WO2016000191A1 - 传输拥塞的确定方法及装置 - Google Patents
传输拥塞的确定方法及装置 Download PDFInfo
- Publication number
- WO2016000191A1 WO2016000191A1 PCT/CN2014/081307 CN2014081307W WO2016000191A1 WO 2016000191 A1 WO2016000191 A1 WO 2016000191A1 CN 2014081307 W CN2014081307 W CN 2014081307W WO 2016000191 A1 WO2016000191 A1 WO 2016000191A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- packets
- actual number
- lost
- transmission
- variance
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W28/00—Network traffic management; Network resource management
- H04W28/02—Traffic management, e.g. flow control or congestion control
Definitions
- the present invention relates to the field of wireless communications, and in particular, to a method and apparatus for determining transmission congestion. Background technique
- Transmission congestion means that the packet transmission requirement exceeds the transmission network's ability to transmit data packets. If the packet transmission requirement is not reduced, the transmission system will discard the data packets that cannot be sent, resulting in an increase in system packet loss. In order to avoid transmission congestion caused by excessive transmission rate, thereby increasing the transmission delay and increasing the packet loss rate, the data transmission rate between the data transmitting host and the data receiving host is usually controlled to maintain an appropriate value. Since this control method needs to adjust the transmission network interface wideband according to the transmission congestion condition, it is necessary to determine whether to transmit congestion in real time.
- the method used is: the number of packets sent by the statistical data transmitting host in a preset number of transmission periods and the number of packets received by the data receiving host in the predetermined number of transmission periods; The number and the number of packets are determined to determine the packet loss rate of the data transmission; whether the packet loss rate is greater than a preset threshold; if the packet loss rate is greater than the preset threshold, the transmission congestion is determined.
- the related art has at least the following problems: Since the related technology determines the packet loss rate, if the packet loss rate is greater than the preset threshold, the transmission congestion is determined. That is to say, the related art considers the packet loss rate to be caused by transmission congestion. However, in the process of data transmission, in addition to transmission congestion can cause packet loss, transmission errors can also result in packet loss. If the system has a packet loss rate due to a transmission error, the transmission congestion determined according to the packet loss rate is inaccurate. In addition, when the packet loss rate due to the transmission error is higher than the preset threshold, the packet loss rate caused by the transmission error is determined to be caused by the transmission congestion, so that the transmission network interface bandwidth is reduced, resulting in reduced transmission. The throughput of the network. Summary of the invention
- embodiments of the present invention provide a method and apparatus for determining transmission congestion.
- the technical solution is as follows:
- a method for determining transmission congestion includes:
- the first linear relationship between the actual number of packets sent and the actual number of packets lost during non-transmission congestion is determined according to the actual number of packets sent and the number of actual packets lost in each transmission cycle, and the actual number of packets and the actual number of packets lost according to each transmission cycle.
- Determining a second linear relationship between the actual number of packets sent and the actual number of packets lost during transmission congestion determining, according to the first linear relationship, the first theoretical packet loss corresponding to the actual number of packets sent in each transmission period, and according to each transmission cycle The actual number of lost packets and the first theoretical number of lost packets determine a first parameter; determining, according to the second linear relationship, a second theoretical packet loss number corresponding to the actual number of packets sent per transmission period, and actually losing according to each transmission cycle The number of packets and the second theoretical packet loss determine a second parameter; determining whether to transmit congestion according to a comparison result of the first parameter and the second parameter.
- the actual number of packets sent and the actual number of packets lost during non-transmission congestion determined according to the actual number of packets sent and the actual number of packets lost per transmission period The first linear relationship is:
- ⁇ is the number of lost packets when the transmission is not congested
- ⁇ ' is the number of lost packets when the transmission is congested
- 5 is the number of packets sent
- ⁇ is the congestion threshold, which is the packet loss rate caused by the transmission error.
- the first variance is taken as the first parameter
- the second variance is determined according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period, and the second variance is used as the second parameter.
- the determining the first variance according to the actual number of lost packets and the first theoretical packet loss in each transmission period Includes:
- the first variance is determined by the following formula according to the actual number of lost packets and the first theoretical packet loss in each transmission period:
- VNCMin The second variance is determined according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period, and the second variance is determined by the following formula according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period:
- WVCMVi is the first variance
- VCM ⁇ is the second variance
- 5 is the actual number of packets sent per transmission period
- R is the actual number of packets received per transmission period, 7. for each transmission cycle
- the actual number of lost packets, N is the default value.
- the determining the first variance according to the actual number of lost packets and the first theoretical packet loss in each transmission period include: The first variance is determined by the following formula according to the actual number of lost packets and the first theoretical packet loss in each transmission period: Determining the second variance according to the actual number of lost packets and the second theoretical packet loss in each transmission period, including:
- the second variance is determined by the following formula according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period:
- VCM Z is the second variance, 5 ; is the actual number of packets sent per transmission cycle, R ; is the actual number of packets received per transmission cycle, 7; for each transmission cycle
- the actual number of lost packets, N is a preset value, and is widely used as the weight of the first packet of the first transmission period and the actual number of packets lost when calculating the second variance.
- the second aspect further provides a determining device for transmitting congestion, where the device includes: an obtaining module, configured to acquire an actual number of packets sent and a number of actual packets lost in a preset number of transmission periods; and a first determining module, configured to The actual number of packets sent and the actual number of packets lost during each transmission period determine the first linear relationship between the actual number of packets sent and the actual number of packets lost during non-transmission congestion;
- a second determining module configured to determine, according to the actual number of packets sent and the number of actual packets lost in each transmission period, a second linear relationship between the actual number of packets sent and the actual number of packets lost during transmission;
- a third determining module configured to determine, according to the first linear relationship, a first theoretical packet loss number corresponding to an actual number of packets sent by each transmission period;
- a fourth determining module configured to determine a first parameter according to an actual number of lost packets and a first theoretical number of lost packets in each transmission period
- a fifth determining module configured to determine, according to the second linear relationship, a second theoretical packet loss number corresponding to the actual number of packets sent in each transmission period
- a sixth determining module configured to determine a second parameter according to the actual number of lost packets in each transmission period and the second theoretical lost packet number
- a seventh determining module configured to determine, according to a comparison result of the first parameter and the second parameter, whether to transmit congestion.
- the actual number of packets sent and the actual number of packets lost during non-transmission congestion determined according to the actual number of packets sent and the actual number of packets lost in each transmission period
- the first linear relationship is:
- ⁇ is the number of lost packets when the transmission is not congested
- ⁇ ' is the number of lost packets when the transmission is congested
- 5 is the number of packets sent
- ⁇ is the congestion threshold, which is the packet loss rate caused by the transmission error.
- the fourth determining module is configured to use the actual number of lost packets according to each transmission period
- the first theoretical packet loss number determines a first variance, and the first variance is used as a first parameter
- the sixth determining module is configured to determine a second variance according to the actual number of lost packets and the second theoretical packet loss in each transmission period, and use the second variance as the second parameter.
- the fourth determining module is configured to use the actual number of lost packets and the first theoretical loss according to each transmission period.
- the number of packets determines the first variance by the following formula:
- the sixth determining module is configured to determine the second variance according to an actual packet loss number and a second theoretical packet loss number of each transmission period by using the following formula:
- WVCMVi is the first variance
- VCM ⁇ is the second variance
- 5 is the actual number of packets sent per transmission cycle
- R is the actual number of packets received per transmission cycle
- 7. is the actual number of each transmission cycle
- N is the default value.
- the fourth determining module is configured to use the actual number of lost packets and the first theoretical loss according to each transmission period.
- the number of packets determines the first variance by the following formula:
- the sixth determining module is configured to determine the second variance according to an actual packet loss number and a second theoretical packet loss number of each transmission period by using the following formula:
- VCM Z is the second variance, 5 ; is the actual number of packets sent per transmission cycle, R ; is the actual number of packets received per transmission cycle, 7. is for each transmission cycle
- the actual number of lost packets, N is a preset value, and is widely used as the weight of the first packet of the first transmission period and the actual number of packets lost when calculating the second variance.
- the number of actual packet loss may be due to packet loss and transmission error caused by transmission congestion.
- the first linearity of non-transmission congestion is determined by the actual number of packets sent according to the preset number of transmission periods and the actual number of packets lost. Relationship and the second linear relationship when transmitting congestion, according to the actual number of lost packets Determining whether to transmit congestion by comparing the first parameter and the second parameter determined according to the theoretical number of lost packets according to the first linear relationship and the second linear relationship, so that the determined transmission congestion is more accurate, and thus the appropriate transmission can be controlled.
- the network interface bandwidth thereby increasing the throughput of the transmission network.
- FIG. 1 is a schematic diagram of an implementation environment involved in a method for determining transmission congestion according to an embodiment of the present invention
- FIG. 2 is a schematic diagram of a relationship between the number of packets sent and the number of packets lost according to another embodiment of the present invention
- FIG. 3 is a flowchart of a method for determining transmission congestion according to another embodiment of the present invention.
- FIG. 5 is a non-congestion fitting line indicating a first linear relationship and a second linear relationship according to another embodiment of the present invention; Congestion fitting straight line;
- FIG. 6 is a schematic structural diagram of a determining apparatus for transmitting congestion according to another embodiment of the present invention
- FIG. 7 is a schematic structural diagram of a determining apparatus for transmitting congestion according to another embodiment of the present invention.
- FIG. 1 it shows a schematic diagram of an implementation environment involved in a method for determining transmission congestion provided by an embodiment of the present invention.
- the implementation environment is a data transmission system including a data transmission host 101 and a data receiving host 102.
- the data sending host 101 and the data receiving host 102 are connected by a transmission network, and the transmission network may be a wired network, or For wireless networks.
- the data transmission system needs to control the data transmission rate between the data transmitting host 101 and the data receiving host 102 to maintain an appropriate value to adjust the transmission network interface bandwidth according to the data transmission rate.
- the data sending host 101 and the data receiving host 102 are not specifically limited in the embodiment of the present invention.
- the data sending host 101 includes, but is not limited to, an RNC (Radio Network Controller), and the data receiving host 102 includes, but is not limited to, a Node B.
- RNC Radio Network Controller
- the cause of packet loss on the transmission network is transmission error, in addition to transmission congestion.
- the method for determining transmission congestion provided by the embodiment of the present invention is to solve the problem that the allocating packets are determined to be caused by transmission congestion when the congestion is transmitted according to the packet loss rate in the related art, thereby causing the determined transmission congestion to be inaccurate.
- a new method for determining transmission congestion is proposed.
- FIG. 2 shows a schematic diagram of the relationship between the number of packets sent and the number of packets lost. It can be seen from FIG. 2 that when there is no packet loss caused by transmission congestion in the system, only when there is a packet loss caused by a transmission error, that is, when the transmission is not congested, the function of the number of packets and the number of packets lost is a straight line passing through the origin; At the same time, when there is packet loss caused by packet loss and transmission error caused by transmission congestion, that is, when congestion is transmitted, the function of the number of packets and the number of packets lost is a straight line with a fixed slope of 1.
- the intersection of the dotted line and the horizontal axis between the non-transmission congestion and the transmission congestion is a congestion threshold, and when the number of packets in one transmission period is less than the congestion threshold, transmission congestion is not caused; When the number of packets sent in a transmission period is greater than the congestion threshold, transmission congestion may result.
- the method according to the embodiment of the present invention obtains the actual number of packets sent by the preset number of transmission periods and the actual number of lost packets according to the relationship between the number of packets sent by the non-transmission congestion and the transmission congestion and the number of packets lost.
- the actual number of packets sent per transmission cycle and the actual number of packets lost determine the non-transmission congestion
- the first linear relationship between the number of outgoing packets and the actual number of lost packets, according to the actual number of packets sent and the actual number of lost packets in each transmission cycle, the second between the actual number of packets sent and the actual number of packets lost during transmission congestion is determined.
- the linear relationship determines whether to transmit congestion according to a comparison result between the first packet and the second parameter between the actual packet loss number and the theoretical packet loss number obtained according to the first linear relationship and the second linear relationship.
- the specific method of determining the transmission congestion is described in detail in the following embodiments.
- the embodiment of the present invention provides a method for determining transmission congestion, in conjunction with the schematic diagram of the implementation environment shown in FIG. 1 and the relationship between the number of packets and the number of packet loss shown in FIG. 2 and corresponding contents. Referring to FIG. 3, the method process provided by the embodiment of the present invention includes:
- the first linear relationship between the actual number of packets sent and the number of actual packets lost according to the actual number of packets sent and the number of actual packets lost per transmission period is:
- ⁇ is the number of lost packets when the transmission is not congested
- ⁇ ' is the number of lost packets when the transmission is congested
- 5 is the number of packets sent
- ⁇ is the congestion threshold, which is the packet loss rate caused by the transmission error.
- determining the first parameter according to the actual number of lost packets and the number of the first theoretical packet loss in each transmission period including:
- Determining the second parameter according to the actual number of lost packets in each transmission period and the second theoretical number of lost packets including: determining a second variance according to the actual number of lost packets and the number of second theoretical packet loss in each transmission period, and using the second variance as the first Two parameters.
- determining the first variance according to the actual number of lost packets in each transmission period and the first theoretical packet loss including:
- the first difference is determined by the following formula:
- the second variance is determined according to the actual number of lost packets in each transmission period and the second theoretical number of lost packets, including: determining the number according to the actual number of lost packets and the second theoretical packet loss according to each transmission period by using the following formula
- VCM ⁇ is the second variance, 5 ; is the actual number of packets sent per transmission period, R ; is the actual number of packets received per transmission period, 7. is the actual number of each transmission cycle
- N is the default value.
- determining the first variance according to the actual number of lost packets and the number of the first theoretical packet loss in each transmission period including: According to the actual number of lost packets in each transmission cycle and the number of first theoretical packet loss, the following formula is determined by the following formula.
- the second variance is determined according to the actual number of lost packets in each transmission period and the second theoretical number of lost packets, including: determining the second variance according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period by using the following formula:
- VCM Z is the second variance, 5 ; is the actual number of packets sent per transmission cycle, R ; is the actual number of packets received per transmission cycle, 7. is for each transmission cycle
- the actual number of lost packets, N is the default value, which is the weight of the number of packets sent in the first transmission cycle and the actual number of packets lost in the calculation of the second variance.
- the method provided by the embodiment of the present invention may cause packet loss caused by packet loss and transmission error caused by transmission congestion in the actual number of lost packets, and is determined by the actual number of packets sent according to the preset number of transmission periods and the actual number of packets lost.
- the comparison result of the parameters determines whether the transmission is congested, so that the determined transmission congestion is more accurate, so that the allocation of the appropriate transmission network interface bandwidth can be controlled, thereby improving the throughput of the transmission network. Referring to FIG.
- an embodiment of the present invention provides a method for determining transmission congestion.
- the first variance determined according to the actual number of lost packets and the first theoretical packet loss according to each transmission period, and the actual number of lost packets and the second theoretical packet loss according to each transmission period are determined.
- the second variance is a first parameter and a second parameter, respectively, so as to determine whether to transmit congestion according to the comparison result of the first variance and the second variance to describe the method for determining transmission congestion provided by the embodiment of the present invention.
- the method flow provided by the embodiment of the present invention includes:
- the manner of obtaining the actual number of packets to be sent and the number of actual packets to be transmitted in a predetermined number of transmission periods is not specifically limited in the embodiment of the present invention.
- the data sending host counts the number of packets sent when it reaches a transmission period.
- the data receiving host counts the number of received data packets when it reaches a transmission period.
- the data transmission system determines and stores the actual number of packets sent according to the number of data packets sent by the data transmitting host in one transmission cycle, and according to the number of data packets sent by the data transmission host in one transmission cycle and the data receiving host
- the number of packets received during a transmission cycle determines the actual number of packets lost.
- the actual number of packets sent and the actual number of packets lost during a transmission cycle can be obtained.
- the actual number of packets and the actual number of packets lost in a predetermined number of transmission periods can be obtained.
- the embodiment of the present invention does not limit the specific number of preset values. When it is implemented, it can be determined as needed. However, in order to ensure that the law between the number of packets sent and the number of packets lost according to the actual number of packets sent by the preset number of transmission periods and the number of packets actually lost is relatively accurate, the larger the number of preset values, the better.
- the function relationship between the number of packets sent and the number of packets lost is usually a straight line, that is, a linear relationship between the number of packets sent and the number of packets lost, and is transmitted.
- the linear relationship between the number of packets sent and the number of packets lost during congestion and non-transmission congestion is different.
- the first linear relationship between the actual number of packets sent and the actual number of packets lost during non-transmission congestion may be determined according to the actual number of packets sent and the actual number of packets lost in each transmission cycle, and according to each transmission.
- the actual number of packets sent by the cycle and the actual number of packets lost determine the second linear relationship between the actual number of packets sent and the actual number of packets lost during transmission congestion.
- the first linear relationship and the second linear relationship may be represented by a straight line as shown in FIG. 2.
- FIG. 5 shows a non-linear relationship between the actual number of packets sent and the number of actual packets lost during non-transmission congestion determined according to the actual number of packets sent and the actual number of packets lost per transmission period.
- a congestion fitting straight line and a congestion fitting straight line indicating a second linear relationship between the actual number of packets sent and the actual number of packets lost during transmission congestion determined according to the actual number of packets sent per transmission period and the actual number of packets lost.
- the point in the elliptical dotted line indicates the actual number of packets sent and the actual number of lost packets in the preset number of transmission cycles.
- the straight line passing through the origin shown by the solid line indicates the first linear relationship determined according to the actual number of packets sent per actual transmission period and the actual number of lost packets when non-transmission congestion; the straight line with a slope of 1 indicated by the dotted line indicates that When transmission congestion occurs, the second linear relationship is determined according to the actual number of packets sent per transmission period and the actual number of packets lost.
- the first linear relationship and the second linear relationship shown in FIG. 5 can be used to obtain a first linear relationship between the actual number of packets sent and the actual number of packets lost during non-transmission congestion:
- r is the number of lost packets when non-transmission congestion
- r' is the number of lost packets when congestion is transmitted
- 5 is the number of packets sent
- ⁇ is the congestion threshold, which is caused by transmission errors. Packet loss rate.
- the first linear relationship between the actual number of packets sent and the actual number of packets lost during non-transmission congestion is determined first, and then the second linearity between the actual number of packets sent and the actual number of packets lost during transmission congestion is determined.
- the relationship is exemplified by an example.
- the second linear relationship between the actual number of packets sent and the actual number of packets lost during transmission congestion may be determined first, and then the actual number of packets sent and the actual number of packets transmitted during non-transmission congestion are determined.
- the first linear relationship between the number of lost packets That is to say, the embodiment of the present invention does not define the order of determining the first linear relationship and the second linear relationship.
- the first theoretical packet loss corresponding to the actual number of packets sent per transmission period may be determined according to the first linear relationship.
- the manner of determining the first theoretical packet loss number corresponding to the actual number of packets sent by each transmission period according to the first linear relationship includes, but is not limited to, substituting the actual number of packets sent for each transmission period into the first linear relationship, thereby obtaining The number of first theoretical packet loss corresponding to the actual number of packets sent per transmission period.
- the actual number of packets sent per transmission period can be substituted into 5 in the formula (1), thereby obtaining the first theory corresponding to the actual number of packets issued per transmission period.
- the number of lost packets is ⁇ .
- the number of the first theoretical packet loss is consistent with the number of transmission periods.
- the method provided by the embodiment of the present invention is exemplified by using the first variance as the first parameter.
- the first variance can be determined according to the actual number of lost packets in each transmission cycle and the first theoretical number of lost packets.
- the first parameter may also be another parameter, for example, the first parameter includes, but is not limited to, the first standard difference, and the like, which is not specifically limited in the embodiment of the present invention.
- the first variance can be calculated without considering the actual number of lost packets and the first theoretical packet loss in each transmission cycle.
- the weight of the time can also be given to the actual number of lost packets in each transmission cycle and the first theoretical number of lost packets has a weight when calculating the first variance. Therefore, before determining the first variance according to the actual number of lost packets and the first theoretical number of lost packets in each transmission period, it is also required to determine the actual number of lost packets in each transmission period and the first theoretical number of lost packets in calculating the first variance. Whether it has a certain weight.
- the method for determining whether the actual number of packet loss in each transmission period and the number of the first theoretical packet loss have a certain weight when calculating the first variance is not specifically limited in the embodiment of the present invention.
- the manner of determining the actual number of lost packets in each transmission period and the weight of the first theoretical packet loss in calculating the first variance is also not specifically limited in the embodiment of the present invention. In a specific implementation, it is required to combine an algorithm used to assign the actual number of lost packets per transmission period and the first theoretical packet loss number when calculating the weight of the first variance.
- the first case the weight of the actual packet loss and the number of the first theoretical packet loss in the calculation of the first variance are not considered, and the actual number of packets lost in each transmission period and the number of -theoretical packet loss are determined.
- the first variance Specifically, when determining the first variance according to the actual number of lost packets and the number of -theoretical packet loss in each transmission period, including but not limited to, by using the following formula:
- N since W is a constant, in order to simplify the calculation, N may not be considered. That is, you can round off N and simplify the first variance formula in equation (3) to:
- Equation (5) In the process of calculating VNC in equation (5), if it is treated as a variable, then equation (5) is a quadratic function, and equation (5) is a quadratic function with an open upward. Therefore, WVC has a minimum Value.
- the quadratic function minimum formula (4ac-b 2 ) /4a can be used to find the minimum value of VNC.
- the minimum value VNCM of the first variance VNC is obtained by ( 4ac-b 2 ) /4a «including but not limited to: VNCMin (6)
- the second case Consider the actual number of lost packets in each transmission cycle and the weight of the first theoretical packet loss when calculating the first difference.
- the manner of determining the first variance according to the actual number of lost packets and the number of the first theoretical packet loss in each transmission period includes but is not limited to: determining the actual number of lost packets and the number of first theoretical packet loss in each transmission period. Different weights are calculated when calculating the first variance, the weight of the first variance is calculated according to the actual number of lost packets and the number of first theoretical packet loss in each transmission period, the actual number of lost packets in each transmission period, and the first theory The number of lost packets determines the first variance.
- a filtering method is used, that is, determining the most recent transmission period.
- the actual number of lost packets and the first theoretical number of lost packets are calculated when the first variance is calculated.
- the actual number of lost packets in the last transmission period of the most recent transmission period and the first theoretical number of lost packets are calculated when the first variance is calculated.
- the weight is * (1- a ); the actual number of lost packets in the last n-1 transmission periods of the most recent transmission period and the first theoretical number of lost packets are "* ( 7- " ) when calculating the first variance.
- the embodiments of the present invention do not limit the specific numerical values. For example, "may be 0.3.
- the first variance is determined according to the actual number of lost packets in each transmission period, the number of first theoretical packet loss, and the actual number of packet loss per transmission period and the number of first theoretical packet loss in calculating the first variance. This includes, but is not limited to, the following formula:
- VNCF' is the first variance
- ⁇ is the actual number of packets lost per transmission cycle
- Si is the actual number of packets sent per transmission cycle
- N is the number of preset number of transmission cycles.
- N is a constant, in order to simplify the calculation, N may not be considered. That is, you can round off and simplify the first variance formula in equation (7) to:
- equation (8) is a quadratic function, and equation (8) is a quadratic function of the opening upward, therefore, VNC' exists.
- Minimum value Consistent with the principle of the process of solving the first variance in the first case described above, the quadratic function minimum formula (4ac-b A 2) /4a can also be used to find the minimum value VNCM «' of VNC' as:
- VNCMin' s q' xL q SLM '
- SLM, SLM, _, x(l-cf)+5, x ⁇ , ⁇
- the second variance when the second variance is determined according to the actual number of lost packets and the second theoretical packet loss in each transmission cycle, the actual number of packets lost in each transmission cycle and the second theory can be considered.
- the weight of the packet loss when calculating the second variance may also have different weights when calculating the second variance regardless of the actual number of lost packets in each transmission period and the second theoretical packet loss. Therefore, before determining the second variance according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period, it is also required to determine whether the actual number of lost packets and the number of second theoretical packets in each transmission cycle are calculated when calculating the second variance. Have a certain weight.
- the method for determining whether the actual number of packet loss and the number of the second theoretical packet loss in each of the transmission periods have a certain weight in calculating the second variance is not specifically limited in the embodiment of the present invention.
- the manner of determining the weight of the actual packet loss and the number of the second theoretical packet loss in the calculation of the second variance is also not limited in the embodiment of the present invention.
- an algorithm used in combination with the actual number of lost packets per transmission period and the number of second theoretical lost packets in calculating the weight of the second variance is used. And set.
- the manner of assigning the actual number of lost packets and the number of second theoretical packet loss in each transmission period in calculating the second variance should be It is consistent with the manner in which the actual number of lost packets and the number of first theoretical lost packets are given in the calculation of the first variance.
- the weight of the actual packet loss and the number of the second theoretical packet loss in the calculation of the second variance are not considered, and the second packet is determined according to the actual number of packets lost in each transmission cycle and the number of second theoretical packets. variance.
- the second variance is determined according to the actual number of lost packets and the second theoretical number of lost packets in each transmission cycle, it can be implemented by a formula corresponding to the formula (3).
- the second variance including but not limited to, by the following formula:
- N since w is a constant, in order to simplify the calculation, and also to correspond to the formula (5) when determining the first variance, N may not be considered. That is, you can round off N and simplify equation (20) to:
- VCMin Case 2: Consider the actual number of lost packets for each transmission cycle and the weight of the second theoretical packet loss when calculating the second variance.
- the manner of determining the second variance according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period includes, but is not limited to: determining the actual number of lost packets in each transmission period, that is, the number of second theoretical packet loss. Calculating the second variance has different weights, according to the actual number of lost packets and the number of second theoretical packet loss in each transmission cycle, the weight of the second variance, the actual number of lost packets in each transmission cycle, and the second theoretical packet loss Determine the second variance.
- the actual number of lost packets and the number of second theoretical lost packets given to each transmission period are also used to calculate the weight of the second variance.
- a filtering method that is, determining the actual number of lost packets in the last transmission period and the weight of the second theoretical packet loss when calculating the second variance is ff; the actual number of lost packets in the previous transmission period of the most recent transmission period and the number The weight of the theoretical packet loss is ff * ( l- a ) when calculating the second variance; the actual number of lost packets and the second theoretical packet loss in the last n-1 transmission cycles of the most recent transmission cycle are calculated in the second
- the weight of the variance is ( l- a ) n - at this time, according to the actual number of lost packets in each transmission cycle, the number of second theoretical packet loss, and the actual number of lost packets per transmission cycle and the number of second theoretical packet loss
- the weight at the second variance determines the second
- equation (24) is a quadratic function with respect to (1-)*A, therefore, VC' has a minimum value.
- the minimum value VCM « of VC' can also be obtained by using the quadratic function minimum formula (4ac-b 2 ) /4a as:
- Equation (31) and (32) are calculated by taking the first variance and the second variance, respectively, in steps 403 and 404, respectively.
- the second variance may also be calculated first, and then the first variance is calculated. That is, the embodiment of the present invention does not calculate the first variance and the second party. The order of the differences is limited.
- the first variance and the second variance are obtained through steps 403 and 404, it is possible to determine whether to transmit congestion based on the comparison result of the first variance and the second variance.
- the manner of determining whether to transmit congestion according to the comparison result of the first variance and the second variance including but not limited to: if the first variance is smaller than the second variance, determining non-transmission congestion; if the first variance is not less than the second The variance is determined to be transmission congestion.
- comparing the first variance and the second variance including but not limited to selecting the minimum value of the first variance obtained in step 403 and step 404 and the minimum value VCMiVi of the second variance or the first variance
- the minimum value WVCMz and the minimum variance VCMin ' of the second variance are implemented.
- the second variance is also used to determine the value determined by the first case when comparing the first variance with the second variance; If the first variance is determined using the method described in the second case, the second variance also uses the value determined by the second case when comparing the first variance with the second variance.
- the experimental results show that the first variance and the second variance obtained by comparing the first case provided by the embodiment of the present invention are compared to determine that the transmission congestion is 80% higher than the accuracy of the transmission congestion determined by the related art;
- the first variance and the second variance obtained by the second case provided by the embodiment of the present invention are compared to determine that the transmission congestion is 85% higher than the accuracy of the transmission congestion determined by the related art. Therefore, the method provided by the embodiment of the present invention can relatively accurately determine whether to transmit congestion.
- the above process determines whether to transmit congestion only by the relationship between the minimum value VNCMin and VCMin of VNC and the relationship between the minimum values WVCMz and VCM n' of WVC' and W.
- WVC and other corresponding values of WVC' and W may also be used to determine The embodiment of the present invention does not specifically limit the transmission congestion.
- the method provided by the embodiment of the present invention may cause packet loss caused by packet loss and transmission error caused by transmission congestion in the actual number of lost packets, and is determined by the actual number of packets sent according to the preset number of transmission periods and the actual number of packets lost.
- the comparison result of the parameters determines whether the transmission is congested, so that the determined transmission congestion is more accurate, so that the allocation of the appropriate transmission network interface bandwidth can be controlled, thereby improving the throughput of the transmission network. Referring to FIG.
- an embodiment of the present invention provides a determining apparatus for transmitting congestion, which may be used to perform the method for determining transmission congestion provided by the embodiment shown in FIG. 3 or FIG. 4, where the apparatus includes: an acquiring module 601, And obtaining the actual number of packets sent by the preset number of transmission periods and the actual number of packets to be lost.
- the first determining module 602 is configured to determine the actual number of packets to be sent and the actual number of packets when the non-transmission congestion is determined according to the actual number of packets sent and the number of actual packets lost in each transmission period. The first linear relationship between the number of lost packets;
- the second determining module 603 is configured to determine, according to the actual number of packets sent and the number of actual packets lost in each transmission period, a second linear relationship between the actual number of packets sent and the actual number of packets lost during transmission congestion;
- the third determining module 604 is configured to determine, according to the first linear relationship, a first theoretical packet loss number corresponding to the actual number of packets sent by each transmission period;
- the fourth determining module 605 is configured to determine the first parameter according to the actual number of lost packets and the first theoretical packet loss in each transmission period;
- a fifth determining module 606, configured to determine, according to the second linear relationship, a second theoretical packet loss number corresponding to the actual number of packets sent by each transmission period;
- a sixth determining module 607 configured to determine a second parameter according to the actual number of lost packets and the second theoretical packet loss in each transmission period;
- the seventh determining module 608 is configured to determine whether to transmit congestion according to a comparison result of the first parameter and the second parameter.
- the first linear relationship between the actual number of packets sent and the number of actual packets lost during non-transmission congestion determined according to the actual number of packets sent and the number of actual packets lost in each transmission period is:
- T' S - (l - 0) x A
- ⁇ is the number of lost packets when the transmission is not congested
- ⁇ ' is the number of lost packets when the transmission is congested
- 5 is the number of packets sent
- ⁇ is the congestion threshold, which is the packet loss rate caused by the transmission error.
- the fourth determining module 605 is configured to determine a first variance according to the actual number of lost packets and the number of first theoretical packet loss in each transmission period, and use the first variance as the first parameter;
- the sixth determining module 607 is configured to determine a second variance according to the actual number of lost packets and the second theoretical packet loss in each transmission period, and use the second variance as the second parameter.
- the fourth determining module 605 is configured to determine the first variance according to the actual number of lost packets and the first theoretical packet loss in each transmission period by using the following formula:
- the sixth determining module 607 is configured to determine the second variance according to the actual packet loss number and the second theoretical packet loss number of each transmission period by using the following formula:
- VCM ⁇ is the second variance, 5 ; is the actual number of packets sent per transmission period, R ; is the actual number of packets received per transmission period, 7; is the actual number of each transmission cycle
- the number of lost packets, N is the default value.
- the fourth determining module 605 is configured to use the actual number of lost packets according to each transmission period and the first The theoretical packet loss number is determined by the following formula:
- the sixth determining module 607 is configured to determine the second variance according to the actual packet loss number and the second theoretical packet loss number of each transmission period by using the following formula:
- VCM Z is the second variance, 5 ; is the actual number of packets sent per transmission cycle, R ; is the actual number of packets received per transmission cycle, 7. is for each transmission cycle
- the actual number of lost packets, N is a preset value, and is widely used as the weight of the first packet of the first transmission period and the actual number of packets lost when calculating the second variance.
- the device provided by the embodiment of the present invention may cause packet loss caused by packet loss and transmission error caused by transmission congestion in the actual number of packet loss, and is determined by the actual number of packets sent according to the preset number of transmission periods and the actual number of packets lost.
- the comparison result of the parameters determines whether the transmission is congested, so that the determined transmission congestion is more accurate, so that the allocation of the appropriate transmission network interface bandwidth can be controlled, thereby improving the throughput of the transmission network. Referring to FIG.
- an embodiment of the present invention provides a determining apparatus for transmitting congestion, which may be used to perform a method for determining transmission congestion provided by the embodiment shown in FIG. 3 or FIG.
- the apparatus includes at least one processor 701 and a memory 702 that stores instructions executable by the processor 701. among them:
- the processor 701 is configured to obtain the actual number of packets sent by the preset number of transmission periods and the actual number of packets to be lost. The processor 701 is further configured to determine, according to the actual number of packets sent and the number of actual packets lost in each transmission period. The first linear relationship between the actual number of packets sent and the number of actual packets lost during transmission congestion, and the number of actual packets and the actual number of packets lost during transmission congestion are determined according to the actual number of packets sent and the number of actual packets lost during each transmission cycle. Bilinear relationship
- the processor 701 is further configured to determine, according to the first linear relationship, the first theoretical packet loss number corresponding to the actual number of packets sent in each transmission period, and determine the first according to the actual number of packet loss and the first theoretical packet loss in each transmission period.
- the processor 701 is further configured to determine, according to the second linear relationship, the second theoretical packet loss number corresponding to the actual number of packets sent in each transmission period, and determine the second number according to the actual number of lost packets and the second theoretical packet loss number of each transmission period.
- the processor 701 is further configured to determine whether to transmit congestion according to a comparison result between the first parameter and the second parameter.
- the first linear relationship between the actual number of packets sent and the number of actual packets lost according to the actual number of packets sent and the number of actual packets lost in each transmission period is:
- T' S - (l - 0) x A
- ⁇ is the number of lost packets when the transmission is not congested
- ⁇ ' is the number of lost packets when the transmission is congested
- 5 is the number of packets sent
- ⁇ is the congestion threshold, which is the packet loss rate caused by the transmission error.
- the processor 701 is further configured to determine a first variance according to an actual number of lost packets and a first theoretical number of lost packets in each transmission period, where the first variance is used as the first parameter;
- the processor 701 is further configured to determine a second variance according to the actual number of lost packets and the second theoretical number of lost packets in each transmission period, and use the second variance as the second parameter.
- the processor 701 is further configured to determine the second variance according to the actual packet loss number and the second theoretical packet loss number of each transmission period by using the following formula:
- WVCMVi is the first variance
- VCM ⁇ is the second variance
- 5 is the actual number of packets sent per transmission period
- R is the actual number of packets received per transmission period, 7. for each transmission cycle
- the actual number of lost packets, N is the default value.
- the processor 701 is further configured to determine the first variance according to an actual packet loss number and a first theoretical packet loss number of each transmission period by using the following formula:
- the processor 701 is further configured to determine the second variance according to the actual packet loss number and the second theoretical packet loss number of each transmission period by using the following formula:
- VCM Z is the second variance, 5 ; is the actual number of packets sent per transmission cycle, R ; is the actual number of packets received per transmission cycle, 7. is for each transmission cycle
- the actual number of lost packets, N is a preset value, and is widely used as the weight of the first packet of the first transmission period and the actual number of packets lost when calculating the second variance.
- the device provided by the embodiment of the present invention may exist due to transmission congestion due to the actual number of lost packets.
- the first linear relationship between the non-transmission congestion and the second linear relationship when the congestion is transmitted is determined by the actual number of packets sent according to the preset number of transmission periods and the actual number of packets lost. Determining whether to transmit congestion according to the comparison between the actual number of packet loss and the first parameter and the second parameter determined according to the first linear relationship and the second linear relationship, so that the determined transmission congestion is more accurate, thereby Controls the allocation of the appropriate transmission network interface bandwidth, which in turn increases the throughput of the transmission network.
- the determining apparatus for transmitting congestion provided by the foregoing embodiment is only exemplified by the division of the foregoing functional modules when determining the transmission congestion.
- the foregoing functions may be allocated by different functional modules according to requirements.
- the internal structure of the device is divided into different functional modules to perform all or part of the functions described above.
- the determining apparatus for the transmission congestion provided by the foregoing embodiment is in the same concept as the method for determining the congestion of the transmission. The specific implementation process is described in detail in the method embodiment, and details are not described herein again.
- a person skilled in the art may understand that all or part of the steps of implementing the above embodiments may be completed by hardware, or may be instructed by a program to execute related hardware, and the program may be stored in a computer readable storage medium.
- the storage medium mentioned may be a read only memory, a magnetic disk or an optical disk or the like.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
本发明实施例提供了一种传输拥塞的确定方法及装置,属于无线通信领域。由于实际丢包数中可能同时存在由于传输拥塞导致的丢包和传输错误导致的丢包,因此,本发明提供的方法通过根据预设数值个传输周期的实际发包数和实际丢包数确定非传输拥塞时的第一线性关系以及传输拥塞时的第二线性关系后,根据实际丢包数和第一线性关系确定第一理论丢包数以及根据实际丢包数和第二线性关系确定第二丢包数,从而根据实际丢包数分别与第一理论丢包数及第二理论丢包数确定的第一参数和第二参数确定是否传输拥塞,使得确定的传输拥塞更加准确,从而可以控制分配合适的传输网络接口带宽,进而提高传输网络的吞吐率。
Description
说 明 书 传输拥塞的确定方法及装置 技术领域
本发明涉及无线通信领域, 特别涉及一种传输拥塞的确定方法及装置。 背景技术
传输拥塞是指数据包发送需求超过了传输网络传输数据包的能力,如果不 降低数据包发送需求, 将会导致传输系统丟弃无法发送的数据包, 从而导致系 统丟包增加。为了避免因传输速率过高而造成传输拥塞,从而加剧传输的时延、 增加丟包率, 通常控制数据发送主机与数据接收主机之间的数据传输速率保持 合适的值。 由于该种控制方式需要根据传输拥塞状况来调整传输网络接口宽 带, 因此, 需要实时确定是否传输拥塞。
相关技术在确定传输拥塞时, 所釆用的方法为: 统计数据发送主机在预设 数值个传输周期内的发包数及数据接收主机在该预设数值个传输周期内的收 包数; 根据发包数和收包数确定数据传输的丟包率; 判断丟包率是否大于预设 阔值; 如果丟包率大于预设阔值, 则确定传输拥塞。
在实现本发明的过程中, 发明人发现相关技术至少存在以下问题: 由于相关技术在确定丟包率后, 如果丟包率大于预设阔值, 则确定传输拥 塞。 也就是说, 相关技术将丟包率全部认为是传输拥塞导致的。 然而, 在数据 传输过程中, 除传输拥塞能导致丟包外, 传输错误也会导致丟包。 如果系统因 存在传输错误导致的丟包率, 则根据丟包率确定的传输拥塞不准确。 另外, 当 由于传输错误导致的丟包率高于预设阔值时,会误将传输错误导致的丟包率确 定为是由传输拥塞引起的, 因此会控制降低传输网络接口带宽, 导致降低传输 网络的吞吐率。
发明内容
为了解决现有技术存在的问题, 本发明实施例提供了一种传输拥塞的确定 方法及装置。 所述技术方案如下:
第一方面, 提供了一种传输拥塞的确定方法, 所述方法包括:
获取预设数值个传输周期的实际发包数和实际丟包数;
根据每个传输周期的实际发包数和实际丟包数确定非传输拥塞时实际发 包数和实际丟包数之间的第一线性关系, 并根据每个传输周期的实际发包数和 实际丟包数确定传输拥塞时实际发包数和实际丟包数之间的第二线性关系; 根据所述第一线性关系确定每个传输周期的实际发包数对应的第一理论 丟包数, 并根据各个传输周期的实际丟包数及第一理论丟包数确定第一参数; 根据所述第二线性关系确定每个传输周期的实际发包数对应的第二理论 丟包数, 并根据各个传输周期的实际丟包数及第二理论丟包数确定第二参数; 根据所述第一参数与所述第二参数的比较结果确定是否传输拥塞。
结合第一方面, 在第一方面的第一种可能的实现方式中, 根据每个传输周 期的实际发包数和实际丟包数确定的非传输拥塞时实际发包数和实际丟包数 之间的第一线性关系为:
Τ = Ξ χ θ
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发 包数和实际丟包数之间的第二线性关系为:
Τ' = Ξ - (\ - θ) χ Α
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数, 5为发包 数, Α为拥塞阔值, 为传输错误导致的丟包率。
结合第一方面或第一方面的第一种可能的实现方式,在第一方面的第二种 可能的实现方式中, 所述根据各个传输周期的实际丟包数及第一理论丟包数确 定第一参数, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差,将所述
第一方差作为第一参数;
所述根据各个传输周期的实际丟包数及第二理论丟包数确定第二参数, 包 括:
根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差,将所述 第二方差作为第二参数。
结合第一方面的第二种可能的实现方式, 在第一方面的三种可能的实现方 式中, 所述根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定所 述第一方差:
W x∑( - ?;·) ∑(H )
VCMin:
N 其中, WVCMVi为第一方差, VCM^为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值。
结合第一方面的第二种可能的实现方式, 在第一方面的三种可能的实现方 式中, 所述根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定所 述第一方差:
所述根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差, 包 括:
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7;为每个传输周期的实际丟包数, N 为预设数值, 广为第 个传输周期的发包数和实际丟包数在计算所述第 二方差时的权重。
第二方面, 还提供了一种传输拥塞的确定装置, 所述装置包括: 获取模块, 用于获取预设数值个传输周期的实际发包数和实际丟包数; 第一确定模块, 用于根据每个传输周期的实际发包数和实际丟包数确定非 传输拥塞时实际发包数和实际丟包数之间的第一线性关系;
第二确定模块, 用于根据每个传输周期的实际发包数和实际丟包数确定传 输拥塞时实际发包数和实际丟包数之间的第二线性关系;
第三确定模块, 用于根据所述第一线性关系确定每个传输周期的实际发包 数对应的第一理论丟包数;
第四确定模块, 用于根据各个传输周期的实际丟包数及第一理论丟包数确 定第一参数;
第五确定模块, 用于根据所述第二线性关系确定每个传输周期的实际发包 数对应的第二理论丟包数;
第六确定模块, 用于根据各个传输周期的实际丟包数及第二理论丟包数确 定第二参数;
第七确定模块, 用于根据所述第一参数与所述第二参数的比较结果确定是 否传输拥塞。
结合第二方面, 在第二方面的第一种可能的实现方式中, 根据每个传输周 期的实际发包数和实际丟包数确定的非传输拥塞时实际发包数和实际丟包数 之间的第一线性关系为:
T = S x 0
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发 包数和实际丟包数之间的第二线性关系为:
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数, 5为发包 数, Α为拥塞阔值, 为传输错误导致的丟包率。
结合第二方面或第二方面的第一种可能的实现方式,在第二方面的第二种 可能的实现方式中, 所述第四确定模块, 用于根据各个传输周期的实际丟包数 及第一理论丟包数确定第一方差, 将所述第一方差作为第一参数;
所述第六确定模块, 用于根据各个传输周期的实际丟包数及第二理论丟包 数确定第二方差, 将所述第二方差作为第二参数。
结合第二方面的第二种可能的实现方式, 在第二方面的第三种可能的实现 方式中, 所述第四确定模块, 用于根据各个传输周期的实际丟包数及第一理论 丟包数通过如下公式确定所述第一方差:
VNCMin
N ;
其中, WVCMVi为第一方差, VCM^为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值。
结合第二方面的第二种可能的实现方式, 在第二方面的第四种可能的实现 方式中, 所述第四确定模块, 用于根据各个传输周期的实际丟包数及第一理论 丟包数通过如下公式确定所述第一方差:
2
i-l ^.
ail- a\ ^S^ x ail- a\ ^T- il- a) ^ 5- Tt
(=1
VNCMin'- (l- a) 5
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值, 广为第 个传输周期的发包数和实际丟包数在计算所述第 二方差时的权重。
本发明实施例提供的技术方案的有益效果是:
由于实际丟包数中可能同时存在由于传输拥塞导致的丟包和传输错误导 致的丟包, 通过根据预设数值个传输周期的实际发包数和实际丟包数确定非传 输拥塞时的第一线性关系以及传输拥塞时的第二线性关系后,根据实际丟包数
与根据第一线性关系和第二线性关系得到的理论丟包数确定的第一参数与第 二参数的比较结果来确定是否传输拥塞, 使得确定的传输拥塞更加准确, 进而 可以控制分配合适的传输网络接口带宽, 从而提高传输网络的吞吐率。 附图说明
为了更清楚地说明本发明实施例中的技术方案, 下面将对实施例描述中所 需要使用的附图作简单地介绍, 显而易见地, 下面描述中的附图仅仅是本发明 的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下, 还可以根据这些附图获得其他的附图。
图 1是本发明一实施例提供的一种传输拥塞的确定方法所涉及的实施环境 示意图;
图 2是本发明另一实施例提供的一种发包数与丟包数之间的关系示意图; 图 3是本发明另一实施例提供的一种传输拥塞的确定方法的流程图; 图 4是本发明另一实施例提供的一种传输拥塞的确定方法的流程图; 图 5是本发明另一实施例提供的一种表示第一线性关系的非拥塞拟合直线 以及表示第二线性关系的拥塞拟合直线;
图 6是本发明另一实施例提供的一种传输拥塞的确定装置的结构示意图; 图 7是本发明另一实施例提供的一种传输拥塞的确定装置的结构示意图。 具体实施方式
为使本发明的目的、 技术方案和优点更加清楚, 下面将结合附图对本发明 实施方式作进一步地详细描述。
如图 1所示, 其示出了本发明实施例提供的一种传输拥塞的确定方法所涉 及的实施环境示意图。 如图 1所示, 该实施环境为一数据传输系统, 该数据传 输系统包括数据发送主机 101和数据接收主机 102。 其中, 数据发送主机 101 和数据接收主机 102通过传输网络连接, 该传输网络可以为有线网络, 也可以
为无线网络。 该数据传输系统需要控制数据发送主机 101和数据接收主机 102 之间的数据传输速率保持合适的值, 以根据数据传输速率来调整传输网络接口 宽带。 为了保证数据发送主机 101和数据接收主机 102之间的数据传输速率保 持在合适的值, 以避免因传输拥塞而导致数据包丟失的情况发生, 需要实时检 测数据发送主机 101和数据接收主机 102之间是否传输拥塞。
关于数据发送主机 101和数据接收主机 102的具体类型, 本发明实施例不 作具体限定。具体实施时,数据发送主机 101包括但不限于可以为 RNC ( Radio Network Controller, 无线网络控制器), 数据接收主机 102包括但不限于可以 为 Node B。
通常, 造成传输网络丟包的原因除有传输拥塞外, 还有传输错误。 为了解 决相关技术中根据丟包率确定是否传输拥塞时将所有的丟包均确定为由传输 拥塞造成的, 从而导致确定的传输拥塞不准确的问题, 本发明实施例提供的传 输拥塞的确定方法综合考虑传输拥塞和传输错误导致的传输丟包,提出一种新 的传输拥塞的确定方法。
一般地, 传输拥塞导致的丟包与传输错误导致的丟包, 其发包数与丟包数 的规律不同。 如图 2所示, 其示出了一种发包数与丟包数之间的关系示意图。 由图 2可知, 当系统不存在传输拥塞导致的丟包, 仅存在传输错误导致的丟包 时, 亦即非传输拥塞时, 发包数与丟包数的函数是一条过原点的直线; 当系统 同时存在传输拥塞导致的丟包和传输错误导致的丟包时, 亦即传输拥塞时, 发 包数与丟包数的函数是一条斜率固定为 1的直线。 在图 2所示的直线中, 表示 非传输拥塞与传输拥塞之间的虚线与横轴的交点为拥塞阔值, 当一个传输周期 内的发包数小于拥塞阔值时, 不会导致传输拥塞; 当一个传输周期内的发包数 大于拥塞阔值时, 可能会导致传输拥塞。
综上, 根据非传输拥塞和传输拥塞时发包数与丟包数的关系, 本发明实施 例提供的方法, 通过获取预设数值个传输周期的实际发包数和实际丟包数, 并 根据获取到的每个传输周期的实际发包数和实际丟包数确定非传输拥塞时实
际发包数和实际丟包数之间的第一线性关系,根据获取到的每个传输周期的实 际发包数和实际丟包数确定传输拥塞时实际发包数和实际丟包数之间的第二 线性关系,从而根据实际丟包数与根据第一线性关系和第二线性关系得到的理 论丟包数之间的第一参数与第二参数的比较结果来确定是否传输拥塞。 具体的 传输拥塞的确定方法详见下述各个实施例。 结合图 1所示的实施环境示意图和图 2所示的发包数与丟包数之间的关系 示意图以及对应的内容, 本发明实施例提供了一种传输拥塞的确定方法。 参见 图 3, 本发明实施例提供的方法流程包括:
301: 获取预设数值个传输周期的实际发包数和实际丟包数。
302: 根据每个传输周期的实际发包数和实际丟包数确定非传输拥塞时实 际发包数和实际丟包数之间的第一线性关系, 并根据每个传输周期的实际发包 数和实际丟包数确定传输拥塞时实际发包数和实际丟包数之间的第二线性关 系。
其中,根据每个传输周期的实际发包数和实际丟包数确定的非传输拥塞时 实际发包数和实际丟包数之间的第一线性关系为:
T = S x 0
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发 包数和实际丟包数之间的第二线性关系为:
Τ' = Ξ - (\ - θ) χ Α
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数, 5为发包 数, Α为拥塞阔值, 为传输错误导致的丟包率。
303: 根据第一线性关系确定每个传输周期的实际发包数对应的第一理论 丟包数, 并根据各个传输周期的实际丟包数及第一理论丟包数确定第一参数。
304:根据第二线性关系确定每个传输周期的实际发包数对应的第二理论 丟包数, 并根据各个传输周期的实际丟包数及第二理论丟包数确定第二参数。
305:根据第一参数与第二参数的比较结果确定是否传输拥塞。 可选地, 根据各个传输周期的实际丟包数及第一理论丟包数确定第一参 数, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差,将第一 方差作为第一参数;
根据各个传输周期的实际丟包数及第二理论丟包数确定第二参数, 包括: 根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差,将第二 方差作为第二参数。
可选地, 根据各个传输周期的实际丟包数及第一理论丟包数确定第一方 差, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定第 一方差:
VNCMin
其中, WVCM 为第一方差, VCM^为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值。
可选地, 根据各个传输周期的实际丟包数及第一理论丟包数确定第一方 差, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定第
I N
ail- a
ί二 1
VNCMin' - all - a) S-'
根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差, 包括: 根据各个传输周期的实际丟包数及第二理论丟包数通过如下公式确定第 二方差:
VCMin' =
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值, 广为第 个传输周期的发包数和实际丟包数在计算第二方 差时的权重。
本发明实施例提供的方法, 由于实际丟包数中可能同时存在由于传输拥塞 导致的丟包和传输错误导致的丟包,通过根据预设数值个传输周期的实际发包 数和实际丟包数确定非传输拥塞时的第一线性关系以及传输拥塞时的第二线 性关系后,根据实际丟包数与根据第一线性关系和第二线性关系得到的理论丟 包数确定的第一参数与第二参数的比较结果来确定是否传输拥塞,使得确定的 传输拥塞更加准确, 从而可以控制分配合适的传输网络接口带宽, 进而提高传 输网络的吞吐率。 参见图 4, 结合上述实施例, 本发明实施例提供了一种传输拥塞的确定方 法。 在本发明实施例中, 以根据各个传输周期的实际丟包数及第一理论丟包数 确定的第一方差和根据各个传输周期的实际丟包数及第二理论丟包数确定的
第二方差分别为第一参数和第二参数,从而根据第一方差与第二方差的比较结 果确定是否传输拥塞来对本发明实施例提供的传输拥塞的确定方法进行举例 说明。 本发明实施例提供的方法流程包括:
401: 获取预设数值个传输周期的实际发包数和实际丟包数。
为了根据数据发送主机与数据接收主机之间的实际发包数和实际丟包数 对数据发送主机和数据接收主机组成的数据传输系统的发包数和丟包数之间 的规律进行分析, 需要获取预设数值个传输周期的实际发包数和实际丟包数。
关于获取预设数值个传输周期的实际发包数和实际丟包数的方式, 本发明 实施例不作具体限定。 具体实施时, 为了便于分析系统的实际发包数与实际丟 包数之间的关系,数据发送主机在到达一个传输周期时统计发送的数据包的个 数。 同理, 数据接收主机在到达一个传输周期时统计接收的数据包的个数。 因 此,数据传输系统根据数据发送主机在一个传输周期内发送的数据包的个数确 定并存储实际发包数, 并根据数据发送主机在一个传输周期内发送的数据包的 个数以及数据接收主机在一个传输周期内接收的数据包的个数确定实际丟包 数。 此时, 便可以获得一个传输周期内的实际发包数和实际丟包数。 同理, 釆 用同样的方式, 可以获得预设数值个传输周期的实际发包数和实际丟包数。
其中, 本发明实施例不对预设数值的具体数量进行限定。 具体实施时, 可 以根据需要确定。 然而, 为了保证根据获取到的预设数值个传输周期的实际发 包数和实际丟包数确定的发包数和丟包数之间的规律比较准确, 该预设数值的 数量越大越好。
402: 根据每个传输周期的实际发包数和实际丟包数确定非传输拥塞时实 际发包数和实际丟包数之间的第一线性关系, 并根据每个传输周期的实际发包 数和实际丟包数确定传输拥塞时实际发包数和实际丟包数之间的第二线性关 系。
仍然如图 2所示, 由于当传输拥塞和非传输拥塞时, 发包数与丟包数之间 的函数关系通常为直线, 也就是说发包数与丟包数之间为线性关系, 且在传输
拥塞和非传输拥塞时发包数与丟包数之间的线性关系不同。 为了确定数据传输 系统是否传输拥塞, 可以根据每个传输周期的实际发包数和实际丟包数确定非 传输拥塞时实际发包数和实际丟包数之间的第一线性关系, 并根据每个传输周 期的实际发包数和实际丟包数确定传输拥塞时实际发包数和实际丟包数之间 的第二线性关系。 其中, 第一线性关系和第二线性关系可以通过如图 2所示的 直线形式表示出来。
如图 5所示, 其示出了一种表示根据每个传输周期的实际发包数和实际丟 包数确定的非传输拥塞时实际发包数和实际丟包数之间的第一线性关系的非 拥塞拟合直线以及表示根据每个传输周期的实际发包数和实际丟包数确定的 传输拥塞时实际发包数和实际丟包数之间的第二线性关系的拥塞拟合直线。 其 中, 椭圓虚线中的点则表示获取到的预设数值个传输周期的实际发包数和实际 丟包数。 实线所示的过原点的直线即表示当非传输拥塞时, 根据每个传输周期 的实际发包数和实际丟包数确定的第一线性关系; 虚线所示的斜率为 1的直线 即表示当传输拥塞时,根据每个传输周期的实际发包数和实际丟包数确定的第 二线性关系。
其中, 结合图 5所示的第一线性关系和第二线性关系, 可以得到非传输拥 塞时实际发包数和实际丟包数之间的第一线性关系为:
Τ = Ξ χ θ ( 1 )
传输拥塞时实际发包数和实际丟包数之间的第二线性关系为:
Τ' = Ξ - (\ - θ) χ Α ( 2 )
其中, 公式(1 )和公式(2 ) 中, r为非传输拥塞时的丟包数, r'为传输 拥塞时的丟包数, 5为发包数, Α为拥塞阔值, 为传输错误导致的丟包率。
需要说明的是, 该步骤仅以先确定非传输拥塞时实际发包数和实际丟包数 之间的第一线性关系,后确定传输拥塞时实际发包数和实际丟包数之间的第二 线性关系为例进行了举例说明。 具体实施时, 也可以先确定传输拥塞时实际发 包数和实际丟包数之间的第二线性关系,后确定非传输拥塞时实际发包数和实
际丟包数之间的第一线性关系。 也就是说, 本发明实施例不对确定第一线性关 系和第二线性关系的顺序进行限定。
403: 根据第一线性关系确定每个传输周期的实际发包数对应的第一理论 丟包数, 并根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差。
在确定非传输拥塞时实际发包数和实际丟包数之间的第一线性关系后,便 可以根据该第一线性关系确定每个传输周期的实际发包数对应的第一理论丟 包数。 关于根据该第一线性关系确定每个传输周期的实际发包数对应的第一理 论丟包数的方式, 包括但不限于将每个传输周期的实际发包数代入该第一线性 关系中, 从而得到每个传输周期的实际发包数对应的第一理论丟包数。 例如, 如果用公式(1 )表示第一线性关系, 则可以将每个传输周期的实际发包数代 入公式( 1 )中的 5中, 从而得到每个传输周期的实际发包数对应的第一理论丟 包数 Γ。 其中, 由于每个传输周期的实际发包数均对应一个第一理论丟包数, 因此, 第一理论丟包数的个数与传输周期的个数一致。
另外, 由于方差能够表征数据之间的离散程度, 因此,在本发明实施例中, 以第一方差为第一参数对本发明实施例提供的方法进行举例说明。 为了确定各 个传输周期的实际丟包数与第一理论丟包数之间的关系, 可以根据各个传输周 期的实际丟包数及第一理论丟包数确定第一方差。
当然, 在该实施例中, 仅以第一方差为第一参数进行了举例说明。 在具体 实施时, 第一参数还可以为别的参数, 例如, 第一参数包括但不限于为第一标 准差, 等等, 本发明实施例对此不作具体限定。
其中, 在根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差 时, 可以不考虑各个传输周期的实际丟包数及第一理论丟包数在计算第一方差 时的权重, 也可以赋予各个传输周期的实际丟包数及第一理论丟包数在计算第 一方差时具有权重。 因此, 在根据各个传输周期的实际丟包数及第一理论丟包 数确定第一方差之前,还需要确定各个传输周期的实际丟包数与第一理论丟包 数在计算第一方差时是否具有一定的权重。
关于确定各个传输周期的实际丟包数及第一理论丟包数在计算第一方差 时是否具有一定的权重的方式, 本发明实施例不作具体限定。 另外, 关于确定 各个传输周期的实际丟包数及第一理论丟包数在计算第一方差时的权重的方 式, 本发明实施例同样不作具体限定。 具体实施时, 需要结合赋予每个传输周 期的实际丟包数及第一理论丟包数在计算第一方差时的权重时所釆用的算法 而定。
综上, 在根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差 时, 包括但不限于有如下两种情况:
第一种情况: 不考虑各个传输周期的实际丟包数及第一理论丟包数在计算 第一方差时的权重,根据各个传输周期的实际丟包数及第 - -理论丟包数确定第 一方差。 具体地, 在根据各个传输周期的实际丟包数及第- -理论丟包数确定第 一方差时, 包括但不限于通过如下公式来实现:
( 3 ) 其中,
τ· = 5; - (4 )
在公式(3 )和公式(4 )中, 为第一方差, 7为每个传输周期的实际 丟包数, 5;为每个传输周期的实际发包数, 为每个传输周期的实际收包数, Ν 为预设数值个传输周期的个数。
具体计算时, 由于 W为常数, 为了简化计算, 可以不考虑 N。 也就是说, 可以舍去 N, 而将公式(3 ) 中的第一方差公式简化为:
(5 )
二 1 二 1 _j '—二 i '—二 1
在计算公式(5 ) 中的 VNC的过程中, 如果将 视为变量, 则公式(5 )是 关于 的二次函数, 且公式(5 )为开口向上的二次函数, 因此, WVC存在最小
值。 为了在后续通过将第一方差和第二方差进行比较来确定是否传输拥塞时能 够选择一个合适的值, 可以利用二次函数最小值公式(4ac-b2 ) /4a求出 VNC的 最小值 VNCM «, 并通过将第一方差的最小值与第二方差的最小值进行比较来 实现。 其中, 通过( 4ac-b2 ) /4a求出第一方差 VNC的最小值 VNCM «包括但不限 于为: VNCMin (6)
第二种情况: 考虑各个传输周期的实际丟包数及第一理论丟包数在计算第 一方差时的权重。 在这种情况下, 根据各个传输周期的实际丟包数及第一理论 丟包数确定第一方差的方式包括但不限于: 确定各个传输周期的实际丟包数及 第一理论丟包数在计算第一方差时具有不同的权重,根据各个传输周期的实际 丟包数及第一理论丟包数在计算第一方差时的权重、各个传输周期的实际丟包 数及第一理论丟包数确定第一方差。
另外, 在本发明实施例中, 在赋予各个传输周期的实际丟包数及第一理论 丟包数在计算第一方差时的权重时釆用 "滤波的方法, 即确定最近一个传输周 期的实际丟包数及第一理论丟包数在计算第一方差时的权重为 最近一个传 输周期的上一个传输周期的实际丟包数及第一理论丟包数在计算第一方差时 的权重为 * (1- a ); 最近一个传输周期的上 n-1个传输周期的实际丟包数及 第一理论丟包数在计算第一方差时的权重为 " * ( 7- " ) 。 其中, 本发明实施 例不对 "的具体数值进行限定。 例如, "可取 0.3。
在公式(7) 中, VNCF'为第一方差, Γ,·为每个传输周期的实际丟包数, Si 为每个传输周期的实际发包数, 为每个传输周期的实际收包数, N为预设数 值个传输周期的个数。
在计算公式(8)中的 VNC'的过程中, 如果将 视为变量, 则公式(8)是 关于 的二次函数, 且公式(8) 为开口向上的二次函数, 因此, VNC'存在最 小值。 同上述第一种情况中求解第一方差的过程的原理一致, 也可以利用二次 函数最小值公式(4ac-bA2) /4a, 求出 VNC'的最小值 VNCM«'为:
VNCMin' =sq'xLq SLM'
则 Sq, (13)
t-l i-l
又 (14)
ί-1
Lq!_1 =^^(ΐ~ a) Tt
(15)
t-\ i _-1
i=l (16)
Sqt = Sq,^ x(\-a) + St xa
( 17)
Lqt = Lql_l x(l-a) xa
(18)
SLM, = SLM,_, x(l-cf)+5,x Τ,χα
( 19) 将公式( 17 )至公式( 19 )代入公式( 13 )并通过迭代计算的方法便可以 计算出 VNCM«'。
404: 根据第二线性关系确定每个传输周期的实际发包数对应的第二理论 丟包数, 并根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差。
该步骤的原理同上述步骤 403中确定第一方差的原理一致。 具体内容可以 参数上述步骤 403中的内容。
因此, 同上述确定第一方差的原理一致, 在根据各个传输周期的实际丟包 数及第二理论丟包数确定第二方差时, 可以考虑各个传输周期的实际丟包数及 第二理论丟包数在计算第二方差时的权重,也可以不考虑各个传输周期的实际 丟包数及第二理论丟包数在计算第二方差时具有不同的权重。 因此, 在根据各 个传输周期的实际丟包数及第二理论丟包数确定第二方差之前,还需要确定各 个传输周期的实际丟包数及第二理论丟包数在计算第二方差时是否具有一定 的权重。
关于确定各个传输周期的实际丟包数及第二理论丟包数在计算第二方差 时是否具有一定的权重的方式, 本发明实施例不作具体限定。 另外, 关于确定 各个传输周期的实际丟包数及第二理论丟包数在计算第二方差时的权重的方 式, 本发明实施例同样不作具体限定。 具体实施时, 需要结合赋予每个传输周 期的实际丟包数及第二理论丟包数在计算第二方差时的权重时所釆用的算法
而定。
需要说明的是, 为了便于后续可以根据第一方差和第二方差确定是否传输 拥塞,赋予各个传输周期的实际丟包数及第二理论丟包数在计算第二方差时的 权重的方式应该与赋予各个传输周期的实际丟包数及第一理论丟包数在计算 第一方差时的权重的方式一致。
综上, 在根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差 时, 包括但不限于有如下两种情况:
第一种情况: 不考虑各个传输周期的实际丟包数及第二理论丟包数在计算 第二方差时的权重,根据各个传输周期的实际丟包数及第二理论丟包数确定第 二方差。 具体地, 在根据各个传输周期的实际丟包数及第二理论丟包数确定第 二方差时, 可以通过与公式(3 )对应的公式来实现。 具体地, 在确定第二方 差时, 包括但不限于通过如下公式实现:
在计算公式(21 ) 中的 VC的过程中, 如果将( 1- ) *A视为变量, 则公式 ( 21 )是关于 (1- ) *A的二次函数, 且公式(21 ) 为开口向上的二次函数, 因此, VC存在最小值。 为了在后续可以通过将第一方差和第二方差进行比较来 确定是否传输拥塞, 同样需要选择一个 VC的合适的值。 为了与第一方差对应,
仍然选择 VC的最小值 VCM «。 因此, 可以利用二次函数最小值公式(4ac-b2 ) /4a, 求出 VC的最小值^¾^«为:
∑(H)
=l
VCMin = 第二种情况: 考虑各个传输周期的实际丟包数及第二理论丟包数在计算第 二方差时的权重。 在这种情况下, 根据各个传输周期的实际丟包数及第二理论 丟包数确定第二方差的方式包括但不限于: 确定各个传输周期的实际丟包数即 第二理论丟包数在计算第二方差时具有不同的权重,根据各个传输周期的实际 丟包数及第二理论丟包数在计算第二方差时的权重、各个传输周期的实际丟包 数及第二理论丟包数确定第二方差。
另外, 在本发明实施例中, 为了与确定第一方差的过程一致, 在赋予各个 传输周期的实际丟包数及第二理论丟包数在计算第二方差时的权重时, 也釆用 a滤波的方法, 即确定最近一个传输周期的实际丟包数及第二理论丟包数在计 算第二方差时的权重为 ff ; 最近一个传输周期的上一个传输周期的实际丟包数 及第二理论丟包数在计算第二方差时的权重为 ff * ( l- a ) ; 最近一个传输周 期的上 n-1个传输周期的实际丟包数及第二理论丟包数在计算第二方差时的权 重为 ( l- a ) n- 此时, 根据各个传输周期的实际丟包数、 第二理论丟包数以及每个传输周 期的实际丟包数及第二理论丟包数在计算第二方差时的权重确定第二方差, 包 括但不限于通过如下公式实现:
在公式(23 ) 中, 为第二方差。
具体计算时, 由于 w为常数, 为了简化计算, 也为了与步骤 403中确定第
一方差时的第二种情况一致, 可以不考虑 N, 而将公式(23) 简化为:
i-l
=∑ ail- a) 7.- 5,.- 1-^ xA
=
在计算公式(24) 中的 VC'的过程中, 如果将(1- ) *A视为变量, 则 公式(24)是关于 (1- ) *A 的二次函数, 因此, VC'存在最小值。 为了与上 述求解第一方差的过程一致, 也可以利用二次函数最小值公式(4ac-b2) /4a求 出 VC'的最小值 VCM«,为:
VCMin' = ^α(\- -7.)2 - - a\ IS- -Tt
i=l i=l ' (25) 其中, 在计算 VCM^,的过程中, 为了简化计算过程, 可令
SLqt =^^(ΐ~ α 1(厂^)2 (26)
i=l
t 、i— 1
SDqt =K - aj ^St-Ttj (27)
i=l
t-l i_x
SDqt = Voril- a) (5, -ΤΛ
H , ^ (30)
SLqt = SLql_l x(l-a) + (5· -T()xa
(31)
SDqt = SDqtx l a) + (Ξ「7 )χα
(32)
将公式(31)和(32)代入公式 (28) 并通过迭代计算的方法便可以计算 出 VCMin'。 需要说明的是, 在本发明实施例中, 仅以步骤 403和步骤 404分别先计算第 一方差后计算第二方差为例进行了举例说明。 具体实施时, 也可以先计算第二 方差, 后计算第一方差。 也就是说, 本发明实施例不对计算第一方差和第二方
差的顺序进行限定。
405: 根据第一方差与第二方差的比较结果确定是否传输拥塞。
由于通过步骤 403和步骤 404求得了第一方差和第二方差, 因此, 可以根据 第一方差与第二方差的比较结果确定是否传输拥塞。 关于根据第一方差与第二 方差的比较结果确定是否传输拥塞的方式, 包括但不限于: 如果第一方差小于 第二方差, 则确定非传输拥塞; 如果第一方差不小于第二方差, 则确定为传输 拥塞。其中,在将第一方差和第二方差进行比较时, 包括但不限于选择步骤 403 和步骤 404中获得的第一方差的最小值 和第二方差的最小值 VCMiVi或者 第一方差的最小值 WVCMz 和第二方差的最小值 VCMin '来实现。
具 ϋ也, 可令 BR = VNCMin - VCMin , 若 SR <0, 贝1 j确定 传输拥塞; BR≥0 , 贝1 J 确定传输拥塞。 者令 = VNCMin'-VCMin' ,若 SR <0,则确定非传输拥塞; BR≥0 , 则确定传输拥塞。
综上, 在将第一方差与第二方差进行比较来确定是否传输拥塞时, 如果由 于在步骤 403和步骤 404中确定第一方差和第二方差时有两种情况。 因此, 如果 确定第一方差时釆用第一种情况所述的方法, 则在将第一方差和第二方差进行 比较时, 第二方差也釆用通过第一种情况确定的值; 如果确定第一方差时釆用 第二种情况所述的方法, 则在将第一方差和第二方差进行比较时, 第二方差也 釆用通过第二种情况确定的值。
另外, 通过实验结果得到, 通过本发明实施例提供的第一种情况求得的第 一方差和第二方差进行比较从而确定的传输拥塞比相关技术确定的传输拥塞 的准确率高 80%; 本发明实施例提供的第二种情况求得的第一方差和第二方差 进行比较从而确定的传输拥塞比相关技术确定的传输拥塞的准确率高 85 %。 因 此, 通过本发明实施例提供的方法能够比较准确的确定是否传输拥塞。
需要说明的是, 上述过程仅以 VNC和 的最小值 VNCMin和 VCMin之间的关 系或者 WVC'和 W的最小值 WVCMz 和 VCM n'之间的关系来确定是否传输拥塞。 在具体实施时, 也可以釆用 WVC和 或者 WVC'和 W的其它相对应的值来确定
是否传输拥塞, 本发明实施例对此不作具体限定。
本发明实施例提供的方法, 由于实际丟包数中可能同时存在由于传输拥塞 导致的丟包和传输错误导致的丟包,通过根据预设数值个传输周期的实际发包 数和实际丟包数确定非传输拥塞时的第一线性关系以及传输拥塞时的第二线 性关系后,根据实际丟包数与根据第一线性关系和第二线性关系得到的理论丟 包数确定的第一参数与第二参数的比较结果来确定是否传输拥塞,使得确定的 传输拥塞更加准确, 从而可以控制分配合适的传输网络接口带宽, 进而提高传 输网络的吞吐率。 参见图 6, 本发明实施例提供了一种传输拥塞的确定装置, 该装置可用于 执行上述图 3或图 4所示实施例提供的传输拥塞的确定方法, 该装置包括: 获取模块 601,用于获取预设数值个传输周期的实际发包数和实际丟包数; 第一确定模块 602, 用于根据每个传输周期的实际发包数和实际丟包数确 定非传输拥塞时实际发包数和实际丟包数之间的第一线性关系;
第二确定模块 603, 用于根据每个传输周期的实际发包数和实际丟包数确 定传输拥塞时实际发包数和实际丟包数之间的第二线性关系;
第三确定模块 604, 用于根据第一线性关系确定每个传输周期的实际发包 数对应的第一理论丟包数;
第四确定模块 605, 用于根据各个传输周期的实际丟包数及第一理论丟包 数确定第一参数;
第五确定模块 606, 用于根据第二线性关系确定每个传输周期的实际发包 数对应的第二理论丟包数;
第六确定模块 607, 用于根据各个传输周期的实际丟包数及第二理论丟包 数确定第二参数;
第七确定模块 608, 用于根据第一参数与第二参数的比较结果确定是否传 输拥塞。
可选地,根据每个传输周期的实际发包数和实际丟包数确定的非传输拥塞 时实际发包数和实际丟包数之间的第一线性关系为:
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发 包数和实际丟包数之间的第二线性关系为:
T' = S - (l - 0) x A
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数, 5为发包 数, Α为拥塞阔值, 为传输错误导致的丟包率。
可选地, 第四确定模块 605, 用于根据各个传输周期的实际丟包数及第一 理论丟包数确定第一方差, 将第一方差作为第一参数;
第六确定模块 607, 用于根据各个传输周期的实际丟包数及第二理论丟包 数确定第二方差, 将第二方差作为第二参数。
可选地, 第四确定模块 605, 用于根据各个传输周期的实际丟包数及第一 理论丟包数通过如下公式确定第一方差:
N 、2
£5,.^ ¾2 - | ^5^ 7.
i=l
VNCMin = i=1 i=1 N
i=l
7 = 5; - ;
第六确定模块 607, 用于根据各个传输周期的实际丟包数及第二理论丟包 数通过如下公式确定第二方差:
∑(H)
VCMin = 1=1
N ;
其中, WVCMVi为第一方差, VCM^为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7;为每个传输周期的实际丟包数, N 为预设数值。
可选地, 第四确定模块 605, 用于根据各个传输周期的实际丟包数及第一
理论丟包数通过如下公式确定第一方差:
I N
∑Τ · a[\ - a
i=l
VNCMin' - ail - a) S;'
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值, 广为第 个传输周期的发包数和实际丟包数在计算所述第 二方差时的权重。
本发明实施例提供的装置, 由于实际丟包数中可能同时存在由于传输拥塞 导致的丟包和传输错误导致的丟包,通过根据预设数值个传输周期的实际发包 数和实际丟包数确定非传输拥塞时的第一线性关系以及传输拥塞时的第二线 性关系后,根据实际丟包数与根据第一线性关系和第二线性关系得到的理论丟 包数确定的第一参数与第二参数的比较结果来确定是否传输拥塞,使得确定的 传输拥塞更加准确, 从而可以控制分配合适的传输网络接口带宽, 进而提高传 输网络的吞吐率。 参见图 7, 本发明实施例提供了一种传输拥塞的确定装置, 该装置可用于 执行图 3或图 4所示实施例提供的传输拥塞的确定方法。 该装置包括至少一个 处理器 701和存储处理器 701可执行的指令的存储器 702。 其中:
处理器 701, 用于获取预设数值个传输周期的实际发包数和实际丟包数; 处理器 701, 还用于根据每个传输周期的实际发包数和实际丟包数确定非
传输拥塞时实际发包数和实际丟包数之间的第一线性关系, 并根据每个传输周 期的实际发包数和实际丟包数确定传输拥塞时实际发包数和实际丟包数之间 的第二线性关系;
处理器 701, 还用于根据第一线性关系确定每个传输周期的实际发包数对 应的第一理论丟包数, 并根据各个传输周期的实际丟包数及第一理论丟包数确 定第一参数;
处理器 701, 还用于根据第二线性关系确定每个传输周期的实际发包数对 应的第二理论丟包数, 并根据各个传输周期的实际丟包数及第二理论丟包数确 定第二参数;
处理器 701, 还用于根据第一参数与第二参数的比较结果确定是否传输拥 塞。
可选地,根据每个传输周期的实际发包数和实际丟包数确定的非传输拥塞 时实际发包数和实际丟包数之间的第一线性关系为:
Τ = Ξ χ θ
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发 包数和实际丟包数之间的第二线性关系为:
T' = S - (l - 0) x A
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数, 5为发包 数, Α为拥塞阔值, 为传输错误导致的丟包率。
可选地, 处理器 701, 还用于根据各个传输周期的实际丟包数及第一理论 丟包数确定第一方差, 将第一方差作为第一参数;
处理器 701, 还用于根据各个传输周期的实际丟包数及第二理论丟包数确 定第二方差, 将第二方差作为第二参数。
可选地, 处理器 701, 还用于根据各个传输周期的实际丟包数及第一理论 丟包数通过如下公式确定第一方差:
i=l i=l
VNCMin -
∑s,2
i=l
处理器 701, 还用于根据各个传输周期的实际丟包数及第二理论丟包数通 过如下公式确定第二方差:
VCMin:
N 其中, WVCMVi为第一方差, VCM^为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值。
可选地, 处理器 701, 还用于根据各个传输周期的实际丟包数及第一理论 丟包数通过如下公式确定第一方差:
[ N
ail - a
VNCMin' - ail - a) S-
处理器 701, 还用于根据各个传输周期的实际丟包数及第二理论丟包数通 过如下公式确定第二方差:
VCMin' =
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7.为每个传输周期的实际丟包数, N 为预设数值, 广为第 个传输周期的发包数和实际丟包数在计算所述第 二方差时的权重。
本发明实施例提供的装置, 由于实际丟包数中可能同时存在由于传输拥塞
导致的丟包和传输错误导致的丟包,通过根据预设数值个传输周期的实际发包 数和实际丟包数确定非传输拥塞时的第一线性关系以及传输拥塞时的第二线 性关系后,根据实际丟包数与根据第一线性关系和第二线性关系得到的理论丟 包数确定的第一参数与第二参数的比较结果来确定是否传输拥塞,使得确定的 传输拥塞更加准确, 从而可以控制分配合适的传输网络接口带宽, 进而提高传 输网络的吞吐率。 需要说明的是: 上述实施例提供的传输拥塞的确定装置在确定传输拥塞 时, 仅以上述各功能模块的划分进行举例说明, 实际应用中, 可以根据需要而 将上述功能分配由不同的功能模块完成, 即将装置的内部结构划分成不同的功 能模块, 以完成以上描述的全部或者部分功能。 另外, 上述实施例提供的传输 拥塞的确定装置与传输拥塞的确定方法实施例属于同一构思, 其具体实现过程 详见方法实施例, 这里不再赘述。
本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通 过硬件来完成, 也可以通过程序来指令相关的硬件完成, 所述的程序可以存储 于一种计算机可读存储介质中, 上述提到的存储介质可以是只读存储器, 磁盘 或光盘等。
以上所述仅为本发明的较佳实施例, 并不用以限制本发明, 凡在本发明的 精神和原则之内, 所作的任何修改、 等同替换、 改进等, 均应包含在本发明的 保护范围之内。
Claims
1、 一种传输拥塞的确定方法, 其特征在于, 所述方法包括:
获取预设数值个传输周期的实际发包数和实际丟包数;
根据每个传输周期的实际发包数和实际丟包数确定非传输拥塞时实际发包 数和实际丟包数之间的第一线性关系, 并根据每个传输周期的实际发包数和实 际丟包数确定传输拥塞时实际发包数和实际丟包数之间的第二线性关系;
根据所述第一线性关系确定每个传输周期的实际发包数对应的第一理论丟 包数, 并根据各个传输周期的实际丟包数及第一理论丟包数确定第一参数; 根据所述第二线性关系确定每个传输周期的实际发包数对应的第二理论丟 包数, 并根据各个传输周期的实际丟包数及第二理论丟包数确定第二参数; 根据所述第一参数与所述第二参数的比较结果确定是否传输拥塞。
2、 根据权利要求 1所述的方法, 其特征在于, 根据每个传输周期的实际发 包数和实际丟包数确定的非传输拥塞时实际发包数和实际丟包数之间的第一线 性关系为:
Τ = Ξ χ θ
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发包 数和实际丟包数之间的第二线性关系为:
T' = S - (l - 0) x A
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数 5为发包数, Α为拥塞阔值, 为传输错误导致的丟包率。
3、 根据权利要求 1或 2所述的方法, 其特征在于, 所述根据各个传输周期 的实际丟包数及第一理论丟包数确定第一参数, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差, 将所述
第一方差作为第一参数;
所述根据各个传输周期的实际丟包数及第二理论丟包数确定第二参数, 包 括:
根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差, 将所述 第二方差作为第二参数。
4、 根据权利要求 3所述的方法, 其特征在于, 所述根据各个传输周期的实 际丟包数及第一理论丟包数确定第一方差, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定所述 第一方差:
7; = 5; - ;
所述根据各个传输周期的实际丟包数及第二理论丟包数确定第二方差, 包 根据各个传输周期的实际丟包数及第二理论丟包数通过如下公式确定所述 第二方差:
∑(H)
VCMin:
N 其中, WVCM^为第一方差, VCM^为第二方差, 5;为每个传输周期的实际发 包数, R;为每个传输周期的实际收包数, 7为每个传输周期的实际丟包数, N为 预设数值。
5、 根据权利要求 3所述的方法, 其特征在于, 所述根据各个传输周期的实 际丟包数及第一理论丟包数确定第一方差, 包括:
根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定所述 第一方差:
根据各个传输周期的实际丟包数及第二理论丟包数通过如下公式确定所述 第二方差:
-1 , 、2
VCMin' ^a - a) 5. - 7;. ^ail- a)1 (St -Τ\
(=1
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, R;为每个传输周期的实际收包数, 7;为每个传输周期的实际丟包数, N 为预设数值, a(l- «广为第 个传输周期的发包数和实际丟包数在计算所述第二 方差时的权重。
6、 一种传输拥塞的确定装置, 其特征在于, 所述装置包括:
获取模块, 用于获取预设数值个传输周期的实际发包数和实际丟包数; 第一确定模块, 用于根据每个传输周期的实际发包数和实际丟包数确定非 传输拥塞时实际发包数和实际丟包数之间的第一线性关系;
第二确定模块, 用于根据每个传输周期的实际发包数和实际丟包数确定传 输拥塞时实际发包数和实际丟包数之间的第二线性关系;
第三确定模块, 用于根据所述第一线性关系确定每个传输周期的实际发包 数对应的第一理论丟包数;
第四确定模块, 用于根据各个传输周期的实际丟包数及第一理论丟包数确
定第一参数;
第五确定模块, 用于根据所述第二线性关系确定每个传输周期的实际发包 数对应的第二理论丟包数;
第六确定模块, 用于根据各个传输周期的实际丟包数及第二理论丟包数确 定第二参数;
第七确定模块, 用于根据所述第一参数与所述第二参数的比较结果确定是 否传输拥塞。
7、 根据权利要求 6所述的装置, 其特征在于, 根据每个传输周期的实际发 包数和实际丟包数确定的非传输拥塞时实际发包数和实际丟包数之间的第一线 性关系为:
Τ = Ξ χ θ
根据每个传输周期的实际发包数和实际丟包数确定的传输拥塞时实际发包 数和实际丟包数之间的第二线性关系为:
T' = S - (l - 0) x A
其中, Γ为非传输拥塞时的丟包数, Γ'为传输拥塞时的丟包数 5为发包数, Α为拥塞阔值, 为传输错误导致的丟包率。
8、 根据权利要求 6或 7所述的装置, 其特征在于, 所述第四确定模块, 用 于根据各个传输周期的实际丟包数及第一理论丟包数确定第一方差, 将所述第 一方差作为第一参数;
所述第六确定模块, 用于根据各个传输周期的实际丟包数及第二理论丟包 数确定第二方差, 将所述第二方差作为第二参数。
VCMin:
N 其中, WVCM^为第一方差, VCM^为第二方差, 5;为每个传输周期的实际发 包数, R;为每个传输周期的实际收包数, 为每个传输周期的实际丟包数, N为 预设数值。
10、 根据权利要求 8 所述的装置, 其特征在于, 所述第四确定模块, 用于 根据各个传输周期的实际丟包数及第一理论丟包数通过如下公式确定所述第一
所述第六确定模块, 用于根据各个传输周期的实际丟包数及第二理论丟包 数通过如下公式确定所述第二方差:
VCMin' = ^αίΐ- a) 5. -7;. ^αίΐ- a) (St -Tt\
i=l
式中, WVCMZ 为第一方差, VCMZ 为第二方差, 5;为每个传输周期的实际 发包数, 为每个传输周期的实际收包数, ;为每个传输周期的实际丟包数, N
为预设数值, a(l- «广为第 个传输周期的发包数和实际丟包数在计算所述第. 方差时的权重。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2014/081307 WO2016000191A1 (zh) | 2014-06-30 | 2014-06-30 | 传输拥塞的确定方法及装置 |
CN201480035957.8A CN105340318B (zh) | 2014-06-30 | 2014-06-30 | 传输拥塞的确定方法及装置 |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
PCT/CN2014/081307 WO2016000191A1 (zh) | 2014-06-30 | 2014-06-30 | 传输拥塞的确定方法及装置 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016000191A1 true WO2016000191A1 (zh) | 2016-01-07 |
Family
ID=55018276
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2014/081307 WO2016000191A1 (zh) | 2014-06-30 | 2014-06-30 | 传输拥塞的确定方法及装置 |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN105340318B (zh) |
WO (1) | WO2016000191A1 (zh) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018100291A1 (fr) | 2016-12-01 | 2018-06-07 | Roquette Freres | Procédé de fabrication d'une composition de chewing gum avec pates à mâcher sans cuisson |
CN112383937A (zh) * | 2020-08-20 | 2021-02-19 | 佳格科技(浙江)股份有限公司 | 无线通信质量检测方法及系统 |
Families Citing this family (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109688064B (zh) * | 2018-12-19 | 2022-11-11 | 北京达佳互联信息技术有限公司 | 数据传输方法、装置、电子设备和存储介质 |
CN112422356B (zh) * | 2019-08-22 | 2022-07-29 | 腾讯科技(深圳)有限公司 | 确定网络状态的方法及装置 |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6424624B1 (en) * | 1997-10-16 | 2002-07-23 | Cisco Technology, Inc. | Method and system for implementing congestion detection and flow control in high speed digital network |
CN102468941A (zh) * | 2010-11-18 | 2012-05-23 | 华为技术有限公司 | 网络丢包处理方法及装置 |
CN102752076A (zh) * | 2012-06-20 | 2012-10-24 | 华为技术有限公司 | 数据发送的控制方法及装置及计算机系统 |
CN103152278A (zh) * | 2013-01-31 | 2013-06-12 | 北京星网锐捷网络技术有限公司 | 拥塞确定方法、装置和网络设备 |
-
2014
- 2014-06-30 WO PCT/CN2014/081307 patent/WO2016000191A1/zh active Application Filing
- 2014-06-30 CN CN201480035957.8A patent/CN105340318B/zh active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6424624B1 (en) * | 1997-10-16 | 2002-07-23 | Cisco Technology, Inc. | Method and system for implementing congestion detection and flow control in high speed digital network |
CN102468941A (zh) * | 2010-11-18 | 2012-05-23 | 华为技术有限公司 | 网络丢包处理方法及装置 |
CN102752076A (zh) * | 2012-06-20 | 2012-10-24 | 华为技术有限公司 | 数据发送的控制方法及装置及计算机系统 |
CN103152278A (zh) * | 2013-01-31 | 2013-06-12 | 北京星网锐捷网络技术有限公司 | 拥塞确定方法、装置和网络设备 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018100291A1 (fr) | 2016-12-01 | 2018-06-07 | Roquette Freres | Procédé de fabrication d'une composition de chewing gum avec pates à mâcher sans cuisson |
CN112383937A (zh) * | 2020-08-20 | 2021-02-19 | 佳格科技(浙江)股份有限公司 | 无线通信质量检测方法及系统 |
Also Published As
Publication number | Publication date |
---|---|
CN105340318A (zh) | 2016-02-17 |
CN105340318B (zh) | 2019-02-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Zhang et al. | Novel quick start (QS) method for optimization of TCP | |
JP6077047B2 (ja) | ネットワーク中で通信するためのシステムおよび方法 | |
US9596281B2 (en) | Transport accelerator implementing request manager and connection manager functionality | |
CN107624232B (zh) | 在通信系统中控制下行链路吞吐量的设备和方法 | |
US8351331B2 (en) | Resource allocation framework for wireless/wired networks | |
US10715442B2 (en) | Congestion control | |
WO2016086551A1 (zh) | 一种基于改进的wred的拥塞控制方法和装置 | |
WO2011133624A1 (en) | Congestion window control based on queuing delay and packet loss | |
WO2015007099A1 (zh) | 数据包传输方法及装置 | |
WO2018112877A1 (zh) | 路径计算和访问请求分发方法、装置及系统 | |
US8570864B2 (en) | Kernel awareness of physical environment | |
KR20140134678A (ko) | 네트워크 트래픽에서의 도착간 지연의 감소 | |
KR101141160B1 (ko) | 네트워크 장치를 위한 버퍼 제어 방법 | |
CN110730469B (zh) | 一种基于扩展卡尔曼无线网络带宽预测方法及其拥塞控制 | |
WO2016000191A1 (zh) | 传输拥塞的确定方法及装置 | |
US20240098155A1 (en) | Systems and methods for push-based data communications | |
US10952102B2 (en) | Method and apparatus for controlling data transmission speed in wireless communication system | |
US9843526B2 (en) | Pacing enhanced packet forwarding/switching and congestion avoidance | |
CN102752076B (zh) | 数据发送的控制方法及装置及计算机系统 | |
US11190430B2 (en) | Determining the bandwidth of a communication link | |
Lu et al. | FAMD: A Flow-Aware Marking and Delay-based TCP algorithm for datacenter networks | |
US11258716B1 (en) | System and method for optimizing dynamic multi-stream network connections | |
KR101348620B1 (ko) | 무선 근거리통신망에서 a-mpdu 전송 방법 | |
Nalini et al. | To improve the performance of wireless networks for resizing the buffer | |
WO2019124290A1 (ja) | 送信データ量制御装置、方法および記録媒体 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 201480035957.8 Country of ref document: CN |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14896656 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 14896656 Country of ref document: EP Kind code of ref document: A1 |