CN113206724B - Predictable deterministic scheduling method and device suitable for quasi-dynamic link - Google Patents
Predictable deterministic scheduling method and device suitable for quasi-dynamic link Download PDFInfo
- Publication number
- CN113206724B CN113206724B CN202110287747.XA CN202110287747A CN113206724B CN 113206724 B CN113206724 B CN 113206724B CN 202110287747 A CN202110287747 A CN 202110287747A CN 113206724 B CN113206724 B CN 113206724B
- Authority
- CN
- China
- Prior art keywords
- transmission
- frame data
- target transmission
- credit value
- moment
- 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.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/0001—Systems modifying transmission characteristics according to link quality, e.g. power backoff
- H04L1/0002—Systems modifying transmission characteristics according to link quality, e.g. power backoff by adapting the transmission rate
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention provides a predictable deterministic scheduling method and a predictable deterministic scheduling device suitable for a quasi-dynamic link, wherein the link grade of a communication link is determined by acquiring distance information and delay information of the communication link; obtaining a CBS parameter at the target transmission moment according to the link level at the target transmission moment and the first reserved bandwidth; determining a credit value of a target transmission moment CBS; determining a target transmission queue for transmitting frame data at a target transmission time in a plurality of transmission queues; sending a switching rate notification to other equipment to enable the transmission rate of a communication link between the current equipment and the other equipment to be converted into the transmission rate required by the target transmission moment; and when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time. Therefore, the link transmission is scheduled by the predictable deterministic scene, and the stability and reliability of the data transmission of the quasi-dynamic system can be improved.
Description
Technical Field
The invention belongs to the technical field of communication, and particularly relates to a predictable deterministic scheduling method and device suitable for a quasi-dynamic link.
Background
With the progress and development of deterministic network technology, the application scenario of deterministic network is no longer limited to fixed network system, and the field of dynamic cooperative work is continuously expanded. For example, a real-time distributed cooperation system is built among satellite nodes, and real-time information is transmitted by using high-speed inter-satellite links; unmanned aerial vehicle formation perceives time delay information difference through the distance between each other, thereby completing cooperative communication or observation tasks.
The deterministic Network technology needs support of a deterministic Network protocol, and is currently most widely applied to a Time Sensitive Network (TSN), wherein the shaping and scheduling technology of frame data is the most critical, and the distribution and delay size of data streams are determined. The traditional time-sensitive network scheduling technology is mainly established on the time and bandwidth reliability of a fixed network optical fiber communication link, and introduces a corresponding scheduling IEEE802.1Qav traffic shaping technology on the basis of ensuring a time synchronization protocol to provide lower time delay and jitter for high-priority data streams.
In the quasi-dynamic link communication process, the positions of dynamic nodes such as satellites, unmanned aerial vehicles and vehicles are constantly changed and are easily interfered by external factors, and constant speed and stable transmission delay cannot be kept. Conventional TSN networks are not suitable for this new link scenario. If the traditional TSN scheduling technology is still adopted, data loss among dynamic nodes can be caused, and therefore reliability and stability of the whole distributed cooperation system are affected.
Disclosure of Invention
In a first aspect, the present invention provides a predictable deterministic scheduling method applicable to a quasi-dynamic link, which is applied to each device in a dynamic network, and includes:
s1, aiming at the current equipment, obtaining distance information and delay information of communication links established between other equipment and the current equipment at the target transmission moment;
s2, determining a link level of the communication link based on the distance information and the delay information;
each link level corresponds to a transmission rate, and each transmission rate corresponds to a first reserved bandwidth;
s3, according to the link level and the first reserved bandwidth of the target transmission moment, changing each parameter in the shaper CBS based on the credit value at the last transmission moment to obtain the CBS matched with the transmission rate required by the target transmission moment;
the CBS is the CBS of frame data in a transmission queue specified in a communication protocol of a communication link;
s4, determining the credit value of the CBS at the target transmission time according to each parameter in the CBS at the target transmission time;
s5, based on the credit value and the grade of the transmission queue, determining a target transmission queue for transmitting frame data at a target transmission time in a plurality of transmission queues;
s6, sending a switching rate notification to the other device, so that the other device performs parameter setting, and the transmission rate of the communication link between the current device and the other device is converted to the transmission rate required by the target transmission time;
and S7, when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time.
Optionally, the parameters of the CBS include a credit decrease rate, a credit increase rate, a credit lower bound, and a credit upper bound, and the step of S3 includes:
obtaining the CBS of the last transmission moment;
calculating the lower bound of the credit value of the target transmission moment by using a preset lower bound calculation formula of the credit value;
calculating the credit value upper bound of the target transmission moment by using a preset credit value upper bound calculation formula;
changing the credit value of the CBS at the last transmission moment into a first reserved bandwidth value at the target transmission moment;
changing the rate of reduction of the credit value of the CBS at the last transmission moment into the difference between the transmission rate required by the target transmission moment and the first reserved bandwidth;
and changing the lower bound of the credit value into the lower bound of the credit value of the target transmission time in the CBS of the last transmission time, and changing the upper bound of the credit value into the upper bound of the credit value of the target transmission time.
Optionally, the step of S5 includes:
aiming at a plurality of transmission queues, when the credit value of frame data sent by the transmission queues is a negative value, determining a target transmission queue according to the grade of the transmission queues from high to low;
when the credit value of the frame data sent by the transmission queue is not a negative value, determining the transmission queue as a target transmission queue;
and transmitting the frame data sent by the target transmission queue at the target transmission time.
Optionally, the step of S5 includes:
step a: judging whether the level of frame data currently transmitted by a port is highest;
step b: if the level of the frame data currently being transmitted by the port is the highest, judging whether the credit value of the frame data is reduced to the lower bound of the credit value after the transmission of the frame data is finished, and if the level of the frame data currently being transmitted by the port is not the highest, judging whether a transmission queue with the highest level is an empty queue;
step c: if the credit value of the frame data is reduced to the lower bound of the credit value after the transmission of the frame data is finished, keeping the credit value unchanged and returning to the step a;
d, if the transmission queue with the highest level is an empty queue, judging whether the credit value of the transmission queue with the highest level is a negative value, if so, increasing the credit value according to the credit value increasing rate, returning to the step a, and if not, setting the credit value to zero and returning to the step a;
and e, if the transmission queue with the highest level is not an empty queue, judging whether the credit value of the transmission queue with the highest level reaches the upper limit of the credit value, if so, keeping the credit value unchanged and returning to the step a.
Optionally, after the step of S2 and before the step of S3, the predictable deterministic scheduling method further includes:
determining a first reserved bandwidth corresponding to a link level at a target transmission moment;
judging whether the transmission rate of the communication link at the target transmission moment is the same as the transmission rate at the last transmission moment;
if the transmission rate of the communication link at the target transmission moment is the same as the transmission rate at the last transmission moment, acquiring a request for accurately determining a first reserved bandwidth at the target transmission moment;
and re-determining the first reserved bandwidth based on the difference value between the first reserved bandwidth and the second reserved bandwidth at the last transmission moment to obtain the accurately determined first reserved bandwidth.
Optionally, after the step of obtaining whether to accurately determine the first reserved bandwidth at the target transmission time if the transmission rate of the communication link at the target transmission time is the same as the transmission rate at the previous transmission time, the predictable deterministic scheduling method further includes:
when the request is a request which does not need to accurately determine the first reserved bandwidth at the current moment, determining the lower limit of the transmission rate of the frame data with the lowest priority based on the preset tolerant time delay of the frame data with the lowest priority;
the frame data of the highest level is stored in the transmission queue with the highest priority, and the frame data of the lowest priority is stored in the transmission queue with the next highest priority;
judging whether the transmission rate based on the transmission of the lowest priority frame data is the same as the lower limit of the transmission rate of the lowest priority frame data, if not, determining the target transmission rate for transmitting the highest level frame data based on the lower limit of the transmission rate of the lowest priority frame data;
and re-estimating the first reserved bandwidth based on the target transmission rate.
Optionally, after the step of re-estimating the first reserved bandwidth based on the target transmission rate, the predictable deterministic scheduling method further includes:
sending a switching rate notification to other equipment to enable the other equipment to carry out parameter setting so that the transmission rate of the communication link is converted into a target transmission rate when the current equipment transmits the highest-level frame data;
and when the target transmission time for transmitting the highest-level frame data is reached, transmitting the highest-level frame data at the target transmission rate.
Optionally, before the step of S1, the method for predictable deterministic scheduling further includes:
for each frame data in the transmission queue with the highest grade and the next highest grade, executing the steps from S1 to S4 on the target transmission time of the frame data to obtain the credit value of each frame data;
searching a lowest credit value in the credit values of each frame data, and determining the target transmission time of the frame data with the lowest credit value as the lowest point time;
compressing the lowest point moment into a frame, and sending the frame to other equipment so as to enable the other equipment to switch the transmission rate of the communication link to be the transmission rate matched with the lowest credit value when the lowest point moment arrives;
scheduling the frame data in the transmission queue except for the highest level and the second highest level by using a priority selection algorithm, determining the transmission rate of a communication link and sending the transmission rate to other equipment;
and when the lowest point moment arrives, transmitting the frame data with the lowest credit value in the transmission queue at the transmission rate of the lowest point moment and transmitting the frame data at the transmission rate corresponding to the target transmission moment.
In a second aspect, the present invention provides a predictable deterministic scheduling apparatus suitable for quasi-dynamic links, which is applied to each device in a dynamic network, and includes:
the link distance and delay detection module is used for,
aiming at the current equipment, obtaining distance information and delay information of communication links established between other equipment and the current equipment at the target transmission moment;
the dynamic bandwidth configuration module is configured to,
determining a link level of the communication link based on the distance information and delay information;
each link level corresponds to a transmission rate, and each transmission rate corresponds to a first reserved bandwidth;
according to the link level and the first reserved bandwidth at the target transmission moment, changing various parameters in the CBS at the last transmission moment to obtain the CBS matched with the transmission rate required by the target transmission moment;
the CBS is the CBS of frame data in a transmission queue specified in a communication protocol of a communication link;
the credit shaping algorithm control module is configured to,
determining a credit value of the CBS at the target transmission moment according to various parameters in the CBS at the target transmission moment;
the transmission selection module is used for selecting the transmission,
determining a target transmission queue for transmitting frame data at a target transmission time in the plurality of transmission queues based on the credit value and the grade of the transmission queue;
the transmission control module is used for controlling the transmission of the data,
sending a switching rate notification to other equipment to enable the other equipment to carry out parameter setting, so that the transmission rate of a communication link between the current equipment and the other equipment is converted into the transmission rate required by the target transmission moment;
and when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time.
The invention provides a predictable deterministic scheduling method suitable for a quasi-dynamic link, which determines the link grade of a communication link by acquiring the distance information and the delay information of the communication link; obtaining a CBS parameter at the target transmission moment according to the link level at the target transmission moment and the first reserved bandwidth; determining a credit value of a target transmission moment CBS; determining a target transmission queue for transmitting frame data at a target transmission time in a plurality of transmission queues; sending a switching rate notification to other equipment to enable the transmission rate of a communication link between the current equipment and the other equipment to be converted into the transmission rate required by the target transmission moment; and when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time. Therefore, the link transmission is scheduled by the predictable deterministic scene, and the stability and reliability of the data transmission of the quasi-dynamic system can be improved.
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Drawings
Fig. 1 is a flowchart of a predictable deterministic scheduling method applicable to a quasi-dynamic link according to an embodiment of the present invention;
FIG. 2 is a block diagram of a satellite network provided by an embodiment of the present invention;
FIG. 3 is a diagram of link distance variation provided by an embodiment of the present invention
FIG. 4 is a schematic diagram of a dynamic CBS update process provided by an embodiment of the present invention;
FIG. 5 is a flowchart of a scenario adjustment policy provided by an embodiment of the present invention;
fig. 6 is a schematic diagram illustrating transmission rate switching during frame data transmission according to an embodiment of the present invention;
fig. 7 is a structural diagram of a predictable deterministic scheduling apparatus suitable for a quasi-dynamic link according to an embodiment of the present invention;
fig. 8 is a block diagram of another predictable deterministic scheduling apparatus suitable for a quasi-dynamic link according to an embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to specific examples, but the embodiments of the present invention are not limited thereto.
Example one
Before describing the present invention, some of the expertise involved with the present invention is first described.
TSN is a set of protocol standards proposed by the IEEE802.1 working group in 2012, which defines the transport mechanism for time-sensitive class data over deterministic ethernet. The method mainly comprises key protocols such AS IEEE802.1 AS (synchronization), IEEE802.1Qav (scheduling), IEEE802.1Qbv (scheduling) and the like.
Quasi-dynamic link:
the traditional deterministic network link takes a ground system as a standard, the distance between nodes is fixed, and meanwhile, the transmission rate is not changed. When the distance between network nodes is not fixed any more, the distance changes and the change of the link rate is influenced, but the change is not random and can be predicted and sensed, the change of the distance can be theoretically described by a corresponding mathematical function, and the new link form is called a quasi-dynamic link in the text.
Predictable certainty:
the certainty of the current TSN technology means that the time delay of data after being processed and forwarded by a scheduling system is determined, and does not exceed a certain upper bound value, and in combination with a synchronization technology, the range of data receiving time points can be defined, which is the biggest difference from the existing ethernet. Taking the AVB protocol data frame as an example, the delay of the A type data frame in a 7-hop node is not more than 2ms, and the delay of the B type data frame in the 7-hop node is not more than 50 ms.
AVB：
AVB is a TSN standard precursor. Two different data streams are mainly specified, namely an A-type data stream and a B-type data stream, the A-type data stream is an audio stream, the B-type data stream is a video stream, the IEEE802.1 Qat protocol specifies that the A-type priority is higher than the B-type priority, the sum of reserved bandwidths of the A-type and the B-type does not exceed 75% of the total bandwidth, and the remaining 25% of the reserved bandwidth transmits a Best Effort data frame, namely a BE frame (Best Effort). The class a and class B data frames are generally referred to as time sensitive network data streams.
In the ieee802.1qav protocol, it is established how to determine the traffic class of a time-sensitive flow using priority information. The priority class is mainly divided into two mechanisms, one is a strict priority algorithm, and the other is a credit shaping based CBS algorithm. The strict priority is that queue scheduling is carried out on data except time-sensitive data streams, eight paths of queues are specified in a protocol, each path of queue has a corresponding priority value, and an algorithm finishes scheduling work according to the priority value. The shaping CBS algorithm based on the credit value guarantees that every path of traffic can get an opportunity in the network data traffic transmission process, and can still guarantee that the time sensitive stream has the minimum time delay scheduling.
As shown in fig. 1, the predictable deterministic scheduling method applicable to quasi-dynamic links provided by the present invention is applied to each device in a dynamic network, and includes:
s1, aiming at the current equipment, obtaining distance information and delay information of communication links established between other equipment and the current equipment at the target transmission moment;
the target transmission time is the time when the data frame is transmitted, and the time is not arrived or just arrived.
S2, determining the link grade of the communication link based on the distance information and the delay information;
each link level corresponds to a transmission rate, and each transmission rate corresponds to a first reserved bandwidth;
s3, according to the link level and the first reserved bandwidth of the target transmission moment, changing each parameter in the shaper CBS based on the credit value at the last transmission moment to obtain the CBS matched with the transmission rate required by the target transmission moment;
the CBS is the CBS of frame data in a transmission queue specified in a communication protocol of a communication link; parameters of the CBS include a credit decrease rate, a credit increase rate, a credit lower bound, and a credit upper bound.
S4, determining the credit value of the CBS at the target transmission time according to each parameter in the CBS at the target transmission time;
s5, based on credit value and transmission queue grade, in multiple transmission queues, determining target transmission queue for transmitting frame data at target transmission time;
s6, sending a switching rate notification to the other device, so that the other device performs parameter setting, and the transmission rate of the communication link between the current device and the other device is converted to the transmission rate required by the target transmission time;
and S7, when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time.
Referring to fig. 2, taking a satellite network as an example, the satellite communication distance between the different-orbit links conforms to a certain periodic sine-cosine function. In order to have general reference, referring to fig. 3, the present invention takes a sine curve as an example to represent the variation trend of the satellite link distance, and takes a third-order rate as an example to be divided into three rate levels, namely a far rate level, a middle rate level and a near rate level, so as to form a satellite quasi-dynamic platform link. The same is true for other scenarios.
The basic working process of the CBS algorithm comprises the following steps: when data arrive in a queue adopting a CBS algorithm, if no other data are transmitted at a port at the moment, the queue is immediately transmitted, the speed of the corresponding credit at sendSlope is reduced, and if the credit is negative at the moment, the port cannot transmit data frames of the queue; if other data frames are transmitted at the port at the moment, waiting for the data in the queue, increasing the credit at the rate of the idleSlope, if the waiting data is not in the queue at the moment and the credit value is greater than 0, setting the credit to 0, and if the credit is less than 0, increasing the credit to 0 at the rate of the idleSlope, and repeating the process. The CBS algorithm essentially belongs to the token bucket algorithm. The parameters include, sendSlope: a credit reduction rate; idleSlope: a credit increase rate; logcredit: a credit value lower bound; hicredit: credit value upper bound.
The node equipment of the quasi-moving platform acquires time information and link distance parameters of links by using a distance sensing system, and determines a link rate level according to the time information and the link distance parameters. For a ground system, a scheduling mechanism is established on the basis of fixed network optical fiber and bandwidth reliability, and the transmission rate is generally unchanged; and under the scene of the quasi-moving platform, the distance between the nodes is continuously changed, but the change trend accords with a certain curve rule, a predictable periodic motion track and a multi-stage periodic change of the speed), the inherent parameters (including idleSlope, sendSlope, logcredit, hicredbit and the like) of the CBS are changed simultaneously according to a corresponding dynamic adjustment strategy, then the credit value is updated by adopting the changed CBS parameters, different adjustments are made according to the credit values under different conditions, and finally, whether the corresponding queue has the transmission qualification or not is determined by combining a priority transmission selection algorithm and is submitted to a bottom link.
The invention provides a predictable deterministic scheduling method suitable for a quasi-dynamic link, which determines the link grade of a communication link by acquiring the distance information and the delay information of the communication link; obtaining a CBS parameter at the target transmission moment according to the link level at the target transmission moment and the first reserved bandwidth; determining a credit value of a target transmission moment CBS; determining a target transmission queue for transmitting frame data at a target transmission time in a plurality of transmission queues; sending a switching rate notification to other equipment to enable the transmission rate of a communication link between the current equipment and the other equipment to be converted into the transmission rate required by the target transmission moment; and when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time. Therefore, the link transmission is scheduled by the predictable deterministic scene, and the stability and reliability of the data transmission of the quasi-dynamic system can be improved.
Example two
As an alternative embodiment of the present invention, the step of S3 includes:
s31, obtaining the CBS of the last transmission time;
s32, calculating the lower bound of the credit value of the target transmission time by using a preset lower bound calculation formula of the credit value;
s33, calculating the credit value upper bound of the target transmission time by using a preset credit value upper bound calculation formula;
s34, changing the credit value of the last transmission time CBS into the value of the first reserved bandwidth of the target transmission time;
s35, changing the rate of reduction of the credit value of the CBS at the last transmission moment into the difference between the transmission rate required by the target transmission moment and the first reserved bandwidth;
s36, the CBS at the previous transmission time changes the lower bound of the credit value to the lower bound of the credit value at the target transmission time, and changes the upper bound of the credit value to the upper bound of the credit value at the target transmission time.
Wherein, the link transmission rate switches the transmission rate at the target transmission time, and the switching time T is the following formula
Currenttime is the current time, k is the predicted time parameter, and tautransferIs the maximum time in transmission, n is the predicted time parameter, τlFor low priority blocking time, m is a predicted time parameter, τtThe process to recover from logredit to 0 is the time spent in the shaping process.
The lower bound of the credit value is calculated by the formula:
loCredit＝(RA-R0)×MA/R0
RAidleSlope, R for class A data frames0Is the port transfer rate, MAIs the maximum length of a class a data frame. The credit value upper bound calculation formula is as follows:
hiCredit＝RA×M0/R0
RAis an idleSlope, R of a class A data frame0Is the port transfer rate (i.e., link rate), M0Representing the maximum data frame length of BE class
EXAMPLE III
As an alternative embodiment of the present invention, the step of S5 includes:
s51, aiming at a plurality of transmission queues, when the credit value of the frame data sent by the transmission queues is a negative value, determining a target transmission queue according to the grade of the transmission queues from high to low;
s52, when the credit value of the frame data sent by the transmission queue is not negative, determining the transmission queue as the target transmission queue;
s53, the frame data sent from the destination transmission queue is transmitted at the destination transmission time.
Example four
As an alternative embodiment of the present invention, referring to fig. 4 in combination with embodiment three, the step of S5 includes:
step a: judging whether the level of frame data currently transmitted by a port is highest;
step b: if the level of the frame data currently being transmitted by the port is the highest, judging whether the credit value of the frame data is reduced to the lower bound of the credit value after the frame data transmission is finished, and if the level of the frame data currently being transmitted by the port is not the highest, judging whether a transmission queue with the highest level is an empty queue;
step c: if the credit value of the frame data is reduced to the lower bound of the credit value after the transmission of the frame data is finished, keeping the credit value unchanged and returning to the step a;
d, if the transmission queue with the highest level is an empty queue, judging whether the credit value of the transmission queue with the highest level is a negative value, if so, increasing the credit value according to the credit value increasing rate, returning to the step a, and if not, setting the credit value to zero and returning to the step a;
and e, if the transmission queue with the highest level is not an empty queue, judging whether the credit value of the transmission queue with the highest level reaches the upper limit of the credit value, if so, keeping the credit value unchanged and returning to the step a.
After the reserved bandwidth is initialized, a rate grade is determined by a distance sensing system, algorithm parameters are dynamically adjusted according to the rate grade, and when the rate is switched, corresponding parameters are adjusted in an equal proportion to maintain the original characteristics of the link (or the bandwidth ratio is properly reduced and increased to meet different scene requirements). When the port transmission queue is an A-type data frame (taking the A-type data frame as an example for explanation), the A-type credit value credit is decreased according to the sendSlope rate, and is kept unchanged after being decreased to a lower bound credit according to different scene additional parameters; when the port transmission queue is not the A-type queue and the A-type queue is an empty queue, if the credit value is negative, the credit value credit of the A-type queue is increased progressively according to the rate of the idleSlope, and parameters are increased according to the scene requirement; if the credit value is greater than 0, directly setting the credit value to 0; when the current transmission queue is not in the class A and the queue has data waiting for transmission, the credit value credit is continuously increased and is kept unchanged after reaching the upper limit credit. (for protecting the characteristic of link shaping, the following design is made, when the credit value reaches hicredit or logredit, the current data frame of the queue is sent out, the queue is immediately switched to the current queue for data transmission, so as to avoid overlong waiting time of the high priority queue)
EXAMPLE five
As an alternative embodiment of the present invention, after the step of S2 and before the step of S3, the predictable deterministic scheduling method further comprises:
the method comprises the following steps: determining a first reserved bandwidth corresponding to a link level at a target transmission moment;
step two: judging whether the transmission rate of the communication link at the target transmission moment is the same as the transmission rate at the last transmission moment;
step three: if the transmission rate of the communication link at the target transmission moment is the same as the transmission rate at the last transmission moment, acquiring a request for accurately determining a first reserved bandwidth at the target transmission moment;
step four: and re-determining the first reserved bandwidth based on the difference value between the first reserved bandwidth and the second reserved bandwidth at the last transmission moment to obtain the accurately determined first reserved bandwidth.
The first reserved bandwidth is a bandwidth of a medium access control layer, the second reserved bandwidth is a bandwidth of a physical layer, and the physical layer has overhead, so the first reserved bandwidth is usually smaller than the second reserved bandwidth. The embodiment can improve the accuracy of predicting and improving the first reserved bandwidth.
EXAMPLE six
As an alternative embodiment of the present invention, referring to fig. 5, after the step of obtaining a request whether to accurately determine a first reserved bandwidth at a target transmission time if a transmission rate of a communication link at the target transmission time is the same as a transmission rate at a previous transmission time, the method for predictable deterministic scheduling further includes:
the method comprises the following steps: when the request is a request for not accurately determining the first reservation bandwidth at the current moment, determining the lower limit of the transmission rate of the frame data with the lowest priority based on the preset tolerant time delay of the frame data with the lowest priority;
the frame data of the highest level is stored in the transmission queue with the highest priority, and the frame data of the lowest priority is stored in the transmission queue with the next highest priority;
step two: judging whether the transmission rate based on the transmission of the frame data with the lowest priority is the same as the lower limit of the transmission rate of the frame data with the lowest priority or not, and if not, determining the target transmission rate of the frame data with the highest priority based on the lower limit of the transmission rate of the frame data with the lowest priority;
step three: and re-estimating the first reserved bandwidth based on the target transmission rate.
EXAMPLE seven
As an optional embodiment of the present invention, after the step of re-estimating the first reserved bandwidth based on the target transmission rate, the predictable deterministic scheduling method further includes:
the method comprises the following steps: sending a switching rate notification to other equipment to enable the other equipment to carry out parameter setting so that the transmission rate of a communication link is converted into a target transmission rate when the current equipment transmits the highest-level frame data;
step two: when a target transmission time at which the highest-level frame data is transmitted is reached, the highest-level frame data is transmitted at a target transmission rate.
Example eight
As an alternative embodiment of the present invention, before the step of S1, the method for predictable deterministic scheduling further comprises:
the method comprises the following steps: for each frame data in the transmission queue with the highest grade and the next highest grade, executing the steps from S1 to S4 on the target transmission time of the frame data to obtain the credit value of each frame data;
step two: searching a lowest credit value in the credit values of each frame data, and determining the target transmission time of the frame data with the lowest credit value as the lowest point time;
step three: compressing the lowest point moment into a frame, and sending the frame to other equipment so as to enable the other equipment to switch the transmission rate of the communication link to be the transmission rate matched with the lowest credit value when the lowest point moment arrives;
step four: scheduling the frame data in the transmission queue except for the highest level and the second highest level by using a priority selection algorithm, determining the transmission rate of a communication link and sending the transmission rate to other equipment;
step five: and when the lowest point moment arrives, transmitting the frame data with the lowest credit value in the transmission queue at the transmission rate of the lowest point moment and transmitting the frame data at the transmission rate corresponding to the target transmission moment.
It can be understood that, the distance change of the quasi-dynamic link may cause the change of the physical layer reception noise ratio, and in order to fully utilize the channel capacity, a multi-rate transmission mechanism may be introduced on the quasi-dynamic link, and a corresponding link transmission rate is adopted in a specific transmission distance interval. If the link is transmitted in a burst mode, each basic transmission unit independently completes receiving demodulation, information such as modulation rate and the like can be given at a preamble of a physical layer, and a receiver can receive corresponding rate. High speed quasi-dynamic transmission links typically employ a continuous transmission scheme. At this time, because the distance between the transceiving nodes (devices) is generally very long, for example, in a space satellite link, in order to achieve transceiving coordination, both sides calculate link delay according to an appointed time and then respectively switch link rates under the condition that the transceiving nodes are kept to have the same time reference.
For a network with a deterministic requirement, the link rate switching time is in an unstable state, and data transmission cannot be performed during the switching period, thereby affecting deterministic service flow. The invention provides a new mechanism, which enables a sending node to predict the most suitable time point for switching according to the current queuing and scheduling body and enables the influence of switching action on service flow to be minimized. The method comprises the following three points:
the first point is as follows: in order to execute switching at the target transmission time selected by the sending node, the sending node can send a notification frame to the receiving node according to the original rate before switching, and then sends a subsequent data frame according to the new rate; and ensures that the receiving node has successfully completed the received mode switch when the subsequent data frame reaches the receiving node. The notification frame needs to be assembled before the switching time of the sending end and sent to the receiving end, so that the receiving end can synchronize the switching time.
According to the switching time node and the condition that the credit value is positive and negative, the CBS algorithm scheduling characteristic is utilized, when the credit value is decreased, the data frame is transmitted, and when the credit value is negative, a port cannot transmit the queue scheduled by the CBS algorithm, the notification frame is started to be constructed when the credit value is 0, and the switching time after the preset time is loaded into the notification frame through the time stamp and is transmitted to the receiving node. When the credit value is 0 and the link is transmitting high priority data frames, the credit value is decremented to negative. When the credit value is negative, a notification frame can be constructed and enqueued; when the credit value is 0 and the link has no high-priority data frame, a notification frame can be directly constructed and sent to the receiving node. In summary, inserting data frames with a credit value of 0 before the determination of the switching time (target transmission time) has minimal impact on high priority data.
And a second point: under the premise, the sending node can select the most suitable rate switching point according to the arrival and departure conditions of the current priority services.
The embodiment adopts a dynamic switching strategy, and the most suitable switching time is determined by the data frame condition on the transmission port after the CBS scheduling. Determining theoretical switching time according to the distance change grade between the dynamic nodes, if a data frame happens to be transmitted at the theoretical switching time point, and data transmission cannot be performed during the switching period according to the regulation, therefore, a protection mechanism shifts the switching time point backwards, performs rate switching and bandwidth configuration in the period after the data transmission of the port is completed, and when the data has a small time gap from the theoretical switching point after the transmission is completed and the port has no data transmission in the period, the protection mechanism shifts the switching time point forwards to complete the switching operation in advance; if the theoretical switching time point node has no frame data transmission, the transmission operation is already completed before or after, so the switching point can perform rate switching in principle as long as the data transmission is not affected. Overall, the most suitable rate switching point fluctuates above and below the theoretical switching timing. For the above procedure, for the data situation after scheduling, the influence of the switching time on the high-priority data stream is not considered, and for this reason, this embodiment uses the CBS queue information to predict the lowest credit value, (the prediction of the lowest credit value depends on the current queue information), and uses the lowest credit value and the current queue transmission situation to determine the optimal switching time, where the influence of the switching on the CBS traffic scheduling is the lowest, and the switching time T is the time of the switching time
Wherein CurrentTime is the current time, k, n, m are the time coefficients of the prediction,
the process of recovering from logredit to 0 is the time spent in the shaping process, τlFor low priority blocking time, τtransferIs the maximum time in transmission. The expression is as follows
And a third point: in order to minimize the influence on the data buffered in the FIFO, the protection bandwidth may be increased by adjusting parameters if necessary, and the data is not allowed to be transmitted within a certain time range near the switching node, or a judgment is made before the switching node to determine whether the data frame can be transmitted before the switching node. When a transmitting end determines theoretical switching time, setting a maximum data frame length time interval or a transmission time table before the switching time, starting a protection state once a data frame enters the frame length interval or exceeds the time of the time table, prohibiting the transmission of the data frame after a port finishes transmitting data, and starting a transmission mode with a new rate after the switching operation is finished; of course, the set time interval or schedule may also be shortened as much as possible, the maximum bandwidth utilization rate is obtained, only whether the data frame to be transmitted can be completed before the switching time point needs to be judged, if the data frame can be completed, the data transmission is kept, the port transmission is closed, and the link data transmission is resumed after the switching operation is completed.
The strategy for determining the switching opportunity node in the second point is influenced by the current scheduling output port, and the bandwidth dynamic configuration in the scheduling is related to the switching opportunity, so that the two are related and influenced with each other. In the third point, the transmission switching strategy is established on the basis of sacrificing a certain time delay, and because the distance of the quasi-dynamic link is long and the factors influencing the time delay are complex, if retransmission is caused by data loss, the paid time cost is far longer than the time occupied by the strategy. The switching scheme mentioned in the third point is therefore most reliable in certain scenarios.
Taking fig. 6 as an example to illustrate, for example, a link adopts third-order rate change, and for case 1, a data frame is transmitted at a time point of theoretical switching, so that a protection mechanism shifts a transmission time point backward, and performs rate switching and bandwidth configuration after the data frame is completely transmitted; for case 2, the theoretical switching time point has no data transmission, and the corresponding transmission is completed before or after, so the switching point can switch the rate in principle as long as the data transmission is not affected. In the figure, the thin dotted line is the switching time point without protection, and the thick dotted line is the switching time point after the protection mechanism is introduced.
Example nine
As shown in fig. 7, the predictable deterministic scheduling apparatus applicable to quasi-dynamic links provided by the present invention is applied to each device in a dynamic network, and the apparatus includes:
the link distance and delay detection module is used for acquiring distance information and delay information of communication links established between other equipment and the current equipment at the target transmission moment aiming at the current equipment;
the dynamic bandwidth configuration module is configured to,
determining a link level of the communication link based on the distance information and the delay information;
each link level corresponds to a transmission rate, and each transmission rate corresponds to a first reserved bandwidth;
according to the link level and the first reserved bandwidth at the target transmission moment, changing various parameters in the CBS at the last transmission moment to obtain the CBS matched with the transmission rate required by the target transmission moment;
the CBS is the CBS of frame data in a transmission queue specified in a communication protocol of a communication link;
the credit shaping algorithm control module is configured to,
determining a credit value of the CBS at the target transmission moment according to various parameters in the CBS at the target transmission moment;
the transmission selection module is used for selecting the transmission,
determining a target transmission queue for transmitting frame data at a target transmission time in the plurality of transmission queues based on the credit value and the grade of the transmission queue;
the transmission control module is used for controlling the transmission of the data,
sending a switching rate notification to other equipment to enable the other equipment to carry out parameter setting, so that the transmission rate of a communication link between the current equipment and the other equipment is converted into the transmission rate required by the target transmission moment;
and when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time.
As shown in fig. 8, another predictable deterministic scheduling apparatus suitable for a quasi-dynamic link according to the present invention corresponds to the method in the eighth embodiment, and the apparatus includes:
the credit deformation algorithm control module is used for executing the step of obtaining the credit value of each frame data by the target transmission time of the frame data aiming at each frame data in the transmission queue with the highest grade and the next highest grade to obtain the credit value of each frame data;
the process of obtaining the credit value of the frame data is not described herein again.
The best credit value unit is used for searching the lowest credit value in the credit values of each frame data;
the link port monitoring unit is used for monitoring the transmission state of the link port;
the detection switching unit is used for determining the target transmission time of the frame data with the lowest credit value as the lowest point time according to the transmission state of the link port;
the detection notification unit is used for compressing the lowest point moment into a frame and sending the frame to the sending module;
the sending module is used for sending the lowest point moment to other equipment so as to enable the other equipment to switch the transmission rate of the communication link to be the transmission rate matched with the lowest credit value when the lowest point moment arrives;
the priority selection algorithm control module is used for scheduling the frame data in the transmission queue except for the highest level and the second highest level by using a priority selection algorithm, determining the transmission rate of a communication link and sending the transmission rate to other equipment;
and when the lowest point moment arrives, transmitting the frame data with the lowest credit value in the transmission queue at the transmission rate of the lowest point moment and transmitting the frame data at the transmission rate corresponding to the target transmission moment according to the frame data.
The above process is the same as that of the eighth embodiment, and is not described again here.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.
Claims (9)
1. A predictable deterministic scheduling method applicable to quasi-dynamic links, applied to each device in a dynamic network, is characterized by comprising the following steps:
s1, aiming at the current equipment, obtaining distance information and delay information of communication links established between other equipment and the current equipment at the target transmission moment;
s2, determining a link level of the communication link based on the distance information and the delay information;
each link level corresponds to a transmission rate, and each transmission rate corresponds to a first reserved bandwidth;
s3, according to the link level and the first reserved bandwidth of the target transmission moment, changing each parameter in the shaper CBS based on the credit value at the last transmission moment to obtain the CBS matched with the transmission rate required by the target transmission moment;
the CBS is the CBS of frame data in a transmission queue specified in a communication protocol of a communication link;
s4, determining the credit value of the CBS at the target transmission time according to each parameter in the CBS at the target transmission time;
s5, based on the credit value and the grade of the transmission queue, determining a target transmission queue for transmitting frame data at a target transmission time in a plurality of transmission queues;
s6, sending a switching rate notification to the other device, so that the other device performs parameter setting, and the transmission rate of the communication link between the current device and the other device is converted to the transmission rate required by the target transmission time;
and S7, when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time.
2. The predictable deterministic scheduling method of claim 1 wherein the parameters of the CBS include a credit decrease rate, a credit increase rate, a credit lower bound, and a credit upper bound, the step of S3 comprising:
obtaining the CBS of the last transmission moment;
calculating the lower bound of the credit value of the target transmission moment by using a preset lower bound calculation formula of the credit value;
calculating the credit value upper bound of the target transmission moment by using a preset credit value upper bound calculation formula;
changing the credit value increasing rate of the last transmission time CBS into the value of the first reserved bandwidth of the target transmission time;
changing the rate of reduction of the credit value of the CBS at the last transmission moment into the difference between the transmission rate required by the target transmission moment and the first reserved bandwidth;
and changing the lower bound of the credit value into the lower bound of the credit value of the target transmission time in the CBS of the last transmission time, and changing the upper bound of the credit value into the upper bound of the credit value of the target transmission time.
3. The predictable deterministic scheduling method of claim 1 wherein the step of S5 comprises:
aiming at a plurality of transmission queues, when the credit value of frame data sent by the transmission queues is a negative value, determining a target transmission queue according to the grade of the transmission queues from high to low;
when the credit value of the frame data sent by the transmission queue is not a negative value, determining the transmission queue as a target transmission queue;
and transmitting the frame data sent by the target transmission queue at the target transmission time.
4. The predictable deterministic scheduling method of claim 3 wherein the step of S5 comprises:
step a: judging whether the level of frame data currently transmitted by a port is highest;
step b: if the level of the frame data currently being transmitted by the port is the highest, judging whether the credit value of the frame data is reduced to the lower bound of the credit value after the transmission of the frame data is finished, and if the level of the frame data currently being transmitted by the port is not the highest, judging whether a transmission queue with the highest level is an empty queue;
step c: if the credit value of the frame data is reduced to the lower bound of the credit value after the transmission of the frame data is finished, keeping the credit value unchanged and returning to the step a;
d, if the transmission queue with the highest level is an empty queue, judging whether the credit value of the transmission queue with the highest level is a negative value, if so, increasing the credit value according to the credit value increasing rate, returning to the step a, and if not, setting the credit value to zero and returning to the step a;
and e, if the transmission queue with the highest level is not an empty queue, judging whether the credit value of the transmission queue with the highest level reaches the upper limit of the credit value, if so, keeping the credit value unchanged and returning to the step a.
5. The predictable deterministic scheduling method of claim 1 wherein after the step of S2 and before the step of S3, the predictable deterministic scheduling method further comprises:
determining a first reserved bandwidth corresponding to a link level at a target transmission moment;
judging whether the transmission rate of the communication link at the target transmission moment is the same as the transmission rate at the last transmission moment;
if the transmission rate of the communication link at the target transmission moment is the same as the transmission rate at the last transmission moment, acquiring a request for accurately determining a first reserved bandwidth at the target transmission moment;
and re-determining the first reserved bandwidth based on the difference value between the first reserved bandwidth and the second reserved bandwidth at the last transmission moment to obtain the accurately determined first reserved bandwidth.
6. The predictable deterministic scheduling method of claim 5, wherein after the step of obtaining the request whether to accurately determine the first reserved bandwidth at the target transmission time if the transmission rate of the communication link at the target transmission time is the same as the transmission rate at the last transmission time, the predictable deterministic scheduling method further comprises:
when the request is a request which does not need to accurately determine the first reserved bandwidth at the current moment, determining the lower limit of the transmission rate of the frame data with the lowest priority based on the preset tolerant time delay of the frame data with the lowest priority;
the frame data of the highest level is stored in the transmission queue with the highest priority, and the frame data of the lowest priority is stored in the transmission queue with the next highest priority;
judging whether the transmission rate based on the transmission of the lowest priority frame data is the same as the lower limit of the transmission rate of the lowest priority frame data, if not, determining the target transmission rate for transmitting the highest level frame data based on the lower limit of the transmission rate of the lowest priority frame data;
and re-estimating the first reserved bandwidth based on the target transmission rate.
7. The predictable deterministic scheduling method of claim 6 wherein after the step of re-estimating the first reserved bandwidth based on the target transmission rate, the predictable deterministic scheduling method further comprises:
sending a switching rate notification to other equipment to enable the other equipment to carry out parameter setting so that the transmission rate of the communication link is converted into a target transmission rate when the current equipment transmits the highest-level frame data;
and when the target transmission time for transmitting the highest-level frame data is reached, transmitting the highest-level frame data at the target transmission rate.
8. The predictable deterministic scheduling method of claim 1 wherein prior to the step of S1, the predictable deterministic scheduling method further comprises:
for each frame data in the transmission queue with the highest grade and the next highest grade, executing the steps from S1 to S4 on the target transmission time of the frame data to obtain the credit value of each frame data;
searching a lowest credit value in the credit values of each frame data, and determining the target transmission time of the frame data with the lowest credit value as the lowest point time;
compressing the lowest point moment into a frame, and sending the frame to other equipment so as to enable the other equipment to switch the transmission rate of the communication link to be the transmission rate matched with the lowest credit value when the lowest point moment arrives;
scheduling the frame data in the transmission queue except for the highest level and the second highest level by using a priority selection algorithm, determining the transmission rate of a communication link and sending the transmission rate to other equipment;
and when the lowest point moment arrives, transmitting the frame data with the lowest credit value in the transmission queue at the transmission rate of the lowest point moment and transmitting the frame data at the transmission rate corresponding to the target transmission moment.
9. A predictable deterministic scheduling apparatus for quasi-dynamic links, applied to each device in a dynamic network, the apparatus comprising:
the link distance and delay detection module is used for,
aiming at the current equipment, obtaining distance information and delay information of communication links established between other equipment and the current equipment at the target transmission moment;
the dynamic bandwidth configuration module is configured to,
determining a link level of the communication link based on the distance information and delay information;
each link level corresponds to a transmission rate, and each transmission rate corresponds to a first reserved bandwidth;
according to the link level and the first reserved bandwidth at the target transmission moment, changing various parameters in the CBS at the last transmission moment to obtain the CBS matched with the transmission rate required by the target transmission moment;
the CBS is the CBS of frame data in a transmission queue specified in a communication protocol of a communication link;
the credit shaping algorithm control module is configured to,
determining a credit value of the CBS at the target transmission moment according to various parameters in the CBS at the target transmission moment;
the transmission selection module is used for selecting the transmission,
determining a target transmission queue for transmitting frame data at a target transmission time in a plurality of transmission queues based on the credit value and the grade of the transmission queue;
the transmission control module is used for controlling the transmission of the data,
sending a switching rate notification to other equipment to enable the other equipment to carry out parameter setting, so that the transmission rate of a communication link between the current equipment and the other equipment is converted into the transmission rate required by the target transmission moment;
and when the target transmission time is reached, transmitting the frame data sent by the target transmission queue according to the transmission rate required by the target transmission time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110287747.XA CN113206724B (en) | 2021-03-17 | 2021-03-17 | Predictable deterministic scheduling method and device suitable for quasi-dynamic link |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110287747.XA CN113206724B (en) | 2021-03-17 | 2021-03-17 | Predictable deterministic scheduling method and device suitable for quasi-dynamic link |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113206724A CN113206724A (en) | 2021-08-03 |
| CN113206724B true CN113206724B (en) | 2022-03-01 |
Family
ID=77025437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110287747.XA Active CN113206724B (en) | 2021-03-17 | 2021-03-17 | Predictable deterministic scheduling method and device suitable for quasi-dynamic link |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113206724B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114285804A (en) * | 2021-12-28 | 2022-04-05 | 北京和利时系统工程有限公司 | Method, device and medium for controlling data transmission |
| CN114900476B (en) * | 2022-05-09 | 2023-06-30 | 中国联合网络通信集团有限公司 | Data transmission method, device, network equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7881194B1 (en) * | 2005-08-10 | 2011-02-01 | Sawyer Law Group, P.C. | Method for pacing multi-rate time sensitive traffic on ethernet and bridge connected networks |
| CN108173614A (en) * | 2017-12-08 | 2018-06-15 | 同济大学 | A kind of time synchronization and dispatching method of vehicle-mounted Ethernet |
| CN111327540A (en) * | 2020-02-25 | 2020-06-23 | 重庆邮电大学 | Deterministic scheduling method for industrial time-sensitive network data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018166576A1 (en) * | 2017-03-13 | 2018-09-20 | Siemens Aktiengesellschaft | Method and device for time-controlled data transmission in a tsn |
| US10764201B2 (en) * | 2017-11-28 | 2020-09-01 | Dornerworks, Ltd. | System and method for scheduling communications |
-
2021
- 2021-03-17 CN CN202110287747.XA patent/CN113206724B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7881194B1 (en) * | 2005-08-10 | 2011-02-01 | Sawyer Law Group, P.C. | Method for pacing multi-rate time sensitive traffic on ethernet and bridge connected networks |
| CN108173614A (en) * | 2017-12-08 | 2018-06-15 | 同济大学 | A kind of time synchronization and dispatching method of vehicle-mounted Ethernet |
| CN111327540A (en) * | 2020-02-25 | 2020-06-23 | 重庆邮电大学 | Deterministic scheduling method for industrial time-sensitive network data |
Non-Patent Citations (1)
| Title |
|---|
| 智能汽车车载TSN网络参数优化配置研究;吴源;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20200115;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113206724A (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP1475985A2 (en) | Traffic scheduling apparatus and method for a base station in a mobile communication system | |
| KR100234212B1 (en) | The congestion control based on adaptive speed control in packet network | |
| US7161907B2 (en) | System and method for dynamic rate flow control | |
| US20020167955A1 (en) | Packet transfer device and packet transfer method adaptive to a large number of input ports | |
| US6894974B1 (en) | Method, apparatus, media, and signals for controlling packet transmission rate from a packet source | |
| CN113206724B (en) | Predictable deterministic scheduling method and device suitable for quasi-dynamic link | |
| US8837490B2 (en) | Systems and methods for dynamically adjusting QoS parameters | |
| US8400920B2 (en) | Method for optimizing the transfer of information in a telecommunication network | |
| EP2050199B1 (en) | Expedited communication traffic handling apparatus and methods | |
| US20130286834A1 (en) | Traffic management apparatus for controlling traffic congestion and method thereof | |
| US6947445B1 (en) | Available bandwidth control mechanism | |
| JPH06224941A (en) | System for controlling network-access | |
| CN1539224A (en) | Method for supporting non-linear, highly scalable increase-dccrease congestion control scheme | |
| US20030138253A1 (en) | Dynamic wavelength management method in OBS networks | |
| CN112738843B (en) | High-rate 5G uplink congestion control method for determining communication delay | |
| CN113543209B (en) | Token scheduling-based congestion control method and device | |
| CN114500394B (en) | Congestion control method for differentiated services | |
| EP1344354B1 (en) | Selecting data packets | |
| JP2001211207A (en) | Packet transmission method, packet transmitter and band ensuring method | |
| US20060080459A1 (en) | Communication system and method | |
| Jeong et al. | A fair scheduling algorithm for wireless packet networks | |
| US7324524B2 (en) | Pseudo synchronous machine | |
| JP7207177B2 (en) | Packet processor and network system | |
| CN115865810B (en) | Credit value flow scheduling system and method in time sensitive network | |
| EP1162791B1 (en) | Available bandwidth control mechanism |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |