WO2011100888A2

WO2011100888A2 - Method and apparatus for scheduling transmission of data streams over a common communication link

Info

Publication number: WO2011100888A2
Application number: PCT/CN2011/070878
Authority: WO
Inventors: Lipin Yang; Lei Xie; Fang Yu
Original assignee: Huawei Technologies Co., Ltd.
Priority date: 2011-02-01
Filing date: 2011-02-01
Publication date: 2011-08-25
Also published as: CN103329608B; CN103329608A; WO2011100888A3

Abstract

A method and an apparatus for scheduling transmission of data streams over a common communication link are disclosed. The method includes the following steps:for each of the data streams, determining a first amount of data in the data stream for an incoming time slot; for each of the data streams, determining, based on an available data amount that is provided to the data streams for the incoming time slot and the first amounts of data in the data streams for the incoming time slot according to priority levels of the data streams, a second amount of data in the data stream; and transmitting, during the incoming time slot, the second amounts of data in the plurality of data streams over the common communication link according to the priority levels of the data streams.

Description

METHOD AND APPARATUS FOR SCHEDULING TRANSMISSION OF DATA STREAMS OVER A COMMON COMMUNICATION LINK

FIELD OF THE INVENTION

[0001] The present invention relates to the field of communication technologies, and in particular, to a method and an apparatus for scheduling transmission of multiple data streams over a common communication link.

BACKGROUND

[0002] Multi-service traffic support is one of the objectives of network convergence. The concept of a converged network - one that combines voice, data, and other signal transmissions into a single, higher-speed network interface - has been around for several decades.

[0003] In the converged network, what is needed is a communication node (such as a base station, a radio network controller (R C), a switch or a router) that is capable of transmitting data in a plurality of data streams having different delay requirements and different priority levels over a common communication link.

[0004] Many solutions have been developed to deal with the transmission of multiple data streams having different delay requirements and different priority levels over the common communication link. An example of the solutions is in a method disclosed by Niyato, D., Diamond, J., and Hossain, E., in "On optimizing token bucket parameters at the network edge under generalized processor sharing (GPS) scheduling" (Global Telecommunications Conference, GLOBECOM '05. IEEE, Volume: 2, 2005), in which an optimization scheme is formulated to obtain parameters (e.g. bucket size and token generation rate) with the objective of minimizing a delay bound for a particular traffic source (i.e. a data stream). The method can be used iteratively to obtain optimized delay bounds for multiple traffic sources.

[0005] However, the above-mentioned method can only be used to minimize the delay bounds of the data streams and it cannot guarantee the optimum delay bounds of the data streams. SUMMARY OF THE INVENTION

[0006] Accordingly, embodiments of the present invention provide a method and apparatus for scheduling transmission of a plurality of data streams over a common communication link.

[0007] In an embodiment of the present invention, a method is provided for scheduling a transmission of a plurality of data streams over a common communication link. The method includes:

for each of the data streams, determining a first amount of data in the data stream for an incoming time slot, wherein the first amount of data is an amount of data in the data stream that will reach a delay before the end of a time slot following the incoming time slot, and wherein the delay applied to the data stream is no larger than a maximum delay value of the data stream;

for each of the data streams, determining, based on an available data amount that is provided to the data streams for the incoming time slot and the first amounts of data in the data streams for the incoming time slot according to priority levels of the data streams, a second amount of data in the data stream, wherein the second amount of data is an amount of data in the data stream that will actually be transmitted in the incoming time slot; and

transmitting, during the incoming time slot, the second amounts of data in the plurality of data streams over the common communication link according to the priority levels of the data streams.

[0008] Another embodiment of the present invention provides an apparatus for scheduling transmission of a plurality of data streams over a common communication link. The apparatus includes:

a receiver comprising a plurality of receiving units, each receiving unit being configured to receive a data stream from a data source;

a processor connected to the receiver, configured to:

for each of the data streams, determine a first amount of data in the data stream for an incoming time slot, wherein the first amount of data is an amount of data in the data stream that will reach a delay before the end of a time slot following the incoming time slot, and wherein the delay applied to the data stream is no larger than a maximum delay value of the data stream;

for each of the data streams, determine, based on an available data amount that is provided to the data streams for the incoming time slot and the first amounts of data in the data streams for the incoming time slot according to priority levels of the data streams, a second amount of data in the data stream, wherein the second amount of data is an amount of data in the data stream that will actually be transmitted in the incoming time slot; and

a transmitter connected to the processor, configured to:

transmit, during the incoming time slot, the second amount of data of the plurality of data streams over the common communication link according to the priority levels of the data streams.

[0009] In yet another embodiment of the present invention, a computer program product for scheduling transmission of a plurality of data streams over a common communication link is provided. The computer program product includes a computer-readable storage medium storing program code thereon for causing a computerized controlling device to execute a process that includes:

transmitting, during the incoming time slot, the second amounts of data in the plurality of data streams over the common communication link according to the priority levels of the data streams. [0010] In the above method, apparatus and computer program product for scheduling transmission of a plurality of data streams over a common communication link, a delay is applied to each of the data streams, and the delay is no larger than a maximum delay value (i.e. delay requirement) of that data stream. An amount of data in each of the data streams that will reach the delay before the end of a next time slot following an incoming time slot is transmitted over the common communication link during the incoming time slot according to priority levels of the data streams, so delay requirements of the data streams are guaranteed at full steam.

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The disclosed aspects will hereinafter be described in conjunction with the appended drawings, provided to illustrate and not to limit the disclosed aspects, wherein like designations denote like elements, and in which:

[0012] Fig. 1 is a flowchart illustrating a method for scheduling transmission of a plurality of data streams over a common communication link according to an embodiment of the present invention;

[0013] Fig. 2 is a flowchart illustrating a method for scheduling transmission of the data streams over the common communication link according to a specific implementation of the method of Fig. 1 ;

[0014] Fig. 3A is an example of the data transmission according to an embodiment of the present invention;

[0015] Fig. 3B illustrates an implementation of the method for scheduling transmission of the data streams over the common communication link with a token bucket algorithm according to an embodiment of the present invention;

[0016] Fig. 3C is another example of the data transmission according to an embodiment of the present invention.

[0017] Fig. 4 is a flowchart illustrating a method for scheduling transmission of the data streams over the common communication link according to another specific implementation of the method of Fig. 1 ;

[0018] Fig. 5 is a flowchart illustrating a method for scheduling transmission of the data streams over the common communication link according to yet another specific implementation of the method of Fig. 1 ; and

[0019] Fig. 6 is a block diagram of an apparatus for scheduling transmission of the data streams over the common communication link according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

[0020] Various aspects of the present invention are now described with reference to the drawings. In the following descriptions, numerous specific details are set forth in order to provide a thoroughly explanation of the present invention. It may be evident, however, that the present invention may be practiced without adhering to these specific details.

[0021] Fig. 1 is a flowchart illustrating a method for scheduling a transmission of a plurality of data streams over a common communication link according to an embodiment of the present invention.

[0022] The method may be implemented in a communication node such as a base station, a radio network controller (R C), a switch, or a router.

[0023] As shown in Fig. 1, at block S10, for each of the data streams, a first amount of data in the data stream is determined for an incoming time slot (for example, a time slot T_j). The first amount of data is the amount of data in that data stream that will reach a delay before the end of a next time slot following the incoming time slot (for example, a time slot T_j+i). The delay applied to the data stream is no larger than a maximum delay value of the data stream.

[0024] At block S12, for each of the data streams, a second amount of data in the data stream is determined. The second amount of data is an amount of data that will actually be transmitted in the incoming time slot T_j over the common communication link. The second amount of data is determined based on an available data amount that is provided to the data streams for the incoming time slot T_j and the first amounts of data in the data streams according to priority levels of the data streams.

[0025] At block S14, the second amount of data in each of the data streams is transmitted during the incoming time slot T_j over the common communication link according to priority levels of the data streams.

[0026] In the above method, each of the data streams is applied with a delay that is no larger than a maximum delay value of that data stream, and data of each of the data streams that will reach the delay before the end of the next time slot T_j+i following the incoming time slot T_j is transmitted over a common communication link during the incoming time slot T_j according to orders from high to low priority levels of the data streams, so that delay requirements of the data streams are fully fulfilled.

[0027] The available data amount may be determined based on length of the incoming time slot T_j and a maximum serving rate provided to the data streams in the incoming time slot T_j. The maximum serving rate may be a maximum data amount per unit time provided to the data streams in the incoming time slot T_j. A duration of the incoming time slot T_j may be equal to a most stringent delay bound, which is no larger than the smallest value of the maximum delay values of the data streams.

[0028] A delay applied to one of the data streams may be a multiple of the most stringent delay bound. For example, the delay applied to a data stream may be calculated according to the following equation: di=Floor(Di/d)*d,

where di is the delay applied to the i^th data stream, Di is the maximum delay value of the i^th data stream, d is the most stringent delay bound, and Floor(Di/d) is the largest integer no larger than Di/d.

[0029] Or, the delay applied to a data stream may be calculated by the following equation:

d,=2^J*d,

where di is the delay applied to the i^th data stream, d is the most stringent delay bound, and j is an integer which causes 2^J*d to be closest to and no larger than the maximum delay value of the i^th data stream.

[0030] The calculation of the second amount of data in the block S12 may include calculating, based on the available data amount and the first amounts of data in the data streams according to priority levels of the data streams, a third amount of data amount in each of the data streams, wherein the third amount of data is a maximum amount of data in each of the data streams that can be transmitted in the incoming time slot T_j over the common communication link, and calculating, based on the first data amount and the third data amount of each of the data streams for the incoming time slot T_j, the second data amount of each of the data streams for the incoming time slot T_j.

[0031] Further, before the block S 10, the method may include obtaining a delay offset of at least one data stream of the data streams based on that a leftover capacity in each time slot is substantially the same, and calculating an adjusted delay for each of the at least one data stream based on the delay applied to each of the at least one data stream and the obtained delay offset of each of the at least one data stream, and the block S 10 may further include calculating an adjusted first amount of data in each of the at least one data stream for the incoming time slot T_j, which is an amount of data of each of the at least one data stream that will reach the adjusted delay before the end of a next time slot T_j+i flowing the incoming time slot T_j, so that a larger contiguous leftover bandwidth is created and occurrence of bandwidth shortage in some time slots in the future is avoided.

[0032] In addition to the blocks from S 10 to SI 4, the method may further include, if the sum of the first amounts of data in the data streams for the incoming time slot is less than the available data amount for the incoming time slot, and after the transmission is finished, sending, over the common communication link during the incoming time slot T_j, data of at least one data stream of the plurality of data streams that doesn't reach its applied delay before the end of the next time slot T_j+i following the incoming time slot T_j; or sending, over the common communication link, data of at least one elastic data stream, whose delay is not as critical as that of other data streams, so that bandwidth is used well while delay requirements of the data streams are guaranteed.

[0033] In addition to the blocks from S 10 to S 14, the method may further include: when an emergency data stream arrives for transmission in the incoming time slot T_j, stopping transmitting of data of a current data stream and sending data of the emergency data stream over the common communication link in the incoming time slot T_j, so that the delay requirement of the emergency data stream is guaranteed.

[0034] The method may further include: after the transmission of the data of the emergency data stream is finished, if amount of data that has been transmitted in the incoming time slot T_j is less than the available data amount, continuing to send during the incoming time slot T_j data of the current data stream and data of the data streams whose priority levels are lower than that of the current data stream according to priority levels of the data streams, so that the delay requirements of the data streams are guaranteed at full steam while the delay requirement of the emergency data stream is guaranteed.

[0035] The method may further include, if not all data of the emergency data stream has been transmitted in the incoming time slot T_j, executing the blocks from S10 to S14 for each time slot following the incoming time slot T_j for the emergency data stream and the data streams, until all data of the emergency data stream has been transmitted, the priority level of the emergency data stream being set to be higher than those of the data streams, so that the delay requirement of the emergency data stream is guaranteed at full steam.

[0036] In addition to the blocks from S10 to S14, the method may further include, when a new data stream arrives in the incoming time slot and the new data stream has data that will reach a maximum delay applied to the new data stream before the end of the next time slot following the incoming time slot, determining whether transmitting of data of a high priority data stream of the data streams whose priority level is higher than that of the new data stream data is finished; if not, computing a first amount of data in the new data stream for the incoming time slot; re-determining, based on leftover data amount that is provided to the data streams in the incoming time slot when the transmitting of data of the high priority data stream in the incoming time slot is finished and the first amount of data in each of the high priority data streams, the new data stream and low priority data streams of the data streams for the incoming time slot, the second amount of data in each of the new data stream and the low priority data streams for the incoming time slot according to priority levels of the new data stream and the low priority data streams, wherein the low priority data streams are those data streams whose priority levels are lower than that of the new data stream; and sending, during the incoming time slot, data of the new data stream and the low priority data streams over the common communication link based on the re-determined second amount of data in each of the new data stream and the low priority data streams for the incoming time slot according to the priority levels of the new data stream and the low priority data streams, after the transmission of data of the high priority data stream in the incoming time slot is finished, so that the delay requirement of the new data stream is guaranteed. [0037] The method may further include stopping transmitting of data of a current data stream of the data streams whose data is being transmitted over the common communication link in the incoming time slot, if the detecting is confirmative; and sending data of the new data stream over the common communication link in the incoming time slot.

[0038] The method may further include continuing to send during the incoming time slot data of at least one data stream of the current data stream and low priority level data streams of the data streams whose priority levels are lower than that of the current data stream according to priority levels of the at least one data stream, after the transmission of data of the new data stream in the incoming time slot is finished, if amount of data that has been transmitted in the incoming time slot after the transmission of data of the new data stream in the incoming time slot is finished is less than the available data amount.

[0039] The method may further include executing the blocks from S10 to S 14 for each time slot following the incoming time slot for the new data stream and the data streams.

[0040] In communication networks, data is generally transmitted in manner of packets and a packet includes a constant amount of data. For purpose of explanation, it is assumed below that data is transmitted in manner of packets, and the data amount is represented by the number of packets.

[0041] Fig. 2 is a flowchart illustrating a process for scheduling a transmission of data streams over a common communication link according to an implementation of the method of the present invention. The process may be executed in a communication node such as a base station, a radio network controller (R C), a switch or a router. It is assumed that there are n data streams Y; (i=l, 2,..., n) that need to be transmitted over the common communication link. Each of the data streams has a maximum delay value Di, and orders from high to low priority levels of the n data streams are Yi>Y₂>.. ·≥Υ_η·

[0042] As shown in Fig. 2, at a step SI 10, the most stringent delay bound D_mi_n is determined. D_mi_n is a value that is no larger than the smallest value of the maximum delay values DI, D2,..., D_n-i and D_n of the n data streams.

[0043] At a step SI 20, for each data stream Yi a delay di is calculated, di is an actual delay to be applied to the data stream Yi and the calculation is based on the most stringent delay bound D_mi_n and the maximum delay value Di of the data stream Yi. Specifically,

=\c- * r> where ki is an integer and

, Floor ( Di/D_mi_n) is the largest integer that is not larger than Di/D_mi_n, and mi is the largest integer such that 2^mi*r> < =D

[0044] At a step S 130, the duration of each time slot T_j =l, 2,...) for transmitting data is set to be equal to Dmin.

[0045] Before each time slot T_j comes, the following steps S 140-S170 are executed.

[0046] At the step S 140, the maximum serving rate R used for the n data streams is determined. In this implementation of the present invention, the maximum serving rate R is represented by maximum number of packets being transmitted per unit time.

[0047] At the step SI 50, data stored in a buffer for each data stream Yi are checked to determine the number of packets PRi of each data stream Yi that will reach the delay di before the end of a next time slot T_j+i following the incoming time slot T_j.

[0048] At the step SI 60, the maximum number of packets Pimax of each data stream Yi that can be transmitted in the incoming time slot T_j is determined based on the D_min, the R and the PRi according to orders from high to low priority levels of the n data streams. Specifically,

Pimax -Min ( Pi_max> PRi ) , P3max =

P2max "Min ( P_2max, PR₂ ) , P₄max = P3max "Min ( P_3max> PR₃ ) ,..., and Pnmax = Pn-lmax "Min

( Pn-imax> PRn-i ) . Min ( ) is an operator of taking a minimum value from all values inside the brackets, and R*D_mm is an available data amount that is provided to the n data streams in the incoming time slot T_j.

[0049] At the step SI 70, the number of packets pi of each data stream Yi that will actually be transmitted in the incoming time slot T_j is calculated based on the PRi and the Pimax. Specifically, pi= Min ( Pimax > PRi )» p₂= Min ( P_2max> PR₂ ) > p₃= Min ( P_3max>

PR₃ ) , ..., p_n= Min ( Pnmax, PRn) .

[0050] During each time slot T_j, the following step SI 80 is executed. [0051] At the step S I 80, the n data streams are scheduled for transmission over the common communication link during the incoming time slot T_j according to orders from high to low priority levels of the n data streams, in which pi packets of each data stream Yi that will reach the delay di before the end of the next time slot T_j+i are transmitted.

[0052] Specifically, if the common communication link can transmit packets of only one data stream at a time, the pi packets of each data stream Yi are transmitted during the incoming time slot T_j according to orders from high to low priority levels of the n data streams, i.e., the pi packets of the data stream Yi are transmitted first, the p₂ packets of the data stream Y₂ are then transmitted after the pi packets of the data stream Y_{l s} the p₃ packets of the data stream Y₃ are next transmitted after the p₂ packets of the data stream Y₂, ... , and so on.

[0053] If the common communication link can transmit packets of two or more data streams at a time, the packets of every two or more data streams are transmitted at a time during the incoming time slot T_j according to orders from high to low priority levels of the n data streams. For example, if the common communication link can transmit packets of two data streams at a time, pi and p₂ packets of the data streams Yi and Y₂ are first transmitted during the incoming time slot T_j, p₃ and p₄ packets of the data streams Y₃ and Y₄ are then transmitted after the transmission of the pi and p₂ packets of the data streams Yi and Y₂ is finished, and so forth.

[0054] In the method shown in Fig. 2, each of the n data streams is applied with a delay that is no larger than the maximum delay value of that data stream and data of each of the n data streams that will reach the applied delay before the end of the next time slot T_j+i following the incoming time slot T_j is transmitted over the common communication link during the incoming time slot T_j according to orders from high to low priority levels of the n data streams, so delay requirements of the n data streams are guaranteed at full steam.

[0055] An illustrative example for scheduling transmission of data streams over a common communication link according to an implementation of the present invention is shown in Fig. 3A. In the example shown in Fig. 3A, four data streams Yi-Y₄ are to be transmitted over a common communication link, and orders from high to low priority levels of the four data streams are Yi>= Y₂>= Y₃>= Y₄. The most stringent delay bound is D_mi_n and the duration of each time slot is equal to D_mi_n. The delay di-d₄ applied respectively to the data streams Yi-Y₄ are

and d₄=2⁴*D_mm. Packets of the data stream Yi are transmitted during each time slot. Packets of the data stream Y₂ are transmitted during each of time slots l ^s2¹ (1=1, 2, 3,...)· Packets of the data stream Y₃ are transmitted during each of time slots 1*2² (1=1, 2, 3,...)· Packets of the data stream Y₄ are transmitted during each of time slots 1*2⁴ (1=1, 2, 3,...)·

[0056] An illustrative implementation of the method for scheduling transmission of data streams over a common communication link with a token bucket algorithm is shown in Fig. 3B. As shown in Fig. 3B, after the number of packets pi of each data stream Yi that will actually be transmitted in the incoming time slot T_j is obtained, the token bucket of each data stream Yi is placed into pi tokens for the incoming time slot T_j and then packets stored in a queue of each data stream Yi are scheduled during the incoming time slot T_j to transmit over the common communication link based on pi tokens for each data stream Yi according to orders from high to low priority levels of the n data streams.

[0057] It is evident that since the maximum number of packets Pi_max of each data stream Yi is calculated according to orders from high to low priority levels of the n data streams, when high priority data streams of the n data streams have too many packets that will reach their delays before the end of the next time slot T_j+i, it is possible that packets of low priority data streams are not transmitted during the incoming time slot T_j. For example, assuming that PRi =0.4* R*D_mm, PR₂ =0.3* R*D_mm, and PR₃ =0.3* R*D_mm, P¾+ PR₂+ PR₃= R*D_mm. In this case, Pi_max = R*D_mm,

P₂max 0.6*R*D_mm, P₃max 0.3 *R*D_mm, and P₄max P₅max P₆max · · · Pnmax 0, and thus pi = 0.4*R*D_min, p₂ = 0.3*R*D_min, p₃ = 0.3*R*D_min, and p₄ = p₅ = p₆ =...=p_n =0. That is, packets of the data streams from Y₄ to Y_n are not transmitted during the incoming time slot T_j even if the data streams from Y₄ to Y_n may have packets that will reach the delay before the end of the next time slot T_j+1. The packets that should be transmitted but are not transmitted during the incoming time slot T_j may be discarded at the end of the incoming time slot T_j. Or, the packets that should be transmitted but are not transmitted during the incoming time slot T_j are discarded if they are not transmitted in the next time slot T_j+1.

[0058] As a first modification of the present invention, those skilled in the art will understand that if the sum from i to p_n is less than the maximum number of packets R* D_mi_n that can be transmitted in the incoming time slot T_j, it is possible that additional packets of at least one data stream of the n data streams that does not reach its delay before the end of the next time slot T_j+i or additional packets of at least one elastic data stream whose delay is not as critical as the n data streams can be transmitted during the incoming time slot T_j after all the data packets pi to p_n are transmitted in the incoming time slot T_j, and the maximum amount of additional packets that can be transmitted during the incoming time slot T_j is equal to a difference between R* D_mi_n and the sum from pi to p_n, so that bandwidth is used well while delay requirements of the n data streams are guaranteed and occurrence of bandwidth shortage in some time slots in the future is avoided.

[0059] Further, it is obvious that the maximum serving rate R used for the n data streams may be variable with time slots or constant for each time slot. If the maximum serving rate R used for the n data streams is variable with the time slots, it is necessary to determine the maximum serving rate R for each time slot. If the maximum serving rate R used for the n data streams is the same and constant for each time slot, it is not necessary to determine the maximum serving rate R once for each time slot, and the maximum serving rate R may be only determined once for all time slots, so that operation load required for the method is reduced.

[0060] As a second modification of the present invention, those skilled in the art will understand that the number of packets pi of each data stream Yi that will actually be transmitted in the incoming time slot T_j may also be calculated directly based on the D_mi_n, the R and the PR; according to orders from high to low priority levels of the n data streams. Specifically, pi= Min ( R^D_mi_n, PRi ) , p₂= Min ( R*D_mi_n- p_{1 ?} PR₂ ) , p₃= Min ( R*D_min- p p₂, PR₃ ), p_n= Min ( R*D_min- p p₂-..., p_n-1 » PR_n) , so that operation load required for the method is reduced.

[0061] Further, as a third modification of the present invention, those skilled in the art will understand that the delay actually applied to at least one of the n data streams may be adjusted to make the transmitting time of the data of at least one data stream to be advanced, so that leftover capacities in each time slot are substantially the same to create a larger contiguous leftover bandwidth and avoid occurrence of bandwidth shortage in some time slots in the future. Specifically, delay offset DO_g (g=l,2,...,zz, 1<=ζζ<=η) of each of zz data streams of the n data streams is obtained based on that leftover capacities in each time slot are substantially the same after the above step S120, preferably the delay offset DO_g of the data stream Y_g being multiple of the most stringent delay bound D_mi_n and less than the maximum delay value D_g of the data stream Y_g; adjusted delay d'_g of each of the zz data streams that is actually applied to each of the zz data streams is calculated before the above step SI 50 based on the delay d_g and the delay offset DO_g of each of the zz data streams, i.e., d'_g = d_g - DO_g; and at the above step SI 50, data stored in the buffer for each of the zz data streams are checked to calculate the number of packets PR_g of each of the zz data streams that will reach the adjusted delay d'_g before the end of the next time slot Tj+i following the incoming time slot T_j.

[0062] Another illustrative example for scheduling transmission of data streams over a common communication link according to the third modification of the present invention is shown in Fig. 3C. As shown in Fig. 3C, compared to Fig. 3A, the delay actually applied to the data streams Y₃ and Y₄ is adjusted based on that the leftover capacities in each time slot are substantially the same. It is apparent from Fig. 3C that delay offset D0₃ of the data streamY₃ is D_mi_n and delay offset D0₄ of the data stream Y₃ is 15*D_mi_n. The adjusted delay d'₃ of the data stream Y₃ is d'₃ = d₃ - D0₃=2²* Dmin - D_min, and the adjusted delay d'₄ of the data stream Y₄ is d'₄ = d₄ - D0₄=2⁴* Dmin - * D_min- It can be seen from Fig. 3C that compared to Fig. 3A, the leftover capacity in each time slot is substantially the same, and thus larger contiguous leftover bandwidth is created and occurrence of bandwidth shortage in some time slots in the future is avoided.

[0063] Fig. 4 is a flowchart illustrating a method for scheduling transmission of data streams over a common communication link according to another implementation of the present invention. The method may be executed in a communication node such as a base station, a radio network controller (RNC), a switch or a router. This implementation deals with such case where an emergency data stream EM with maximum delay value D_EM arrives in a time slot T_p when packets of the n data streams are being transmitted over the common communication link during the time slot T_p.

[0064] At a step S200, it is judged whether or not the emergency data stream EM has any packets that will reach the maximum delay value D_EM before the end of the next time slot T_p+i following the time slot T_p.

[0065] If the emergency data stream EM has at least one packet that will reach the maximum delay value DEM before the end of the next time slot T_p+1, at a step S210, the number of packets PREM of the emergency data stream EM that will reach the maximum delay value DEM before the end of the time slot T_p+i is determined by detecting the packets stored in the buffer for the emergency data stream EM.

[0066] At a step S220, transmission of packets of a current data stream Y_s (l<=s<=n) during the time slot T_p is stopped. It is assumed that the current data stream Y_s is a data stream whose data packets are being transmitted over the common communication link during the time slot T_p and transmission of data packets of the data streams from Yi to Y_s-i (whose priority levels are higher than that of the current data stream Y_s) during the time slot T_p has been finished.

[0067] At a step S230, PEM packets of the emergency data stream EM that will reach the maximum delay value DEM before the end of the next time slot T_p+i are transmitted over the common communication link during the time slot T_p. PEM = Min (Ps max-pals, PREM), and pal_s is the amount of data packets of the current data stream Y_s that has been transmitted in the time slot T_p.

[0068] At a step S240, transmission of pp_h (h=s, s+1 , s+2, ... ,n) packets of each of the data streams from Y_s to Y_n that will reach their delays d_s to d_n, respectively, before the end of the next time slot T_p+i over the common communication link during the time slot T_p are scheduled according to orders from high to low priority levels of the data streams from Y_s to Y_n, after the PEM packets of the emergency data stream EM are transmitted, if the number of packets that have been transmitted in the time slot T_p after the PEM packets of the emergency data stream EM are transmitted in the time slot T_p is less than the maximum number of packets R*D_mi_n that can be transmitted in the time slot T_p. Specifically, pp_s = Min (P_{s max} - pal_s - p_EM, PR_S - pal_s), pps+i = Min (Ps max - pal_s - PEM - pp_s, PR_s+i), pp_S+2 = Min (P_{s max} - pal_s - p_EM - pp_s - pp_s+i, PR_s+2), ... , pp_n = Min (P_{s max} - pal_s - PEM - pp_s - pp_s+i - ... - ppn-i , PRO-

[0069] At a step S250, if not all packets of the emergency data stream EM have been transmitted in the time slot T_p or if the emergency data stream EM has at least one packet that will reach the maximum delay value DEM before the end of the next time slot T_p+i, the emergency data stream EM is set to have the highest priority level than the n data streams, and for these data streams including the emergency data stream EM and the n data streams, the above steps S110-S130 are executed for next time slot T_p+i following the time slot T_p and the above steps S140-S180 are executed repeatedly for each time slot following the time slot T_p, until all packets of the emergency data stream EM are transmitted.

[0070] In the method shown in Fig. 4, when the emergency data stream EM arrives in the time slot T_p, if the emergency data stream EM has a packet that will reach the maximum delay value D_EM before the end of the next time slot T_p+i following the time slot T_p, the packet of the emergency data stream EM are transmitted immediately during the time slot T_p, and compared to packets of the n data streams, packets of the emergency data stream EM are transmitted first during each time slot after the time slot T_p until all packets of the emergency data stream EM are transmitted, so the delay requirement of the emergency data stream EM is guaranteed.

[0071] Further, as a fourth modification of the present invention, those skilled in the art will understand that packets of the emergency data stream EM that don't reach its delay before the end of the time slot T_e+i may also be transmitted over the common communication link during the time slot T_e after transmitting of the packets of the n data streams in the time slot T_e has been finished, if the emergency data stream EM doesn't have any packet that will reach its delay before the end of the time slot T_e+i and the maximum number of packets that can be transmitted over the common communication link in the time slot T_e is larger than the sum of the amount of packets of the n data streams that are transmitted actually in the time slot T_e, so that bandwidth is used well while the delay requirements of the emergency data stream is guaranteed and occurrence of bandwidth shortage in some time slots in the future is avoided.

[0072] Fig. 5 is a flowchart illustrating a method for scheduling transmission of data streams over a common communication link according to a further implementation of the present invention. The method may be executed in the communication node such as a base station, a radio network controller (R C), a switch or a router. This implementation deals with a case where a new data stream Ynew with a maximum delay value D_NEW arrives in a time slot T_v when packets of the n data streams are being transmitted over the common communication link during the time slot T_v. It is assumed that priority level of the new data stream Y_new is lower than that of the data stream Y_m of the n data streams and higher than that that of the data stream Y_m+i of the n data streams.

[0073] At a step S300, it is judged whether or not the new data stream Y_new has any packets that will reach the delay D_new before the end of the next time slot T_v+i following the time slot T_v.

[0074] If the new data stream Y_new has no packet that will reach the delay D_new before the end of the time slot T_v+i , at a step S310, for the n+1 data streams including the new data stream Y_new and the n data streams, the above steps S110-S130 are executed for a next time slot T_v+i following the time slot T_v and the above steps S140-S180 are executed repeatedly for each time slot following the time slot T_v.

[0075] If the new data stream Y_new has at least one packet that will reach the delay D_new before the end of the time slot T_v+i, at a step S320, packets stored in the buffer for the new data stream Y_new are checked to calculate the number of packets PR_new of the new data stream Y_new that will reach the delay D_new before the end of the time slot T_v+i .

[0076] At a step S330, it is determined whether transmitting of packets of the data stream Y_m in the time slot T_v has been finished.

[0077] If the transmitting of packets of the data stream Y_m in the time slot T_v has not been finished, at a step S340, the maximum number of packets P'_{z m}ax of each of the new data stream Y_new and the data streams from Y_m+i to Y_n can be transmitted in the time slot T_v is re-calculated based on P_m max, PR_m, PRnew, P m+i PRn according to orders from high to low priority levels of the new data stream Y_new and the data streams from Y_m+i to Y„. Specifically, P'_new max = P_m max - Min (P_{m max}, PR_m), P'_m+i max

P new max^— Min (P _{new max}, PR_new), P _m+_{2 ma}x P m+1 max^— Min (P _m+j _max, PR_m+j) , . . . , P nmax^— P n-lmax^— Min (P n-lmax» PRn-ΐ)·

[0078] At a step S350, the number of packets p'_z of each of the new data stream Y_new and the data streams from Y_m+i to Y_n that will actually be transmitted in the time slot T_v is re-determined based on PR_m, PR_new, PR_m+i , . . . , PR_n and P'_new max, P'm+i max,

P'm+2 max, - · · , P'nmax- Specifically, p'new⁼ Min ( P'new max' PRnew) ' p'm+l⁼ Min( P '_m+i _max, PR_m+l ) , p'_m+2= Min ( P'_m+2 max' PRm+₂ ) ' ..., p'n= Min ( P'nmax' PRn) . [0079] At step S360, the new data stream Y_new and the data streams from Y_m+i to Y_n are scheduled during the time slot T_v according to orders from high to low priority levels of the new data stream Y_new and the data streams from Y_m+i to Y_n, to transmit over the common communication link p'_z packets of each of the new data stream Y_new and the data streams from Y_m+i to Y_n that will reach the respective delay before the end of the time slot T_v+i, after the transmission of packets of the data stream Y_m in the time slot T_v has been finished.

[0080] On the other hand, if the transmission of packets of the data stream Y_m in the time slot T_v has been finished, at a step S370, a current data stream Y_a of the data streams from Y_m+i to Y_n whose packets are being transmitted during the time slot T_v is determined.

[0081] At a step S380, the transmission of packets of the current data stream Y_a in the time slot T_v is stopped and the number of packets p_ai of the current data stream Y_a that have been transmitted to in the time slot T_v is calculated.

[0082] At a step S390, p"_new packets of the new data stream Y_new are transmitted over the common communication link in the time slot T_v. Specifically, p" _new =Min

(Pa-1 max" Pal? PR-new)-

[0083] At a step S400, the data streams from Y_a to Y_n are scheduled in the time slot T_v according to orders from high to low priority levels of the data streams from Y_a to Y_n, to transmit over the common communication link p"i packets of each of the data streams from Y_a to Y_n, after the transmission of the packets of the new data stream Y_new has been finished in the time slot T_v, if the amount of packets that have been transmitted in the time slot T_v after the packets of the new data stream are transmitted in the time slot T_v is less than the maximum number of packets R*D_mi_n that can be transmitted in the time slot T_v. Specifically, p"_a = Min (P_a-i _max- p_ai- p"_new,

PRa-Pal), P"a+1 = Min (P_a-i _max- p_al- p"„_ew-p"a, PR-a+l), P"a+2 = Min (P_a-i _max- p_ai- P"new-P"a-P"a+1 , PR-a+2), . . . , p"n = Min (P_a-i _max- p_ai- p"_new-p"a-p"a+l-. · ·" P"n-1 , PRn) »

[0084] In the method shown in Fig. 5, when the new data stream Y_new arrives in the time slot T_v, packets of the new data stream Y_new are transmitted during the time slot T_v or each time slot after the time slot T_v according to order of priority level of the new data stream Y_new with respect to priority levels of the n data streams, so the delay requirement of the new data stream Y_new is guaranteed at full steam. [0085] Further, as a fifth modification of the present invention, those skilled in the art will understand that packets of the new data stream Y_new is transmitted during the time slot T_v only after transmitting of corresponding packets of the data streams from Yi to Y_n in the time slot T_v has been finished when the sum from pi to p_n for the time slot T_v is less than the maximum number of packets R* D_mi_n that can be transmitted in the time slot T_v. Specifically, the maximum number of packets P_new max of the new data stream Y_new that can be transmitted in the time slot T_v is first calculated as P_new max = Pn max -Min (P_{n max}, PR_n), the number of packets p_new of the new data stream Y_new that can actually be transmitted in the time slot T_v is then calculated as p_new = Min (P_new max, PRnew), and p_new packets of the new data stream Y_new are transmitted finally during the time slot T_v after transmitting of packets of the data stream Y_n during the time slot T_v is scheduled, so that the delay requirements of the n data streams are guaranteed first.

[0086] It is evident that when a data stream of the n data stream ends in a time slot T_ee, for the remainder n-1 data streams, the above steps S110-S130 are executed for a next time slot T_ee+i following the time slot T_ee and the above steps S140-S180 are executed repeatedly for each time slot after the time slot T_ee.

[0087] Based on the descriptions of the preceding embodiments, those skilled in the art may understand that the present invention may be implemented by hardware, or by software running on a hardware platform. The technical solution of the present invention may be embodied by a software program product which comprises a nonvolatile storage medium storing program codes thereon for executing the method provided in the embodiments of the present invention by a processing device (such as a personal computer, a server, or a network device). The storage medium can be, for example, a Compact Disk Read-Only Memory (CD-ROM), a USB disk, or a mobile hard disk.

[0088] Fig. 6 illustrates an apparatus 400 for scheduling transmission of a plurality of data streams over a common communication link according to an embodiment of the present invention. At least a portion of the apparatus 400 may be located at a communication node such as a base station, a radio network controller (RNC), a switch or a router.

[0089] The apparatus 400 includes a receiver 410, a processor 420 and a transmitter 430.

[0090] The receiver 410 comprises a plurality of receiving units 410i, 410₂,„„ , 410_n, each configured to receive one of a plurality of data streams from a data source.

[0091] The processor 420 is configured to:

for each of the data streams, determine a first amount of data in the data stream for an incoming time slot, wherein the first amount of data is an amount of data in the data stream that will reach a delay before the end of a time slot following the incoming time slot, and wherein the delay applied to the data stream is no larger than a maximum delay value of the data stream; and

for each of the data streams, determine, based on an available data amount that is provided to the data streams for the incoming time slot and the first amounts of data in the data streams for the incoming time slot according to priority levels of the data streams, a second amount of data in the data stream, wherein the second amount of data is an amount of data in the data stream that will actually be transmitted in the incoming time slot.

[0092] The transmitter 430 is configured to transmit, during the incoming time slot, the second amount of data of the plurality of data streams over the common communication link according to the priority levels of the data streams.

[0093] Preferably, the processor 420 may further be configured to determine the delay applied to each of the data streams as a multiple of a most stringent delay bound, wherein the most stringent delay bound is no larger than the smallest value of the maximum delay values of the data streams.

[0094] In an implementation of the present invention, the processor 420 may determine the first amount of data in each of the data streams for an incoming time slot in a manner as shown in the above step SI 50, determine the delay applied to each of the data streams in a manner as shown in the above step SI 20, and determine the most stringent delay bound as shown in the above step SI 10.

[0095] In an implementation of the present invention, the processor 420 may determine the second amount of data in each of the data streams for the incoming time slot in a manner as shown in the above steps SI 60 and SI 70. In another implementation of the present invention, the processor 420 may determine the second amount of data in each of the data streams for the incoming time slot as described in the above second modification of the present invention.

[0096] In the apparatus 400, each of the multiple data streams is applied a delay that is not larger than a maximum delay value of that data stream and data of each of the multiple data streams that will reach the applied delay before the end of a next time slot following an incoming time slot is transmitted over a common communication link during the incoming time slot according to orders from high to low priority levels of the multiple data streams, so delay requirements of the multiple data streams are guaranteed at full steam.

[0097] The processor 420 may further be configured to calculate, based on the available data amount and the first amount of data in each of the data streams for the incoming time slot according to the priority levels of the data streams, a third amount of data in each of the data streams for the incoming time slot, which is a maximum amount of data of each of the data streams that can be transmitted in the incoming time slot over the common communication link, and calculate, based on the first amount of data and the third amount of data in each of the data streams for the incoming time slot, the second amount of data amount in each of the data streams for the incoming time slot. The available data amount may be determined based on the duration of the time slot and a maximum serving rate provided to the data streams in the incoming time slot. The maximum serving rate may be maximum data amount per unit time provided to the data streams in the incoming time slot. In an implementation of the present invention, the processor 420 may calculate the third amount of data and the second amount of data in each of the data streams for the incoming time slot in a manner as shown in the steps SI 60 and SI 70.

[0098] The processor 420 may further be configured to obtain a delay offset of each of at least one data stream of the data streams based on that leftover capacities in each time slot is substantially the same, and calculate an adjusted delay for each of the at least one data stream based on the delay applied to each of the at least one data stream and the obtained delay offset of each of the at least one data stream, and the processor 420 is further configured to calculate the adjusted first amount of data in each of the at least one data stream for the incoming time slot, which is an amount of data in each of the at least one data stream that will reach the adjusted delay before the end of the next time slot, so that a larger contiguous leftover bandwidth is created and occurrence of bandwidth shortage in some time slots in the future is avoided. In an implementation of the present invention, the processor may determine the delay offset and calculate the adjusted delay in a manner as shown in the above third modification of the present invention.

[0099] The transmitter 430 may further be configured to send, over the common communication link during the incoming time slot, data of at least one data stream of the data streams that doesn't reach the delay applied to the data stream before the end of the next time slot following the incoming time slot or send, over the common communication link during the incoming time slot, data of at least one elastic data stream whose delay is not as critical as the data streams, after the transmission of the second amounts of data of the data streams is finished, if sum of the first data amount of the data streams for the time slot is less than the available data amount, so that bandwidth is used well while delay requirements of the data streams are guaranteed. In an implementation of the present invention, the transmitter 430 may send data in a manner as shown in the above first modification of the present invention.

[00100] When an emergency data stream arrives in the incoming time slot, the transmitter 430 may be further configured to stop, transmitting of data of a current data stream of the data streams in the incoming time slot and send data of the emergency data stream over the common communication link in the incoming time slot, so that the delay requirement of the emergency data stream is guaranteed. In an implementation of the present invention, the transmitter 430 may send data of the emergency data stream in a manner as shown in the above steps S200-S230.

[00101] The transmitter 430 may further be configured to continue to send, during the incoming time slot, data of at least one data stream of the current data stream and data streams of the data streams whose priority levels are lower than that of the current data stream according to priority levels of the at least one data stream, after the transmission of data of the emergency data stream in the incoming time slot is finished, if the amount of data that has been transmitted in the incoming time slot after the transmission of data of the emergency data stream in the incoming time slot is finished is less than the available data amount, so that the delay requirements of the data streams are guaranteed at full steam while the delay requirement of the emergency data stream is guaranteed. In an implementation of the present invention, the transmitter 430 may continue to send data in a manner as shown in the above step [00102] The transmitter 430 may further be configured to, if not all data of the emergency data stream has been transmitted in the incoming time slot, transmit in each time slot following the incoming time slot data in the emergency data stream and the data streams in the same manner as for the incoming time slot, until all data of the emergency data stream is transmitted, wherein the priority level of the emergency data stream is set to be higher than that of the data streams, so that the delay requirement of the emergency data stream is guaranteed at full steam.

[00103] The receiver 410 may further be configured to receive a new data stream in the incoming time slot, and the new data stream may have data that will reach a maximum delay within the next time slot following the incoming time slot. The processor 420 may further be configured to calculate the first amount of data of the new data stream for the incoming time slot, if the transmission of data of the high priority data stream whose priority level is higher than that of the new data stream has not been finished. The processor 420 may further be configured to re-determine, based on leftover data amount that is provided to the data streams in the incoming time slot when the transmission of data of the high priority data stream in the incoming time slot is finished and the first data amount of each of the high priority data stream, the new data stream and low priority data streams of the data streams for the incoming time slot, the second data amount of each of the new data stream and the low priority data streams for the incoming time slot according to priority levels of the new data stream and the low priority data streams, the low priority data streams being those data streams whose priority levels are lower than that of the new data stream. The transmitter 430 may further be configured to send, during the incoming time slot, data of the new data stream and the low priority data streams over the common communication link based on the re-determined second amounts of data in each of the new data stream and the low priority data streams for the incoming time slot according to the priority levels of the new data stream and the low priority data streams, after the transmission of data of the high priority data stream is finished. Thus, the delay requirement of the new data stream is guaranteed at full steam. In an implementation of the present invention, the processor 420 may calculate the first amount of data in the new data stream in a manner as shown in the above step S320, the processor 420 may re-determine the first amount of data in a manner as shown in the steps S340 and S350, and the transmitter 430 may send data in a manner as shown in the above step S360.

[00104] The transmitter 430 may further be configured to stop, if the transmission of data of the high priority data stream has been finished, transmission of data of a current data stream of the data streams whose data is being transmitted over the common communication link in the incoming time slot, and send data of the new data stream over the common communication link in the incoming time slot. In an implementation of the present invention, the transmitter 430 may stop transmitting of data and send data of the new data stream in a manner as shown in the steps S370-S390.

[00105] The transmitter 430 may further be configured to continue to send during the incoming time slot data of at least one data stream of the current data stream and low priority level data streams of the data streams whose priority levels are lower than that of the current data stream according to priority levels of the at least one data stream, after the transmission of data of the new data stream in the incoming time slot is finished, if the amount of data that has been transmitted in the incoming time slot after the transmission of data of the new data stream in the incoming time slot is finished is less than the available data amount. In an implementation of the present invention, the transmitter 430 may continue to send data in a manner as shown in the step S400.

[00106] The processor 420 and the transmitter 430 may further be configured to execute, respectively, the calculating, the determining and the transmission for each time slot following the incoming time slot for the new data stream and the data streams.

[00107] The apparatus 400 may further comprise a non-transient memory device 440, configured to store a computer program for executing a process according to the method of the present invention. The memory device may further be configure to store data of the plurality of data streams that are scheduled to be transmitted in the incoming time slot and/or time slots following the incoming time slot.

[00108] A communication node may be made based on the apparatus as shown in Fig. 6. The communication node may be a base station, a radio network controller (R C), a switch or a router. [00109] What has been described above includes examples of one or more embodiments. It is, of course, not possible to describe every conceivable combination of components or methodologies for purpose of describing the aforementioned embodiments, but one of ordinary skill in the art may recognize that many further combinations and permutations of various embodiments are possible. Accordingly, the described embodiments are intended to embrace all such alterations, modifications and variations that fall before the end of the spirit and scope of the appended claims. Furthermore, to the extent that the term "includes" is used in either the detailed description or the claims, such term is intended to be inclusive in a manner similar to the term "comprising" as "comprising" is interpreted when employed as a transitional word in a claim.

Claims

CLAIMS What is claimed is:

1. A method for scheduling a transmission of a plurality of data streams over a common communication link, comprising:

2. The method according to claim 1, wherein determining the second amount of data comprises:

calculating, based on the available data amount and the first amount of data in each of the data streams for the incoming time slot according to the priority levels of the data streams, a third amount of data in each of the data streams, wherein the third amount of data is a maximum amount of data in each of the data streams that can be transmitted in the incoming time slot over the common communication link; and

calculating, based on the first amount of data and the third amount of data in each of the data streams for the incoming time slot, the second amount of data in each of the data streams for the incoming time slot.

3. The method according to claim 1, wherein the available data amount is determined based on a duration of the time slot and a maximum serving rate provided to the data streams in the incoming time slot.

4. The method according to claim 1, wherein the delay applied to a data stream is a multiple of a most stringent delay bound of the data streams, and the most stringent delay bound is no larger than the smallest value of the maximum delay values of the data streams.

5. The method according to claim 1, wherein before determining the first amount of data for each of the data streams, the method further comprises:

determining a delay offset of each of at least one data stream of the data streams; and

calculating an adjusted delay of each of the at least one data stream based on the delay applied to each of the at least one data stream and the obtained delay offset of each of the at least one data stream, wherein after the adjusted delay is applied to each of the at least one data stream, leftover capacity in each time slot is substantially a constant,

and wherein calculating the first amount of data further comprises:

calculating an adjusted first amount of data in each of the at least one data stream for the incoming time slot, wherein the adjusted first amount of data is an amount of data of each of the at least one data stream that will reach the adjusted delay before the end of the next time slot.

6. The method according to claim 3, wherein the duration of the time slot is equal to the most stringent delay bound.

7. The method according to claim 1, wherein the delay applied to a data stream is calculated according to the following equation:

d Floor^Af d wherein di is the delay applied to the i data stream, Di is the maximum delay value of the i^th data stream, d is the most stringent delay bound, and Floor(Di/d) is the largest integer no larger than Di/d.

8. The method according to claim 1, wherein the delay applied to each of the data streams is calculated by the following equation:

d,=2^J*d

wherein di is the delay applied to the i^th data stream, d is the most stringent delay bound, and j is an integer which causes 2^J*d to be closest to and no larger than the maximum delay value of the i^th data stream.

9. The method according to claim 1, further comprising:

sending, over the common communication link during the time slot, data of at least one data stream of the data streams that doesn't reach the delay before the end of the next time slot, after the transmission of the second amounts of data is finished, if the sum of the first data amount of the data streams for the time slot is less than the available data amount, or

sending, over the common communication link during the time slot, data of at least one elastic data stream whose delay is not as critical as the data streams, after the transmission of the second amounts of data is finished, if the sum of the first data amount of the data streams for the time slot is less than the available data amount.

10. The method according to claim 1, further comprising:

stopping transmission of data of a current data stream of the data streams whose data is being transmitted over the common communication link in the incoming time slot, when an emergency data stream arrives in the incoming time slot; and

sending data of the emergency data stream over the common communication link in the time slot.

11. The method according to claim 10, further comprising:

continuing to send, during the incoming time slot, data of at least one data stream of the current data stream and low priority level data streams of the data streams whose priority levels are lower than that of the current data stream according to priority levels of the at least one data stream, after the transmission of data of the emergency data stream in the incoming time slot is finished, if amount of data that has been transmitted in the time slot after the transmission of data of the emergency data stream in the incoming time slot is finished is less than the available data amount.

12. The method according to claim 10, further comprising:

if not all data of the emergency data stream has been transmitted in the incoming time slot, repeating the calculating, the determining and the transmitting for each time slot following the incoming time slot for the emergency data stream and the data streams, until all data of the emergency data stream is transmitted, wherein priority level of the emergency data stream is set to be higher than those of the data streams.

13. The method according to claim 1, further comprising:

when a new data stream arrives in the incoming time slot and the new data stream has data that will reach a maximum delay of the new date stream before the end of the next time slot, determining whether the transmission of data of a high priority data stream of the data streams whose priority level is higher than that of the new data stream data is finished;

computing the first amount of data in the new data streams for the incoming time slot, if the transmission of the data of a high priority data stream of the data streams whose priority level is higher than that of the new data stream data is not finished; re-determining, based on leftover data amount that is provided to the data streams in the incoming time slot when the transmission of data of the high priority data stream in the time slot is finished and the first data amount of each of the high priority data stream, the new data stream and low priority data streams of the data streams for the time slot, the second data amount of each of the new data stream and the low priority data streams for the incoming time slot according to priority levels of the new data stream and the low priority data streams, wherein the low priority data streams are those data streams whose priority levels are less than that of the new data stream; and

sending, during the incoming time slot, data of the new data stream and the low priority data streams over the common communication link based on the re-determined second amount of data in each of the new data stream and the low priority data streams for the incoming time slot according to the priority levels of the new data stream and the low priority data streams, after the transmission of data of the high priority data stream in the incoming time slot is finished.

14. The method according to claim 13, further comprising:

stopping transmitting of data of a current data stream whose data is being transmitted over the common communication link in the incoming time slot, if transmission of the data of a high priority data stream of the data streams whose priority level is higher than that of the new data stream data is finished; and

sending data of the new data stream over the common communication link in the incoming time slot.

15. The method according to claim 14, further comprising:

continuing to send during the incoming time slot data of at least one data stream of the current data stream and low priority level data streams of the data streams whose priority levels are lower than that of the current data stream according to priority levels of the at least one data stream, after the transmission of data of the new data stream in the incoming time slot is finished, if amount of data that has been transmitted in the incoming time slot after the transmission of data of the new data stream in the incoming time slot is finished is less than the available data amount.

16. The method according to claim 13, further comprising:

executing the calculating, the determining and the transmitting for each time slot following the incoming time slot for the new data stream and the data streams.

17. An apparatus for scheduling a transmission of a plurality of data streams over a common communication link, comprising:

a processor connected to the receiver, configured to:

for each of the data streams, determine a first amount of data in the data stream for an incoming time slot, wherein the first amount of data is an amount of data in the data stream that will reach a delay before the end of a time slot following the incoming time slot, and wherein the delay applied to the data stream is no larger than a maximum delay value of the data stream; for each of the data streams, determine, based on an available data amount that is provided to the data streams for the incoming time slot and the first amounts of data in the data streams for the incoming time slot according to priority levels of the data streams, a second amount of data in the data stream, wherein the second amount of data is an amount of data in the data stream that will actually be transmitted in the incoming time slot; and

and a transmitter connected to the processor, configured to:

18. The apparatus according to claim 17, wherein

the processor is further configured to calculate, based on the available data amount and the first amount of data in each of the data streams for the incoming time slot according to the priority levels of the data streams, a third amount of data in each of the data streams for the incoming time slot, wherein the third amount of data is a maximum amount of data in each of the data streams that can be transmitted in the incoming time slot over the common communication link, and calculate, based on the first amount of data and the third amount of data in each of the data streams for the incoming time slot, the second amount of data amount in each of the data streams for the time slot.

19. The apparatus according to claim 17, wherein the processor is further configured to determine the available data amount based on a duration of the time slot and a maximum serving rate provided to the data streams in the incoming time slot.

20. The apparatus according to claim 17, wherein

the processor is further configured to determine the delay applied to each of the data streams as a multiple of a most stringent delay bound, wherein the most stringent delay bound is no larger than the smallest value of the maximum delay values of the data streams.

21. The apparatus according to claim 17, wherein the processor is further configured to:

determine a delay offset of each of at least one data stream of the data streams, and calculate an adjusted delay for each of the at least one data stream based on the delay applied to each of the at least one data stream and the obtained delay offset of each of the at least one data stream, wherein when the adjusted delay is applied to each of the at least one data stream, a leftover capacity in each of the time slots is substantially the same,

and wherein the processor is further configured to calculate an adjusted first amount of data in each of the at least one data stream for the incoming time slot, wherein the adjusted first amount of data is an amount of data of each of the at least one data stream that will reach the adjusted delay before the end of the next time slot.

22. The apparatus according to claim 17, wherein

the transmitter is further configured to send, over the common communication link during the incoming time slot, data of at least one data stream of the data streams that doesn't reach the applied delay before the end of the time slot , or send data of at least one elastic data stream whose delay is not as critical as the data streams, after the transmission is finished, if a sum of the first data amount of the data streams for the incoming time slot is less than the available data amount .

23. The apparatus according to claim 17, wherein the apparatus is a base station, a radio network controller, a switch or a router.

24. A computer program product comprising a computer readable storage medium storing program code thereon for use by a communication node for scheduling transmission of a plurality of data streams over a common communication link, the program code comprising instructions for executing a method that comprises: for each of the data streams, determining a first amount of data in the data stream for an incoming time slot, wherein the first amount of data is an amount of data in the data stream that will reach a delay before the end of a time slot following the incoming time slot, and wherein the delay applied to the data stream is no larger than a maximum delay value of the data stream;