.2kbps, 4.75bps).

Form 4 provides the meaning of all transformat combination, just the transformat combination user data that is used for transmitting.

Transmission channel	TrCh1	?TrCh2	?TrCh3	?TrCh4	?TrCh5	Meaning
							TFI	????0	????0	????0	????0	????0	Do not send data
????1	????0	????0	????0	????0	Send quiet frame
						????2		????1	????1	????0	????0	Send 12.2kbps pattern voice data
????3	????2	????0	????0	????0	Send 7.4kbps pattern voice data
						????4		????3	????0	????0	????0	Send 4.75kbps pattern voice data
????0	????0	????0	????1	????0	Send grouped data
						????0		????0	????0	????0	????1	Send signaling data
????0	????0	????0	????1	????1	Send signaling and grouped data
						????1		????0	????0	????1	????0	Send quiet frame and send grouped data
????1	????0	????0	????0	????1	Send quiet frame and send signaling data
						????1		????0	????0	????1	????1	Send quiet frame and send signaling and grouped data
????2	????1	????1	????1	????0	Send 12.2kbps pattern voice data and send grouped data
						????2		????1	????1	????0	????1	Send 12.2kbps pattern voice data and send signaling data
????2	????1	????1	????1	????1	Send 12.2kbps pattern voice data and send signaling and grouped data
						????3		????2	????0	????1	????0	Send 7.4kbps pattern voice data and send grouped data
????3	????2	????0	????0	????1	Send 7.4kbps pattern voice data and send signaling data
						????3		????2	????0	????1	????1	Send 7.4kbps pattern voice data and send signaling and grouped data
????4	????3	????0	????1	????0	Send 4.75kbps pattern voice data and send grouped data
						????4		????3	????0	????0	????1	Send 4.75kbps pattern voice data and send signaling data
????4	????3	????0	????1	????1	Send 4.75kbps pattern voice data and send signaling and grouped data

Form 4

The pattern that comprises when mode select signal 9 is (12.2kbps, 4.75bps), making the wave point functional unit to handle in the Speech frame coded sequence 2 is the voice data of 12.2kbps and 4.75bps, if in needs are dispatched to the output of transmission channel signaling data 5 and grouped data 6, do not have valid data, and combinations of transport formats (2,1,1,0,0) be effective combination again, the wave point functional unit can sending mode be the Speech frame of 12.2kbps; If being dispatched at needs has valid data to need to transmit in the transmission channel signaling data 5, there are not valid data in the grouped data, and because the restriction combinations of transport formats (2 of physical channel bandwidth, 1,1,0,1) is invalid combination, and combinations of transport formats (4,3,0,0,1) not being subjected to the bandwidth constraints of physical channel is effective combination, and at this moment, wave point functional unit sending mode simultaneously is Speech frame and the signaling of 4.75kbps.

The pattern that comprises when mode select signal 9 is (12.2kbps, 4.75bps), making the wave point functional unit to handle in the Speech frame coded sequence 2 is the voice data of 12.2kbps and 4.75bps, if this moment, the configuration of physical channel changed, thereby make the bandwidth reduction of channel make combinations of transport formats (2,1,1,0,0) becomes invalid combination, and combinations of transport formats (3,2,0,0,0) and (4,3,0,0,0) not being subjected to the bandwidth constraints of physical channel is effective combination, and at this moment, wave point functional unit sending mode is the Speech frame of 4.75kbps when only voice data being arranged, send to the AMR voice encryption device again simultaneously and comprise (7.4kbps, 4.75bps) mode select signal 9, behind the voice data that becomes 7.4kbps and 4.75bps in Speech frame coded sequence 2, the wave point functional unit can sending mode be the Speech frame of 7.4kbps when only voice data being arranged.

Fig. 3 is that the number of adaptive multi-rate (AMR) pattern that comprises of mode select signal 9 is the processing schematic diagram of 3 coding module, digital voice signal 1 is simultaneously to 100,101 and 102 outputs, the structure of the counterpart in Fig. 1 speech processing capacity signal (Overview of audio processingfunctions) of 100 same universal mobile telecommunications system (UMTS) standard TS26.071 (or TS26.171) is basic identical, difference is the coding mode index signal 17 of having drawn among Fig. 2 and having outputed to the speech coding module, what the voice activation detection module was exported is that voice activation detects sign (Voice Activity Detectorflag) 10,100 speech coding module output speech coding frame 12, what 101 and 102 speech coding modules were exported is respectively speech coding frame 19 and 21, speech coding frame 12, every frame bit number of 19 and 21 depends on coding mode index signal 17 respectively, 18 and 15, coding mode index signal 17,3 tunnel outputs of 1 pattern 1 tunnel that 18 and 15 is coding mode control modules resolves into 3 patterns in the mode select signal 9.Discontinuous transmission and the many speed of operational module output adaptive (AMR) frame type signal 11 are to multi-path voice code multiplexing module and speech coding module, 11 indications of adaptive multi-rate (AMR) frame type signal: whether information bit 14 is efficient adaptive multi-speed coding frames, the pattern of self-adaptive multi-speed coding frame, the pattern of described self-adaptive multi-speed coding frame is the pattern of Speech frame or the pattern of quiet frame, when the pattern of self-adaptive multi-speed coding frame was quiet frame, information bit 14 was that mourning in silence of comfort noise sending module output detected frame 13.101 and 102 speech coding modules are many rate speech of output adaptive coded frame 19 and 21 and adaptive multi-rate separately (AMR) frame type signal 16 and 20 respectively, and they represent the pattern of adaptive multi-rate speech coding frame 19 and 21 respectively.Multi-path voice code multiplexing module will be merged into Speech frame coded sequence 2 with adaptive multi-rate speech coding frame 19 and 21 from the information bit 14 of discontinuous transmission and operational module, the method that merges is: when adaptive multi-rate (AMR) frame type signal 11 indication information bits 14 were invalid, the pattern that Speech frame coded sequence 2 is set was designated as invalid Speech frame coded sequence; When adaptive multi-rate (AMR) frame type signal 11 indication information bits 14 were quiet frame, the pattern that Speech frame coded sequence 2 is set was designated as quiet frame, information bit 14 is put in the Speech frame coded sequence of Speech frame coded sequence 2; When adaptive multi-rate (AMR) frame type signal 11 indication information bits 14 are non-quiet frame, information bit 14,19 and 21 together is put in the Speech frame coded sequence of Speech frame coded sequence 2, the pattern that Speech frame coded sequence 2 is set simultaneously is designated as the indicated pattern of adaptive multi-rate (AMR) frame type signal 11,16 and 20.

---travelling carriage transmits the embodiment of adaptive multi-rate (AMR) coded frame with discontinuous manner

The generation of travelling carriage of buffer memory and the embodiment of method that transmits adaptive multi-rate (AMR) Speech frame are arranged referring to Fig. 4, the difference of Fig. 4 and Fig. 1 has been to have more push-up storage (FIFO) and has read control assembly with the transport format combination selection push-up storage and substituted the transport format combination selection parts among Fig. 1.The output Speech frame coded sequence 2 of coding module is to push-up storage (FIFO), push-up storage (FIFO) buffer memory Speech frame coded sequence, and output storage state flags 25, storage state flags 25 indications: whether the Speech frame coded sequence that is not read is arranged, and the number of these Speech frame coded sequences that are not read.The transport format combination selection push-up storage reads control assembly output reading order 26, and reading order 26 makes push-up storage (FIFO) output Speech frame coded sequence 3.

For the delay of speech is controlled within certain limit, the transport format combination selection push-up storage reads control assembly can check whether the number of the Speech frame coded sequence of the middle storage of push-up storage (FIFO) that storage state flags 25 provides surpasses limit value, when surpassing, can define the Speech frame coded sequence of overtime storage, with reading order 26 taking-up from push-up storage (FIFO) of overtime Speech frame coded sequence is abandoned then, can be determined above-mentioned limit value by the degree that postpones control.

The configuration of form 5 an expression embodiment exemplary transmission channel in the course of the work, provide transformat sign (TFI) the pairing transformat (TF) and the Transmission Time Interval (TTI) of the attribute and the parameter, particularly each transmission channel of each transmission channel.

Transmission channel		?1	?2	??3	?4	?5
							Chnnel coding		?Turbo	Convolution	Convolution	?Turbo	?Turbo
Code rate		?1/3	?1/2	??1/2	?1/3	?1/3
							??TFI	????0	?1×0，20ms	?0，20ms	??0，20ms	?0×120，20ms	?0×120，20ms
????1	?1×39，20ms	?1×103，20ms	??1×60，20ms	?1×120，20ms	?1×120，20ms
						????2		?1×81，20ms	?1×87，20ms		?2×120，20ms	?2×120，20ms
????3	?1×61，20ms	?1×53，10ms			?1×30，10ms
						????4		?1×42，10ms
????5	?1×39，10ms

Form 5

The pattern that comprises when mode select signal 9 is (12.2kbps, 4.75bps), make that what be stored in that Speech frame coded sequence in the push-up storage (FIFO) comprises in the wave point functional unit is the voice data of 12.2kbps and 4.75bps pattern, there are valid data to need to transmit in the transmission channel signaling data 5 if be dispatched at needs, there are not valid data in the grouped data, and because the restriction combinations of transport formats (2 of physical channel bandwidth, 1,1,0,1) is invalid combination, the wave point functional unit is being used combinations of transport formats (0,0,0,0,1) in one 20 milliseconds Transmission Time Interval, send signaling data after, the wave point functional unit can be with (4,3,0,0,0) be the speech coding frame that is buffered of 4.75kbps at continuous 2 10 milliseconds Transmission Time Interval sending mode, previous in 2 is delayed because of seizing, and back one is not delayed, like this, the voice data that tried to be the first by signaling data and suspend transmission has obtained the transmission that postpones, and does not influence the transmission of follow-up Speech frame simultaneously again.If above-mentioned what seized is the start frame of mourning in silence, the method that available (5,0,0,0,0) sends this start frame of mourning in silence that is buffered makes it the transmission that obtains postponing.

For example the overtime limit value of push-up storage (FIFO) is set to the travelling carriage of 10 20 milliseconds time span, its wave point functional unit can't send radio frames when switching, when switching is finished the wave point functional unit and can be sent radio frames, 8 Speech frame coded sequences in push-up storage (FIFO), have been deposited between transfer period, a same example of these Speech frame coded sequences equally is that the pattern that mode select signal 9 comprises is (12.2kbps, 4.75bps) the result, the wave point functional unit can be with (4,3,0,0,1) the self adaptation speech coding frame that to send a pattern be 4.75kbps and begin to send the signaling message of the individual delayed adaptation coded frame of 16 (8+4+2+1+1) from present frame, then with (4,3,0,0,0) the delay Speech frame coded sequence of the time of depositing in push-up storage (FIFO) above 20 milliseconds that sends the self adaptation speech coding frame of buffer memory between transfer period and cause thus at continuous 15 10 milliseconds Transmission Time Interval afterwards just can the recovery pattern be 20 milliseconds of transmissions once of the speech coding frame of 12.2kbps.

---the embodiment of the many speed of travelling carriage delayed adaptation (AMR) decoding

Referring to Fig. 5, the reception information source control rate processing device (Rx SCR handler) of travelling carriage interrelates by push-up storage (FIFO) and wave point functional unit, 32,33 and 34 is respectively the information bit of push-up storage (FIFO) to each received frame of reception information source control rate processing device (the Rx SCR handler) output of travelling carriage, pattern indication and type of receipt, 37,38 and 39 is respectively the information bit of wave point functional unit to each received frame of push-up storage (FIFO) output, pattern indication and type of receipt, pattern is indicated the adaptive multi-rate that provides received frame, and (AMR pattern, type of receipt are as shown in Table 6.Each adaptive multi-rate (AMR) received frame that push-up storage (FIFO) is sent the wave point functional unit is buffer memory according to the order of sequence, during buffer memory above-mentioned information bit, pattern indication and type of receipt is preserved together.Indicate at buffer status sign 30: whether adaptive multi-rate (AMR) received frame that is not read is arranged, and the number of these adaptive multi-rate that is not read (AMR) received frames.The received frame that is not read that reception information source control rate processing device (RxSCR handler) does not go immediately to receive when a voice call begins reads and deciphers, but when the number of the received frame of depositing in the push-up storage (FIFO) that is not read has reached the lower limit that is provided with for this buffer area of push-up storage (FIFO), the reception information source control rate processing device (SCR handler) of travelling carriage just begins per 20 milliseconds and therefrom reads a self-adaptive multi-speed coding frame, and the method that reads is to send reading order 31 to push-up storage (FIFO).Push-up storage (FIFO) output receives lost condition sign 35, and it provides the number of the travelling carriage reception lost frames (type is NO_DATA_SPEECH) that active is added under voice (SPEECH) pattern.When receiving value that lost condition sign 35 provides show that the reception lost frames number of depositing in the push-up storage (FIFO) that initiatively adds has surpassed the predetermined restricted value under voice (SPEECH) pattern, the wave point functional unit is no longer exported the reception lost frames that add under voice (SPEECH) pattern toward push-up storage (FIFO).If when push-up storage (FIFO) the output received frame, the wave point functional unit sends the reception lost frames that initiatively the add deletion that delete command 36 is deposited in push-up storage (FIFO) to push-up storage (FIFO) to the wave point functional unit under voice (SPEECH) pattern with the per 20 milliseconds speed that surpass 1 frame.

Form 7 has provided a series of operations to received frame that travelling carriage postpones to decipher, in this example, the lower limit of this buffer area of push-up storage (FIFO) is 4, and its prescribed limits value of adding reception lost frames number under voice (SPEECH) pattern is 2.Listed the processing to received frame in first 20 milliseconds to the 8th the 20 milliseconds time periods of each unit among Fig. 5 in the form.

Type of receipt	The information bit content
		SPEECH_GOOD	Do not have and detect wrong Speech frame
SPEECH_BAD	Have and detect wrong Speech frame
		SID_FIRST	The start frame of mourning in silence that begins during the sign comfort noise
SID_UPDATE	Do not have and detect the wrong renewal frame of mourning in silence
		SID_BAD	Have and detect the wrong renewal frame of mourning in silence
NO_DATA_SID	The wave point functional unit is not received Useful Information under the comfort noise pattern, initiatively the reception lost frames that add
		NO_DATA_SPEECH	There is being under the voice mode wave point functional unit not receive Useful Information, initiatively the reception lost frames that add

Form 6

The order of 20 milliseconds of units	Frame that the wave point functional unit receives and type	The frame of buffer memory in the push-up storage	The frame that AMR SCR controller reads in	Note
					????1	The first frame SPEECH_GOOD	First frame	Do not have	Travelling carriage is in the SPEECH pattern when initial, the wave point functional unit writes push-up storage to first frame, and Rx SCR handler does not read frame
Empty
	Empty
Empty
	????2	The second frame SPEECH_GOOD	First frame	Do not have			Travelling carriage still is in the SPEECH pattern, the wave point functional unit writes push-up storage to second frame, and Rx SCR handler does not still read frame
Second frame
			Empty
Empty
			????3		The 3rd frame SPEECH_GOOD	First frame		Do not have	Travelling carriage still is in the SPEECH pattern, the wave point functional unit writes push-up storage to the 3rd frame, and Rx SCR handler does not still read frame
Second frame
	The 3rd frame
Empty
	????4	The 4th frame NO_DATA_SPEECH		Second frame		First frame	Travelling carriage still is in the SPEECH pattern, the wave point functional unit writes the 4th frame, number in the push-up storage reaches 4, and Rx SCR handler begins per 20 milliseconds of ground and reads frame
The 3rd frame
			The 4th frame
Empty
			Empty
????5				The 5th frame SPEECH_GOOD the 6th frame SPEECH_GOOD	The 3rd frame			Second frame	Travelling carriage still is in the SPEECH pattern, the wave point functional unit writes the 5th and the 6th frame, the reception lost frames that add under the 4th frame-SPEECH pattern in the deletion push-up storage
	The 5th frame
		The 6th frame
	Empty
		Empty

Continuous table

The order of 20 milliseconds of units	Frame that the wave point functional unit receives and type	The frame of buffer memory in the push-up storage	The frame that AMR SCR controller reads in	Note
					????6	The 7th frame NO_DATA_SPEECH	The 5th frame	The 3rd frame	Travelling carriage still is in the SPEECH pattern, and the wave point functional unit writes the 7th frame
The 6th frame
	The 7th frame
Empty
	????7	The 8th frame NO_DATA_SPEECH	The 6th frame	The 5th frame			Travelling carriage still is in the SPEECH pattern, and the wave point functional unit writes the 8th frame
The 7th frame
			The 8th frame
Empty
			????8		The 9th frame NO_DATA_SPEECH	The 7th frame		The 6th frame	Travelling carriage still is in the SPEECH pattern, the reception lost frames that add under two SPEECH patterns have been arranged in the push-up storage, no longer write reception lost frames the 9th frame that adds under the SPEECH pattern
The 8th frame
	Empty
	????9	The tenth frame SPEECH_GOOD the 11 frame SPEECH_GOOD		The 11 frame		The tenth frame	The wave point functional unit is written to the tenth and the 11 frame the reception lost frames front of adding under the SPEECH pattern; The reception lost frames that add under the 7th frame in the deletion push-up storage-SPEECH pattern, Rx SCR handler reads the tenth frame
The 8th frame
			Empty
			????10	The 12 frame SPEECH_GOOD the 13 frame SPEECH_GOOD	The 12 frame			The 11 frame	The wave point functional unit writes the 12 and the 13 frame, the reception lost frames that add under the 8th frame-SPEECH pattern in the deletion push-up storage, Rx SCR handler reads the 11 frame
The 13 frame
	Empty
	????11	The 14 frame SPEECH_GOOD the 15 frame SPEECH_GOOD			The 13 frame	The 12 frame	The wave point functional unit writes the 14 and the 15 frame, because the front does not have the reception lost frames that add under the SPEECH pattern, so there is not deletion action, Rx SCR handler reads the 12 frame
The 14 frame
			The 15 frame
Empty

Form 7

---network side generates and transmits the embodiment of adaptive multi-rate (AMR) coded frame

Referring to Fig. 6, shown in embodiment wherein, code pattern converter (TC) is made code conversion to speech code signal 41, produce Speech frame coded sequence 42, each Speech frame coded sequence comprises several adaptive multi-rate (AMR) speech coding frame or 1 adaptive multi-rate (AMR) quiet frame or 1 free of data (NO_DATA) type frame, each frame in the Speech frame coded sequence 42 not only comprises the information bit of adaptive multi-rate (AMR) core frames, also comprises encoder Cyclic Redundancy Check, frame type, quality indication and pattern indication.The Speech frame coded sequence 42 that code pattern converter (TC) produces forms Iu user plane frame 43 earlier and is input to radio network controller (RNC) after Iu interface function cell processing, after Iu interface function cell processing, be reduced into Speech frame coded sequence 42 again, output to the transport format combination selection unit then, to indicate this frame be Speech frame or quiet frame or free of data (NO_DATA) type frame to the frame type of every frame in the Speech frame coded sequence 42, and the indication of the pattern of every frame has provided the coding mode of this frame in the Speech frame coded sequence 42.Transport format combination selection module in the radio network controller (RNC) is input with Speech frame coded sequence 42, grouping busihess data 44 and signaling data 45, the transport format combination selection module is selected a transformat combination, determines to transmit the transformat and the corresponding output transmission block of speech according to this transformat combination.When the frame in the Speech frame coded sequence 42 is Speech frame or quiet frame rather than free of data (NO_DATA) type frame, its corresponding transmission block is adaptive multi-rate (AMR) coded frame 46, this adaptive multi-rate (AMR) coded frame 46 is 1 or 1 quiet frame wherein in several adaptive multi-rate (AMR) Speech frame in the Speech frame coded sequence 42, and the corresponding transmission block of grouping busihess data 44 and signaling data 45 is respectively packet data transmission piece 48 and signaling data transmission block 47.Other unit of radio network controller (RNC) is mapped to voice data transmission piece 46, packet data transmission piece 48 and signaling data transmission block 47 on the physical channel and sends.Radio network controller (RNC) sends core net-radio access network interface (Iu) user plane rate controlled frame 49 to media gateway (MGW) output of core net, sequence number sign (RFCI n indicator) territory of a plurality of RAB stream combination indications in core net-radio access network interface (Iu) user plane rate controlled frame 49, a plurality of one to one RAB stream combination indications (RFCI) have been specified, the rate controlled unit of media gateway (MGW) produces the mode select signal 40 of indication a plurality of adaptive multi-rate (AMR) speech coding pattern according to these RAB stream combination indications (RFCI-RAB sub-Flow Combination Indicator), output mode selects signal 40 to code pattern converter (TC), the pattern of a plurality of adaptive multi-rate (AMR) coded frame that should comprise in the speech coding frame sequence 42 of designated code code converter (TC) output.

It is to be provided by RAB stream combination indication (RFCI) that core net-radio access network interface (Iu) uses the combination of any RAB stream, be comprised in the Iu user plane frame 43 in stage of communication RAB stream combination indication (RFCI), RAB stream combination indication (RFCI) has provided the structure of core net-radio access network interface (Iu) user plane frame 43.

Form 8 is that radio network controller (RNC) transmits the example of Speech frame coded sequence to the employed RAB stream combination of core net (RFC) and RAB stream combination indication (RFCI) value thereof, needs the part that is sent out when they also are Iu UP initialization in core net-radio access network interface user plane (Iu UP) initialization procedure.

In the example of form 9, provided: utilize sequence number sign (RFCI n indicator) media gateway (MGW) in core net of the RAB stream combination indication of core net-radio access network interface (Iu) user plane rate controlled frame 49 to send indication with the RFC corresponding relation, the rate controlled part is determined a plurality of adaptive multi-rate (AMR) coding mode according to this indication, output contains the mode select signal 40 of described a plurality of adaptive multi-rate (AMR) coding mode to code pattern converter (TC), code pattern converter (TC) is done the conversion of sign indicating number type according to the type of the speech coding signal of receiving 41, and the suitable RAB stream combination indication (RFCI) of core net-radio access network interface (Iu) functional unit selection comes the frame in the voice-over frame coded sequence 42.

RFCI (RAB stream combination indication)	RAB stream 1	RAB stream 2	RAB stream 3	RAB stream 4	RAB stream 5	RAB stream 6	RAB stream 7	RAB stream 8	RAB stream 9
										????0	????0	????0	????0	????0	????0	????0	????0	????0	????0
????1	????39	????0	????0	????0	????0	????0	????0	????0	????0
										????2	????42	????53	????0	????0	????0	????0	????0	????0	????0
????3	????61	????87	????0	????0	????0	????0	????0	????0	????0
										????4	????81	????103	????60	????0	????0	????0	????0	????0	????0
????5	????42	????53	????0	????61	????87	????0	????0	????0	????0
										????6	????42	????53	????0	????81	????103	????60	????0	????0	????0
????7	????61	????87	????0	????81	????103	????60	????0	????0	????0
										????8	????42	????53	????0	????61	????87	????0	????81	????103	????60

Form 8

The type of speech coding digital signal 41	The indication of the RFCI sequence number in the Iu UP rate controlled frame 49	Mode select signal 40 mode designated	Frame in the Speech frame coded sequence 42	Hu user plane frame 43 employed RFCI	Note
						The ITU of 64kbps is the A rule G.711, and sample rate is 8kHZ	0，1，2，3，4，5，6，7，8	??4.75kbps， ??7.4kbps， ??12.2kbps	??4.75kbps， ??7.4kbps， ??12.2kbps	????8
7.4kbps the AMR speech coding frame of pattern	0，1，2，3，5	??4.75kbps， ??7.4kbps	??4.75kbps， ??7.4kbps	????5
						7.4kbps the AMR speech coding frame of pattern	0，1，3，4，7	??7.4kbps， ??12.2kbps	??7.4kbps	????3	Because the pattern of input has only 7.4kbps, convert the quality that 12.2kbps can't improve speech to
Quiet frame	0，1，2，3，4，5，6，7，8	??4.75kbps， ??7.4kbps， ??12.2kbps	Quiet frame	????1	Quiet frame does not need conversion
						Free of data frame (NO_DATA)	0，1，3，4，7	??7.4kbps， ??12.2kbps	Free of data frame (NO_DATA)	????0	The free of data frame does not need conversion

Form 9

---network side transmits the embodiment of adaptive multi-rate (AMR) coded frame in the mode of being interrupted

Referring to Fig. 7, the scheme of Fig. 7 is the improvement of Fig. 6, and the difference of it and Fig. 6 has been to have more push-up storage (FIFO) and has read control assembly with the transport format combination selection push-up storage and substituted the transport format combination selection parts among Fig. 6.In Speech frame coded sequence 42 image patterns 6 of code pattern converter output, be input to radio network controller (RNC) through forming Iu user plane frame 43 after the Iu interface function cell processing earlier, after Iu interface function cell processing, be reduced into Speech frame coded sequence 42 again, output to push-up storage (FIFO) then, push-up storage (FIFO) buffer memory Speech frame coded sequence, and output storage state flags 51, storage state flags 51 indications: whether the Speech frame coded sequence that is not read is arranged, and the number of these Speech frame coded sequences that are not read.The transport format combination selection push-up storage reads control assembly output reading order 52, and reading order 52 makes push-up storage (FIFO) output Speech frame coded sequence 50.

Near the buffer area lower limit time corresponding that the delay of speech is controlled at the purpose travelling carriage, after storage state flags 51 provided the indication of the buffer area lower limit that outnumbers the purpose travelling carriage of Speech frame coded sequence of storage, the transport format combination selection push-up storage read control assembly and can select the Speech frame coded sequences that it is read and abandoned from push-up storage (FIFO) with reading order 52.

The configuration of form 10 an expression embodiment exemplary transmission channel in the course of the work, provide transformat sign (TFI) the pairing transformat (TF) and the Transmission Time Interval (TTI) of the attribute and the parameter, particularly each transmission channel of each transmission channel.

Transmission channel		?1	?2	??3	?4	?5
							Chnnel coding		?Turbo	Convolution	Convolution	?Turbo	?Turbo
Code rate		?1/3	?1/2	??1/2	?1/3	?1/3
							?TFI	????0	?1×0，20ms	?0，20ms	??0，20ms	?0×120，20ms	?0×120，20ms
????1	?1×39，20ms	?1×103，20ms	??1×60，20ms	?1×120，20ms	?1×120，20ms
						????2		?1×81，20ms	?1×87，20ms		?2×120，20ms	?2×120，20ms
????3	?1×61，20ms	?1×53，10ms
						????4		?1×42，10ms

Form 10

With the example of form 8 as RAB stream combination (RFC) and RAB stream combination indication (RFCI) value thereof, the sequence number sign (RFCI n indicator) of indicating when the RAB stream combination of core net-radio access network interface (Iu) user plane rate controlled frame 49 is (0,1,2,4,6), the pattern that the mode select signal 40 that makes the rate controlled unit export comprises is (12.2kbps, 4.75bps), core net use value in core net-radio access network interface (Iu) user plane frame is that 6 RFCI comes transporting speech frame coded sequence 42, what radio network controller (RNC) was stored in that Speech frame coded sequence in the push-up storage (FIFO) comprises is the voice data of 12.2kbps and 4.75bps pattern, there are valid data to need to transmit in the transmission channel signaling data 5 if be dispatched at needs, there are not valid data in the grouped data, and because the restriction combinations of transport formats (2 of physical channel bandwidth, 1,1,0,1) is invalid combination, the wave point functional unit is with combinations of transport formats (0 like this, 0,0,0,1) in one 20 milliseconds Transmission Time Interval, send signaling data after, the wave point functional unit can be with (4,3,0,0,0) be the Speech frame of 4.75kbps at continuous 2 10 milliseconds Transmission Time Interval sending mode, make, tried to be the first by signaling data and the voice data that suspends transmission has obtained the transmission that postpones, do not influence the transmission of follow-up Speech frame simultaneously again.

When the buffer area length of target MS is 4 frames, radio network controller can't send radio frames when switching, radio network controller has been received 8 Speech frame coded sequences altogether from core net when between transfer period, radio network controller is determined with 5 (4+1) as timeout parameter, the number that provides the Speech frame coded sequence of storage when storage state flags 51 abandons overtime sequence greater than 5 the time, a same example of these Speech frame coded sequences is the same, be (12.2kbps, 4.75bps) pattern, the wave point functional unit can be with (4 after switching was finished, 3,0,0,0) pattern that sends in the speech coding frame sequence between transfer period at continuous 5 10 milliseconds Transmission Time Interval is the frame of 4.75kbps, next sending the delay Speech frame coded sequence that the interior resting period of push-up storage (FIFO) surpasses 20 milliseconds with several Transmission Time Intervals of 10 milliseconds again, then just can the recovery pattern be that the Speech frame of 12.2kbps has sent.

---network side transmits the embodiment of adaptive multi-rate (AMR) coded frame that is delayed

The embodiment of Fig. 7 indication also can be used for core net and send the processing that band postpones the speech coding frame sequence of sign to radio network controller (RNC), postpone to indicate extension field (spareextension) lining that can be placed on core net-radio access network interface (Iu) user plane frame, the speech coding frame sequence that band postpones sign is to be caused by the adaptive multi-rate speech coding frame that is delayed that the transmitting terminal travelling carriage sends, for example, the transmitting terminal travelling carriage causes switching the adaptive multi-rate speech coding frame that will send after finishing in these Speech frame coded sequences that are delayed owing to having accumulated a plurality of Speech frame coded sequences in the buffer area that switches in transmission.The speech coding frame sequence that band postpones sign always arrives radio network controller (RNC) with other coded frame sequences together in 20 milliseconds, so, radio network controller (RNC) does not send the AMR coded frame of an independent band delay sign in 20 milliseconds time, come the speech coding frame of these band delay signs of scheduling on transmission channel in the mode that sends a plurality of frames in each 20 milliseconds transmitting time.

The configuration of form 11 an expression embodiment exemplary transmission channel in the course of the work, provide transformat sign (TFI) the pairing transformat (TF) and the Transmission Time Interval (TTI) of the attribute and the parameter, particularly each transmission channel of each transmission channel.

Form 12 is the scheduling speech coding frame done in the course of the work of embodiment examples to transmission channel.

Transmission channel		?1	?2	???3
					Chnnel coding		?Turbo	Convolution	Convolution
Code rate		?1/3	?1/2	???1/2
					??TFI	????0	?1×0，20ms	?0，20ms	???0，20ms
????1	?1×39，20ms	?1×103，20ms	???1×60，20ms
				????2		?1×81，20ms	?1×87，20ms
????3	?1×61，20ms	?1×53，10ms
				????4		?1×42，10ms

Form 11

Speech coding frame sequence in the push-up storage	Employed TFI	Note
			(4.75kbps, postpone sign) (4.75kbps)	????(4，3，0) ????(4，3，0)
(4.75kbp postpones sign) (4.75kbps, 7.4kbps)	????(4，3，0) ????(4，3，0)	In form 11, can't find the transformat of 10 milliseconds 7.4kbps pattern, can only use the 4.75kbps pattern instead without the 7.4kbps pattern
			(4.75kbps postpones sign) (4.75kbps postpones sign)	????(4，3，0) ????(4，3，0)

Form 12

Claims (18)

1. method that in travelling carriage, generates and transmit adaptive multi-rate (AMR) coded frame, in the voice call process, it is 20 milliseconds Speech frame that adaptive multi-rate (AMR) encoder obtains length by the sampling to the input voice signal, each Speech frame contains the sample of 20 milliseconds of speeches, to 20 milliseconds of Speech frame codings, what described coding produced is adaptive multi-rate (AMR) coded frame that meets universal mobile telecommunications system (UMTS) standard, need the wave point functional unit to send the voice data of adaptive multi-rate (AMR) coded frame form, the wave point functional unit sends described voice data by the transformat of each transmission channel, it is characterized in that:

Send coded command to adaptive multi-rate (AMR) encoder, described coded command has been specified a plurality of adaptive multi-rate (AMR) pattern, effective Speech frame coded sequence that adaptive multi-rate (AMR) encoder is produced 20 milliseconds of Speech frames coding according to described coded command, it is the set of a plurality of adaptive multi-rate (AMR) speech coding frame, it is an adaptive multi-rate (AMR) coded frame of mourning in silence, when described Speech frame coded sequence was a plurality of adaptive multi-rate (AMR) speech coding frame, the pattern of its adaptive multi-rate (AMR) coded frame was consistent with the adaptive multi-rate in the described coded command (AMR) pattern;

To wave point functional unit output Speech frame coded sequence;

In the wave point functional unit, send adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence to the process of transmission channel, comprise, in this Speech frame coded sequence, select an adaptive multi-rate (AMR) coded frame and select a transformat combination, the transformat combination of picking out, comprised the transformat that transmits all class bits of adaptive multi-rate (AMR) coded frame of picking out, adaptive multi-rate (AMR) coded frame of picking out has been dispatched on the transmission channel with this transformat combination.

2. according to the method for claim 1, it is characterized in that:

Each adaptive multi-rate (AMR) pattern in the described coded command is corresponding with a speech transmissions format combination in the transport format combination set of travelling carriage, and described speech transmissions format combination is the transformat combination that comprises all class bit transformats of a kind of adaptive multi-rate (AMR) pattern-coding frame.

3. according to the method for claim 2, it is characterized in that, during the radio frames of travelling carriage time-out transmission voice-over data, the Speech frame coded sequence from adaptive multi-rate (AMR) encoder is carried out buffer memory, after the described EOT end of timeout, adaptive multi-rate (AMR) coded frame in transmitting the Speech frame coded sequence is in the process of transmission channel, in the mode of per 20 a milliseconds of scheduling above adaptive multi-rate (AMR) coded frame, handle the part or all of Speech frame coded sequence that is buffered to transmission channel.

4. according to the method for claim 3, it is characterized in that described travelling carriage refers to that the sub-district is between transfer period during suspending the transmission radio frames.

5. according to the method for claim 3, it is characterized in that described travelling carriage suspends to send and refers to during the radio frames during the time-out voice data transmission during to transfer channel of other logic channel that priority scheduling is higher than the speech priority of logical channels.

6. according to method any in the claim 3 to 5, it is characterized in that, adaptive multi-rate (AMR) coded frame that described per 20 milliseconds of scheduling (do not comprise one) more than one is to dispatch the transmission channel of a complete adaptive multi-rate (AMR) coded frame to 10 millisecond Transmission Time Interval (TTI) to the mode of transmission channel.

7. the method for adaptive multi-rate (AMR) coded frame that receives of the processing of a travelling carriage, travelling carriage receives adaptive multi-rate (AMR) coded frame from the transmission channel that transmits speech, adaptive multi-rate (AMR) coded frame that adaptive multi-rate (AMR) decoder is received travelling carriage is characterized in that by the per 20 milliseconds of decodings outputs of finishing an adaptive multi-rate (AMR) coded frame of the order after arrive first:

Be that a voice call is set up the placement adaptive multi-rate, (AMR) lower limit of coded frame buffer area and this buffer area, receive the adaptive multi-rate of described voice call for the first time from travelling carriage, (AMR) coded frame begins, with the adaptive multi-rate of receiving, (AMR) the speech coding frame is stored in the buffer area, adaptive multi-rate in buffer area, when (AMR) number of coded frame reaches described lower limit, adaptive multi-rate, (AMR) decoder begins adaptive multi-rate that travelling carriage is received, (AMR) coded frame decoding

When travelling carriage is in voice (SPEECH) pattern, check that adaptive multi-rate (AMR) speech coding frame increases in per 20 milliseconds of buffer areas number: when described number less than 1 the time, initiatively add adaptive multi-rate (AMR) reception lost frames; When described number greater than 1 and buffering area in when having under voice (SPEECH) pattern reception lost frames that initiatively add, delete described under voice (SPEECH) pattern the reception lost frames of interpolation initiatively.

8. according to the method for claim 7, it is characterized in that:

Set up the number limits value of the reception lost frames that under voice (SPEECH) pattern, initiatively add of described buffer area, when the number of the reception lost frames that initiatively add under voice in the buffer area (SPEECH) pattern reaches described number limits value, under voice (SPEECH) pattern, no longer add receiving lost frames.

9. according to each method in the claim 7 to 8, it is characterized in that travelling carriage sends the message of the lower limit that contains described buffer area to network.

10. method that generates and transmit adaptive multi-rate (AMR) coded frame, in the voice call process, code pattern converter in core net (TC) becomes adaptive multi-rate (AMR) coded frame with the speech coding conversion of signals of input, described adaptive multi-rate (AMR) coded frame is placed on core net-radio access network interface (Iu) user plane frame and sends to radio network controller (RNC), radio network controller (RNC) is dispatched to adaptive multi-rate (AMR) coded frame on the transmission channel, it is characterized in that:

Send coded command to code pattern converter, described coded command has been specified a plurality of adaptive multi-rate (AMR) pattern, code pattern converter is by described coded command, speech coding signal to input carries out encoding operation, the Speech frame coded sequence of the effective 20 milliseconds of length speeches that produced, it is the set of some (more than or equal to 1) individual adaptive multi-rate (AMR) speech coding frame, it is an adaptive multi-rate (AMR) coded frame of mourning in silence, it is a free of data (NO_DATA) type frame, when described Speech frame coded sequence was many (greater than 1) individual adaptive multi-rate (AMR) speech coding frame, the pattern of its adaptive multi-rate (AMR) coded frame was consistent with the adaptive multi-rate in the described coded command (AMR) pattern;

With all adaptive multi-rate (AMR) coded frame in core net-radio access network interface (Iu) user plane frame voice-over frame coded sequence, send core net-radio access network interface (Iu) user plane frame of described voice-over frame coded sequence to radio network controller (RNC);

In radio network controller (RNC), dispatch adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence to the process of transmission channel, comprise, in this Speech frame coded sequence, select an adaptive multi-rate (AMR) coded frame and select a transformat combination, the transformat combination of picking out, comprised the transformat that transmits all class bits of adaptive multi-rate (AMR) coded frame of picking out, adaptive multi-rate (AMR) coded frame of picking out has been dispatched on the transmission channel with this transformat combination.

11. the method according to claim 10 is characterized in that:

Radio network controller (RNC) sends core net-radio access network interface (Iu) user plane rate controlled frame to core net, sequence number sign (RFCI n indicator) territory of a plurality of RAB stream combination indications in core net-radio access network interface (Iu) user plane rate controlled frame is corresponding one by one with a plurality of RAB stream combination indications (RFCI), these RAB stream combination indications (RFCI) are corresponding with adaptive multi-rate (AMR) pattern again, after core net is received described core net-radio access network interface (Iu) user plane rate controlled frame, to the coded command that code pattern converter sends, a plurality of adaptive multi-rate (AMR) pattern of appointment in the coded command is exactly described RAB stream combination indication (RFCI) pairing adaptive multi-rate (AMR) voice mode.

12. method according to claim 11, it is characterized in that, radio network controller (RNC) is set up buffer area, the Speech frame coded sequence that the Iu user plane frame of receiving during the transmission of time-out radio frames by the core net transmission is carried is buffered in this buffer area, after the described EOT end of timeout, adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence that scheduling is buffered adopts the mode of per 20 milliseconds of scheduling adaptive multi-rate (AMR) coded frame that (do not comprise) more than to transmission channel in the process of transmission channel.

13. the method according to claim 12 is characterized in that, is by the minizone within the radio network controller (RNC) between transfer period during described time-out radio frames sends.

14. the method according to claim 12 is characterized in that, is between radio network controller (RNC) between transfer period during described time-out radio frames sends.

15. method according to claim 12; it is characterized in that, during described time-out radio frames sends be the time-out voice data transmission of other logic channel during that be higher than the speech priority of logical channels by priority scheduling to transfer channel caused fail to send adaptive multi-rate (AMR) coded frame in the Speech frame coded sequence that those core net-radio access network interfaces (Iu) user plane frame carries during.

16. method according to claim 11, it is characterized in that, the Iu user plane frame that described core net sends to the voice-over frame coded sequence of radio network controller (RNC) comprises the delay sign of this Speech frame coded sequence, when postponing sign indication Speech frame coded sequence is when being delayed, and radio network controller (RNC) (does not comprise one) with per 20 milliseconds of scheduling more than one, and adaptive multi-rate (AMR) coded frame is dispatched this Speech frame coded sequence to the mode of transmission channel.

17. according to each method in the claim 12 to 16, it is characterized in that, adaptive multi-rate (AMR) coded frame that described per 20 milliseconds of scheduling (do not comprise one) more than one is to dispatch the transmission channel of a complete adaptive multi-rate (AMR) coded frame to 10 millisecond Transmission Time Interval (TTI) to the mode of transmission channel.

18. according in the claim 12 to 16 each method, it is characterized in that:

After the buffer area lower limit that outnumbers the purpose travelling carriage of the Speech frame coded sequence of buffer memory, begin to abandon the Speech frame coded sequence of buffer memory.

Images