JP2010130226A - Radio communication apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a radio communication apparatus for accurately controlling accumulation amount of packets without adverse effects of difference in clock to the transmission side, variation in delay time of receiving packet, and jitter or the like and for enabling handover to the radio communication network of different kind without lowering of reproducing quality and real time property. <P>SOLUTION: When performing handover to a second radio communication network 16 from a first radio communication network 15 and when start of preparation for handover is determined by obtaining previously the preparation time until the handover and respective delay times in the first radio communication network 15 and the second radio communication network 16, reproducing rate is controlled, based on the receiving packet data and reproducing packet data in a jitter buffer, so that the forecasted accumulation amount of packets of the jitter buffer becomes the target accumulation amount of packets based on the respective delay times in the first radio communication network 15 and the second radio communication network 16. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a radio communication apparatus capable of performing handover between different radio communication networks.

  In recent years, the Internet Engineering Task Force (IETF) has developed an IP mobility technology that enables seamless movement by enabling handover between a plurality of different wireless communication networks such as a mobile phone network and a wireless LAN in order to realize a ubiquitous environment. Is being considered. Specific protocols in this IP mobility technology include Mobile IPv4 and Mobile IPv6 (hereinafter collectively referred to as Mobile IP) that support movement of individual communication terminals, and support movement in units of networks. There is NEMO (Network Mobility).

  In mobile IP and NEMO, when a mobile node (MN) is handed over, the home agent (HA) to which the MN belongs is assigned the care-of address (CoA: Care of) of the IP address of the handover destination wireless communication network. By registering as (Address), communication with a counterpart node (CN: Correspondent Node) that is a communication partner is enabled via the wireless communication network of the handover destination.

  By the way, when a real-time application such as VoIP is executed through a wireless communication network, the allowable bandwidth of the wireless communication path changes depending on the propagation environment such as fading, and the change in the allowable bandwidth is in accordance with the change of the allowable bandwidth. The arrival interval of packets received by the communication terminal also changes.

  For this reason, generally, a jitter buffer is provided in a communication terminal, and received packets are temporarily accumulated in the jitter buffer, and then packets are read out from the jitter buffer and played back at intervals according to the application. That is, a packet reproduction interval shift due to arrival interval shift (jitter) is absorbed to prevent a decrease in reproduction quality such as reproduction sound quality. Also, if the jitter is large and there are no packets in the jitter buffer and silence occurs, or if a large amount of packets are received in a short time and the packets do not fit in the jitter buffer, the playback speed is increased. It changes, discards received packets, and changes the size of the jitter buffer.

  On the other hand, the downlink absolute delay time of the packet received by the communication terminal, that is, the time (delay time) required until the packet transmitted from the partner communication terminal is received via the wireless communication network depends on the wireless communication network. Different. For this reason, in the case of a wireless communication device in which a communication terminal moves, when handover is performed to a different wireless communication network, for example, if the downlink absolute delay time of the handover destination is longer than the downlink absolute delay time of the handover source, A packet reception idle time is generated by the difference.

  In such a case, for example, when the packet reading interval from the jitter buffer is set to a constant interval according to the application and the packet is reproduced at a constant reproduction speed, the last packet received from the handover source wireless communication network is If the above reception idle time is longer than the time required for reading from the jitter buffer, that is, the jitter buffer standard delay time equivalent to the standard packet accumulation amount of the jitter buffer by the application, the long time is the jitter The packet in the buffer becomes empty. As a result, since the packet is not played back at least during this time period, the sound becomes silent and the playback quality deteriorates.

  14A and 14B are diagrams for explaining the control method of the jitter buffer in this case. FIG. 14A shows the number of packets received by the jitter buffer per unit time, and FIG. 14B shows the packets from the jitter buffer. FIG. 14C shows the packet accumulation amount TBP (sec) in the jitter buffer. FIG. 15 shows the packet flow in this case, where “transmission” is the packet transmission timing by the counterpart communication terminal, “reception” is the reception timing of the packet received by the jitter buffer of the wireless communication device, “Reproduction” indicates packet reproduction timing (packet read timing from the jitter buffer) by the wireless communication apparatus. Here, in each of the handover source radio communication network A and the handover destination radio communication network B, it is shown that there is no fluctuation (shift in arrival interval) in the received packet.

  As apparent from FIGS. 14 and 15, the downlink absolute delay time TddnB in the handover destination radio communication network B is longer than the downlink absolute delay time TddnA in the handover source radio communication network A, and (TddnB-TddnA) Is longer than the jitter buffer standard delay time TBPs corresponding to the standard packet accumulation amount of the jitter buffer, if the packets in the jitter buffer are continuously reproduced at the standard reproduction speed Vs, TBPn = {(TddnB-TddnA) -TBPs} During the time, the packet is not reproduced. In addition, in this case, when a packet is received from the wireless communication network B that is the handover destination, the packet is immediately reproduced, so that jitter cannot be absorbed.

  For example, if the packet reception status is monitored and the packet cannot be received at the standard reception interval, reading of the packet from the jitter buffer, that is, the packet A jitter buffer control method for controlling the reproduction speed has been proposed (see, for example, Patent Document 1).

  FIG. 16 is a diagram for explaining the jitter buffer control method disclosed in Patent Document 1. FIGS. 16A to 16C are similar to FIGS. 14A to 14C, respectively. The number of received packets per unit time, the playback speed V, and the packet accumulation amount TBP in the jitter buffer are shown. FIG. 17 shows the packet flow in this case.

  In FIG. 16 and FIG. 17, the packet is received at the reception interval until then, as in the case of handover from the wireless communication network A having the downlink absolute delay time TddnA to the wireless communication network B having the downlink absolute delay time TddnB longer than TddnA. Is not received, the packet playback speed V in the jitter buffer is gradually reduced from the standard playback speed Vs as the reception interval increases. Control is performed so that the playback speed V is gradually increased to the standard playback speed Vs in accordance with the packet accumulation amount TBP in the buffer.

JP 2006-238445 A

  However, in the jitter buffer control method disclosed in Patent Document 1, when a packet cannot be received at the previous reception interval, the reproduction speed V of the packet currently accumulated in the jitter buffer is gradually decreased. It is only controlled like this. For this reason, when the downlink absolute delay time TddnB at the handover destination is relatively long, there is a concern that the reproduction speed is greatly changed to a low speed and the reproduction quality is deteriorated. Therefore, for example, in the case of VoIP, the playback speed V greatly changes from the speed of the original voice, which is the standard playback speed Vs, so that the quality of the playback sound is greatly reduced and it is difficult for the user to hear.

  FIGS. 16 and 17 show control examples in the case where the packet in the jitter buffer is not emptied and no silence is generated, but how much the packet reception interval is actually vacant. Is unknown. For this reason, depending on the amount of packets stored in the jitter buffer and the downlink absolute delay time TddnB at the handover destination, there is a concern that the packets in the jitter buffer become empty and silence may occur. In order to prevent the occurrence of silence or the like, it is conceivable to increase the standard packet accumulation amount of the jitter buffer. However, if this is done, for example, in the case of VoIP, there is a delay in the packet reproduction from the counterpart terminal, so that the reproduction quality and real-time property are deteriorated.

  As a wireless communication device that solves such a problem, the present inventors, for example, prior to the start of handover, set the packet playback speed to a low playback speed that is lower than the standard playback speed, so that at the start of handover. A wireless communication apparatus has been developed in which the jitter buffer packet storage amount is set to a desired target packet storage amount that does not become empty during handover. According to this wireless communication apparatus, since the packet accumulation amount of the jitter buffer does not become empty at the time of handover, it is possible to effectively prevent the occurrence of silence and improve the reproduction quality and real-time property.

  However, according to further experimental studies by the present inventors, even if the packet playback speed is changed to a low playback speed before the start of handover as described above, the amount of packets stored in the jitter buffer is In addition to this, for example, there is a difference between the clock on the transmission side and the clock on the reception side, or when the packet delay time varies depending on the state of the communication path. For this reason, even if a low playback speed period is set based on the target packet accumulation amount, the packet accumulation amount cannot be accurately controlled, and the target packet accumulation amount may not be obtained at the start of handover. It has been found.

  As a solution to this, it is possible to monitor the actual packet storage amount in the jitter buffer along with the change to the low playback speed, and return the playback speed from the low playback speed to the normal playback speed when the target packet storage amount is reached. It is done. However, the packet accumulation amount temporarily fluctuates due to jitter. For this reason, if it is detected that the target packet accumulation amount has been reached due to temporary fluctuations in the packet accumulation amount due to jitter, the packet accumulation amount cannot be accurately controlled.

FIG. 18 shows an example of control in this case. (A) is a reception sequence number of a packet received by the jitter buffer, (b) is a reproduction speed V, and (c) is a packet accumulation amount TBP in the jitter buffer. Each is shown. As shown in FIG. 18, when the reproduction speed V is set to a low reproduction speed V _L lower than the standard reproduction speed Vs and the packet accumulation amount TBP in the jitter buffer is controlled to the target packet accumulation amount TBPd, the jitter buffer substantially reduces the packet. In the period T1 during reception at a predetermined timing, the packet accumulation amount TBP continuously and gradually increases.

After that, when a period (dwell period) T2 in which no packet is temporarily received due to the occurrence of jitter occurs, the packet accumulation amount TBP is gradually reduced during that dwell period T2 because the playback speed V is the low playback speed V _L. Decrease. Thereafter, when a large number of packets that should originally be received in the retention period T2 are received in a short time after the retention period T2 has elapsed, the packet accumulation amount TBP increases rapidly. For this reason, depending on the retention period T2, when the packet accumulation amount TBP far exceeds the target packet accumulation amount TBPd, the reproduction speed V returns from the low reproduction speed V _L to the standard reproduction speed Vs. However, the fact that the packet accumulation amount TBP is larger than necessary means that the call delay time is unnecessarily long, and the call quality is further deteriorated. In addition, when the packet accumulation amount TBP suddenly increases, if the target packet accumulation amount TBPd is far exceeded, packets exceeding the maximum capacity of the jitter buffer are discarded and not reproduced, resulting in poor reproduction quality. There is concern about inviting.

Further, as described above, prior to the start of the handover, the packet reproduction rate V is changed to the low reproduction rate V _{L so} that the packet accumulation amount TBP in the jitter buffer is controlled to the target packet accumulation amount TBPd. However, as shown in FIG. 19 (a), if the playback speed V is kept constant until the start of handover thereafter, the standard playback speed Vs remains constant due to the clock difference between the transmission and reception and the packet delay time fluctuation. As shown in FIG. 19B, when the amount of accumulated packets decreases, there is a concern that the packets in the jitter buffer are emptied and silence is generated between the start and completion of the handover. .

In FIG. 19, when the packet accumulation amount TBP is smaller than the target packet accumulation amount TBPd according to the fluctuation of the packet accumulation amount until the handover starts, the reproduction speed V is changed to the low reproduction speed V _L , If you increase than packet accumulated amount TBPd, the playback speed V as a fast high reproduction speed V _H than the normal reproduction speed Vs, it is conceivable to maintain the target packet accumulated amount TBPd. However, when the reproduction speed V is changed to the low reproduction speed V _L and the actual packet accumulation amount TBP of the jitter buffer is monitored and increased to the target packet accumulation amount TBPd, the same problem as described in FIG. 18 occurs. I am concerned that it will occur.

Also, if the playback speed V is changed to the high playback speed V _H and the actual packet accumulation amount TBP in the jitter buffer is monitored and reduced to the target packet accumulation amount TBPd, the same problem may occur. . FIG. 20 shows an example of control in this case. (A) is a reception sequence number of a packet received by the jitter buffer, (b) is a reproduction speed V, and (c) is a packet accumulation amount TBP in the jitter buffer. Each is shown.

As shown in FIG. 20, when the reproduction speed V is set to the high reproduction speed V _H , the packet accumulation amount TBP continuously decreases during the period T1 in which the jitter buffer receives packets at almost predetermined timing. After that, when the retention period T2 occurs, the packet accumulation amount TBP decreases rapidly because the reproduction speed V is the high reproduction speed V _H in the retention period T2, and the target packet accumulation amount TBPd is reduced in the decrease process. If it falls below that, the playback speed V returns from the high playback speed V _H to the standard playback speed Vs, and the decrease in the packet accumulation amount TBP becomes gradual. Thereafter, when a large number of packets that should originally be received in the retention period T2 are received in a short time after the retention period T2 has elapsed, the packet accumulation amount TBP increases rapidly. For this reason, depending on the retention period T2, after the playback speed V returns to the standard playback speed Vs, the packet storage amount TBP far exceeds the target packet storage amount TBPd, and the packet storage amount cannot be sufficiently reduced. There is a concern that the call delay time cannot be suppressed to the minimum necessary, leading to a decrease in reproduction quality.

  Therefore, an object of the present invention made in view of such a point is to accurately control the packet accumulation amount without being affected by a clock difference from the transmission side, a variation in delay time of a received packet, jitter, etc. An object of the present invention is to provide a wireless communication apparatus capable of performing handover to a different wireless communication network without degrading real-time performance.

The invention of a wireless communication device according to claim 1 that achieves the above object is as follows:
A wireless communication unit that performs wireless communication by connecting to a first wireless communication network and a second wireless communication network different from the first wireless communication network;
An execution unit that executes an application of a real-time communication system via the wireless communication unit;
A communication quality acquisition unit for acquiring communication quality of a radio link in the first wireless communication network while executing the application by connecting to the first wireless communication network;
A determination unit that determines whether to start preparation for handover from the first wireless communication network to the second wireless communication network based on the communication quality acquired by the communication quality acquisition unit;
When the determination unit determines the start of handover preparation during execution of the application, an estimation unit that estimates a handover preparation time until the handover is started based on the communication quality acquired by the communication quality acquisition unit;
When the determination unit determines the start of handover preparation, the measurement unit that measures respective delay times in the first wireless communication network and the second wireless communication network,
The execution unit is
A jitter buffer for accumulating received packets;
A jitter buffer monitoring unit that monitors received packet information including a sequence number and a reception time of a packet stored in the jitter buffer, and monitors reproduction packet information including a sequence number and a reproduction time of the packet read from the jitter buffer; ,
A jitter buffer control unit for controlling the playback speed of packets from the jitter buffer;
When the determination unit determines the start of handover preparation, the predicted packet storage amount of the jitter buffer is calculated based on the received packet information and the reproduction packet information by the jitter buffer monitoring unit, and the predicted packet storage amount is Jitter buffer accumulation for controlling the reproduction speed by the jitter buffer controller so as to be a target packet accumulation amount based on a delay time in the first radio communication network and the second radio communication network measured by the measurement unit And a quantity control unit.

The invention according to claim 2 is the wireless communication apparatus according to claim 1,
The jitter buffer accumulation amount control unit
For each packet transmission interval by the application, based on the received packet information for a certain period and the current time, calculate a scheduled reception sequence number of a packet that is considered to be currently received, and at the start of the application Based on the reproduction packet information and the current time, a reproduction scheduled sequence number of a packet considered to be currently reproduced is calculated, and the predicted packet is calculated from a difference between the reception scheduled sequence number and the reproduction scheduled sequence number. Calculate the accumulated amount,
It is characterized by this.

The invention according to claim 3 is the wireless communication device according to claim 1 or 2,
The jitter buffer accumulation amount control unit
Controlling the reproduction speed by the jitter buffer control unit so that the predicted packet accumulation amount becomes the target packet accumulation amount until the handover preparation time estimated by the estimation unit elapses.
It is characterized by this.

The invention according to claim 4 is the wireless communication apparatus according to any one of claims 1 to 3,
The jitter buffer accumulation amount control unit
When the handover from the first wireless communication network to the second wireless communication network is completed, the playback speed by the jitter buffer control unit is controlled so that the predicted packet storage amount becomes the standard packet storage amount by the application. ,
It is characterized by this.

The invention according to claim 5 is the wireless communication apparatus according to any one of claims 1 to 4,
The jitter buffer accumulation amount control unit
The delay time in the first wireless communication network is compared with the delay time in the second wireless communication network, and the delay time in the second wireless communication network is a predetermined time than the delay time in the first wireless communication network. If it is longer, the playback speed by the jitter buffer control unit is slowed down.
It is characterized by this.

  When performing a handover from the first wireless communication network to the second wireless communication network, the wireless communication device according to the present invention previously prepares for the handover, and sets each of the first wireless communication network and the second wireless communication network. When the delay time is acquired and the start of handover preparation is determined, the estimated packet accumulation amount of the jitter buffer is determined in the first wireless communication network and the second wireless communication network based on the received packet information and the reproduction packet information in the jitter buffer. The playback speed is controlled so that the target packet accumulation amount is based on each delay time. This makes it possible to accurately control the amount of packet accumulation without being affected by clock differences from the transmission side, fluctuations in the delay time of received packets, jitter, etc., and different wireless communication networks without degrading playback quality and real-time performance. Handover to is possible.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a diagram showing a schematic configuration of a communication network that can be used by a wireless communication apparatus according to an embodiment of the present invention. In FIG. 1, it is assumed that a wireless communication device (MN) 11 that is a mobile node performs a call using VoIP that is an application of a real-time communication system with a counterpart communication terminal (CN) 12 that is an opposite node. The wireless communication device 11 can be handed over between the first wireless communication network 15 and the second wireless communication network 16. The first wireless communication network 15 and the second wireless communication network 16 are coupled to the Internet 17.

  Here, the first wireless communication network 15 is composed of, for example, a wireless LAN, and the second wireless communication network 16 is composed of, for example, a cdma2000 mobile phone network, and the delay time (downlink absolute delay time) in the first wireless communication network 15 is This is shorter than the delay time (downlink absolute delay time) in the second wireless communication network 16. In FIG. 1, reference numeral 15 a indicates an access point of the first wireless communication network 15, and reference numeral 16 a indicates a base station of the second wireless communication network 16.

  For example, the partner communication terminal 12 includes a personal computer to which a handset 12a is connected and a softphone as a telephone function unit is installed, and is connected to the Internet 17 via an Internet service provider (not shown).

  Further, SIP (Session Initiation Protocol) servers 21 and 22 for controlling communication are connected to the first wireless communication network 15 and the second wireless communication network 16, respectively. Further, a home agent (HA) 23 that transfers a received packet addressed to the wireless communication device 11 to the wireless communication network to which the wireless communication device 11 is connected and a SIP server 24 that controls communication are connected to the Internet 17. Has been.

  In the communication network shown in FIG. 1, the home address used in the wireless communication network to which the wireless communication apparatus 11 originally belongs is registered in the HA 23, and the IP address of the handover destination wireless communication network is taken care of (CoA: care) at the time of handover. of address) allows handover between different wireless communication networks. Note that such IP mobility technology is well known in the above-described mobile IP and NEMO, and thus detailed description thereof is omitted here.

  In the following description, the wireless communication device 11 registers the IP address of the first wireless communication network 15 in the HA 23 as a care-of address (first wireless CoA), and communicates with the counterpart communication terminal 12 via the first wireless communication network 15. It is assumed that handover is performed from the communication state to the second wireless communication network 16. At this time, even if the IP address changes due to handover, multihoming (temporarily registering two Care of Addresses in the HA 23) is used so that the call can be continued without causing packet loss.

  FIG. 2 is a functional block diagram showing a schematic configuration of radio communication apparatus 11 according to the present embodiment shown in FIG. The wireless communication device 11 includes a first wireless I / F (interface) 31 corresponding to the first wireless communication network 15, a second wireless I / F 32 corresponding to the second wireless communication network 16, and a real-time system such as VoIP. A telephone function unit 33 that constitutes an execution unit that executes an application, a communication processing unit 34 that controls connection to the first wireless communication network 15 and the second wireless communication network 16, and a first wireless communication network 15 and a second wireless communication. A wireless information acquisition unit 35 that acquires wireless information of the communication network 16 and a handover control unit 36 that controls handover between the first wireless communication network 15 and the second wireless communication network 16 are provided.

  The communication processing unit 34 constitutes a wireless communication unit that performs wireless communication. Between the telephone function unit 33 and the counterpart communication terminal 12, the first wireless communication network 15 or the second wireless communication network 16 is used. While making a call, the connection of the first wireless I / F 31 or the second wireless I / F 32 is controlled so as to communicate with the HA 23 under the control of the handover control unit 36.

  The wireless information acquisition unit 35 acquires (measures) the communication quality of the corresponding first wireless communication network 15 and second wireless communication network 16 from the first wireless I / F 31 and the second wireless I / F 32 as wireless information. Then, the acquired communication quality is supplied to the handover control unit 36. Here, as the communication quality, for example, an RSSI (Received Signal Strength Indicator) representing a radio state is acquired. Note that the wireless state of the second wireless communication network 16 that is not used for a call is acquired (measured) by receiving broadcast information transmitted from the base station 16a, for example. Therefore, in the present embodiment, the radio information acquisition unit 35 constitutes a communication quality acquisition unit that acquires the communication quality (radio state) of the radio link.

  Based on the communication quality from the radio information acquisition unit 35, the handover control unit 36 determines whether or not to schedule handover, that is, whether or not to start preparation for handover. The time until the start of the handover is calculated, and the handover process is executed according to the handover start time. In addition, when the handover control unit 36 determines the schedule of the handover, the handover control unit 36 supplies the information to the telephone function unit 33, and the downlink absolute delay time in the first wireless communication network 15 that is the handover source and the handover, Handover information including the absolute downlink delay time in the second wireless communication network 16 is acquired and supplied to the telephone function unit 33. Further, the handover control unit 36 supplies handover start information and handover completion information to the telephone function unit 33 in accordance with the handover process.

  FIG. 3 is a functional block diagram showing a schematic configuration of the telephone function unit 33 of the wireless communication apparatus 11 shown in FIG. The telephone function unit 33 is composed of, for example, a soft phone, and has a button input unit 41, a screen display unit 42, a microphone 43, an encoder 44, a packet transmission unit 45, a packet reception unit 46, a jitter buffer, as in the configuration of a known soft phone. 47, a decoder 48, a speaker 49, a jitter buffer monitoring unit 50, a jitter buffer control unit 51, a SIP control unit 52, and an overall control unit 53 for controlling the overall operation.

  The overall control unit 53 acquires user operation information via the button input unit 41 and the screen display unit 42, and controls the overall operation based on the acquired information. The SIP control unit 52 controls SIP procedures for starting and ending a call. During a call, the audio data acquired from the microphone 43 is encoded by the encoder 44 which is an encoding unit, and the encoded data is sent from the packet transmission unit 45 to an RTP (Real-time Transport Protocol) packet for communication. The data is transmitted to the partner communication terminal 12 via the processing unit 34.

  The RTP packet from the partner communication terminal 12 received by the packet receiving unit 46 via the communication processing unit 34 is once taken into the jitter buffer 47 and read out. The read packet is read by the decoder 48. The payload portion is decoded and output as reproduced sound from the speaker 49.

  In the present embodiment, the jitter buffer monitoring unit 50 receives the received packet information including the sequence number and reception time of the packet received by the jitter buffer 47 and the reproduction including the sequence number and reproduction time of the packet reproduced from the jitter buffer 47. Monitor packet information.

  The telephone function unit 33 further includes a handover information acquisition unit 55 and a packet accumulation amount control unit 56. The handover information acquisition unit 55 monitors the presence / absence of handover schedule determination information from the handover control unit 36 at regular intervals. When the handover schedule determination information is acquired, the handover information acquisition unit 55 further receives the handover source and handover destination information from the handover control unit 36. The handover information including each downlink absolute delay time is acquired, and the acquired handover information is supplied to the packet accumulation amount control unit 56. When the handover information acquisition unit 55 receives the handover start information and the handover completion information from the handover control unit 36, the handover information acquisition unit 55 notifies the packet accumulation amount control unit 56 to that effect.

  When a call (application) is started, the packet accumulation amount control unit 56 obtains reproduction packet information including the sequence number and reproduction time of the packet reproduced first from the jitter buffer monitoring unit 50, and the jitter buffer 47 uses the packet Is received from the jitter buffer monitoring unit 50, the received packet information including the sequence number of the received packet and the reception time. Then, when the handover information is input from the handover information acquisition unit 55, the packet accumulation amount control unit 56, based on the handover information and the reproduction packet information and the reception packet information, the predicted packet accumulation amount of the jitter buffer 47 And the reproduction speed of packets from the jitter buffer 47 is controlled via the jitter buffer control unit 51 so that the predicted packet accumulation amount becomes a target packet accumulation amount that does not cause silence during handover.

  Further, the packet accumulation amount control unit 56 does not perform the above-described reproduction packet information and reception packet until the handover information is notified from the handover information acquisition unit 55 even after the control of the target packet accumulation amount based on the handover information is finished. Based on the information, the packet reproduction speed is controlled via the jitter buffer control unit 51 so that the predicted packet accumulation amount of the jitter buffer 47 is maintained within the allowable range of the target packet accumulation amount. Further, when the packet accumulation amount control unit 56 receives a handover completion notification from the handover information acquisition unit 55, the predicted packet accumulation amount of the jitter buffer 47 is determined by the application standard based on the reproduction packet information and the reception packet information. The packet reproduction speed is controlled via the jitter buffer controller 51 so that the packet accumulation amount is within the allowable range.

  Hereinafter, the operation of radio communication apparatus 11 according to the present embodiment will be described.

  First, the operation of the handover control unit 36 will be mainly described. The handover control unit 36 determines a handover schedule based on the communication qualities acquired from the first wireless I / F 31 and the second wireless I / F 32 via the wireless information acquisition unit 35, respectively. For example, when a call is made by forming a wireless link with the first wireless communication network 15, the communication quality acquired from the first wireless I / F 31 is lower than the handover schedule determination threshold, and the second wireless I / F When the communication quality of F32 is equal to or higher than the handover schedule determination threshold, the handover control unit 36 determines the handover schedule to the second wireless communication network 16, that is, the start of handover preparation.

  When it is determined to start preparation for handover, the information is supplied to the telephone function unit 33, and the time until the start of handover, that is, until a Registration Request (Binding Update in NEMO) is transmitted as handover request information. The handover preparation time Ts (sec), which is the time of the above, is calculated, and the handover process is executed according to the handover start time. When the start of handover preparation is determined, the downlink absolute delay time Tddn1 (sec) in the currently used wireless communication network (here, the first wireless communication network 15) and the handover destination wireless communication network ( Here, the downlink absolute delay time Tddn2 (sec) in the second wireless communication network 16) is acquired, and the information is supplied to the telephone function unit 33 as handover information. Therefore, in the present embodiment, the handover control unit 36 determines whether to start preparation for handover, an estimation unit that estimates handover preparation time, and the first radio communication network 15 and the second radio communication. A measuring unit for measuring each delay time in the network 16 is configured.

  Next, a method for acquiring the handover preparation time Ts, the handover source downlink absolute delay time Tddn1, and the handover destination downlink absolute delay time Tddn2 by the handover control unit 36 will be described.

(How to obtain handover preparation time Ts)
For example, as shown in FIGS. 4A and 4B, the handover preparation time Ts is calculated based on a unit time change rate ΔRs (slope) of the radio state (Rs) for determining communication quality. Here, the rate of change ΔRs can be measured and acquired when the radio state falls below the handover schedule determination threshold and the handover schedule is determined, but in this embodiment, the handover schedule determination is performed during the call. The change rate average value ΔRsrms from the time point to a predetermined time before is acquired.

For this reason, the handover control unit 36 calculates the change rate ΔRs (t) of the unit time (Δt) of the radio state in the currently used radio communication network according to the following formula at a predetermined timing, A plurality of change rates ΔRs (t) up to 2 seconds before) are held in the memory, and when a handover schedule is determined, an average change rate ΔRsrms up to a predetermined time held at that time is calculated. Here, it is assumed that the wireless state is gradually getting worse.
[Equation 1]
ΔRs (t) = | {Rs (t) −Rs (t−Δt)} / Δt |

  Thereafter, the handover control unit 36 determines whether or not the calculated change rate average value ΔRsrms is smaller than a preset change rate threshold value Rsref. As a result, in the case of ΔRsrms ≦ Rsref, that is, when the wireless state changes slowly, as shown in FIG. 4A, the handover preparation time Ts is set to a preset standard time Tref (for example, 5 sec). To do.

  On the other hand, when ΔRsrms> Rsref, that is, when the radio state changes rapidly, for example, Ts = Tref (Rsref / ΔRsrms) is calculated, and the higher the rate of change ΔRsrms, the higher the handover preparation time Ts. Is set shorter than the standard time Tref. FIG. 4B shows a case where ΔRsrms> Rsref and the handover preparation time Ts is set to approximately half the standard time Tref (2.5 sec).

(Acquisition method of absolute delay time Tddn1, Tddn2)
The handover source downlink absolute delay time Tddn1 and the handover destination downlink absolute delay time Tddn2 are acquired by, for example, any one of first to fourth absolute delay time acquisition methods described below.

(A) First Absolute Delay Time Acquisition Method When the handover control unit 36 determines a handover schedule, for example, the HA 23 that is time-synchronized with the wireless communication device 11 via the first wireless I / F 31 that is performing communication. On the other hand, transmission of a measurement packet having a transmission time stamp is requested, and thereby, the measurement packet is transmitted from the HA 23 to both the first wireless communication network 15 and the second wireless communication network 16. The wireless communication device 11 receives the measurement packet transmitted from the HA 23 via the corresponding first wireless I / F 31 and second wireless I / F 32, respectively, and based on the reception time and the time stamp of the measurement packet. Then, the downstream absolute delay times Tddn1 and Tddn2 of the corresponding network are measured. When the absolute downlink delay time of the handover source wireless communication network can be measured from the received packet during a call, transmission of the measurement packet to the wireless communication network can be omitted.

(B) Second Absolute Delay Time Acquisition Method When the handover control unit 36 determines a handover schedule, for example, the HA 23 that is time-synchronized with the wireless communication device 11 via the first wireless I / F 31 that is performing communication. As a result, the HA 23 sends a measurement packet to both the first wireless communication network 15 and the second wireless communication network 16 in the same manner as in the first absolute delay time acquisition method. Then, the downstream absolute delay times Tddn1 and Tddn2 of the corresponding network are measured.

(C) Third Absolute Delay Time Acquisition Method When the handover control unit 36 determines a handover schedule, the first wireless I / F 31 is transmitted from the wireless communication device 11 to the HA 23 that is time-synchronized with the wireless communication device 11. And via the second wireless I / F 32, the measurement packets such as PING and RTCP are transmitted from both the first wireless communication network 15 and the second wireless communication network 16, and the reply is received. Downlink absolute delay times Tddn1 and Tddn2 are measured.

  In the above (a) to (c), since the network between the counterpart communication terminal (CN) 12 and the HA 23 is not switched, the absolute delay time between them is not considered.

(D) Fourth Absolute Delay Time Acquisition Method When the handover control unit 36 determines a handover schedule, the downlink absolute delay time of each wireless communication network is acquired using the handover technique studied in IEEE 802.21. To do. In IEEE 802.21 (Media Independent Handover (MIH)), as a handover technique between different types of wireless communication networks (WiFi, WiMAX, mobile phones, etc.), a means for controlling handover (in FIG. 2, the handover control unit 36) is an MIH user. It is considered that MIHF (MIH Function) acquires wireless information of a communication device based on a request from the MIH user and provides it to the MIH user. It is also considered that an MIH user acquires information from an information server in a connected network through MIHF in his / her terminal.

  FIG. 5 is a diagram for explaining the fourth absolute delay time acquisition method. In FIG. 5, a first wireless communication network 15 and a second wireless communication network 16 are connected to a backbone network 60 constituting the Internet 17 of FIG. 1 together with other wireless communication networks, and a delay time is added to the backbone network 60. A measurement server 61 to be measured is connected. A first information server 62 is connected to the first wireless communication network 15, and a second information server 63 is connected to the second wireless communication network 16. The partner communication terminal 12 is connected to the backbone network 60 via the provider 65.

  The first information server 62 uses the one-way network delay reference time Tn1 from the measurement server 61 to the access point 15a as a reference for delay time measurement, and the wireless communication apparatus 11 connected from the access point 15a to the access point 15a. Holds up and down radio delay reference times Trup1, Trdn1. Similarly, the second information server 63 uses a one-way network delay reference time Tn2 from the measurement server 61 to the base station 16a as a reference for delay time measurement, and a radio connected from the base station 16a to the base station 16a. The upper and lower radio delay reference times Trup2 and Trdn2 up to the communication device 11 are held.

  Here, the network delay reference times Tn1 and Tn2 are transmitted / received between the access point 15a and the measurement server 61, and between the base station 16a and the measurement server 61, respectively (such as PING and RTCP). The round trip time is measured, and the round trip time is halved.

  The upper and lower radio delay reference times Trup1 and Trdn1 in the first radio communication network 15 send packets from the access point 15a to the radio communication device 11 connected to the access point 15a and time-synchronized with the access point 15a. The wireless communication apparatus that has received the packet records the received time and sends back the packet, thereby calculating each of the upstream and downstream delay times.

  Similarly, the upper and lower radio delay reference times Trup2 and Trdn2 in the second radio communication network 16 send packets from the base station 16a to the radio communication device 11 connected to the base station 16a and time-synchronized with the base station 16a. The wireless communication device that has received the packet records the received time and sends back the packet to calculate each of the upstream and downstream delay times.

  When connecting to the first wireless communication network 15, the handover control unit 36 transmits the network delay reference time Tn1 and the wireless delay reference time from the first information server 62 connected to the first wireless communication network 15 via MIHF. Get Trdn1 and Trup1. In addition, the handover control unit 36 transmits / receives a packet to / from the other party (here, the other communication terminal 12 that is not time-synchronized with the wireless communication apparatus 11) for which the delay time is to be measured. The round trip time (Tn3 + Trdn3 + Tn3 + Trup3) is measured. From this value, the one-way delay time (Tn3-Tn1) between the partner communication terminal 12 and the measurement server 61 is obtained as follows, and the handover between the wireless communication device 11 and the partner communication terminal 12 is performed. Tn3 + Trdn3 corresponding to the original downlink absolute delay time Tddn1 is calculated.

[Equation 2]
Tn3-Tn1 = {(Tn3 + Trdn3 + Tn3 + Trup3)-(Tn1 + Trdn1 + Tn1 + Trup1)} / 2
Tddn1 = Tn3 + Trdn3 = Tn1 + Trdn1 + (Tn3-Tn1)

  Tn3 + Trup3 corresponding to the handover source upstream absolute delay time Tdup1 between the wireless communication device 11 and the counterpart communication terminal 12 can be obtained by Tdup1 = Tn3 + Trup3 = Tn1 + Trup1 + (Tn3-Tn1). .

  Further, when the handover control unit 36 determines the handover schedule, the handover control unit 36 acquires the network delay reference time Tn2 and the radio delay reference time Trdn2 of the handover destination, and therefore the first information server of the first wireless communication network 15 currently connected to the handover control unit 36. The location information of the wireless communication device 11 is transmitted to the second information server 63 of the second wireless communication network 16 that is the handover destination via 62, and the network delay reference time Tn2 and the wireless delay reference time Trdn2 are returned. Request. As a result, the second information server 63 determines the network delay reference time Tn2 and the radio delay reference time Trdn2 of the base station 16a considered to be connected in consideration of the position information and the number of connected users of each base station. 1 It returns to the wireless communication device 11 via the information server 62.

  The handover controller 36 receives the handover destination network delay reference time Tn2 and the radio delay reference time Trdn2 returned from the second information server 63, and uses the obtained information and the calculated (Tn3-Tn1), In the following manner, Tn4 + Trdn4 corresponding to the handover destination downlink absolute delay time Tddn2 between the wireless communication apparatus 11 and the counterpart communication terminal 12 is calculated.

[Equation 3]
Tddn2 = Tn4 + Trdn4 = (Tn2 + Trdn2) + (Tn3-Tn1)

  The handover control unit 36 requests the second information server 63 to return the wireless delay reference time Trup2, which corresponds to the handover destination upstream absolute delay time Tdup2 between the wireless communication device 11 and the counterpart communication terminal 12. Tn4 + Trup4 can be obtained by Tdup2 = Tn4 + Trup4 = (Tn2 + Trup2) + (Tn3-Tn1).

  The downlink absolute delay times Tddn1 and Tddn2 acquired by any one of the first to fourth absolute delay time acquisition methods described above are the radio communication network together with the downlink absolute delay times acquired in the same manner for other radio communication networks. Each time it is stored in a memory (not shown) in the handover control unit 36, and each time a handover schedule is determined, the corresponding downlink source delay time of the handover source and the handover destination is supplied to the telephone function unit 33. May be.

  When determining the handover schedule, the handover control unit 36 controls the communication processing unit 34 to connect the second wireless I / F 32 to the second wireless communication network 16. Thereafter, when the handover preparation time Ts elapses, the handover control unit 36 transmits a Registration Request (Binding Update in NEMO) to the HA 23 via the second wireless communication network 16 that is the handover destination, and the handover destination to the HA 23. Register the care of address.

  At that time, the handover control unit 36 sets 8 bits of the Registration Request Field of the Registration Request message in the communication processing unit 34 (in NEMO, multiple care of address is used), and the first radio communication network 15 also uses the second radio. The communication network 16 can also communicate.

  After that, when receiving the Registration Reply (Binding Acknowledge in NEMO) which is the handover completion information returned from the HA 23, the handover control unit 36 cancels the registration of the care-of address of the first wireless communication network 15 that is the handover source, After disconnecting the connection, the communication processing unit 34 is controlled to continue the VoIP application via the second wireless communication network 16 that is the handover destination, and the received handover completion information is supplied to the telephone function unit 33.

  Next, the operation of the telephone function unit 33 will be described with reference to the sequence diagram shown in FIG.

  The handover information acquisition unit 55 monitors whether or not there is handover schedule determination information from the handover control unit 36 at regular intervals. If the handover schedule determination information is acquired, the handover information acquisition unit 55 further receives a handover source downlink from the handover control unit 36. The handover information of the absolute delay time Tddn1 and the handover destination downlink absolute delay time Tddn2 is acquired and supplied to the packet accumulation amount control unit 56.

  When the handover information is input from the handover information acquisition unit 55, the packet accumulation amount control unit 56 performs handover based on the input handover information, and the reproduced packet information and the received packet information from the jitter buffer monitoring unit 50. In this case, the jitter buffer control unit 51 changes the reproduction rate of the received packet of the jitter buffer 47 so as to obtain a target packet accumulation amount in which no silence is generated, thereby controlling the predicted packet accumulation amount in the jitter buffer 47.

  Therefore, when acquiring the handover information, the packet accumulation amount control unit 56 calculates the absolute delay time difference Tab (Tab = Tddn2−Tddn1) between the handover source and the handover destination from the handover information, and the calculated absolute delay time difference Tab Is equal to or greater than a certain value (> 0), the jitter buffer control unit 51 changes the reproduction rate of the received packet from the jitter buffer 47 to a low speed so as to absorb the silent period due to the absolute delay time difference Tab. The predicted packet accumulation amount of 47 is set as a target packet accumulation amount that is increased by the absolute delay time difference Tab with respect to the standard packet accumulation amount.

That is, assuming that the standard packet accumulation amount of the jitter buffer 47 is TBPs (sec), the target packet accumulation amount TBPd (sec) in this case is set to TBPd = TBPs + Tab so that the target packet accumulation amount TBPd is obtained. In addition, the playback speed V is changed to a low playback speed V _L that is slower than the standard playback speed Vs. Note that the low playback speed V _L in this case is set to a speed (eg, 80% or 90% of the standard playback speed Vs) that the user does not feel uncomfortable.

  Thereafter, the packet accumulation amount control unit 56 determines whether or not the predicted packet accumulation amount of the jitter buffer 47 has reached the allowable range of the target packet accumulation amount TBPd based on the monitoring result by the jitter buffer monitoring unit 50. In such a case, the playback speed V is returned to the standard playback speed Vs.

Here, the speed control to the low playback speed V _L by the jitter buffer control unit 51 is executed by, for example, one of the first playback speed control method and the second playback speed control method described below.

(A) First playback speed control method TRs is a packet reading interval from the jitter buffer 47 in the case of the standard playback speed Vs, and TR _L is a packet reading interval from the jitter buffer 47 in the case of the low playback speed V _L. Where TR _L = TRs / k. For example, at the standard playback speed Vs, when a low playback speed V _{L is set} to 80% (k = 0.8) of the standard playback speed Vs in a VoIP application that reads and plays back packets in the jitter buffer 47 at 20 msec intervals. the read interval TR _L packets from the jitter buffer 47, _{approximately, TR L = 20 / 0.8 (} msec), and is.

(B) Second playback speed control method When control to the low playback speed _VL is started, the time stamp of the packet played back immediately after that is recorded in combination with the playback time. Subsequent packets are read from the jitter buffer 47 and reproduced at time Tv shown in the following equation. In the following equation, TD is a delay time, and the initial value is 0.
[Equation 5]
Tv = (packet time stamp-first packet time stamp) + (first packet playback time + TD)

Here, when the packet is read from the jitter buffer 47, the [{Vs / (Vs−V _L )} − 1] -th read packet is copied and stored in the memory in the decoder 48, and the copy source packet After the packet is reproduced, the copied packet is read and reproduced at the next reproduction timing. For example, when the low reproduction speed V _{L is set} to 80% of the standard reproduction speed Vs, the sequential four packets P1 to P4 in the jitter buffer 47 are sequentially read and reproduced as shown in FIG. At the same time, the fourth packet P4 is copied, and the copied packet P4 ′ is reproduced at the next reproduction timing after reproducing the copy source packet P4. Thereafter, when the packet P5 is read from the jitter buffer 47, the TD is increased by the reproduction interval by copying. If the [{Vs / (Vs−V _L )} − 1] -th packet to be read has not arrived or has been discarded and is not in the jitter buffer 47, the packet is read at the next playback timing. On the other hand, the same processing is performed.

FIG. 8 is a flowchart showing an operation when the predicted packet accumulation amount of the jitter buffer 47 is controlled to the target packet accumulation amount by the reproduction at the low reproduction speed V _L. When the reproduction rate is changed to the low reproduction rate V _L , the packet accumulation amount control unit 56 updates the packet reception scheduled timing at every packet transmission interval by the application (every 20 msec in the case of VoIP) and The sequence number is calculated (step S81).

  Specifically, the packet accumulation amount control unit 56 stores received packet information acquired for a certain period from the jitter buffer monitoring unit 50, and for each received packet information for a certain period stored for each packet transmission interval. , (Reception sequence number x packet transmission interval-reception time), the average reception timing TR (Av) that is the average of them is calculated, the current time is added to the average reception timing TR (Av), A reception scheduled sequence number that is a sequence number of a packet that is considered to be received is calculated. That is, the scheduled reception sequence number = TR (Av) + current time is calculated.

  In addition, the packet accumulation amount control unit 56, for each packet transmission interval, based on the reproduction packet information acquired from the jitter buffer monitoring unit 50 at the start of the call (application), the transmission interval by the application, and the current time, A reproduction scheduled sequence number that is a sequence number of a packet that is considered to be currently reproduced is calculated (step S82). That is, the scheduled playback sequence number = (first playback sequence number × packet transmission interval × playback speed−first playback time) + current time is calculated. In this equation, the reproduction speed is “1” as a standard (100%). The “first playback sequence number” and “first playback time” are reset when the playback speed is changed.

Then, the packet accumulation amount control unit 56 is regarded as a predicted packet accumulation amount that is the difference between the scheduled reception sequence number calculated in step S81 and the reproduction scheduled sequence number calculated in step S82, that is, currently accumulated. The calculated packet accumulation amount is calculated, and the predicted packet accumulation amount is compared with the target packet accumulation amount (step S83). As a result, if the predicted packet storage amount does not reach the target packet storage amount, the playback speed remains at the low playback speed _VL , the process proceeds to step S81, and the above operation is repeated. If the amount has been reached, the playback speed is returned to the standard raw speed Vs, and the playback process at the low playback speed V _L is terminated.

FIG. 9 is a diagram for explaining the operation of the low-speed playback process. FIG. 9A shows the reception sequence number of the packet received by the jitter buffer 47, FIG. 9B shows the playback speed V, and FIG. c) shows the packet accumulation amount TBP in the jitter buffer 47, respectively. In FIG. 9, in the period T1 during which the jitter buffer 47 receives packets at almost a predetermined timing, the reproduction speed V is a low reproduction speed V _L that is lower than the standard reproduction speed Vs. Gradually increases. Thereafter, when the retention period T2 occurs, no packet is received in the retention period T2, and therefore the packet accumulation amount TBP gradually decreases.

Here, in the present embodiment, for each transmission interval, the scheduled reception sequence number is calculated based on the received packet information, and the scheduled reproduction sequence number is calculated based on the first reproduced packet information. When a certain predicted packet accumulation amount reaches the target packet accumulation amount, the reproduction speed V is returned from the low reproduction speed V _L to the standard reproduction speed Vs. Therefore, even if the actual packet accumulation amount TBP does not reach the target packet accumulation amount TBPd within the retention period T2, if the predicted packet accumulation amount reaches the target packet accumulation amount TBPd, the playback speed V at that time is the standard reproduction. Since the speed returns to Vs, even when a large number of packets that should originally be received in the residence period T2 are received in a short time when the residence period T2 elapses, unlike the case of FIG. The target packet accumulation amount TBPd is not exceeded. As a result, it is possible to prevent the received packet from being discarded while minimizing the call delay time, and to improve the reproduction quality.

  After the packet accumulation amount change control based on the above-described handover information is completed, the packet accumulation amount control unit 56 does not receive a handover start notification from the handover control unit 36 via the handover information acquisition unit 55 until the jitter buffer monitoring unit Based on the monitoring result of the jitter buffer 47 by 50, the change of the reproduction speed V is controlled so that the predicted packet accumulation amount in the jitter buffer 47 is maintained within the allowable range of the target packet accumulation amount TBPd.

  Therefore, as shown in the flowchart of FIG. 10, the packet accumulation amount control unit 56 is acquired from the jitter buffer monitoring unit 50 in the same manner as in step S81 of FIG. 8 until the handover start notification is acquired (step S101). Based on the received packet information, the scheduled reception sequence number considered to be currently received is calculated (step S102), and the first reproduction obtained from the jitter buffer monitoring unit 50 is performed as in step S82 of FIG. Based on the packet information, a reproduction scheduled sequence number considered to be currently reproduced is calculated (step S103). Then, from the difference between the scheduled reception sequence number and the scheduled reproduction sequence number, a predicted packet storage amount that is considered to be currently stored in the jitter buffer 47 is calculated, and (predicted packet storage amount−target packet storage amount) Is less than an allowable error Tm (Tm <0) representing the lower limit value of the allowable range (step S104).

As a result, if the allowable error Tm is not reached, the playback speed V is returned from the standard playback speed Vs to the low playback speed V _L again, and the predicted packet storage is performed in the same manner as in the packet storage amount change control based on the handover information described above. The amount is controlled to be the target packet accumulation amount (step S105), and the process proceeds to step S101. Here, the allowable error Tm is set to an arbitrary number of packets, for example, between −1 and − (the number of delay time difference packets).

On the other hand, in step S104, when (predicted packet accumulation amount−target packet accumulation amount) is equal to or larger than the allowable error Tm, the packet accumulation amount control unit 56 further allows the above difference to represent the upper limit value of the allowable range. It is determined whether or not the error Tp (Tp> 0) is exceeded (step S106). As a result, if less than the allowable error Tp, the process proceeds to step S101, allowable if beyond the error Tp, and change the playback speed V to a high playback speed V _H higher than the standard reproduction speed Vs, the predicted packet accumulation The amount is controlled to be the target packet accumulation amount (step S107), and the process proceeds to step S101. Here, the allowable error Tp is set to an arbitrary number of packets of 1 or more, for example. Further, the high playback speed V _H is set to a speed (eg, 110% or 120% of the standard playback speed Vs) that the user does not feel uncomfortable.

FIG. 11 is a flowchart showing the operation when the predicted packet accumulation amount of the jitter buffer 47 is controlled to the target packet accumulation amount by reproduction at the high reproduction speed V _H in step S107 of FIG. When the playback speed is changed to the high playback speed V _H , the packet accumulation amount controller 56 receives the received packet information for each transmission interval by the application, as in the case of playback at the low playback speed V _L shown in FIG. The scheduled reception sequence number is calculated based on (step S111), and the scheduled reproduction sequence number is calculated based on the first reproduction packet information (step S112).

Then, the packet accumulation amount control unit 56 calculates a predicted packet accumulation amount that is the difference between the calculated reception scheduled sequence number and the reproduction scheduled sequence number, and compares the predicted packet accumulation amount with the target packet accumulation amount. (Step S113). As a result, if the predicted packet accumulation amount still exceeds the target packet accumulation amount, the reproduction speed remains at the high reproduction speed V _H , the process proceeds to step S111, and the above operation is repeated. if it has reached the target packet accumulated amount, to return the playback speed to normal production speed Vs, and ends the playback processing performed by a high playback speed V _H.

In the reproduction process at the high reproduction speed V _H , for example, as in the case of the reproduction process at the low reproduction speed V _L described above, the reading interval of packets from the jitter buffer 47 at the standard reproduction speed Vs is set to TRs, When the readout interval of packets from the jitter buffer 47 at the high reproduction speed V _H is TR _H , TR _H = TRs / k. However, k is set to k = 1.1 when the high playback speed V _H is 110% of the standard playback speed Vs, and is set to k = 1.2 when 120%. Or, conversely, the read interval of packets from the jitter buffer 47 remains the read interval TRs of the standard reproduction speed Vs, and the reproduction speed V is equivalently set to the high reproduction speed V _H by thinning out the read packets. .

12A and 12B are diagrams for explaining the operation of the high-speed playback process. FIG. 12A shows the reception sequence number of the packet received by the jitter buffer 47, FIG. 12B shows the playback speed V, and FIG. c) shows the packet accumulation amount TBP in the jitter buffer 47, respectively. In FIG. 12, during the period T1 during which the jitter buffer 47 receives packets at almost a predetermined timing, the playback speed V is a high playback speed V _H that is faster than the standard playback speed Vs. Decrease gradually. Thereafter, when the retention period T2 occurs, no packet is received in the retention period T2, so the packet accumulation amount TBP decreases rapidly.

Here, in the present embodiment, for each transmission interval, the scheduled reception sequence number is calculated based on the received packet information, and the scheduled reproduction sequence number is calculated based on the first reproduced packet information. When a certain predicted packet accumulation amount reaches the target packet accumulation amount, the reproduction speed V is returned from the high reproduction speed _VH to the standard reproduction speed Vs. Therefore, even if the actual packet accumulation amount reaches the target packet accumulation amount within the retention period T2, if the predicted packet accumulation amount does not reach the target packet accumulation amount, the reproduction speed V will return to the standard reproduction speed Vs. Absent. Thus, unlike the case of FIG. 20, since the packet accumulation amount TBP does not exceed the target packet accumulation amount TBPd after the reproduction speed V returns to the standard reproduction speed Vs, the call delay time is minimized. The reproduction quality can be improved.

As described above, in this embodiment, after the packet accumulation amount change control to the target packet accumulation amount based on the handover information is finished, the predicted packet accumulation amount of the jitter buffer 47 is calculated until the handover is started. Thus, the playback speed V is controlled so that the predicted packet accumulation amount falls within the allowable range of the target packet accumulation amount TBPd. As a result, for example, as shown in FIGS. 13A and 13B, the predicted packet accumulation amount of the jitter buffer 47 is changed to the target packet by the clock difference between transmission and reception, the packet delay time fluctuation, etc. When the accumulated amount TBPd decreases below the allowable error Tm, the reproduction speed V is changed again to the low reproduction speed V _L , and the predicted packet accumulated amount is maintained within the allowable range of the target packet accumulated amount TBPd. Therefore, unlike the case of FIG. 19, since the packet in the jitter buffer 47 is not emptied between the start and completion of the handover, it is possible to more reliably prevent the occurrence of silence.

  Further, upon receiving notification of handover completion from the handover information acquisition unit 55, the packet accumulation amount control unit 56 calculates the predicted packet accumulation amount of the jitter buffer 47, and compares the predicted packet accumulation amount with the standard packet accumulation amount TBPs. The packet accumulation amount TBP is controlled by changing the reproduction speed V so that the predicted packet accumulation amount is within the allowable error Tm2 of the standard packet accumulation amount TBPs.

  That is, similar to the flowchart shown in FIG. 10, based on the received packet information acquired from the jitter buffer monitoring unit 50, the reception scheduled sequence number considered to be currently received is calculated, and the jitter buffer monitoring unit Based on the first playback packet information acquired from 50, the scheduled playback sequence number that is considered to be currently played back is calculated, and from the difference between them, it is considered that it is currently stored in the jitter buffer 47. The predicted packet accumulation amount is calculated, and it is determined whether or not the difference between the predicted packet accumulation amount and the standard packet accumulation amount TBPs is less than an allowable error Tm2 (Tm2 <0) indicating the lower limit value of the allowable range.

As a result, when the error is lower than the allowable error Tm2, the playback speed V is changed to the low playback speed _VL so that the predicted packet storage amount becomes the standard packet storage amount TBPs. Here, the allowable error Tm2 is set to an arbitrary number of packets, for example, between −1 and − (standard number of stored packets).

On the other hand, when the difference between the predicted packet accumulation amount and the standard packet accumulation amount TBPs exceeds the allowable error Tm2, the packet accumulation amount control unit 56 further determines the allowable error Tp2 in which the difference between the two indicates the upper limit value of the allowable range. (Tp2> 0) to determine whether more than, if exceeding the allowable error Tp2, and change the playback speed V to a high playback speed V _H, so that the predicted packet accumulated amount becomes the standard packet accumulated amount TBPs To control. Here, the allowable error Tp2 is set to an arbitrary number of packets of 1 or more, for example.

  As a result, the estimated packet storage amount of the jitter buffer 47 is maintained within the allowable range of the standard packet storage amount TBPs without being affected by jitter, a clock difference between transmission and reception, or a packet delay time fluctuation, etc. even at the handover destination. Since (adjustment) can be performed, reproduction quality and real-time performance can be further improved. The control of the packet accumulation amount after the handover is completed may be performed until the handover schedule is determined in the communication by each wireless communication network, or the control is terminated here and the normal playback control is restored. Good.

  In addition, this invention is not limited only to the said embodiment, Many deformation | transformation or a change is possible. For example, the present invention is not limited to the case of executing a VoIP application, but can be effectively applied to a case of executing a real-time communication system application such as streaming reproduction of multimedia data such as video and music. In this case, the application execution unit may be configured by a multimedia function unit having a similar jitter buffer control function instead of the telephone function unit. The present invention can also be effectively applied when handing over to a network with a short absolute delay time. In this case, the reproduction speed is controlled to be higher than the standard speed so that the predicted packet accumulation amount becomes a target packet accumulation amount that does not cause packet discard in the jitter buffer at the time of handover by determining the start of handover. Thereafter, control is performed in the same manner as in the above embodiment, and the predicted packet accumulation amount is maintained within the allowable range of the target packet accumulation amount. Further, the first wireless communication network and the second wireless communication network are not limited to the combination of the wireless LAN and cdma2000, but are optional in the 3rd generation (3G), the 3rd generation (3.9G), or the IMT-ADVANCED. It should be noted that the present invention can be applied to a combination of different wireless communication networks.

It is a figure which shows schematic structure of the communication network which can use the radio | wireless communication apparatus which concerns on one embodiment of this invention. It is a block diagram which shows schematic structure of the radio | wireless communication apparatus shown in FIG. It is a functional block diagram which shows schematic structure of the telephone function part shown in FIG. FIG. 3 is a diagram for explaining a method for calculating a handover preparation time by the handover control unit shown in FIG. 2. FIG. 3 is a diagram for explaining an example of an absolute delay time acquisition method by a handover control unit shown in FIG. 2. It is a sequence diagram which shows operation | movement of the principal part of the telephone function part shown in FIG. It is a figure for demonstrating an example of the low reproduction speed control of the received packet by the jitter buffer control part shown in FIG. 4 is a flowchart for explaining target packet accumulation amount control at a low reproduction speed by the packet accumulation amount control unit shown in FIG. 3. It is a figure for demonstrating the low-speed reproduction | regeneration processing of the jitter buffer by the flowchart of FIG. 4 is a flowchart for explaining packet accumulation amount control from the end of packet accumulation amount change control to the start of handover based on handover information by the packet accumulation amount control unit shown in FIG. 3. 11 is a flowchart for explaining target packet accumulation amount control at a high reproduction speed in the flowchart of FIG. 10. It is a figure for demonstrating the high-speed reproduction process of the jitter buffer by the flowchart of FIG. It is a figure for demonstrating packet accumulation amount control of the jitter buffer by the flowchart of FIG. It is a figure for demonstrating an example of the control method of the conventional jitter buffer. It is a figure which shows the flow of the packet by the control method shown in FIG. It is a figure for demonstrating the other example of the control method of the conventional jitter buffer. It is a figure which shows the flow of the packet by the control method shown in FIG. It is a figure for demonstrating the low-speed reproduction | regeneration process based on the actual packet accumulation amount of a jitter buffer. It is a figure for demonstrating the malfunction at the time of the assumed handover. It is a figure for demonstrating the high-speed reproduction | regeneration processing based on the actual packet accumulation amount of a jitter buffer.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Wireless communication apparatus 12 Counterparty communication terminal 12a Handset 15 1st wireless communication network 15a Access point 16 2nd wireless communication network 16a Base station 17 Internet 21, 22, 24 SIP server 23 Home agent (HA)
31 First wireless I / F
32 Second wireless I / F
33 Telephone Function Unit 34 Communication Processing Unit 35 Radio Information Acquisition Unit 36 Handover Control Unit 47 Jitter Buffer 50 Jitter Buffer Monitoring Unit 51 Jitter Buffer Control Unit 54 Radio Status Monitoring Unit 55 Handover Information Acquisition Unit 56 Packet Accumulation Amount Control Unit 60 Core Network 61 Measurement server 62 First information server 63 Second information server 65 Provider

Claims (5)

A wireless communication unit that performs wireless communication by connecting to a first wireless communication network and a second wireless communication network different from the first wireless communication network;
An execution unit that executes an application of a real-time communication system via the wireless communication unit;
A communication quality acquisition unit for acquiring communication quality of a radio link in the first wireless communication network while executing the application by connecting to the first wireless communication network;
A determination unit that determines whether to start preparation for handover from the first wireless communication network to the second wireless communication network based on the communication quality acquired by the communication quality acquisition unit;
When the determination unit determines the start of handover preparation during execution of the application, an estimation unit that estimates a handover preparation time until the handover is started based on the communication quality acquired by the communication quality acquisition unit;
When the determination unit determines the start of handover preparation, the measurement unit measures each delay time in the first wireless communication network and the second wireless communication network,
The execution unit is
A jitter buffer for accumulating received packets;
A jitter buffer monitoring unit that monitors received packet information including a sequence number and a reception time of a packet stored in the jitter buffer, and monitors reproduction packet information including a sequence number and a reproduction time of the packet read from the jitter buffer; ,
A jitter buffer control unit for controlling the playback speed of packets from the jitter buffer;
When the determination unit determines the start of handover preparation, the predicted packet storage amount of the jitter buffer is calculated based on the received packet information and the reproduction packet information by the jitter buffer monitoring unit, and the predicted packet storage amount is Jitter buffer storage for controlling the reproduction speed by the jitter buffer control unit so as to be a target packet storage amount based on a delay time in the first wireless communication network and the second wireless communication network measured by the measurement unit A wireless communication apparatus comprising: a quantity control unit; The jitter buffer accumulation amount control unit
For each packet transmission interval by the application, based on the received packet information for a certain period and the current time, calculate a scheduled reception sequence number of a packet that is considered to be currently received, and at the start of the application Based on the reproduction packet information and the current time, a reproduction scheduled sequence number of a packet considered to be currently reproduced is calculated, and the predicted packet is calculated from a difference between the reception scheduled sequence number and the reproduction scheduled sequence number. Calculate the accumulated amount,
The wireless communication apparatus according to claim 1. The jitter buffer accumulation amount control unit
Controlling the reproduction speed by the jitter buffer control unit so that the predicted packet accumulation amount becomes the target packet accumulation amount until the handover preparation time estimated by the estimation unit elapses.
The wireless communication apparatus according to claim 1, wherein the wireless communication apparatus is a wireless communication apparatus. The jitter buffer accumulation amount control unit
When the handover from the first wireless communication network to the second wireless communication network is completed, the playback speed by the jitter buffer control unit is controlled so that the predicted packet storage amount becomes the standard packet storage amount by the application. ,
The wireless communication device according to claim 1, wherein the wireless communication device is a wireless communication device. The jitter buffer accumulation amount control unit
The delay time in the first wireless communication network is compared with the delay time in the second wireless communication network, and the delay time in the second wireless communication network is a predetermined time than the delay time in the first wireless communication network. If it is longer, the playback speed by the jitter buffer control unit is slowed down.
The wireless communication device according to claim 1, wherein the wireless communication device is a wireless communication device.

Images