CN102547852A - Packet scheduling method and device - Google Patents

Abstract

The invention provides a packet scheduling method and device. The method comprises the following steps of: obtaining the business data throughput of a user, and estimating the average throughput of the user according to the business data throughput of the user; and if the average throughput of the user is less than the minimal throughput, improving the scheduling priority of the user so that the user gets service. Through the scheme of the invention, the fairness among users of the system can be improved.

Description

A kind of grouping scheduling method and device

Technical field

The present invention relates to HSDPA (High Speed Downlink Package Access, high-speed downlink packet inserts) dispatching algorithm field, be meant a kind of user fairness and preferential grouping scheduling method and device taken into account especially.

Background technology

In the HSDPA system; Realize the management function of Radio Resource by packet scheduling algorithm; The advanced grouping height algorithm of research be improve the data service throughput, guarantee between the user fairness, satisfy the basic of professional Qos (Quality of Service, service quality).

Current, Proportional Fair algorithm (PF) is a kind of dispatching algorithm of in mobile radio network, widely using, and it can provide good balance for the fairness between system's maximum throughput and user, can satisfy the requirement of non-real-time service well.But it is, therefore, unsatisfactory for the real time business effect of delay sensitive because this algorithm is not considered the time delay of packet.

To this problem, preferential (M-LWDF) dispatching algorithm of maximum weighted time delay has been done further improvement in the prior art, and it is widely used in the systems such as HSDPA, EVDO, 4G.This algorithm has been introduced user Qos parameter and packet time delay; Its main thought is with the time delay of packet data package and how effectively utilizes the channel information balance to take into account; The calculating of its User Priority is not only relevant with the current channel quality of user, and is also relevant with the formation time delay of bag.User i is at n TTI (Transmission Time Interval; Transmission Time Interval) priority computing formula is following: i=1; 2 ... n

Wherein, the Qos parameter of δ representative of consumer, R [n] is the maximum data rate of user at n TTI, and λ is the average throughput of user i, and D [n] is the formation time-delay of user i place bag, and T is the patient maximum delay of user i (abandoning time parameter)

The delay variation that this algorithm is produced cell throughout and base station formation has been done effectively compromise; It is a kind of non-fair algorithm; The subscriber channel condition has better Qos when good, generally have 2-3 second the formation time-delay, yet concerning the user of bad channel conditions; This algorithm can cause these user's data to wrap in base station side has bigger time delay, and the maximum tolerance that surpasses the user when time delay will be abandoned during the time.

The M-LWDF algorithm is the throughput preferred algorithm; It takes all factors into consideration the time delay and the current channel quality information of different user packet data package; But the M-LWDF algorithm is not taken into account user fairness, and the lifting of its throughput performance is a cost to sacrifice user fairness, and the user's that channel status is relatively poor meeting in group is waited in the formation of base station; In case and overtime limits, these groupings will be lost and no longer served.Relatively poor to user's fairness like this.

Summary of the invention

The technical problem that the present invention will solve provides a kind of grouping scheduling method and the device comparatively fair to the user.

For solving the problems of the technologies described above, embodiments of the invention provide a kind of grouping scheduling method, comprising:

Obtain user's business datum throughput, and according to the average throughput of said user's business datum throughput estimating user;

If said user's average throughput then improves said user's dispatching priority less than minimum throughout, make said user obtain service.

Wherein, the said user's of said raising dispatching priority is specially:

Pass through formula:

improve said user's dispatching priority;

Wherein, P _iFor by the priority of the said user i that served, j is the priority that has the user of limit priority, and K is an integer.

Wherein, above-mentioned grouping scheduling method also comprises:

For real time business, the average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout.

Wherein, the average throughput in the formation of base station, waited for is specially less than the real-time dual compensation that at least one said user of minimum throughout carries out priority and data throughout:

Pass through formula: $P_i\left[n\right]=\left\{\begin{array}{cc}-\mathrm{Log}\left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]& \mathrm{If}{\mathrm{\&lambda;}}_i\&GreaterEqual;k\\ -\mathrm{Log}\left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]\&CenterDot;Z_i\left[n\right]& \mathrm{If}{\mathrm{\&lambda;}}_i<k\end{array}\right.$

The average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout;

Wherein, δ _iBe the Qos parameter of user i, R _i[n] is the maximum data rate of user i at n Transmission Time Interval TTI, λ _i[n] is the average throughput of user i, D _i[n] is the formation time-delay of user i place bag, T _iBe the patient maximum delay of user i, k is a minimum throughout,

L _i[n] is the data to be transmitted amount of user i, P _i[n] is the priority of user i at n Transmission Time Interval TTI.Wherein, when each packet scheduling begins, add up current L _i[n], order

After every completion one number of sub-carrier is distributed, upgrade λ i by following formula _[N] and L _i[n]:

${\mathrm{\&lambda;}}_i\left[n\right]={\mathrm{\&lambda;}}_i\left[n\right]+\frac{1}{n_r}\&CenterDot;{\mathrm{\&rho;}}_i\&CenterDot;R_i\left[n\right]$

L _i[n]＝L _i[n]-L _ii

Wherein, R _i[n] is the peak transfer rate of active user i on carrier wave, L _IiBe the data volume of user i actual transmissions on carrier wave, when sending the data of user i on the carrier wave, ρ _i=1, otherwise ρ _i=0.

Embodiments of the invention also provide a kind of packet scheduling apparatus, comprising:

First processing module is used to obtain user's business datum throughput, and according to the average throughput of said user's business datum throughput estimating user;

Second processing module is used for making said user obtain service if said user's average throughput less than minimum throughout, then improves said user's dispatching priority.

Wherein, said second processing module is specifically passed through formula:

improve said user's dispatching priority;

Wherein, P _iFor by the priority of the said user i that served, j is the priority that has the user of limit priority, and K is an integer.

Wherein, above-mentioned packet scheduling apparatus also comprises:

The 3rd processing module is used for for real time business, the average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout.

Wherein, said the 3rd processing module is specifically passed through formula:

$P_i\left[n\right]=\left\{\begin{array}{cc}-\log \left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]& \mathrm{if}{\mathrm{\&lambda;}}_i\&GreaterEqual;k\\ -\log \left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]\&CenterDot;Z_i\left[n\right]& \mathrm{if}{\mathrm{\&lambda;}}_i<k\end{array}\right.$

The average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout;

Wherein, δ _iBe the Qos parameter of user i, R _i[n] is the maximum data rate of user i at n Transmission Time Interval TTI, λ _i[n] is the average throughput of user i, D _i[n] is the formation time-delay of user i place bag, T _iBe the patient maximum delay of user i, k is a minimum throughout,

Be the data to be transmitted amount of user i, P _i[n] is the priority of user i at n Transmission Time Interval TTI.

Wherein, said the 3rd processing module is further used for: when each packet scheduling begins, add up current L _i[n], order

After every completion one number of sub-carrier is distributed, upgrade λ by following formula _i[n] and L _i[n]:

${\mathrm{\&lambda;}}_i\left[n\right]={\mathrm{\&lambda;}}_i\left[n\right]+\frac{1}{n_r}\&CenterDot;{\mathrm{\&rho;}}_i\&CenterDot;R_i\left[n\right]$

L _i[n]＝L _i[n]-L _ii

Wherein, R _i[n] is the peak transfer rate of active user i on carrier wave, L _IiBe the data volume of user i actual transmissions on carrier wave, when sending the data of user i on the carrier wave, ρ _i=1, otherwise ρ _i=0.

The beneficial effect of technique scheme of the present invention is following:

In the such scheme,, and during less than minimum throughout, then improve said user's dispatching priority, make said user obtain service at user's average throughput through the average throughput of estimating user; This method has considered that user's minimum throughout guarantees, so that weigh the loading condition and the priority that improves low throughput users of Subscriber Queue better, has increased the fairness between the user in the system.

Description of drawings

Fig. 1 is the grouping scheduling method first embodiment flow chart of the present invention;

Fig. 2 is the grouping scheduling method second embodiment flow chart of the present invention.

Embodiment

For technical problem, technical scheme and advantage that the present invention will be solved is clearer, be described in detail below in conjunction with accompanying drawing and specific embodiment.

The present invention is directed to existing dispatching algorithm to the relatively poor problem of user's fairness, a kind of grouping scheduling method and the device comparatively fair to the user are provided.

As shown in Figure 1, grouping scheduling method of the present invention comprises:

Step 11 is obtained user's business datum throughput, and according to the average throughput of said user's business datum throughput estimating user;

Step 12 if said user's average throughput then improves said user's dispatching priority less than minimum throughout, makes said user obtain service.

Such scheme of the present invention is through the average throughput of estimating user, and during less than minimum throughout, then improves said user's dispatching priority at user's average throughput, makes said user obtain service; This method has considered that user's minimum throughout guarantees, so that weigh the loading condition and the priority that improves low throughput users of Subscriber Queue better, has increased the fairness between the user in the system.

Wherein, for the user's data that reduces bad channel quality encapsulates the probability that abandons, improve user's fairness, adopted new priority formula that said user's priority is improved in the present embodiment, the user lower to throughput compensates; Specifically, in the above-mentioned steps 12, the dispatching priority that improves said user specifically can for:

Pass through formula:

improve said user's dispatching priority;

Wherein, P _iFor by the priority of the said user i that served, j is the priority that has the user of limit priority, and K is an integer;

Minimum throughout guarantees it is the important component part that QoS (service quality) requires, if each user has minimum throughout to guarantee, all users that satisfy this condition will be served so; To not satisfying the user of this condition, can its dispatching priority be improved, to guarantee the fairness of low throughput users through changing scheduling rule (improving user's dispatching priority) as adopting above-mentioned formula.

In addition, on the basis of the embodiment of above-mentioned dispatching method shown in Figure 1, as shown in Figure 2, above-mentioned dispatching method can further include:

Step 13 for real time business, is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout to the average throughput of in the formation of base station, waiting for;

Can reduce Node B side data stream like this and encapsulate the probability that abandons; Guarantee user fairness; The number of users of maximization real time business under the condition of the minimum throughout that guarantees user grouping; Specifically, can adopt following manner the user of bad channel conditions to be carried out the real-time dual compensation of priority and data throughout:

Promptly the average throughput in the formation of base station, waited for is specially less than the real-time dual compensation that at least one said user of minimum throughout carries out priority and data throughout:

Pass through formula: $P_i\left[n\right]=\left\{\begin{array}{cc}-\mathrm{Log}\left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]& \mathrm{If}{\mathrm{\&lambda;}}_i\&GreaterEqual;k\\ -\mathrm{Log}\left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]\&CenterDot;Z_i\left[n\right]& \mathrm{If}{\mathrm{\&lambda;}}_i<k\end{array}\right.$

The average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout;

Wherein, δ _iBe the Qos parameter of user i, R _i[n] is the maximum data rate of user i at n TTI, λ _i[n] is the average throughput of user i, D _i[n] is the formation time-delay of user i place bag, T _iBe the patient maximum delay of user i, k is a minimum throughout, L _i[n] is the data to be transmitted amount of user i, P _i[n] is the priority of user i at n Transmission Time Interval TTI.

In above-mentioned formula, λ _i[n] and L _i[n] is the statistic about user and formation, need upgrade timely, promptly upgrades after each sub-carrier resources is accomplished distribution and carries out; Compare with original update method of after each packet scheduling finishes, carrying out, the method for real-time update is ageing stronger, can reflect the user better by the time of day of service degree and queuing message, and when packet scheduling, has stronger flexibility;

Specifically, when each packet scheduling begins, add up current L _i[n], order

After every completion one number of sub-carrier is distributed, upgrade λ by following formula _i[n] and L _i[n]:

${\mathrm{\&lambda;}}_i\left[n\right]={\mathrm{\&lambda;}}_i\left[n\right]+\frac{1}{n_r}\&CenterDot;{\mathrm{\&rho;}}_i\&CenterDot;R_i\left[n\right]$

L _i[n]＝L _i[n]-L _ii

Wherein, R _i[n] is the peak transfer rate of active user i on carrier wave, L _IiBe the data volume of user i actual transmissions on carrier wave, when sending the data of user i on the carrier wave, ρ _i=1, otherwise ρ _i=0.

This real time updating method is introduced 2 new parameters simultaneously: the data to be transmitted amount L of user i, and better weigh the loading condition of Subscriber Queue, and adopted the method for real-time update state parameter, have stronger dispatching flexibility.Begin to be lower than the user of minimum throughout for average throughput,, increased the fairness between the user in the system through introducing the priority that a weights factor Z improves low throughput users.

For real time business, this new method of the present invention is not only considered the influence of channel condition and user's time delay, has considered that also user's minimum throughout guarantees; And algorithm has been introduced the information of new measurement service queue load; Adopted the method for real-time update state parameter, had stronger dispatching flexibility, so that weigh the loading condition and the priority that improves low throughput users of Subscriber Queue better; Increase the fairness between the user in the system, helped improving the overall performance of system.

Corresponding with above-mentioned method shown in Figure 1, scheme of the present invention also provides a kind of packet scheduling apparatus, comprising:

First processing module is used to obtain user's business datum throughput, and according to the average throughput of said user's business datum throughput estimating user;

Second processing module is used for making said user obtain service if said user's average throughput less than minimum throughout, then improves said user's dispatching priority.

Wherein, said second processing module is specifically passed through formula: improve said user's dispatching priority;

Wherein, P _iFor by the priority of the said user i that served, j is the priority that has the user of limit priority, and K is an integer.

Further, above-mentioned dispatching device also comprises:

The 3rd processing module is used for for real time business, the average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout.

Particularly, said the 3rd processing module is specifically passed through formula:

$P_i\left[n\right]=\left\{\begin{array}{cc}-\log \left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]& \mathrm{if}{\mathrm{\&lambda;}}_i\&GreaterEqual;k\\ -\log \left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]\&CenterDot;Z_i\left[n\right]& \mathrm{if}{\mathrm{\&lambda;}}_i<k\end{array}\right.$

The average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout;

Wherein, δ _iBe the Qos parameter of user i, R _i[n] is the maximum data rate of user i at n Transmission Time Interval TTI, λ _i[n] is the average throughput of user i, D _i[n] is the formation time-delay of user i place bag, T _iBe the patient maximum delay of user i, k is a minimum throughout,

L _i[n] is the data to be transmitted amount of user i, P _i[n] is the priority of user i at n Transmission Time Interval TTI.

Wherein, Said the 3rd processing module is further used for: when each packet scheduling begins, add up current

After every completion one number of sub-carrier is distributed, upgrade λ by following formula _i[n] and L _i[n]:

${\mathrm{\&lambda;}}_i\left[n\right]={\mathrm{\&lambda;}}_i\left[n\right]+\frac{1}{n_r}\&CenterDot;{\mathrm{\&rho;}}_i\&CenterDot;R_i\left[n\right]$

L _i[n]＝L _i[n]-L _ii

Wherein, R _i[n] is the peak transfer rate of active user i on carrier wave, L _IiBe the data volume of user i actual transmissions on carrier wave, when sending the data of user i on the carrier wave, ρ _i=1, otherwise ρ _i=0.

Need to prove: all concrete implementations all are applicable to corresponding module among this device embodiment in the said method of the present invention; Also can reach identical technique effect, and above-mentioned dispatching method of the present invention can be widely used in the systems such as HSDPA, EVDO, 4G with device.

The above is a preferred implementation of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from principle according to the invention; Can also make some improvement and retouching, these improvement and retouching also should be regarded as protection scope of the present invention.

Claims (10)

1. a grouping scheduling method is characterized in that, comprising:

Obtain user's business datum throughput, and according to the average throughput of said user's business datum throughput estimating user;

If said user's average throughput then improves said user's dispatching priority less than minimum throughout, make said user obtain service.

2. grouping scheduling method according to claim 1 is characterized in that, the said user's of said raising dispatching priority is specially:

Pass through formula:

improve said user's dispatching priority;

Wherein, P _iFor by the priority of the said user i that served, j is the priority that has the user of limit priority, and K is an integer.

3. grouping scheduling method according to claim 1 and 2 is characterized in that, also comprises:

For real time business, the average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout.

4. grouping scheduling method according to claim 3 is characterized in that, the average throughput in the formation of base station, waited for is specially less than the real-time dual compensation that at least one said user of minimum throughout carries out priority and data throughout:

Pass through formula: $P_i\left[n\right]=\left\{\begin{array}{cc}-\mathrm{Log}\left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]& \mathrm{If}{\mathrm{\&lambda;}}_i\&GreaterEqual;k\\ -\mathrm{Log}\left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]\&CenterDot;Z_i\left[n\right]& \mathrm{If}{\mathrm{\&lambda;}}_i<k\end{array}\right.$

The average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout;

Wherein, δ _iBe the Qos parameter of user i, R _i[n] is the maximum data rate of user i at n Transmission Time Interval TTI, λ _i[n] is the average throughput of user i, D _i[n] is the formation time-delay of user i place bag, T _iBe the patient maximum delay of user i, k is a minimum throughout,

L _i[n] is the data to be transmitted amount of user i, P _i[n] is the priority of user i at n Transmission Time Interval TTI.

5. grouping scheduling method according to claim 4 is characterized in that, when each packet scheduling begins, adds up current L _i[n], order

After every completion one number of sub-carrier is distributed, upgrade λ by following formula _i[n] and L _i[n]:

${\mathrm{\&lambda;}}_i\left[n\right]={\mathrm{\&lambda;}}_i\left[n\right]+\frac{1}{n_r}\&CenterDot;{\mathrm{\&rho;}}_i\&CenterDot;R_i\left[n\right]$

L _i[n]＝L _i[n]-L _ii

Wherein, R _i[n] is the peak transfer rate of active user i on carrier wave, Li _iBe the data volume of user i actual transmissions on carrier wave, when sending the data of user i on the carrier wave, ρ _i=1, otherwise ρ _i=0.

6. a packet scheduling apparatus is characterized in that, comprising:

First processing module is used to obtain user's business datum throughput, and according to the average throughput of said user's business datum throughput estimating user;

Second processing module is used for making said user obtain service if said user's average throughput less than minimum throughout, then improves said user's dispatching priority.

7. packet scheduling apparatus according to claim 6; It is characterized in that said second processing module is specifically passed through formula:

improve said user's dispatching priority;

Wherein, P _iFor by the priority of the said user i that served, j is the priority that has the user of limit priority, and K is an integer.

8. according to claim 6 or 7 described packet scheduling apparatus, it is characterized in that, also comprise:

The 3rd processing module is used for for real time business, the average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout.

9. packet scheduling apparatus according to claim 8 is characterized in that, said the 3rd processing module is specifically passed through formula:

$P_i\left[n\right]=\left\{\begin{array}{cc}-\log \left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]& \mathrm{if}{\mathrm{\&lambda;}}_i\&GreaterEqual;k\\ -\log \left({\mathrm{\&delta;}}_i\right)\&CenterDot;\frac{R_i\left[n\right]}{{\mathrm{\&lambda;}}_i\left[n\right]}\&CenterDot;\frac{D_i\left[n\right]}{T_i}\&CenterDot;L_i\left[n\right]\&CenterDot;Z_i\left[n\right]& \mathrm{if}{\mathrm{\&lambda;}}_i<k\end{array}\right.$

The average throughput of in the formation of base station, waiting for is carried out the real-time dual compensation of priority and data throughout less than at least one said user of minimum throughout;

Wherein, δ _iBe the Qos parameter of user i, R _i[n] is the maximum data rate of user i at n Transmission Time Interval TTI, λ _i[n] is the average throughput of user i, D _i[n] is the formation time-delay of user i place bag, T _iBe the patient maximum delay of user i, k is a minimum throughout,

L _i[n] is the data to be transmitted amount of user i, P _i[n] is the priority of user i at n Transmission Time Interval TTI.

10. packet scheduling apparatus according to claim 9 is characterized in that, said the 3rd processing module is further used for: when each packet scheduling begins, add up current L _i[n], order

After every completion one number of sub-carrier is distributed, upgrade λ by following formula _i[n] and L _i[n]:

${\mathrm{\&lambda;}}_i\left[n\right]={\mathrm{\&lambda;}}_i\left[n\right]+\frac{1}{n_r}\&CenterDot;{\mathrm{\&rho;}}_i\&CenterDot;R_i\left[n\right]$

L _i[n]＝L _i[n]-L _ii

Wherein, R _i[n] is the peak transfer rate of active user i on carrier wave, L _IiBe the data volume of user i actual transmissions on carrier wave, when sending the data of user i on the carrier wave, ρ _i=1, otherwise ρ _i=0.

Images