JP2009171529A - Receiver and communication data reordering processing method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To attain reduction of mount scale by reducing the number of timers for monitoring queuing of data that have not yet arrived, concerning an receiver and a communication data reordering processing method. <P>SOLUTION: Based on sequence numbers (TSNs) of received communication data, a TSN missing occurrence determining unit 1-1 determines the occurrence of missing in the sequence numbers; and when the occurrence of missing is determined, a timer 1-51 is started. Each time the occurrence of missing in the sequence numbers is newly determined for communication data received continuously, the timer 1-51 is reset and restarted, and a sequence number of data arriving in the occurrence of missing is stored in a monitor target TSN storage unit 1-52. When unreached data of which the sequence number is younger than the sequence number stored in the storage unit 1-51 do not arrive and the timer 1-51 becomes a timeout, an output control unit 1-3 gives up queuing of the unreached data and reads out the communication data stored in a reordering buffer 1-2, in the order of sequence numbers. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a receiving apparatus and a communication data reordering processing method, and in particular, waits for retransmission data to be retransmitted when radio data is not normally received in mobile communication, and originally transmits communication data including the retransmission data. The present invention relates to a reordering process in which data are rearranged in order, and more specifically, relates to a technique for determining by timer monitoring whether or not to give up waiting for retransmission data in the reordering process.

  Mobile communication devices are required to support multimedia communication, Internet communication, and the like, and high-speed and high-quality communication is required. A hybrid automatic repeat request (HARQ) method is generally employed as a means for compensating for a failure to receive wireless data due to a change in wireless transmission path or temporary deterioration and improving communication quality. The hybrid automatic retransmission (HARQ) method is defined by 3GPP (Third Generation Partnership Project), and performs retransmission control on the assumption that decoding failure data is not discarded on the receiving side and is decoded in combination with retransmission data. .

  When the hybrid automatic retransmission (HARQ) method is used, the order of transmission of communication data on the transmission side and the order of reception of communication data on the reception side are switched (the order is switched by performing retransmission). It is necessary to rearrange in the order of transmission on the side, and it is essential to implement a reordering processing function for performing this rearrangement.

  FIG. 5 shows a functional part of a general reordering process and its outline. The functional unit of the reordering process includes a sequence number (TSN) determination unit 5-1, a reordering buffer 5-2, and an output control unit 5-3. The sequence number (TSN) determination unit 5-1 determines the sequence number (TSN) of the communication data that has arrived, and notifies the reordering buffer 5-2 of the sequence number (TSN).

  The reordering buffer 5-2 is provided with a storage area for communication data for each sequence number (TSN), and determines a storage area according to the sequence number (TSN) notified from the sequence number (TSN) determination unit 5-1. Communication data is stored in the storage area. The output control unit 5-3 reads and outputs the communication data stored in the reordering buffer 5-2 in ascending order of the sequence number (TSN).

  In the example of FIG. 3, originally, communication data should arrive in the order of sequence numbers (TSN) 0, 1, 2, 3, 4, and 5, but sequence numbers (TSN) 0, 3, 5, and 5 are controlled by reproduction control. An example in which communication data arrives in the order of 1, 4 and 2 is shown.

  In the reordering buffer 5-2, the communication data that has arrived is stored in the storage area corresponding to the sequence number (TSN), the communication data of sequence number (TSN) 0 is stored in the storage area of TSN 0, and the sequence number (TSN) 1 Communication data of sequence number (TSN) 2 in the storage area of TSN 2, communication data of sequence number (TSN) 3 in the storage area of TSN 3, and communication of sequence number (TSN) 4 Data is stored in the storage area of TSN4, and communication data of sequence number (TSN) 5 is stored in the storage area of TSN5.

  FIG. 6 shows the format of communication data. The communication data is divided into a header area 6-1 and a user data area 6-2, and the header area 6-1 includes a transmission sequence number (TSN) value. On the transmission side, communication data is transmitted with a sequence number (TSN) incremented at the time of data transmission. On the receiving side, communication data is output in the same order as the transmission order by rearranging the arrived communication data based on this sequence number (TSN).

  However, in the reordering process, when rearranging the arrival communication data based on the sequence number (TSN), the retransmission of communication data by hybrid automatic retransmission (HARQ) control may fail. Note that the number (TSN) is missing.

  When the communication data of the sequence number (TSN) in the order (expected value) to be output has not arrived, the communication data is waited. However, if certain conditions described below are satisfied, reception of unarrived communication data is abandoned and waiting is stopped, and the sequence number (TSN) of unarrived communication data is left as a missing number. Communication data of sequence number (TSN) is output.

  In order to determine the occurrence of the above-described missing numbers, a determination function based on window monitoring and a determination function based on timer monitoring are generally implemented. In the determination by window monitoring, the sequence number (TSN) of the arrived communication data is compared with the missing sequence number (TSN) in the section (window size) for monitoring the unarrived communication data, and the difference is constant. If it is more than the value, the communication data of the missing sequence number (TSN) is determined to be a failure of hybrid automatic retransmission (HARQ) control, and waiting for retransmission data is given up and the next processing is started.

  In the determination by the timer monitoring, when the communication data of the sequence number (TSN) determined to be missing does not arrive within a certain time, it is determined that the hybrid automatic retransmission (HARQ) control has failed, and the waiting for retransmission data reception is given up. Move on to the next process. In the determination by the timer monitoring, if the sequence number (TSN) of the arrived communication data is different from the expected sequence number (TSN), it is determined that a missing number has occurred, and the missing sequence number (TSN) is determined. The communication apparatus measures the non-arrival time of the communication data and waits for the arrival of the communication data by retransmission until the non-arrival time reaches a certain value.

  In that case, since it is expected that the sequence number (TSN) of the arrived communication data will jump (that is, a plurality of missing numbers occur), it is necessary to mount a plurality of timers for determining missing numbers. In other words, for the occurrence of missing numbers by timer monitoring, prepare timers for the number of communication data (window size) for the window width used for window monitoring, and for all sequence numbers (TSN) that may be missing Each is monitored by a timer. This is specified by 3GPP TS 25.321.

  7 to 9 show specific operation examples of the reordering process. On the receiving side, the sequence number (TSN) is checked for the communication data that has arrived. If the sequence number (TSN) of the arrived communication data matches the expected sequence number (TSN) (that is, the value predicted to arrive next), the arrival communication data is transferred to the subsequent stage without waiting. Output.

  If the sequence number (TSN) of the arrival communication data is different from the expected sequence number (TSN), the sequence number (TSN) of the arrival communication data is in the order (slightly higher) while the arrival communication data is stored in the reordering buffer. Wait for the missing communication data until they are arranged without any missing numbers.

  In the example shown in FIG. 7, the communication data with the sequence number (TSN) 0 expected at the beginning arrives, and the communication data is output without waiting. Next, the communication data of sequence number (TSN) 1 is expected. However, since the sequence number (TSN) of the arrived communication data is 3, the communication data of the sequence number (TSN) 3 is temporarily stored in the reordering buffer in order to rearrange the order.

  Here, since the sequence number (TSN) of the communication data that has arrived next is 5, which is different from the expected value 1, the communication data is similarly stored in the reordering buffer. The sequence number (TSN) of the communication data that has arrived next is 1, which matches the expected value 1. Therefore, the arrival communication data is output, and the sequence number (TSN) of the next expected communication data is set to 2.

  In this state, communication data of sequence number (TSN) 3 and sequence number (TSN) 5 is stored in the reordering buffer, but since it is different from the expected sequence number (TSN) 2, no output is performed. Next, the communication data of sequence number (TSN) 4 arrives, but since it is different from the expected sequence number (TSN) 2, it is also stored in the reordering buffer and the communication data of sequence number (TSN) 2 is stored. Make a meeting.

  Finally, when the communication data of sequence number (TSN) 2 arrives, this is the expected sequence number (TSN). Therefore, the communication data is output, and the sequence number (TSN) of the next expected communication data is output. 3. Here, communication data of sequence numbers (TSN) 3 to 5 are stored in the reordering buffer. Since the communication data of sequence number (TSN) 3 stored in the reordering buffer matches the expected value 3, it is output from the reordering buffer. Through similar processing, communication data of sequence numbers (TSN) 4 and 5 are output, and the reordering processing is stopped when the reordering buffer becomes empty.

  FIG. 8 shows an example of operation for determining occurrence of missing sequence numbers (TSN) by conventional timer monitoring. In the operation example of FIG. 8, it is assumed that communication data arrives in the same order as in FIG. In timer monitoring, when a missing sequence number (TSN) occurs and arrival communication data is stored in the reordering buffer, measurement of a waiting time is started for each missing communication data.

  In the operation example of FIG. 8, first, it is determined that a sequence number (TSN) is missing when communication data with a sequence number (TSN) 3 arrives. At that time, the target of the timer monitoring is communication data from the sequence number (TSN) 1 of the expected value to the sequence number (TSN) 2 immediately preceding the sequence number (TSN) 3 of the arrival communication data. .

  Therefore, the timers (TSN1 timer and TSN2 timer) are activated for the sequence numbers (TSN) 1 and 2 that are the targets of timer monitoring (8-1, 8-2). These timers continuously count up the clocks until the corresponding communication data is output. Next, when the communication data of the sequence number (TSN) 5 arrives, the waiting target (expected value) is from the sequence number (TSN) 1 to the sequence number (TSN) 4 immediately before the sequence number (TSN) 5. Become.

  However, since the timers with sequence numbers (TSN) 1 and 2 are already activated, these timer activations are excluded. In addition, since the communication data of the sequence number (TSN) 3 has already arrived, this timer activation is also excluded. Therefore, the timer activated at this time is the timer of the sequence number (TSN) 4 (TSN4 timer), and activates the timer (TSN4 timer) (8-3). The start control of the timer as described above is performed, the waiting time of each unarrived communication data is measured, and when this waiting time exceeds a predetermined value and a timeout occurs, the waiting for the unarrived communication data is given up. Thus, only the communication data that has arrived is output in the sequence number (TSN) order.

  Explaining the flow of the timer start / stop processing for each time 1 to 12 in FIG. 8, at time 1, the expected value is the sequence number (TSN) 0 and the communication data with the sequence number (TSN) 0 has arrived. So the timer is not started. Since no communication data arrives at time 2 and time 3, there is no change.

  At time 4, since the expected value is the sequence number (TSN) 1 and the communication data of the sequence number (TSN) 3 arrives, the sequence number (TSN) 1 and sequence number (TSN) 2 timers are started. At time 5, the activated timer continues counting. Since the communication data of sequence number (TSN) 5 has arrived at time 6, the timers of sequence number (TSN) 1 and sequence number (TSN) 2 continue counting, and the communication data of sequence number (TSN) 3 has already arrived. Therefore, the timer of the sequence number (TSN) 4 is started because it is not subject to timer monitoring and the communication data of the sequence number (TSN) 4 has not arrived.

  At time 7, since communication data with sequence number (TSN) 1 arrives, the timer with sequence number (TSN) 1 stops counting, and the other active timers continue counting. At time 8 and 9, the activated timer continues to count, and at time 10, the arrival of communication data of sequence number (TSN) 4 causes the timer of sequence number (TSN) 4 to stop counting and other activated The timer continues counting. At time 11, the activated timer continues counting, and at time 12, the sequence number (TSN) 2 timer stops counting because the communication data of sequence number (TSN) 2 arrives.

  FIG. 9 shows an operation example in the case of timeout in conventional timer monitoring. In the operation example of FIG. 9, the arrival of communication data is the same as in the operation example shown in FIG. 8, and FIG. 9 shows an example in which the timer of sequence number (TSN) 2 has timed out. In the operation example of FIG. 9, after the communication data of sequence number (TSN) 0 is transmitted as in the case of FIG. 8, the arrival of communication data of sequence number (TSN) 1 is expected.

  Thereafter, when communication data of sequence number (TSN) 3, 5, 1 arrives, communication data of sequence number (TSN) 1 is output, and communication data of sequence number (TSN) 2 is expected to arrive. If the communication data of sequence number (TSN) 4 arrives and the timer of sequence number (TSN) 2 times out (9-1) at that time, the communication data of sequence number (TSN) 2 is given up. The expected value is changed to the next sequence number (TSN) 3 and the communication data of the sequence number (TSN) 3 is output.

  The communication data of the next sequence number (TSN) 4 arrives at the same time as the timeout (7-1) of the sequence number (TSN) 2, and the communication data of the sequence number (TSN) 5 has already arrived and is reordered. Since it is stored in the buffer, communication data of sequence numbers (TSN) 3 to 5 is output when the timer of sequence number (TSN) 2 times out. Note that the communication data of the sequence number (TSN) 2 that arrives late after the timeout is discarded (out of window monitoring).

The reordering processing of communication data and the waiting control of communication data of the missing sequence number (TSN) in which hybrid automatic retransmission (HARQ) control is performed in the above HSDPA (High Speed Downlink Packet Access) communication data transmission method are described below. Patent Document 1 and the like.
JP 2006-33019 A (paragraphs 0008 to 0010 etc.)

  As described above, in order to perform the timer monitoring for waiting for the communication data of the missing sequence number (TSN), it is necessary to mount a counter for measuring the waiting time for each unarrived communication data. Here, assuming that the number of communication data (window size) of the window width for monitoring unarrived communication data is W and the upper limit of the time for waiting for each unarrived communication data is T, the window width is obtained as the total amount of monitoring time. A counter capable of measuring the product of the number W of communication data and the upper limit T of the waiting time, that is, the amount of time W × T, is required.

Usually, since the waiting time is monitored based on the period of TTI (Transmission Time Interval), for example, the number of communication data (window size) W of the window width is 256, and the waiting upper limit time T is When the TTI cycle is 1 ms for 1 second, a counter for measuring quantities up to 256 × 1 ÷ (1 × 10 ⁻³ ) = 256,000 is required. Providing such a counter that counts up to a large number of count values causes a problem that the mounting scale of the apparatus increases.

  In addition, when performing timer control to give up waiting for each unarrived communication data, the number of sequences in the sequence number (TSN) of arrival communication data becomes enormous, and the number of patterns in the sequence array Accordingly, many variations occur in the timer control, and the timer control becomes complicated.

  It is an object of the present invention to reduce the number of timers for monitoring waiting for unarrived communication data to reduce the implementation scale, and to simplify timer control for waiting monitoring for a plurality of unarrived communication data. .

  The communication data reordering process in the receiving apparatus includes a sequence number missing occurrence determination unit that determines occurrence of a missing sequence number of the received communication data based on a sequence number of the received communication data, and the sequence number missing occurrence determination unit. The timer is started when it is determined that the occurrence of a missing sequence number is performed, and whenever the occurrence of a missing sequence number is newly determined for communication data that is subsequently received, the timer is reset and restarted. A waiting time measuring unit that stores the sequence number of the communication data that triggered each start of the timer as a monitoring target sequence number, and a sequence number that is a lower number than the monitoring target sequence number before the timer times out. When all communication data arrives, the reordering buffer If the timer times out without arrival of non-arrival communication data with a sequence number lower than the monitored sequence number, the communication data stored in the An output control unit that gives up waiting for data and reads and outputs the communication data stored in the reordering buffer in the order of the sequence number.

  In the section where unarrived communication data is monitored, there is a time when the arrived communication data is stored in the reordering buffer and a time when the unarrived communication data is awaited. The time waiting for unarrived communication data is the time that was not scheduled on the transmission side, or the time that communication data did not arrive due to the discard of communication data in the hybrid automatic retransmission (HARQ) process. is there.

  As a basic method for determining whether or not to wait for unarrived data, the conventional method of monitoring the waiting time of unarrived communication data by setting the monitoring target as the sequence number (TSN) of unarrived data and the communication data that has already arrived There is a method of monitoring the time when communication data that has already arrived is stored in a reordering buffer with the sequence number (TSN) as a monitoring target.

  In the conventional method, the monitoring target is the sequence number (TSN) of the unarrived data, and the waiting time is measured until the estimated time when the unarrived communication data is estimated to arrive. In the communication data reordering process in the receiving apparatus, the monitoring target is the arrival data sequence number (TSN), the time when the incoming communication data is reordered and buffered is measured, and the waiting is given up based on the measured time. It prompts the output of arrival communication data.

  The communication data reordering process in the disclosed receiving apparatus can be implemented with only one timer without mounting a timer for the window size as in the prior art. Reduction). In addition, since the start control of the timer for waiting monitoring is performed only by occurrence of a missing sequence number (TSN), it can be easily performed. In addition, the accuracy of the determination to give up waiting for the missing sequence number (TSN) is slightly lower than that of the prior art, and it is possible to determine whether to give up waiting for the missing sequence number (TSN) without any practical problem.

  In addition, by implementing timers for the number of occurrences of discontinuous missing sequence numbers (TSN) that are expected in accordance with the usage environment, the scale of implementation is greatly reduced compared to the conventional one, and is almost the same as the conventional one. With this accuracy, it is possible to determine whether to give up waiting for the missing sequence number (TSN), and it is possible to easily improve performance with high expandability.

  FIG. 1 shows an example of the configuration of a reordering processing function unit according to the first embodiment of the present invention. The reordering processing function unit according to the present invention includes a sequence number (TSN) determination unit 1-1, a reordering buffer 1-2, an output control unit 1-3, a sequence number (TSN) loss occurrence determination unit 1-4, and a waiting time. A measurement unit 1-5 is provided.

  The sequence number (TSN) determination unit 1-1 determines the sequence number (TSN) of received (arrived) communication data, and the sequence number (TSN) is missing from the reordering buffer 1-2 and the sequence number (TSN). Notify the determination unit 1-4. The reordering buffer 1-2 is provided with a communication data storage area for each sequence number (TSN), and determines a storage area according to the sequence number (TSN) notified from the sequence number (TSN) determination unit 1-1. The communication data that has arrived in the storage area is stored.

  The output control unit 1-3 reads and outputs the communication data stored in the reordering buffer 1-2 in order of the sequence number (TSN) in accordance with a control signal from a waiting time measurement unit 1-5 described later. .

  The sequence number (TSN) loss occurrence determination unit 1-4 calculates the difference between the sequence number (TSN) of the arrival communication data notified from the sequence number (TSN) determination unit 1-1 and the sequence number (TSN) of the expected value. Based on this, it is determined whether or not the sequence number is missing, and when the loss occurs, the sequence number (TSN) information of the arrival communication data is notified to the waiting time measuring unit 1-5 together with the message indicating the occurrence of the loss.

  The waiting time measurement unit 1-5 includes one timer 1-51 and one monitoring target sequence number (TSN) storage unit 1-52 in the first embodiment, and is missing from the sequence number missing occurrence determination unit 1-4. When the occurrence message is notified, the timer 1-51 is started, and the sequence number (TSN) of the arrival communication data notified from the sequence number missing occurrence determination unit 1-4 is used as the monitoring target sequence number. The number (TSN) storage unit 1-52 stores the number.

  The occurrence of a missing sequence number in newly received communication data is newly determined by the sequence number missing occurrence determination unit 1-4 and is notified to the waiting time measuring unit 1-5 each time the timer 1-51 is notified. Is reset and restarted, and the sequence number of the communication data notified at that time is stored as a new monitoring target sequence number in the monitoring target sequence number (TSN) storage unit 1-52.

  That is, the waiting time from the time of the last missing occurrence determination is measured by one timer 1-51, and the sequence number (TSN) of the arrival communication data at the time of the last missing occurrence determination is stored as a monitoring target sequence number (TSN) storage unit 1-52. And the monitoring target sequence number memorize | stored in the monitoring target sequence number (TSN) memory | storage part 1-52 is notified to the output control part 1-3 with the information of the waiting time by the timer 1-51.

  The output control unit 1-3, when all communication data having a sequence number lower than the monitoring target sequence number arrives and is stored in the reordering buffer 1-2 before the waiting time by the timer 1-51 times out. The communication data stored in the reordering buffer 1-2 is read and output in the order of sequence numbers.

  Further, the output control unit 1-3, when the non-arrival communication data having a sequence number lower than the monitoring target sequence number does not arrive and the waiting time by the timer 1-51 times out, Waiting for the waiting, the communication data stored in the reordering buffer 1-2 is read and output in the order of the sequence number.

  An operation example of the reordering process (timer monitoring) of the first embodiment is shown in FIG. In the operation example of FIG. 2, the arrival order of communication data is assumed to arrive in the order of sequence numbers (TSN) 0, 3, 5, 1, 4, 2 as in the operation example shown in FIG. In FIG. 2, similarly to the case of FIG. 9, after arrival communication data of sequence number (TSN) 0 is output, arrival of communication data of the next sequence number (TSN) 1 is expected.

  Thereafter, one timer 1-51 is started when communication data of sequence number (TSN) 3 arrives. The timer 1-51 is not a timer for the communication data of the sequence numbers (TSN) 1 and 2, but is a timer for monitoring the arrived sequence number (TSN) 3.

  However, this timer 1-51 is not a timer dedicated to the sequence number (TSN) 3, but a single timer common to each sequence number (TSN), and this timer 1-51 is activated to wait. Recognizing that the sequence number (TSN) of the communication data is less than 3, 3 is stored in the monitoring target sequence number (TSN) storage unit 1-52 as the monitoring target sequence number (TSN).

  Next, communication data of sequence number (TSN) 5 arrives. At this time, since the communication data of the sequence number (TSN) subsequent to the sequence number (TSN) 3 that triggered the previous timer arrival has arrived, the activated timer 1-51 is temporarily reset, and the timer monitoring target Is started again as a sequence number (TSN) 5 and wait time measurement is newly started, and it is recognized that the sequence number (TSN) to be waited for by starting the timer is less than 5, and the sequence to be monitored As a number (TSN), 5 is stored in the monitoring target sequence number (TSN) storage unit 1-52.

  Next, when the communication data with the sequence number (TSN) 1 arrives, the communication data with the sequence number (TSN) 0 has already been output and the expected value of the next sequence number (TSN) is 1. The communication data of (TSN) 1 is output and the expected value is set to 2 below, but since the communication data output of the sequence number (TSN) 5 being monitored is not performed, the time by the timer 1-51 Continue measurement.

  Next, the communication data of the sequence number (TSN) 4 arrives, but this sequence number (TSN) 4 does not match the expected value 2, and the communication data output of the sequence number (TSN) 5 being monitored is output. Therefore, the time measurement by the timer 1-51 is continued as it is.

  If the waiting time by the timer 1-51 times out before the communication data of the expected value sequence number (TSN) 2 arrives, an older sequence number (TSN) than the expected value sequence number (TSN) 2 is assumed. ), All communication data stored in the reordering buffer 1-2 are output.

  Here, even if there is a loss other than the sequence number (TSN) 2 between the sequence number (TSN) 5 and the sequence number (TSN) 5 communication data has been output, the sequence number has been reached. Similar to (TSN) 2, the communication data of other missing sequence numbers (TSN) is given up and output.

  As is clear from the above operation example, when the sequence number (TSN) 3 arrives, the present invention recognizes the lack of the sequence number (TSN) 2 and starts timer monitoring. The timer is reset when communication data of sequence number (TSN) 5 arrives, and counting is started again. This is for monitoring including the sequence number (TSN) 4 together with the sequence number (TSN) 2.

  Here, as can be seen by comparing FIG. 9 and FIG. 2, the timing for giving up communication data of sequence number (TSN) 2 is different, and the timing for giving up communication data of sequence number (TSN) 2 is delayed in FIG. . Originally, when the communication data with the sequence number (TSN) 3 arrives, the time that the communication data with the sequence number (TSN) 2 may arrive should be estimated, and the arrival of the communication data should be monitored until that time. However, in the present invention, when the communication data with the sequence number (TSN) 5 arrives, the time measurement up to the estimated time is reset and the time measurement is restarted from the beginning. Even if the monitoring performed with the sequence number (TSN) 3 is originally performed with the sequence number (TSN) 5, there is no problem for the following reason.

The reason is,
(1) Since the order of communication data changes due to retransmission by hybrid automatic retransmission (HARQ) control,
(2) Since the judgment function by window monitoring works,
It is.

  The reason for the above (1) will be described. The monitoring time T by the timer is a value that allows for the time at which communication data arrives delayed by hybrid automatic retransmission (HARQ) control. In the operation example of FIG. 9, when the communication data with the sequence number (TSN) 3 arrives, time monitoring of the arrival of the communication data with the sequence numbers (TSN) 1 and 2 is started.

  However, originally, the communication data of sequence numbers (TSN) 1 and 2 should have arrived before sequence number (TSN) 3, and when the sequence number (TSN) 3 arrives, the sequence number (TSN) By starting the time monitoring of the arrival of the communication data 1 and 2, it is originally lower than the original accuracy. Therefore, even if the communication data arrival of the sequence numbers (TSN) 1 and 2 is performed at the time of arrival of the sequence number (TSN) 5, the accuracy is slightly lowered and does not cause a large error.

  Next, the reason (2) will be described. In the reordering process, in addition to the timer monitoring, there is a function of giving up waiting by determining occurrence of missing numbers by window monitoring. This determination function based on window monitoring uses the difference between the sequence number (TSN) of the expected value and the sequence number (TSN) of the arrival communication data, and when the difference is a value that is more than a fixed value, The communication data arrival of the sequence number (TSN) of the value is given up.

  In a normal state in which communication data is continuously communicated, the judgment based on window monitoring functions effectively. This is because the asynchronous hybrid automatic retransmission (HARQ) control makes it very difficult to estimate the retransmission time, resulting in an ambiguous estimated time, and the timer monitoring is determined based on this ambiguous estimated time. A large estimated time with a margin more than the actual retransmission time by HARQ) control is used. Therefore, in a state where communication data of continuous sequence numbers (TSN) are continuously communicated, the occurrence of missing numbers is determined earlier in window monitoring than in timer monitoring, and the determination by window monitoring becomes effective.

  The generation of missing numbers by timer monitoring, that is, the determination of giving up for waiting is valid when communication data of consecutive sequence numbers (TSN) is completed or when communication data is communicated in bursts. Only in such a case, the praise determination of waiting by timer monitoring is used as an auxiliary, and therefore there is almost no influence even if the accuracy of time measurement in timer monitoring is somewhat reduced.

  In this way, only one timer can be implemented without implementing a timer for the window size as in the prior art, so the implementation scale can be reduced, and waiting for the missing sequence number (TSN) can be reduced. Since the accuracy of the compliment determination is only slightly reduced, it is possible to determine whether to give up a waiting sequence number (TSN) without any practical problem.

  Next, an embodiment in which a plurality of timers are used to improve the accuracy of the determination time for waiting and giving up will be described. In the first embodiment shown in FIG. 1, a single timer is used to determine whether to give up or not. However, as in the second embodiment shown in FIG. The accuracy of the compliment determination time can be improved.

  FIG. 3 shows a configuration example in which two monitoring timers are mounted. In FIG. 3, a sequence number (TSN) determination unit 1-1, a reordering buffer 1-2, and a sequence number (TSN) loss occurrence determination unit 1-4 have the same functions as those described in FIG. The difference from the first embodiment is that in the waiting time measuring unit 3-1, a plurality of timers 3-11 and 3-13 and a plurality of monitored sequence number (TSN) storage units 3-32 corresponding to each timer, 3-34.

  The waiting time measuring unit 3-1 starts different timers 3-11 and 3-13 each time a sequence number (TSN) missing occurrence determination unit 1-4 is notified of the occurrence of a missing sequence number. The sequence numbers of the communication data that triggered the start of the timers 3-11 and 3-13 are stored as monitoring target sequence numbers in the monitoring target sequence number (TSN) storage units 3-32 and 3-34, respectively.

  Before the timers 3-11 and 3-13 time out, the output control unit 3-2 arrives at the reordering buffer 1 when all communication data having a sequence number lower than the monitoring target sequence number related to the timer arrives. -2 is stored in the order of the sequence numbers from the reordering buffer 1-2 in order of the sequence numbers including the monitoring target sequence number.

  When each timer 3-11, 3-13 times out without arrival of non-arrival communication data having a sequence number lower than the monitoring target sequence number related to the timer, the timers 3-11 and 3-13 wait for the non-arrival communication data. Give up and read out the communication data of the lower sequence number including the monitoring target sequence number related to the timer from the reordering buffer 1-2 in the order of the sequence number.

  FIG. 4 shows an operation example of the reordering process (timer monitoring) of the second embodiment. Also in the operation example of FIG. 4, it is assumed that communication data arrives in the order of sequence numbers (TSN) 0, 3, 5, 1, 4, 2 as in the case of the operation example of FIG. In the operation example of FIG. 4, the operation is the same as that of the embodiment of FIG. 2 until the communication data of sequence number (TSN) 3 arrives and the first timer (# 1) is started. Here, the first timer (# 1) and the second timer (# 1) are one and the other of the timers 3-11 and 3-13, respectively.

  Next, when the communication data with the sequence number (TSN) 5 arrives, the sequence number (TSN) 3 to be monitored by the activated first timer (# 1) is compared with the sequence number (TSN) 5 of the arrival communication data. In contrast, it is determined that the missing sequence numbers (TSN) 1 and 2 previously recognized and the new discontinuous sequence number (TSN) 4 are missing, and the second timer (# 2) is started.

  After the communication data of the expected value sequence number (TSN) 1 arrives and is output, before the communication data of the next expected value sequence number (TSN) 2 arrives, the sequence number (TSN) 3 is set. When the first timer (# 1) to be monitored times out, communication data up to the sequence number (TSN) 3 to be monitored is output. In this operation example, the communication data of sequence number (TSN) 2 is given up, the expected value is set to 3 earlier than the operation example of FIG. 2, and the communication data of sequence number (TSN) 3 is output.

  Since the communication data of the sequence number (TSN) 5 that has already arrived has not arrived, the communication data of the sequence number (TSN) 4 has not arrived. Wait for communication data. When the communication data of sequence number (TSN) 4 arrives before the second timer (# 2) times out, the communication data of sequence numbers (TSN) 4 and 5 are output.

  Compared with the embodiment of FIG. 1, the embodiment of FIG. 2 can quickly give up waiting for the sequence number (TSN) 2 and output the communication data of the sequence numbers (TSN) 3 to 5 earlier. The delay can be reduced. That is, as shown in the embodiment of FIG. 1, the sequence number (TSN) of the waiting target is discontinuous as compared with the mode in which the time monitoring is performed with one timer for all the waiting sequence numbers (TSN). Then, the waiting time of each discrete missing sequence number (TSN) is measured by monitoring the waiting time using a separate timer for each discrete missing sequence number (TSN). Thus, it is possible to quickly determine whether to give up waiting for each missing sequence number (TSN), and it is possible to improve the accuracy of determining whether to give up waiting for a missing sequence number (TSN) compared to the first embodiment. This function is effective when communication data is transmitted at a low rate and the scheduling interval of transmitted communication data is large.

  As a configuration example of the second embodiment, a configuration example using two timers has been described. However, when a larger number of discontinuous missing sequence numbers (TSN) are expected to occur, a larger number is generated. In the same manner, it is possible to accurately determine whether or not to wait for each missing sequence number (TSN) by measuring the waiting time for each discrete missing sequence number (TSN) by using the timer. Can do.

  In the second embodiment, when the occurrence of a discontinuous missing sequence number (TSN) occurs more than the number of prepared timers, one of a plurality of timers is used as in the first embodiment. The timer is reset at the time of determining the occurrence of the final missing sequence number (TSN) and restarted, and the sequence number (TSN) of the arrival communication data at the time of determining the final occurrence of missing is the monitoring target sequence paired with the timer The number (TSN) is stored in the storage unit. Then, the monitoring target sequence number stored in the monitoring target sequence number (TSN) storage unit is notified to the output control unit 1-3 together with information on the waiting time by the timer.

  In this way, the implementation scale can be reduced by implementing only the timers for the number of occurrences of expected discontinuous missing sequence numbers (TSNs) without implementing a timer for the window size as in the prior art. At the same time, it is possible to determine whether or not to wait for the missing sequence number (TSN) with almost the same accuracy as in the past.

It is a figure which shows the structural example of the reordering process function part of 1st Embodiment by this invention. It is a figure which shows the operation example of the reordering process (timer monitoring) of 1st Embodiment. It is a figure which shows the structural example of the reordering process function part of 2nd Embodiment by this invention. It is a figure which shows the operation example of the reordering process (timer monitoring) of 2nd Embodiment. It is a figure which shows the function part of a general reordering process, and its outline | summary. It is a figure which shows the format of communication data. It is a figure which shows the operation example of an ordering process. It is a figure which shows the operation example of the loss | deletion generation | occurrence | production determination of the sequence number (TSN) by the conventional timer monitoring. It is a figure which shows the operation example at the time of timeout in the conventional timer monitoring.

Explanation of symbols

1-1 Sequence Number (TSN) Determination Unit 1-2 Reordering Buffer 1-3, 3-2 Output Control Unit 1-4 Sequence Number (TSN) Missing Occurrence Determination Unit 1-5, 3-1 Waiting Time Measurement Unit 1 -51, 3-11, 3-13 Timer 1-52, 3-32, 3-34 Monitoring target sequence number (TSN) storage unit

Claims (4)

In a receiving apparatus that receives communication data transmitted with a sequence number, stores communication data including retransmission data in a reordering buffer, and then reads out and outputs the communication data in order of the sequence number from the reordering buffer.
Based on the sequence number of the received communication data, sequence number missing occurrence determining means for determining occurrence of missing sequence number of received communication data;
A timer is started when the occurrence of a missing sequence number is determined by the sequence number missing occurrence determination means, and each time a new occurrence of a sequence number is newly determined in communication data that is subsequently received, A waiting time measuring means for resetting and starting again, and storing the sequence number of communication data that triggered each start of the timer as a monitoring target sequence number;
When all communication data having a sequence number lower than the monitoring target sequence number arrives before the timer times out, the communication data stored in the reordering buffer is read and output in order of the sequence number, and the monitoring When the timer has timed out without arrival of unarrived communication data with a sequence number lower than the target sequence number, the communication data stored in the reordering buffer is sequenced, giving up waiting for the unarrived communication data. Output control means for reading and outputting in numerical order; and
A receiving device. The waiting time measuring means has a plurality of timers, and each time a sequence number is determined to be missing, a different timer is started, and the sequence number of the communication data that triggered each timer is monitored Store each as a sequence number,
The output control means includes, when the communication data having a sequence number lower than the monitoring target sequence number related to the timer all arrives before each timer times out, including the monitoring target sequence number. The communication data of the younger sequence number is read out from the reordering buffer in the order of the sequence number and output, and the non-arrival communication data of the younger sequence number arrives at each timer than the monitoring target sequence number related to the timer. If time-out occurs, the waiting for the unarrived communication data is abandoned, and the communication data of the lower sequence number including the monitored sequence number related to the timer is read out from the reordering buffer in the order of the sequence number and output. The configuration according to claim 1, further comprising: The receiving device. Communication data reordering processing that receives communication data transmitted with a sequence number, stores communication data including retransmission data in a reordering buffer, and then reads out and outputs the communication data from the reordering buffer in order of the sequence number In the method
Based on the sequence number of the received communication data, the sequence number missing occurrence determination process for determining the occurrence of the missing sequence number of the received communication data,
A timer is started when the occurrence of a missing sequence number is determined in the sequence number missing occurrence determination step, and each time a new occurrence of a sequence number is newly determined in communication data that is subsequently received, Waiting time measurement process to store the sequence number of the communication data that triggered each start of the timer as a monitoring target sequence number, while resetting and starting again,
When all communication data having a sequence number lower than the monitoring target sequence number arrives before the timer times out, the communication data stored in the reordering buffer is read and output in order of the sequence number, and the monitoring When the timer has timed out without arrival of unarrived communication data with a sequence number lower than the target sequence number, the waiting for the unarrived communication data is given up, and the communication data stored in the reordering buffer is sequenced. An output control process that reads and outputs in numerical order; and
A communication data reordering processing method including: The waiting time measurement process uses a plurality of timers, and each time a sequence number is determined to be missing, a different timer is started, and the sequence number of the communication data that triggers the start of each timer is monitored. Remember each as a number,
In the output control process, when all communication data having a sequence number lower than the monitoring target sequence number related to the timer arrives before each timer times out, including the monitoring target sequence number, The communication data of the younger sequence number is read out from the reordering buffer in the order of the sequence number and output, and the non-arrival communication data of the younger sequence number arrives at each timer than the monitoring target sequence number related to the timer. If time-out occurs, the waiting for the unarrived communication data is abandoned, and the communication data of the lower sequence number including the monitored sequence number related to the timer is read out from the reordering buffer in the order of the sequence number and output. The process of claim 3 comprising: Communication data reordering processing methods.

Images