WO2009097744A1 - 基于窗的内容同步的网关和基站及其方法 - Google Patents

基于窗的内容同步的网关和基站及其方法 Download PDF

Info

Publication number
WO2009097744A1
WO2009097744A1 PCT/CN2009/000109 CN2009000109W WO2009097744A1 WO 2009097744 A1 WO2009097744 A1 WO 2009097744A1 CN 2009000109 W CN2009000109 W CN 2009000109W WO 2009097744 A1 WO2009097744 A1 WO 2009097744A1
Authority
WO
WIPO (PCT)
Prior art keywords
packet
window
synchronization control
sequence number
gateway
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2009/000109
Other languages
English (en)
French (fr)
Inventor
Yonggang Wang
He Wang
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Nokia Shanghai Bell Co Ltd
Alcatel Lucent SAS
Original Assignee
Alcatel Lucent Shanghai Bell Co Ltd
Alcatel Lucent SAS
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Alcatel Lucent Shanghai Bell Co Ltd, Alcatel Lucent SAS filed Critical Alcatel Lucent Shanghai Bell Co Ltd
Priority to ES09709304.1T priority Critical patent/ES2524592T3/es
Priority to JP2010544564A priority patent/JP5296102B2/ja
Priority to EP09709304.1A priority patent/EP2239880B1/en
Priority to US12/865,777 priority patent/US8509240B2/en
Priority to KR1020107019394A priority patent/KR101503247B1/ko
Publication of WO2009097744A1 publication Critical patent/WO2009097744A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04HBROADCAST COMMUNICATION
    • H04H20/00Arrangements for broadcast or for distribution combined with broadcast
    • H04H20/65Arrangements characterised by transmission systems for broadcast
    • H04H20/67Common-wave systems, i.e. using separate transmitters operating on substantially the same frequency
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/34Flow control; Congestion control ensuring sequence integrity, e.g. using sequence numbers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W56/00Synchronisation arrangements

Definitions

  • the present invention relates to synchronization techniques in mobile communication single frequency networks, and in particular to a gateway, base station and corresponding method for window based content synchronization technology.
  • the existing content synchronization mechanism is that the gateway determines whether the data packet is segmented and cascaded in layer 2 (L2) according to the fixed air interface transmission rate. After the statistical multiplexing is used, the air interface transmission rate is instantaneously changed, and the gateway cannot determine. Whether data packets are segmented and cascaded. Therefore, content synchronization and other operations need to be separated, and the content synchronization scheme of the self-contained system is designed. Therefore, existing content synchronization mechanisms are not capable of cooperating with statistical multiplexing. In order to work well with statistical multiplexing and to meet the latency/robust requirements of EMBMS, the content synchronization mechanism design should be modified.
  • the present invention proposes an ALU proposal for content synchronization design, namely a "window-based approach" which works well for statistical multiplexing. Summary of the invention
  • the present invention provides a gateway, comprising: a receiving unit, which receives packet data to be sent to a base station; and a window counting unit, configured to count packet data to be sent, and group the number of windows into a window. ; serial number counting unit, used to generate the points to be sent
  • I a serial number of the group; a timestamp setting unit for setting a timestamp of the packet in the entire window, the timestamps of all the packets in the window are associated; and a processing unit that will generate the serial number, the timestamp, and The byte count is included in the packet to be transmitted, the byte count contains the sum of the bytes of the packet of the previous window, and the count of the bytes of all the packets in one window is the same.
  • the processing unit adds a synchronization protocol to the header of the packet to be sent, and the synchronization protocol includes the following three items: a sequence number; a timestamp; a byte count.
  • all packets in a window include the same timestamp, and the timestamp is the time when the first packet of the window is sent from the gateway plus the maximum transmission delay, the maximum processing delay and the delay margin, and the maximum transmission.
  • the delay is the maximum delay of packet transmission in the network from the gateway to all the affiliated base stations.
  • the maximum processing delay is the maximum processing delay required for the packet to be sent from Layer 2 to the air interface at all base stations.
  • the gateway further includes a timer for timing the time when the receiving unit does not receive the data, and a synchronization control packet generating unit, configured to generate the synchronization control packet when the timer expires; wherein the sending unit further sends The resulting synchronization control packet.
  • the synchronization control packet includes: a sequence number, which is a sequence number of a data packet to be sent next; a timestamp, a time when the gateway generates the synchronization control packet plus a maximum transmission delay, a maximum processing delay, and a delay margin ; and the total number of transmitted packet bytes of the window that was terminated before the synchronization control packet.
  • the duration of the timer is set to a predetermined number of repeated transmissions N of the delay margin/synchronization control packet.
  • the gateway further comprises a window setting unit for setting the size of the window and notifying all the base stations to which it belongs.
  • a base station including: a receiving unit, configured to receive packet data from a gateway; and an information extracting unit, configured to extract a sequence number, a timestamp, and a word of the packet from the received packet data. a section counting unit, configured to count received packets according to the extracted sequence number to form a window, and a processing unit configured to process the packet data of the entire window according to the acquired timestamp To set the allowable air interface transmission time of the current window.
  • the allowed air interface transmission time of the current window is set as the time stamp in the synchronization protocol in the data packet in the next window.
  • the base station further includes a synchronization control packet determining unit configured to analyze the received packet data to determine whether the received synchronization control packet is received.
  • the synchronization control packet includes a sequence number, which is a sequence number of a packet to be transmitted next; a time stamp; and a total number of transmitted packet bytes of a window that is terminated before the synchronization control packet.
  • the window counting unit forms a window based on the sequence number in the synchronization control packet.
  • the base station further comprises a packet loss determining unit, configured to determine whether there is packet loss according to the sequence number of the received packet or the sequence number of the synchronization control packet, and if there is packet loss, perform packet loss processing.
  • a packet loss determining unit configured to determine whether there is packet loss according to the sequence number of the received packet or the sequence number of the synchronization control packet, and if there is packet loss, perform packet loss processing.
  • the packet loss judging unit judges whether or not there is a packet loss based on whether the sequence number of the previously received packet and the sequence number of the currently received packet or the sequence number of the synchronization control packet are consecutive.
  • the packet loss processing comprises: regenerating the packet, the number of the regenerated packets is a difference of the discontinuous sequence numbers, and the total number of bytes of the packet is the sum of the bytes of all the received packets of the current window and the next window.
  • the length of each packet to be regenerated may be arbitrarily selected or evenly distributed.
  • the base station further comprises an alarm unit for notifying the gateway of the occurrence of a network transmission problem when a continuous packet of a window size or a predetermined number of repeated transmissions of N consecutive synchronization control packets is lost.
  • a packet data processing method for content synchronization in a gateway comprising the steps of: receiving packet data to be transmitted to a base station; counting packet data to be transmitted, causing a window The number of groups of packets form a window; the sequence number of the packet to be sent is generated; the timestamp of the packet in the entire window is set, the timestamps of all the packets in the window are associated; and the generated sequence number, timestamp And the byte count is included in the packet to be transmitted, the byte count contains the sum of the bytes of the packet of the previous window, and the byte counts of all the packets in one window are the same.
  • the step of including the generated sequence number, time stamp, and byte count in the packet to be sent includes: adding a synchronization protocol to the header of the packet to be sent, the synchronization protocol includes the following three items: : serial number; timestamp; byte count.
  • all packets in a window include the same timestamp, and the timestamp is the time when the first packet of the window is sent from the gateway plus the maximum transmission delay, the maximum processing delay and the delay margin, and the maximum transmission.
  • the delay is the maximum delay of packet transmission in the network from the gateway to all the affiliated base stations.
  • the maximum processing delay is the maximum processing delay required for the packet to be sent from Layer 2 to the air interface at all base stations.
  • the method further comprises the steps of: counting the time at which the receiving unit does not receive the data; and generating the synchronization control packet when the time is full, and transmitting the generated synchronization control packet.
  • the synchronization control packet includes: a sequence number, which is a sequence number of a data packet to be sent next; a timestamp, a maximum transmission delay, a maximum processing delay, and a delay margin for the generation timing of the synchronization control packet; And the total number of transmitted packet bytes of the window that was terminated before the synchronization control packet.
  • the timing duration is set to the predetermined margin of transmission of the delay margin/synchronization control packet ⁇
  • the method further comprises the steps of: setting a size of the window and informing all gateways to which the gateway belongs.
  • a data packet processing method for content synchronization in a base station comprising the steps of: receiving packet data from a gateway; extracting a sequence number and time of the packet from the received packet data. Stamping, byte count information; counting received packets according to the extracted serial number to form a window; and processing the number of packets of the entire window, setting the permission of the current window based on the acquired timestamp Air interface transmission time;
  • the step of setting the allowed air interface transmission time of the current window according to the acquired time stamp comprises: setting the allowed air interface transmission time of the current window as a time stamp in the synchronization protocol in the data packet in the next window.
  • the method further comprises: analyzing the received packet data to determine whether the received synchronization control packet is received.
  • the synchronization control packet includes: a sequence number, a sequence number of a packet to be transmitted next; a timestamp; and a total number of transmitted packet bytes of the window that is terminated before the synchronization control packet.
  • the method further comprises: determining whether there is packet loss according to the sequence number of the received packet or the sequence number of the synchronization control packet, and if there is packet loss, performing packet loss processing.
  • whether or not there is packet loss is judged based on whether the sequence number of the previously received packet and the sequence number of the currently received packet or the sequence number of the synchronization control packet are consecutive.
  • the packet loss processing comprises: regenerating the packet, the number of the regenerated packets is a difference of the discontinuous sequence numbers, and the total number of bytes of the packet is the sum of the bytes of all the received packets of the current window and the next window.
  • the length of each packet to be regenerated may be arbitrarily selected or evenly distributed.
  • the method further comprises the step of notifying the gateway of the occurrence of a network transmission problem when a consecutive packet of a window size or a predetermined number of repeated transmissions of N consecutive synchronization control packets is lost.
  • the solution of the present invention has the following advantages:
  • Figure 1 shows the structure of a network in which the present invention can be implemented
  • FIG. 2 shows a schematic diagram of a synchronization protocol PDU in accordance with the present invention
  • FIG. 3 shows a block diagram of a gateway in accordance with the present invention
  • Figure 4 shows a flow chart of the operation of the gateway
  • Figure 5 is a diagram showing the relationship of a synchronization control packet to a window according to the present invention
  • Figure 6 is a block diagram showing a base station according to the present invention
  • Figure 7 shows a flow chart of the operation of the base station. detailed description
  • Figure 1 shows the structure of a network in which the inventive concept can be practiced.
  • the access gateway GW is connected to a plurality of base stations eNB1, eNB2, and eNB3, and an interface between them is called an S I interface.
  • the plurality of base station eNBs described above are connected in the form of a mesh (dashed line in Fig. 1), and the interface between the base stations eNB is referred to as an X2 interface.
  • a plurality of user equipments UE1 1 to E12, UEs 21 to 23, and UEs 31 to 33 are schematically shown in the respective cells of the base stations eNB1 to 3.
  • Each synchronization protocol PDU on the M1 interface carries the timestamp and byte count information set by the gateway in the synchronization PDU header for packet transmission time indication and packet loss processing.
  • the key terms in the content synchronization protocol according to the present invention are explained one by one below.
  • a window includes a specific number of packets.
  • the size of the window i.e., the number of packets, is pre-configured and is known by each eNB in the SFN.
  • the gateway sends a synchronization control packet, the window is terminated immediately even if the size of the window is not reached.
  • the synchronization control packet indicates that no data packets have arrived for a long time. This will be defined later.
  • T t T ri + T Ma , Delay , which is defined as the time at which the gateway receives the MBMS data packet.
  • T n its corresponding time stamp is represented in Figure 1 as T n .
  • T maxDe ' includes the maximum transmission delay between the gateway and the eNB, the delay margin, and the maximum eNB layer 2 processing time, which is defined as
  • the timestamps of all packets in a window are associated so that the timestamps of other packets can be inferred from the timestamp of one of the packets. For example, you can set the timestamps of all groups in a window to be the same (as shown), equal, or otherwise.
  • the eNB receives a packet for the next window, or synchronizes the control packet, it means that all packets of the previous window can be processed and transmitted.
  • Time of grouping or synchronization control grouping in the next window The stamp indicates that the transmission of all packets of the previous window in the eNB should start on the air interface with the first transmission opportunity or after the indicated time.
  • the allowed transmission time of packet n is not the timestamp of packet n, but the timestamp of the packet in the next window or the timestamp of the control packet immediately following it, so that all eNBs can be guaranteed at the allowed transmission time of the packet of the window.
  • the eNB with the longest delay of the S1 interface can receive at least the data packets or control packets in the next N window, ensuring that the packets of the window are completely received or fully recovered at the time of transmission permission, and synchronization is ensured.
  • the allowed transmission time can also be set to other times according to the timestamp of packet n as long as the time is known to the eNB.
  • All packets in the RLC buffer that allow transmission times earlier than the transport time of the transport block can be processed, i.e., segmented, concatenated, and populated in the transport block.
  • the transport block transmission time can be "reuse cycle start time”.
  • the byte count contains the number of bytes of all packets in the entire previous window sequence.
  • the byte counts of all the packets in the window are the same.
  • the byte count determines the total length of the lost packet.
  • the sequence number (SN) is used to calculate the number of lost packets in the eNB, if any packets are lost.
  • SDU Service Data Unit
  • LI length indication
  • the eMBMS has a Radio Link Control Protocol Data Unit (RLC PDU) format with the principle of "per SUD- LI" so that multiple packets can be recovered in content synchronization.
  • RLC PDU Radio Link Control Protocol Data Unit
  • Proportional equal scheduling can be applied in statistical multiplexing. Synchronous control group
  • the MBMS gateway monitors the status of packets transmitted to the eNB.
  • N, N are set as the number of times of repeatedly transmitting the synchronization packet which is set in advance in order to obtain a reliable transmission synchronization control packet at the S1 interface.
  • the gateway transmits a sync control packet to the eNB.
  • the synchronization control packet may be transmitted N times to the eNB during the time interval to ensure that the synchronization control packet is reliably received in the eNB.
  • Data packets and synchronization control packets can be distinguished by setting flag bits in the packet, or other common methods. Packet loss recovery
  • the window mechanism and the synchronization control packet are designed to recover the out-of-synchronization caused by the moment of loss of the particular packet.
  • the eNB can know that packet loss has occurred by the SN of the next correctly received packet.
  • the synchronization layer puts the pseudo-packet and the allowed transmission time recovered by the next correct packet into the RLC buffer, and all transport blocks containing a part of the dummy packet will be silenced in the air interface.
  • a change in RLC payload may occur, as indicated by the LI (Length Indicator) field.
  • LI Length Indicator
  • the gateway according to the embodiment of the present invention is as shown in FIG. 3, wherein the gateway 100 according to the present invention includes a window setting unit. , 101, is used to set the size of the window and notify all the eNBs to which the gateway 100 belongs to the set window size; the GW window counting unit 102 is configured to count IP packets belonging to one window; the serial number counting unit 103, Counting the IP packets to be sent to generate a sequence number of the IP packet; a timer 104 for timing; a GW receiving unit 105 for receiving an IP packet to be sent; and a GW processing unit 106 for The transmitted IP packet is processed; the synchronization control packet generating unit 107 is configured to generate a synchronization control packet; and the GW transmitting unit 108 is configured to send the IP packet processed by the processing unit 106 or the control packet generated by the synchronization control packet generating unit 107.
  • the gateway transmitting unit 108 is configured to send the IP packet processed by the processing unit 106 or the control packet generated by the
  • the window setting unit 101 of the gateway 100 defines a window size parameter.
  • WINDOW — S I ZE WINDOW — S I ZE and inform the base station.
  • the serial number counting unit 103 counts the serial number SN from zero.
  • the GW window counting unit 102 also starts counting the window WIND0W_C0UNT.
  • the sequence number counting unit 103 increments the sequence number by 1, generates the sequence number of the IP packet, and the GW processing unit 106 pairs the IP.
  • the packet performs some of the following processing, and then the GW transmitting unit 108 transmits it to all base stations to which the gateway 100 belongs by means of IP multicast.
  • the gateway adds a synchronization protocol to each packet header.
  • the synchronization protocol includes three items, one is the sequence number counted by the sequence number counting unit 103, where, for example, the sequence number counting unit 103 goes from 0 to 65535 loop count; second is a timestamp.
  • the timestamp is the time when the first packet of the window is sent from the gateway, plus the maximum transmission delay, the maximum processing delay and the delay margin, and the maximum transmission.
  • the delay is the maximum delay of the packet transmission in the network from the gateway to all the affiliated base stations.
  • the maximum processing delay packet is the maximum processing delay required for all base stations to process from layer 1 to the air interface; the third is the byte count. The sum of the bytes of the packet containing the previous window. The byte counts of all the packets in a window are filled with the same value.
  • the timestamps of the individual packets in the window are set to be the same, ie the packets in one window are marked with the same timestamp, ie the timestamp of the first packet; of course, as before In the above, the time stamps in one window may be different, and as long as they are associated, and the positional relationship is known according to the serial number relationship, the other can be inferred from one.
  • the gateway adds a layer of synchronization protocol to each packet header and then sends it through IP multicast. The destination is all base stations to which the gateway belongs.
  • the GW window counting unit 102 increments by 1. When the window count reaches the window size WIND0W_S IZE, such as 5, it indicates that the next packet is the start of a new window. The window count is cleared, and the GW receiving unit 105 waits to receive the next IP packet.
  • the gateway In order to prevent the next packet received by the gateway from reaching the gateway for a long time, the gateway also sets a timer, such as timer 104, the timer duration is T, T-delay margin/N, when the slave broadcast and the group If a packet sent by the broadcast service center arrives at the gateway and no next packet arrives after the T time, the timer expires, where N is the number as defined above. The gateway then indicates that the current window is terminated by transmitting a synchronization control packet to the base station, at which point the window ends even though the specified window size is not reached. The synchronization control packet is generated by the synchronization control packet generation unit 107.
  • the synchronization control packet contains the following fields: The serial number, the serial number of the next data packet to be sent, the timestamp, the time from the gateway for the synchronization control packet plus the maximum transmission delay, maximum processing delay, and delay margin. Quantity; and the sum of the bytes of all packets of the currently terminated window. Setting the duration T of the timer to T-delay margin /N allows the synchronous control group to send N consecutive bursts within the delay margin to ensure reliable transmission.
  • the synchronization control packet When the next data packet arrives at the gateway, it indicates the start of a new window. All operations are the same. Although the synchronization control packet has a sequence number field, the sequence number is the sequence number of the next data packet to be sent. The synchronization control packet is not in the serial number accumulation, as shown in Figure 5.
  • the gateway cyclically performs the above processing, and transmits the received IP packet or synchronization control packet to the associated base station.
  • FIG. 6 shows a base station in accordance with the present invention.
  • the base station 200 includes: an NB receiving unit 201, configured to receive a packet; an information extracting unit 202, configured to extract a sequence number, a timestamp, and a byte count from the received data; and a synchronization control packet determining unit 203, For determining whether the received packet is a synchronization control packet, the packet loss determination unit 204 is configured to determine whether the packet is lost; the NB window counting unit 205 is configured to count IP packets belonging to one window; the NB processing unit 206, Processing all IP packets belonging to a window, the base station buffer 207, for buffering IP packet data received from the gateway; packet loss processing unit 208. It is configured to perform packet loss processing when the packet loss determining unit 204 determines that there is a packet loss.
  • the base station 200 shown in Fig. 6 operates as shown in Fig. 7. Since the service has been assumed to start with the first packet with sequence number 0, the first window is started. It is of course also possible to send N synchronization control packets before the start of the service data packet to indicate to the base station the start of the window.
  • the NB receiving unit 201 of the base station receives the data packet.
  • the information extracting unit 202 extracts the value in the following fields in the synchronization protocol of the packet: a sequence number; a timestamp, indicating a time when the packet of the previous window is sent in the air interface; The section count, used to recover the number of bytes of the packet that the previous window may have dropped.
  • the packet loss judging unit 204 can judge whether there is any data packet lost before the received packet by using the continuity of the sequence number. If it is lost, it can be judged to be lost according to the difference between the sequence number of the packet and the previous packet. Several groups.
  • window count unit 205 counts the sequence number of the received data packet. If the counter reaches the window size, it indicates that the next data packet is the beginning of a new window. The window is counted by the serial number instead of the received data packet. For example, if the current window starts from 0 and 0, 1, 2, 3, and 6 are received, the current window contains 0, 1, and 2. 3, 4, the next window is 5, 6...
  • the packet loss judging unit 203 Upon receiving the packet of sequence number 6, the packet loss judging unit 203 can judge that the window has lost 2 packets according to the difference between 6 and the sequence number 3 of the received previous packet, and the window size 5, and the next one The window lost a packet. After judging that the packet has been lost, the packet loss processing is performed.
  • the processing unit 206 in the base station processes the packet of the received window, that is, the data packet in the next window is in the synchronization protocol.
  • the timestamp is assigned to the data packet of the current window and is named to allow the air interface transmission time. All data packets of the window are accompanied by the time that the air interface transmission time is allowed to be placed into the Layer 2 (RLC) buffer.
  • the time when the air interface is allowed to be transmitted means that the data packet is filled in the transport block of the air interface protocol frame, and the time at which the transport block is sent out in the air interface must be exactly equal to or later than the time.
  • all packets of the window can be processed by the base station L2 user interface and sent out on the air interface.
  • the grouping of all the windows before the window is not sent out by the air interface, but stays in the L2
  • the synchronization packet judging unit 203 of the base station judges that one or several consecutive synchronization control packets have been received, it indicates that the current window is terminated by the synchronization control packet, and the last packet of the current window is the sequence number -1 in the synchronization control packet. At this point, the grouping within the window is shown in Figure 5. All processing is the same as when the packet of a window is completely received, that is, the base station assigns the time stamp in the synchronization control packet to the data packet of the current window, and names it as the air interface transmission time, all data packets of the window. The time allowed for the air interface transmission is placed in the L2 (RLC) buffer.
  • the packet loss processing unit 208 needs to perform synchronization recovery. deal with.
  • the base station receives a packet of a window and has received the data packet in the next window, or the current window is terminated by the synchronization control packet, the base station counts the sum of the bytes of all the received packets of the current window, and then takes out The number of bytes in the data packet or synchronization control packet in the next window. The difference between the two is the total number of bytes of all packets lost by the current window.
  • the number of packets lost by the current window can be known by the discontinuous sequence number.
  • the recovery method is to regenerate the packet. If the number of packets is equal to 1, the length is the number of bytes of a single packet lost by the determined current window; If the number of packets is greater than 1, the total length is the total number of bytes of all packets lost in the determined current window, and the specific length of each packet may be arbitrarily or evenly distributed.
  • the regenerated packet identifier is false, the data content of the packet is invalid, but the total length is the same as the true packet. L2 knows which packets are recovered false packets. When the air interface transmits, the transport block containing these fake packets does not transmit power. For silence.
  • the base station may further include an alarm unit 209 for notifying the gateway or the O0 that an abnormal network transmission problem occurs in the case where a window-sized continuous packet or N consecutive synchronization control frames are lost.
  • Nok ia In R3-071920 MB SFN schedu l ing and content synchron i za t ion, Nok ia, Nok ia S iemens Network, 2007-10, Nok ia only proposes a content synchronization scheme based on fixed burst length judgment. Its "total count frame" and the synchronization control packet of the present invention Similarly, it also has similar information, including the total number of bytes of the previous burst (in the present invention, the previous window), the timestamp, and the total number of previous bursts. However, in the Nokia scheme, a fixed length burst, such as 1 second or other value, is first defined.
  • the frequency of generation of the "total count frame" is defined based on the burst length.
  • the synchronization processing delay is also affected by the burst length. According to the simulation of the Nokia scheme of the present invention, the burst timing length set to only 0.5 seconds is acceptable in terms of the performance of the synchronization layer processing delay and the number of control packets.
  • the synchronization layer processing delay performance of the ALU scheme is 0.6768 seconds, and the Nok i a is 0.6862 seconds. The difference between the two is 0. 0186 seconds.
  • These two schemes have similar synchronization layer processing delay performance. However, the number of control packets between the two schemes is very different.
  • the ALU scheme only produces 695 control packets, but the Nok i a scheme has 1 1989 control packets, which is 17.25 times that of the ALU scheme. Therefore, from a comprehensive perspective, the ALU scheme is superior to the Nok i a scheme.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Security & Cryptography (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Data Exchanges In Wide-Area Networks (AREA)
  • Communication Control (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Synchronisation In Digital Transmission Systems (AREA)

Description

基于窗的内容同步的网关和基站及其方法
技术领域
本发明涉及移动通信单频网中的同步技术,具体地讲, 涉及一种用于 基于窗的内容同步技术的网关、 基站及相应方法。 背景技术
在 RAN3#57bis和 RAN2#59bis会议上,统计复用被一致同意是 E- MBMS 复用操作的要求。现有的内容同步机制是网关根据固定的空口传输速率来 判断数据分组是否在层 2 (L2)进行分段和级联, 在采用统计复用后, 空 口传输速率是瞬时变化的, 网关无法判断是否数据分组是否分段和级联。 因此需要将内容同步和其他操作独立开, 设计自成系统的内容同步的方 案。 因此, 现有的内容同步机制不能够与统计复用协作。 为了很好地与统 计复用协作并满足 EMBMS的延迟 /鲁棒需求,应该修改内容同步机制设计。
在 [R3- 071020 MBSFN scheduling and content synchronization,
Nokia, Nokia Siemens Network, 2007-10]中, Nokia简单地提出了一种 内容同步方案, 该方案基于固定时间长度的突发机制。 该方案可与统计复 用协作。 但是, 其中并未清楚地描述该方案的一些关键点。 此外, 为了实 现 1 秒的处理延迟要求, 该方法需要较大的控制分组负荷。 另一方面, Nokia的方法中的控制机制并不灵活。 此外, 根据仿真发现, 在 Nokia方 法中仅有一种参数设置选择可应用。
因此, 需要一种能够克服 Nokia方案的上述缺点的内容同步方案。 本发明提出一种针对内容同步设计的 ALU建议, 即 "基于窗的方法", 该方法对于统计复用工作良好。 发明内容
为了克服上述缺点, 本发明提供一种网关, 包括: 接收单元, 接收要 发送到基站的分组数据;窗计数单元,用于对要发送的分组数据进行计数, 使窗大小数目的分组组成一个窗; 序列号计数单元, 用于产生要发送的分
I 组的序列号; 时间戳设置单元, 用于设置整个窗中的分组的时间戳, 该窗 中的所有分组的时间戳都相关联; 以及处理单元, 将所产生的序列号、 时 间戳、 以及字节计数包括在要发送的分组中, 所述字节计数包含前一个窗 的分组的字节总和, 一个窗中的所有分组的字节计数都相同。
优选地,处理单元对要发送的分组的头上加一层同步协议,该同步协 议包含如下三项内容: 序列号; 时间戳; 字节计数。
优选地,一个窗中所有的分组包括相同的时间戳,并且该时间戳是该 窗的第一个分组从网关发送的时刻加上最大传输时延、最大处理时延和延 迟裕量,最大传输时延是分组在从网关到所有所属基站的网络中传输的最 大时延,最大处理时延是分组在所有基站处从层 2处理到从空口发出所需 的最大处理时延。
优选地,该网关还包括定时器,用于对接收单元未接收到数据的时间 进行计时; 以及同步控制分组产生单元, 用于在定时器计时满时产生同步 控制分组; 其中, 发送单元还发送所产生的同步控制分组。
优选地, 同步控制分组包括: 序列号, 为下一个要发送的数据分组的 序列号; 时间戳, 为网关产生该同步控制分组的时刻加上最大传输时延、 最大处理时延和延迟裕量;以及同步控制分组前被终止的窗的总发送分组 字节数。
优选地, 定时器的时长设置为延迟裕量 /同步控制分组的预定重复传 输次数 N。
优选地, 该网关还包括窗设置单元, 用于设置窗的大小, 并告知所有 所属基站。
根据本发明的另一方面, 还提供一种基站, 包括: 接收单元, 用于从 网关接收分组数据; 信息提取单元, 用于从接收到的分组数据中提取分组 的序列号、 时间戳、 字节计数信息; 窗计数单元, 用于根据所提取的序列 号, 对接收的分组进行计数, 以形成一个窗; 以及处理单元, 用于对整个 窗的分组数据进行处理,根据所获取的时间戳来设置当前窗的允许空口传 输时刻。
优选地,将当前窗的允许空口传输时刻设置为下一个窗中的数据分组 中同步协议中的时间戳。 优选地,该基站还包括同步控制分组判断单元,用于对接收的分组数 据进行分析, 以判断接收到的是否是同步控制分组。
优选地,所述同步控制分组包括序列号, 为下一个要发送的分组的序 列号; 时间戳; 以及同步控制分组前被终止的窗的总发送分组字节数。
优选地,在接收到的分组是同步控制分组时, 窗计数单元根据同步控 制分组中的序列号, 形成一个窗。
优选地,该基站还包括分组丢失判断单元,用于根据接收到的分组的 序列号或同步控制分组的序列号, 判断是否存在分组丢失, 如果存在分组 丢失, 则进行分组丟失处理。
优选地,分组丢失判断单元根据前一个接收到的分组的序列号和当前 接收到的分组的序列号或同步控制分组的序列号是否连续,来判断是否存 在分组丢失。
优选地, 分组丢失处理包括: 重新生成分组, 重新生成的分组的个数 为不连续序列号之差,分组的字节总数是当前窗的所有接收到的分组的字 节总和与下一个窗的数据分组或同步控制分组中的字节计数之差。
优选地,在要重新生成的分组的个数大于 1时,可随意选择或平均分 配各个要重新生成分组的长度。
优选地,该基站还包括警报单元,用于在出现一个窗大小的连续分组 或预定重复传输次数 N个连续同步控制分组丢失时,向网关通知出现网络 传输问题。
根据本发明的另一方面,还提供一种在网关中的用于内容同步的分组 数据处理方法, 包括以下步骤: 接收要发送到基站的分组数据; 对要发送 的分组数据进行计数, 使窗大小数目的分组组成一个窗; 产生要发送的分 组的序列号; 设置整个窗中的分组的时间戳, 该窗中的所有分组的时间戳 都相关联; 以及将所产生的序列号、 时间戳、 以及字节计数包括在要发送 的分組中, 所述字节计数包含前一个窗的分组的字节总和, 一个窗中的所 有分组的字节计数都相同。
优选地, 将所产生的序列号、 时间戳、 以及字节计数包括在要发送的 分组中的步骤包括: 对要发送的分组的头上加一层同步协议, 该同步协议 包含如下三项内容: 序列号; 时间戳; 字节计数。 优选地,一个窗中所有的分组包括相同的时间戳,并且该时间戳是该 窗的第一个分组从网关发送的时刻加上最大传输时延、最大处理时延和延 迟裕量,最大传输时延是分组在从网关到所有所属基站的网络中传输的最 大时延,最大处理时延是分组在所有基站处从层 2处理到从空口发出所需 的最大处理时延。
优选地,该方法还包括步骤:对接收单元未接收到数据的时间进行计 时; 以及在计时时长满时产生同步控制分组, 并发送所产生的同步控制分 组。
优选地, 同步控制分组包括: 序列号, 为下一个要发送的数据分组的 序列号; 时间戳, 为该同步控制分组的产生时刻加上最大传输时延、 最大 处理时延和延迟裕量;以及同步控制分组前被终止的窗的总发送分组字节 数。
优选地, 计时时长设置为延迟裕量 /同步控制分组的预定重复传输次 数^
优选地, 该方法还包括步骤: 设置窗的大小, 并告知网关所属所有基 站。
根据本发明的另一方面,还提供一种在基站中的用于内容同步的数据 分组处理方法, 包括以下步骤: 从网关接收分组数据; 从接收到的分组数 据中提取分组的序列号、 时间戳、 字节计数信息; 根据所提取的序列号, 对接收的分组进行计数, 以形成一个窗; 以及对整个窗的分组数椐进行处 理, 根椐所获取的时间戳来设置当前窗的允许空口传输时刻;
优选地 ,根据所获取的时间戳来设置当前窗的允许空口传输时刻的步 骤包括:将当前窗的允许空口传输时刻设置为下一个窗中的数据分组中同 步协议中的时间戳。
优选地, 该方法还包括: 对接收的分组数据进行分析, 以判断接收到 的是否是同步控制分组。
优选地, 所述同步控制分组包括: 序列号, 为下一个要发送的分组的 序列号; 时间戳; 以及同步控制分组前被终止的窗的总发送分组字节数。
优选地,在接收到的分组是同步控制分组时,根据同步控制分组中的 序列号, 形成一个窗。 优选地,该方法还包括:根据接收到的分组的序列号或同步控制分组 的序列号, 判断是否存在分组丢失, 如果存在分组丟失, 则进行分组丢失 处理。
优选地,根据前一个接收到的分组的序列号和当前接收到的分组的序 列号或同步控制分组的序列号是否连续, 来判断是否存在分组丢失。
优选地, 分组丢失处理包括: 重新生成分组, 重新生成的分组的个数 为不连续序列号之差,分组的字节总数是当前窗的所有接收到的分组的字 节总和与下一个窗的数据分组或同步控制分组中的字节计数之差。
优选地,在要重新生成的分组的个数大于 1时,可随意选择或平均分 配各个要重新生成分组的长度。
优选地,该方法还包括步骤:在出现一个窗大小的连续分组或预定重 复传输次数 N 个连续同步控制分组丢失时, 向网关通知出现网络传输问 题。 本发明的方案与现有技术相比, 具有如下优点:
( 1 )在统计复用下也能良好地工作; (2 )具有较好的同步处理延迟 和控制分组负荷的综合性能。 附图说明
从下面结合附图的详细描述中, 本发明的上述特征和优点将更明显, 其中:
图 1示出了可实施本发明的网络的结构;
图 2示出了根据本发明的同步协议 PDU的示意图;
图 3示出了根据本发明的网关的框图;
图 4示出了网关的操作的流程图;
图 5示出了根据本发明的同步控制分组与窗的关系的示意图; 图 6示出了根据本发明的基站的框图; 以及
图 7示出了基站的操作的流程图。 具体实施方式
图 1示出了可以实践本发明思想的网络的结构。 其中, 接入网关 GW与 多个基站 eNBl、 eNB2和 eNB3连接, 它们之间的接口被称为 S I接口。 上述的 多个基站 eNB之间以网格(mesh ) 的形式(图 1中的虚线)连接, 基站 eNB 之间的接口被称为 X2接口。 基站 eNBl ~ 3的各个小区中示意性地显示了多 个用户设备 UE1 1 ~ E12, UE21 ~ 23以及 UE31 ~ 33。
同步协议说明
根据本发明的基于窗的内容同步方法的基本思想如图 2所示。 Ml接 口上的每个同步协议 PDU在同步 PDU头中载有网关设置的时间戳和字节计 数信息, 用于分组传输时间指示和分组丟失处理。 下面对根据本发明的内 容同步协议中的关键术语逐一进行解释。
1 ) 窗
一个窗包括特定数目的分组。 预先配置窗的大小, 即分组的数目, 并 且被 SFN中的每个 eNB知道。 第一个窗开始于会话开始之后的 SN=0分组 处。 在网关发送同步控制分组时, 即使没有达到窗的大小, 也立即结束该 窗。
同步控制分组指示长时间内没有数据分组到达。 这将在后面进行定 义。
2 ) 时间戳
每个窗均设置有时间戳。 对千窗中的第一个分组, Tt, = Tri + TMa,Delay, 其中 被定义为网关接收到 MBMS数据分组的时间。 对于分组 Pn, 在图 1 中将其相应的时间戳表示为 Tn。 "TmaxDe ' 包括网关和 eNB之间的最大传输 延迟、 延迟裕量和最大 eNB 层 2 处理时间 , 被定义为
TwaxDe l ay-TTransDe l ay+Twa rg i n+TprocTi me
一个窗中的所有分组的时间戳是相关联的,以便可以根据其中一个分 组的时间戳而推断出其它分组的时间戳。 例如, 可以将一个窗中的所有分 组的时间戳设置为相同的 (如图所示)、 等差的, 或者其它的。
3 )允许传输时间
一旦 eNB接收到下一个窗的分组,或者同步控制分组,则意味着可以 处理并传输前一窗的所有分组。下一个窗中的分组或同步控制分组的时间 戳指示在 eNB 中前一个窗的所有分组的传输应该以第一个传输机会在空 口上开始或者在所指示的时间之后开始。
分组 n的允许传输时间不是分组 n的时间戳,而是下一个窗中分组的 时间戳或者紧随其后的控制分组的时间戳,这样可以保证在该窗的分组的 允许传输时刻, 所有 eNB, 特别是 S 1接口时延最长的 eNB, 都能至少收到 N个下一个窗中的数据分组或控制分组, 确保在允许传输时刻, 该窗的分 组完全收到或完全恢复, 保证同步。 当然, 也可以根据分组 n的时间戳将 允许传输时间设置为其它的时间, 只要该时间对于 eNB而言可以获知。
4 )分段和串联
可以处理允许传输时间早于传输块传输时间的 RLC緩沖器中的所有 分组, 即分段、 串联并填充在该传输块中。
对于业务复用, 传输块传输时间可以是 "复用周期开始时间"。
5 ) 字节计数
字节计数包含整个前一窗序列中的所有分组的字节数。窗中的所有分 组的字节计数是相同的。 字节计数可以确定丟失分组的总长度。
6 )序列号
序列号 (SN )用于计算 eNB中丢失的分组数, 如果有分组丢失的话。 7 )每个业务数据单元( SDU ) 一个长度指示 ( LI )原理
推荐 eMBMS的具有 "每个 SUD—个 LI " 原理的无线链路控制协议数 据单元(RLC PDU )格式, 从而可以在内容同步中恢复多个分组。
8 )统计复用 数据是相同的, 并且在不同 eNB中是一致的
可在统计复用中应用比例平等调度。 同步控制分组
MBMS 网关监控传送给 eNB的分组的状态。 首先设置 N, N是为了在 S 1 接口获得可靠传输同步控制分组而预先设置的重复传输该同步分组的 次数。 一旦发现 7Lf,„/N时间间隔没有传送数据分组, 则网关向 eNB发送 同步控制分组。 该同步控制分组包含同步头, 该同步头具有下一个数据分组的 SN、 前一窗的总发送分组字节数、 以及新的时间戳值 = T„ + TMa,De y, 其中 ;,被定义为网关产生控制分组的时间。
在 时间间隔期间可以向 eNB发送该同步控制分组 N次, 以便确 保 eNB中可靠地接收到同步控制分组。
可以通过在分组中设置标志位,或者其它常用方法,来区分数据分组 和同步控制分组。 分组丢失恢复
对于正常情况, 不会出现分组丢失, 并且利用内容同步机制在不同 eNB中保持一致是没有问题的。 但是, 对于出现分组丢失的特殊情况, 在 业务复用中将破坏内容同步, 例如丢失了包含新的时间戳的最后分组。 因 为在当前调度周期中, 在所涉及的 eNB中, 与其它 eNB相比每个业务的资 源分配是不同的。
为了解决该问题,窗机制和同步控制分组设计用于恢复由于该特殊分 组丢失时刻而引起的失步。
eNB可以通过下一个正确接收到的分组的 SN而知道发生了分组丢失。 同步层将伪分组以及下一个正确分组所恢复的允许传输时间一起放 入 RLC緩冲器中, 包含伪分组的一部分的所有传输块将在空口中静默。
在 RAN2中, 为了允许 RLC串联多个 RLC SDU , 可能出现 RLC有效载 荷的变化, 这可由 LI (长度指示符)字段指示。 利用 "每个 SDU—个 LI " 的 RLC PDU格式, 可以恢复任何情况的分组丢失, 只要 eNB知道丢失的分 组数、 丢失分组的总大小、 并将它们串联起来。
异常情况
如果一个窗大小的连续分组或 N个连续同步控制帧丢失了,则向网关 或 0&M通知出现异常网络传输问题。 下面, 将参照附图更详细地描述本发明的这些及其它方面。
为了实现接入网关所属的多个基站 eNB之间的内容同步,根据本发明 实施例的网关如图 3所示, 其中,根据本发明的网关 1 00包括窗设置单元 , 101 , 用于设置窗的大小并向该网关 100所属的所有 eNB通知所设置的窗 大小; GW窗计数单元 102, 用于对属于一个窗的 IP分组进行计数; 序列 号计数单元 103, 用于对要发送的 IP分组进行计数, 以产生 IP分组的序 列号; 定时器 104, 用于进行计时; GW接收单元 105, 用于接收要发送的 I P分组; GW处理单元 106, 用于对要发送的 IP分组进行处理; 同步控制 分组产生单元 107, 用于产生同步控制分组; GW发送单元 108 , 用于将处 理单元 106所处理的 I P分组或同步控制分组产生单元 107所产生的控制 分组发送到该网关所属的所有基站; 时间戳设置单元 109, 用于设置时间 戳。 根据本发明的网关 100如图 4所示地进行操作。
首先, 网关 100 的窗设置单元 101 定义了一个窗大小参数
WINDOW— S I ZE,并告知给基站。为了方便起见,假设窗大小 WINDOW— SI ZE=5, 5是指 5个 S1接口上的分组。 并且假定从序列号为 0的第一个分组开始 业务, 从此开始第一个窗。 因此, 序列号计数单元 103从零开始对序列号 SN进行计数。 并且 GW窗计数单元 102也开始对窗 WIND0W_C0UNT进行计 数。 于是, 在 GW接收单元 105接收到从广播与组播业务中心发来的 I P分 组时, 序列号计数单元 103将序列号递增 1 , 产生该 IP分组的序列号, 并由 GW处理单元 106对 IP分组进行以下一些处理,然后 GW发送单元 108 通过 IP组播的方式将其发给该网关 100所属的所有基站。
第一步, 网关给每个分组头上加一层同步协议,该同步协议包含三项 内容, 一是序列号计数单元 103所计数的序列号, 其中, 比如, 序列号计 数单元 103从 0到 65535循环计数; 二是时间戳, 在本发明的实施例中, 时间戳为窗的第一个分组从网关发出的时刻, 加上最大传输时延、 最大处 理时延和延迟裕量,最大传输时延是分组在从网关到所有所属基站的网络 中传输的最大时延,最大处理时延分组在所有基站从层 1处理到从空口发 出的所需最大处理时延;三是字节计数,包含前一个窗的分组的字节总和, 一个窗中的所有分组的字节计数都填同样的值。
在本发明的该实施例中, 窗内各个分组的时间戳设置为相同的, 即在 一个窗中的分组都打上的是相同的时间戳,即第一个分组的时间戳;当然, 如前面所述的, 一个窗内的时间戳可不同, 只要它们相关联, 而根据其序 列号关系可知其位置关系, 即可从一个推知其它。 网关给每个分组头上加一层同步协议后通过 IP组播的方式发送, 目 的地是该网关所属的所有基站。
每次 GW发送单元 108发送了一个分组,则 GW窗计数单元 102递增 1 , 当窗计数达到窗大小 WIND0W_S IZE时, 如 5, 则预示着下一个分组是一个 新的窗的开始, 此时将窗计数清零, 并且 GW接收单元 105等待接收下一 个 I P分组。
为了避免网关接收的下一个分组在很长时间内都没有到达网关,网关 还设置了一个定时器, 例如定时器 104 , 定时器时长为 T, T-延迟裕量 /N, 当从广播与组播业务中心发出的一个分组到达网关之后经过了 T 时间后 还没有下一个分组到来, 则定时器超时, 其中的 N是如上面所定义的数。 于是网关通过向基站发送同步控制分组来表明当前窗终止,此时虽然没有 到达规定的窗大小, 该窗也结束。 同步控制分组由同步控制分组产生单元 107产生。 同步控制分组包含了下面一些字段: 序列号, 为下一个要发送 的数据分组的序列号, 时间戳, 为同步控制分组从网关发出的时间加上最 大传输时延、最大处理时延和延迟裕量; 以及当前被终止的窗的所有分组 的字节总和。 将定时器的时长 T设置为 T-延迟裕量 /N使得该同步控制分 组可以在延迟裕量时间内连发 N个以确保可靠传输。
当下一个数据分组到达网关,则预示着一个新的窗的开始,所有操作 都一样, 虽然同步控制分组里有序列号的字段, 但该序列号是下一个要发 送的数据分组的序列号, 因此同步控制分组不在序列号累计中, 即如图 5 所示。
根据本发明的网关循环进行上述处理, 将接收的 IP分组或同步控制 分组发送到所属的基站。
图 6示出了根据本发明的基站。 根据本发明的基站 200包括: NB接 收单元 201, 用于接收分组; 信息提取单元 202, 用于从接收到的数据中 提取序列号、 时间戳和字节计数; 同步控制分组判断单元 203, 用于判断 所接收的分组是否是同步控制分组; 分组丢失判断单元 204, 用于判断是 否丢失了分组; NB窗计数单元 205, 用于对属于一个窗的 IP分组进行计 数; NB处理单元 206, 用于对属于一个窗的所有 IP分组进行处理, 基站 緩沖器 207 , 用于緩冲从网关接收到的 IP分组数据; 分组丢失处理单元 208 , 用于在分组丢失判断单元 204判断存在分组丢失的情况下进行分组 丢失处理。
如图 6所示的基站 200如图 7所示地进行操作。由于已经假定业务从 序列号为 0的第一个分组开始, 从此开始第一个窗。 当然也可以在业务数 据分组开始前先发 N个同步控制分组来向基站表明窗的起始。
从第一个窗的第一个分组开始, 基站的 NB接收单元 201接收数据分 组。基站每接收一个分组, 则信息提取单元 202将该分组的同步协议中的 下列各项字段中的值取出来: 序列号; 时间戳, 用于指示前一个窗的分组 在空口发送的时间; 字节计数, 用于恢复前一个窗可能丢掉的分组的字节 数。
其中,分组丢失判断单元 204可通过序列号的连续性来判断在接收到 的该分组之前有没有丢失数据分组, 如果丢失了, 则根据该分组与前一个 分组的序列号之差可判断丢失了几个分组。 同时, 窗计数单元 205对接收 到的数据分组的序列号进行计数, 如果计数器达到窗大小, 则预示着下一 个数据分组是一个新的窗的开始。窗是以序列号而不是接收到的数据分组 来计数的, 举个例子, 如当前窗从 0开始, 收到了 0、 1、 2、 3、 6 , 则当 前窗包含了 0、 1、 2、 3、 4 , 下一个窗是 5、 6...。 在接收到序列号为 6的 分组时, 分组丟失判断单元 203根据 6与接收到的前一个分组的序列号 3 之差, 以及窗大小 5 , 可以判断该窗丢失了 2个分组, 而下一个窗丢失了 一个分组。 在判断丟失了分组之后, 进行分组丟失的处理。
当基站每接收完一个窗的分组,并已经接收到了下一个窗中的数据分 组, 基站中的处理单元 206对接收完的窗的分组进行处理, 即将下一个窗 中的数据分组中同步协议中的时间戳赋给当前窗的数据分组,并命名为允 许空口传输时刻,该窗的所有数据分组附带着允许空口传输时刻被放置到 层 2 ( RLC )緩存中。 允许空口传输时刻, 指的是该数据分组被填充到空 口协议帧的传输块中,该传输块在空口被发走的时间必须恰好等于或晚于 该时刻。
当基站每接收完一个窗的分组,并已经接收到了下一个窗中的数据分 组, 则该个窗的所有分组可以被基站 L2用户接口所处理, 并在空口上发 走。 当然, 该个窗之前的所有窗的分组若没有被空口发走, 而留在 L2緩
I I 存中, 则需要依次在空口发走。
接收到同歩控制分组的处理
如果基站的同步分组判断单元 203 判断接收到了一个或数个连续的 同步控制分组, 则预示着当前窗被同步控制分组所终止, 当前窗的最后一 个分组是同步控制分组中的序列号 -1。 此时, 窗内的分组如图 5所示。 所 有的处理与完全接收到一个窗的分组时的处理是一样的,即基站将同步控 制分组中的时间戳赋给当前窗的数据分组, 并命名为允许空口传输时刻, 该窗的所有数据分组附带着允许空口传输时刻被放置到 L2 ( RLC )緩存中。
丟失分组的处理
如果基站接收到一个分组,而分组丢失判断单元 204发现接收到的分 组的序列号与前一个接收到的分组的序列号不连续, 则表明有分组丢失, 需要由分组丢失处理单元 208执行同步恢复处理。基站每接收完一个窗的 分组, 并已经接收到了下一个窗中的数据分组, 或者当前窗是由同步控制 分组终止的, 则基站统计当前窗的所有接收到的分组的字节总和, 然后取 出下一个窗中的数据分组或同步控制分组中的字节计数。两者相减得到的 就是当前窗所丢失的所有分组的字节总数。由不连续的序列号可以知道当 前窗所丟失的分组的个数,恢复的方法是重新生成分组,若分组数等于 1, 则长度为确定的当前窗所丢失的单个分组的字节数; 若分组数大于 1 , 则 总长度为确定的当前窗所丢失的所有分组的字节总数,各个分组具体的长 度可以是随意也可以是平均分配。 重新生成的分组标识为是假的, 分组的 数据内容无效, 只是总长度与真的分组一样, L2 知晓哪些分组是恢复的 假分组,空口传输的时候, 包含这些假分组的传输块不发射功率,为静默。
在本发明中, 该基站还可包括一个警报单元 209, 用于在出现一个窗 大小的连续分组或 N个连续同步控制帧丢失了的情况下,向网关或 0&M通 知出现异常网络传输问题。
ALU和 Nok i a方案的性能比较
在 R3-071920 MB SFN schedu l ing and content synchron i za t ion, Nok i a, Nok ia S iemens Network, 2007- 10中, Nok ia仅提出了基于固定 突发长度判断的内容同步方案。 其 "总计数帧"与本发明的同步控制分组 类似, 同样具有类似的信息, 包括前一突发的总字节数(在本发明中是前 一窗)、 时间戳、 前一突发分组的总数。 但是, 在 Nok i a方案中, 首先定 义了固定长度的突发, 例如 1秒或其它值。 "总计数帧" 的产生频率是根 据该突发长度定义的。 同步处理延迟也受该突发长度的影响。根据本发明 的针对 Nok i a方案的仿真, 在同步层处理延迟和控制分组数的性能方面, 仅设置为 0. 5秒的突发定时长度是可接受的。
在本发明的 ALU 方案中, 窗大小和突发定时长度这两个因素影响性 能。组合这两个参数将更加灵活地控制同步层处理延迟和控制分组数的性 能。 根据仿真, 针对 ALU方案的适当参数设置是窗大小 =5并且突发定时 长度 =0. 1秒。
针对 Nok i a的参数设置是突发定时长度 =0. 5秒。
然后,在上述参数设置下同步层处理延迟和控制分组数性能的比较如 下。
ALU方案的同步层处理延迟性能是 0. 6768秒, Nok i a的是 0. 6582秒。 两者之差是 0. 0186秒。 这两个方案具有类似的同步层处理延迟性能。 但 是两个方案之间的控制分组数有很大的不同。 ALU方案仅产生 695个控制 分组, 但是 Nok i a方案具有 1 1989个控制分组, 是 ALU方案的 17. 25倍。 因此, 从综合角度看, ALU方案优于 Nok i a方案。 工业实用性
本发明的内容同步在统计复用下也能良好工作,并且可满足基本的 1 秒处理延迟的要求, 并且有更少的控制分组负荷。 上面的描述仅用于实现本发明的实施方式,本领域的技术人员应该理 解, 在不脱离本发明的范围的任何修改或局部替换, 均应该属于本发明的 权利要求来限定的范围, 因此, 本发明的保护范围应该以权利要求书的保 护范围为准。

Claims

权 利 要 求 书
1. 一种网关, 包括:
接收单元, 接收要发送到基站的分组数据;
窗计数单元, 用于对要发送的分组数据进行计数, 使窗大小数目的 分组组成一个窗;
序列号计数单元, 用于产生要发送的分组的序列号;
时间戳设置单元, 用于设置整个窗中的分组的时间戳, 该窗中的所 有分组的时间戳都相关联; 以及
处理单元, 将所产生的序列号、 时间戳、 以及字节计数包括在要发 送的分组中, 所述字节计数包含前一个窗的分组的字节总和, 一个窗中 的所有分组的字节计数都相同。
2. 根据权利要求 1 所述的网关, 其中处理单元对要发送的分组的 头上加一层同步协议, 该同步协议包含如下三项内容: 序列号; 时间戳; 字节计数。
3. 根据权利要求 1或 2所述的网关, 其中一个窗中所有的分组包 括相同的时间戳, 并且该时间戳是该窗的第一个分组从网关发送的时刻 加上最大传输时延、 最大处理时延和延迟裕量, 最大传输时延是分组在 从网关到所有所属基站的网络中传输的最大时延, 最大处理时延是分组 在所有基站处从层 2处理到从空口发出所需的最大处理时延。
4. 根据权利要求 1或 2所述的网关, 还包括定时器, 用于对接收 单元未接收到数据的时间进行计时; 以及同步控制分组产生单元, 用于 在定时器计时满时产生同步控制分组; 其中, 发送单元还发送所产生的 同步控制分组。
5. 根据权利要求 4所述的网关, 其中同步控制分组包括: 序列号, 为下一个要发送的数据分组的序列号'; 时间戳, 为网关产生该同步控制 分组的时刻加上最大传输时延、 最大处理时延和延迟裕量; 以及同步控 制分组前被终止的窗的总发送分组字节数。
6. 根据权利要求 5 所述的网关, 其中定时器的时长设置为延迟裕 量 /同步控制分组的预定重复传输次数 N。
7. 根据权利要求 1 所述的网关, 还包括窗设置单元, 用于设置窗 的大小, 并告知所有所属基站。
8. 一种基站, 包括:
接收单元, 用于从网关接收分组数据;
信息提取单元, 用于从接收到的分组数据中提取分组的序列号、 时 间戳、 字节计数信息;
窗计数单元, 用于根据所提取的序列号, 对接收的分组进行计数, 以形成一个窗; 以及
处理单元, 用于对整个窗的分组数据进行处理, 根据所获取的时间 戳来设置当前窗的允许空口传输时刻。
9. 根据权利要求 8 所述的基站, 其中将当前窗的允许空口传输时 刻设置为下一个窗中的数据分组中同步协议中的时间戳。
1 0. 根据权利要求 8或 9所述的基站, 还包括同步控制分组判断单 元, 用于对接收的分组数据进行分析, 以判断接收到的是否是同步控制 分组。
1 1. 根据权利要求 1 0所述的基站, 其中, 所述同步控制分组包括 序列号, 为下一个要发送的分组的序列号; 时间戳; 以及同步控制分组 前被终止的窗的总发送分组字节数。
1 2. 根据权利要求 1 1 所述的基站, 其中, 在接收到的分组是同步 控制分组时, 窗计数单元根据同步控制分组中的序列号, 形成一个窗。
1 3. 根据权利要求 8或 12所述的基站, 还包括分组丢失判断单元, 用于根据接收到的分组的序列号或同步控制分组的序列号, 判断是否存 在分组丢失, 如果存在分组丢失, 则进行分组丟失处理。
14. 根据权利要求 1 3所述的基站, 其中, 分组丟失判断单元根据 前一个接收到的分组的序列号和当前接收到的分组的序列号或同步控制 分组的序列号是否连续, 来判断是否存在分组丢失。
15. 根据权利要求 14 所述的基站, 其中分组丢失处理包括: 重新 生成分组, 重新生成的分组的个数为不连续序列号之差, 分组的字节总 数是当前窗的所有接收到的分组的字节总和与下一个窗的数据分组或同 步控制分组中的字节计数之差。
1 6. 根据权利要求 1 5 所述的基站, 其中在要重新生成的分组的个 数大于 1时, 可随意选择或平均分配各个要重新生成的分组的长度。
1 7. 根据权利要求 1 3所述的基站, 还包括警报单元, 用于在出现 一个窗大小的连续分组或预定重复传输次数 N个连续同步控制分组丢失 时, 向网关通知出现网络传输问题。
1 8. 根据权利要求 1 4 所述的基站, 还包括警报单元, 用于在出现 一个窗大小的连续分组或预定重复传输次数 N个连续同步控制分组丢失 时, 向网关通知出现网络传输问题。
1 9. 一种在网关中的用于内容同步的分组数据处理方法, 包括以下 步骤:
接收要发送到基站的分组数据;
对要发送的分组数据进行计数, 使窗大小数目的分组组成一个窗; 产生要发送的分组的序列号;
设置整个窗中的分组的时间戳, 该窗中的所有分组的时间戳都相关 联; 以及
将所产生的序列号、时间戳、以及字节计数包括在要发送的分组中, 所述字节计数包含前一个窗的分组的字节总和, 一个窗中的所有分组的 字节计数都相同。
2 0. 根据权利要求 1 9 所述的方法, 其中将所产生的序列号、 时间 戳、 以及字节计数包括在要发送的分组中的步骤包括: 对要发送的分组 的头上加一层同步协议, 该同步协议包含如下三项内容: 序列号; 时间 戳; 字节计数。
2 1. 根据权利要求 1 9或 20所述的方法, 其中一个窗中所有的分组 包括相同的时间戳, 并且该时间戳是该窗的第一个分组从网关发送的时 刻加上最大传输时延、 最大处理时延和延迟裕量, 最大传输时延是分组 在从网关到所有所属基站的网络中传输的最大时延, 最大处理时延是分 组在所有基站处从层 2处理到从空口发出所需的最大处理时延。
22. 根据权利要求 19或 20所述的方法, 还包括步骤: 对接收单元 未接收到数据的时间进行计时;以及在计时时长满时产生同步控制分组, 并发送所产生的同步控制分组。
23. 根据权利要求 22 所述的方法, 其中同步控制分组包括: 序列 号, 为下一个要发送的数据分组的序列号; 时间戳, 为该同步控制分组 的产生时刻加上最大传输时延、 最大处理时延和延迟裕量; 以及同步控 制分组前被终止的窗的总发送分组字节数。
24. 根据权利要求 23所述的方法, 其中计时时长设置为延迟裕量 / 同步控制分组的预定重复传输次数 N。
25. 根据权利要求 19 所述的方法, 还包括步骤: 设置窗的大小, 并告知网关所属所有基站。
26. 一种在基站中的用于内容同步的数据分组处理方法, 包括以下 步骤:
从网关接收分组数据;
从接收到的分组数据中提取分组的序列号、时间戳、字节计数信息; 根据所提取的序列号, 对接收的分组进行计数, 以形成一个窗; 以 及
对整个窗的分组数据进行处理, 根据所获取的时间戳来设置当前窗 的允许空口传输时刻。
27. 根据权利要求 26 所述的方法, 根据所获取的时间戳来设置当 前窗的允许空口传输时刻的步骤包括: 将当前窗的允许空口传输时刻设 置为下一个窗中的数据分组中同步协议中的时间戳。
28. 根据权利要求 26或 27所述的方法, 还包括: 对接收的分组数 据进行分析, 以判断接收到的是否是同步控制分组。
29. 根据权利要求 28所述的方法, 其中, 所述同步控制分组包括: 序列号, 为下一个要发送的分组的序列号; 时间戳; 以及同步控制分组 前被终止的窗的总发送分组字节数。
30. 根据权利要求 29 所述的方法, 其中, 在接收到的分组是同步 控制分组时, 居同步控制分组中的序列号, 形成一个窗。
31. 根据权利要求 26或 30所述的方法, 还包括: 根据接收到的分 组的序列号或同步控制分组的序列号, 判断是否存在分组丢失, 如果存 在分组丢失, 则进行分组丟失处理。
32. 根据权利要求 31 所述的方法, 其中, 根据前一个接收到的分 组的序列号和当前接收到的分组的序列号或同步控制分组的序列号是否 连续, 来判断是否存在分组丢失。
33. 根据权利要求 32 所述的方法, 其中分组丢失处理包括: 重新 生成分组, 重新生成的分组的个数为不连续序列号之差, 分组的字节总 数是当前窗的所有接收到的分组的字节总和与下一个窗的数据分组或同 步控制分组中的字节计数之差。
34. 根据权利要求 33所述的方法, 其中在要重新生成的分组的个 数大于 1时, 可随意选择或平均分配各个要重新生成分组的长度。
35. 根据权利要求 31 所述的方法, 还包括步骤: 在出现一个窗大 小的连续分组或预定重复传输次数 N个连续同步控制分组丟失时, 向网 关通知出现网络传输问题。
36. 根据权利要求 32 所述的方法, 还包括步骤: 在出现一个窗大 小的连续分组或预定重复传输次数 N个连续同步控制分组丟失时, 向网 关通知出现网络传输问题。
PCT/CN2009/000109 2008-02-02 2009-01-23 基于窗的内容同步的网关和基站及其方法 Ceased WO2009097744A1 (zh)

Priority Applications (5)

Application Number Priority Date Filing Date Title
ES09709304.1T ES2524592T3 (es) 2008-02-02 2009-01-23 Pasarela y estación de base basadas en la sincronización de contenido de ventana y su procedimiento
JP2010544564A JP5296102B2 (ja) 2008-02-02 2009-01-23 ウィンドウベースのコンテンツ同期化のためのゲートウェイ、基地局および方法
EP09709304.1A EP2239880B1 (en) 2008-02-02 2009-01-23 Gateway and base station based on content synchronization of window and its method
US12/865,777 US8509240B2 (en) 2008-02-02 2009-01-23 Gateway, base station and method for window-based content synchronization
KR1020107019394A KR101503247B1 (ko) 2008-02-02 2009-01-23 윈도우의 콘텐트 동기화 기반 게이트웨이, 기지국 및 방법

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN200810033551.2A CN101500341B (zh) 2008-02-02 2008-02-02 基于窗的内容同步的网关和基站及其方法
CN200810033551.2 2008-02-02

Publications (1)

Publication Number Publication Date
WO2009097744A1 true WO2009097744A1 (zh) 2009-08-13

Family

ID=40947154

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/000109 Ceased WO2009097744A1 (zh) 2008-02-02 2009-01-23 基于窗的内容同步的网关和基站及其方法

Country Status (7)

Country Link
US (1) US8509240B2 (zh)
EP (1) EP2239880B1 (zh)
JP (1) JP5296102B2 (zh)
KR (1) KR101503247B1 (zh)
CN (1) CN101500341B (zh)
ES (1) ES2524592T3 (zh)
WO (1) WO2009097744A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012039349A (ja) * 2010-08-06 2012-02-23 Nec Engineering Ltd 宇宙機器の同期化システム及びこれに用いる宇宙機器

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006105010A1 (en) 2005-03-25 2006-10-05 Neocific, Inc. Methods and apparatus for cellular broadcasting and communication system
CN101998285B (zh) 2005-06-09 2012-12-12 桥扬科技有限公司 用于高功率效率的广播和通信系统的方法和设备
EP1798998B1 (en) 2005-12-14 2011-06-15 Research In Motion Limited Method and apparatus for user equipment directed radio resource control in a UMTS network
CN101132222B (zh) * 2006-08-22 2011-02-16 上海贝尔阿尔卡特股份有限公司 网关设备、通信网络和同步方法
GB0711833D0 (en) * 2007-06-18 2007-07-25 Nokia Siemens Networks Oy A method for providing a plurality of services
WO2011061352A1 (en) * 2009-11-23 2011-05-26 Research In Motion Limited Method and apparatus for state/mode transitioning
US8656170B2 (en) * 2010-05-28 2014-02-18 Cisco Technology, Inc. Protection of control plane traffic against replayed and delayed packet attack
WO2012050838A1 (en) 2010-09-28 2012-04-19 Neocific, Inc. Methods and apparatus for flexible use of frequency bands
WO2012044234A1 (en) 2010-09-29 2012-04-05 Telefonaktiebolaget L M Ericsson (Publ) Determining loss of ip packets
CN102065472B (zh) * 2010-12-09 2013-02-06 重庆邮电大学 基于跨层设计的确定无线网络编码参与分组的方法
WO2012126418A2 (zh) * 2012-05-09 2012-09-27 华为技术有限公司 传输信息的方法和装置
WO2013171745A1 (en) 2012-05-13 2013-11-21 Lacoon Security Ltd. Anti-malware detection and removal systems and methods
WO2014036728A1 (zh) * 2012-09-07 2014-03-13 华为技术有限公司 一种空口传输方法及相关设备、系统
RU2603695C1 (ru) * 2012-10-31 2016-11-27 Телефонактиеболагет Л М Эрикссон (Пабл) РАСПРЕДЕЛЕНИЕ ПАКЕТОВ СИНХРОНИЗАЦИИ ПО ТРАНСПОРТНЫМ ЛИНИЯМ СВЯЗИ Wi-Fi
US20140293825A1 (en) * 2013-03-27 2014-10-02 Broadcom Corporation Timestamping data packets
US9525611B2 (en) 2014-01-27 2016-12-20 Imagine Communications Corp. Transmission system implementing delay measurement and control
CN105450356B (zh) * 2014-09-03 2019-10-11 中兴通讯股份有限公司 丢失数据的恢复处理方法及装置
CN105471825B (zh) * 2014-09-03 2019-08-23 南京中兴软件有限责任公司 丢失数据报文的识别方法及装置
US9910809B2 (en) 2014-12-19 2018-03-06 Intel Corporation High performance interconnect link state transitions
US9526078B1 (en) 2015-01-26 2016-12-20 Sprint Communications Company L.P. Wireless communication system to synchronize a single frequency network
US10313143B1 (en) * 2015-02-23 2019-06-04 Sprint Communications Company L.P. Wireless communication system to provide buffering in a single frequency network
US10212065B2 (en) 2016-10-20 2019-02-19 Gatesair, Inc. Extended time reference generation
US12382406B2 (en) 2020-03-02 2025-08-05 Telefonaktiebolaget Lm Ericsson (Publ) Over the air antenna synchronization in wireless communication network
WO2022012274A1 (zh) * 2020-07-14 2022-01-20 上海朗帛通信技术有限公司 一种副链路中继无线通信的方法和装置
WO2022030454A1 (ja) * 2020-08-03 2022-02-10 京セラ株式会社 通信制御方法
CN115175300B (zh) * 2022-08-17 2023-10-31 上海磐启微电子有限公司 一种蓝牙自组网系统的时钟校准方法

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1110932C (zh) * 1998-11-30 2003-06-04 摩托罗拉公司 通信系统中用于将数据传送到基站的数据网关和方法
CN1866830A (zh) * 2005-07-29 2006-11-22 华为技术有限公司 通信系统中媒体网关实现同步的方法

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FI106498B (fi) * 1999-03-16 2001-02-15 Nokia Mobile Phones Ltd Menetelmä siirtoikkunan valinnassa ja matkaviestin
US6941132B2 (en) * 2000-03-20 2005-09-06 Telefonaktiebolaget L M Ericsson (Publ) Transport of radio network-originated control information
CN1145327C (zh) * 2001-06-05 2004-04-07 华为技术有限公司 通用分组无线业务网络节点向计费网关发送话单的方法
US20050152397A1 (en) * 2001-09-27 2005-07-14 Junfeng Bai Communication system and techniques for transmission from source to destination
KR100547871B1 (ko) * 2001-11-08 2006-02-01 삼성전자주식회사 이동 통신시스템의 기지국 제어기와 기지국간 패킷 데이터재전송 방법 및 장치
US7620071B2 (en) * 2004-11-16 2009-11-17 Intel Corporation Packet coalescing
US7590165B2 (en) * 2005-07-08 2009-09-15 Industrial Technology Research Institute Sub-system based window placement and size decision mechanism and method thereof
CN101039175B (zh) * 2007-03-27 2011-08-24 中兴通讯股份有限公司 Rlc在无线网络控制器间同步的方法及系统

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1110932C (zh) * 1998-11-30 2003-06-04 摩托罗拉公司 通信系统中用于将数据传送到基站的数据网关和方法
CN1866830A (zh) * 2005-07-29 2006-11-22 华为技术有限公司 通信系统中媒体网关实现同步的方法

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012039349A (ja) * 2010-08-06 2012-02-23 Nec Engineering Ltd 宇宙機器の同期化システム及びこれに用いる宇宙機器

Also Published As

Publication number Publication date
EP2239880B1 (en) 2014-09-24
US20110002311A1 (en) 2011-01-06
JP5296102B2 (ja) 2013-09-25
EP2239880A4 (en) 2013-01-02
KR20100113614A (ko) 2010-10-21
EP2239880A1 (en) 2010-10-13
ES2524592T3 (es) 2014-12-10
KR101503247B1 (ko) 2015-03-17
US8509240B2 (en) 2013-08-13
CN101500341A (zh) 2009-08-05
CN101500341B (zh) 2011-02-02
JP2011514705A (ja) 2011-05-06

Similar Documents

Publication Publication Date Title
WO2009097744A1 (zh) 基于窗的内容同步的网关和基站及其方法
US8005085B2 (en) Method for distribution of data packets in a single frequency mobile communication network, an access network node, a base station and a single frequency mobile communication network therefor
CN101741701B (zh) 同步调度方法和装置
CN105898857B (zh) 一种数据同步方法和系统
CN101741538B (zh) 同步调度方法
EP2320599B1 (en) Method and apparatus for recovery processing of synchronously transmitted service data
TWI429221B (zh) 電信系統中廣播/多播資料之分佈
CN101981874B (zh) 在网关和基站之间进行同步的方法及相应的基站
CN101686431B (zh) 同步处理方法和装置
CN101651517A (zh) 突发序列丢失的检测方法和装置
WO2015154523A1 (zh) 丢失数据的恢复处理方法及装置
CN101651605A (zh) 数据发送和接收方法、数据收发系统
WO2011050681A1 (zh) 一种mbms同步的方法、设备和系统
CN101651488A (zh) 一种多媒体广播组播业务网元间的协同方法
CN102056084B (zh) 多媒体广播组播业务系统及其同步方法
Richards et al. Network/content synchronisation for MBSFN transmission in 3GPP networks

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09709304

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 4768/CHENP/2010

Country of ref document: IN

WWE Wipo information: entry into national phase

Ref document number: 12865777

Country of ref document: US

Ref document number: 2010544564

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2009709304

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20107019394

Country of ref document: KR

Kind code of ref document: A