CN105873226A - LTE (Long Term Evolution) fair scheduling method with priority of throughput - Google Patents

Abstract

This invention discloses a LTE fair scheduling method with a priority of throughput. The method comprises the following steps: checking the throughput; updating the priority; and scheduling users with high throughput and users with low throughput. The method provided by the invention can automatically track the speed change of each user in a residential area, dynamically set an evaluation threshold, sort out the users with good channel quality, and improve the scheduling priority thereof; the user with low throughput can be prevented from starving in the manner of fair scheduling; furthermore, the throughput and the fairness of the system are both considered; and thus, the service quality for developing the service can be improved so as to ensure the customer perception.

Description

The LTE equity dispatching method that handling capacity is preferential

Technical field

The present invention relates to the dispatching method of a kind of LTE system, the LTE equity dispatching side that a kind of handling capacity is preferential Method.

Background technology

LTE uses the multiple key technologies such as OFDM, SC-FDMA and MIMO to improve and enhance the air interface technologies of 3G, The time delay of user face and chain of command can be significantly reduced, improve the handling capacity of community.Dispatching algorithm is the most preferably used to be an up Handling capacity key point, for different users, according to the channel instruction that it is current, the many factors such as service quality and dispatch state Carry out RB resource distribution, ensure that the resource of every sub-distribution makes full use of as far as possible, promote user's perception.Limited by frequency spectrum resource Property constraint, the resource scheduling algorithm of LTE seems most important, which determine packet transmission order and available transmission rate.

Such as RR (Round Robin, polling dispatching), PF (Proportion Fair, equitable proportion) and SJF (Short Job First, shortest job first) etc. dispatching method, the most merely emphasize throughput of system or fairness, do not account for two Equilibrium how is realized between person.

Summary of the invention

The goal of the invention of the present invention is to overcome LTE system scheduling of resource efficiency of the prior art low, it is impossible to take into account Throughput of system and the deficiency of fairness, it is provided that the LTE equity dispatching method that a kind of handling capacity is preferential.

To achieve these goals, the present invention is by the following technical solutions:

The LTE equity dispatching method that a kind of handling capacity is preferential, including N number of user { u₁, u₂..., u_N, set u_iIt is each User, u_i∈{u₁, u₂..., u_N}；Comprise the steps:

(1-1) handling capacity is adjusted；

(1-2) priority update；

(1-3) high throughput users scheduling；

(1-4) low throughput users scheduling.

The LTE fairness dispatching method that the handling capacity of the present invention is preferential, the user good by filtering out channel quality, promote Its dispatching priority, and for the highest user of handling capacity, then to prevent it from suffering from hunger by the way of equity dispatching, and then make Dispatching method can take into account throughput of system and fairness simultaneously.

The LTE fairness dispatching method that the handling capacity of the present invention is preferential, can automatically follow the tracks of the speed of each user in community Rate changes, and dynamically arranges assessment threshold, and then the service quality that the business that promotes as much as possible is carried out is to ensure client perception.

As preferably, step (1-1) comprises the steps:

(2-1) set the signal to noise ratio of each user asThe bit error rate of business isDistribute to the RB of each user Resource Block quantity isUtilize formula Calculate the throughput evaluation index of each user

(2-2) a width of B of band shared by each RB resource is set_RB, utilize formulaCalculate each The handling capacity of user

As preferably, step (1-2) comprises the steps:

(3-1) formula is utilizedCalculate handling capacity desired valueAnd Utilize formulaCalculate the nondimensionalization throughput value of each user

(3-2) formula is utilizedCalculate the desired value after nondimensionalizationUtilize formulaCalculate variances sigma_i, utilize formulaCalculate weight coefficient

(3-3)

Utilize formulaCalculate and update the dispatching priority ρ of each user_i。

As preferably, step (1-3) comprises the steps:

(4-1) set k user to meetThe priority of k user is { ρ₁, ρ₂..., ρ_k, calculate minimum excellent First level ρ_a=min (ρ_j), ρ_j∈{ρ₁, ρ₂..., ρ_k}；

(4-2) by { ρ₁, ρ₂..., ρ_kCarrying out descending, scheduler carries out elder generation according to the size of each User Priority Rear scheduling.

As preferably, step (1-4) comprises the steps:

(5-1) for remaining N-k user, its priority is { ρ respectively_k+1, ρ_k+2..., ρ_N, utilize formulaρ_q∈{ρ_k+1, ρ_k+2..., ρ_NCalculate N-k user desired value ρ_b；

(5-2) formula ρ is utilized_N-k=min (ρ_a, ρ_b) calculate ρ_N-k, arrange unified for the priority of remaining N-k user For ρ_N-k, scheduler is according to ρ_N-kRemaining N-k user is scheduling.

Therefore, there is advantages that and take into account throughput of system and fairness, can automatically follow the tracks of community In the speed change of each user, assessment threshold is dynamically set, the service quality that the business that promotes as much as possible is carried out is to ensure visitor Family perception.

Accompanying drawing explanation

Fig. 1 is a kind of flow chart of the present invention；

Fig. 2 is a kind of handling capacity comparison diagram of the present invention and RR, SJF；

Fig. 3 is a kind of packet loss comparison diagram of the present invention and RR, SJF；

Fig. 4 is a kind of fairness comparison diagram of the present invention and RR, SJF.

Detailed description of the invention

The present invention will be further described with detailed description of the invention below in conjunction with the accompanying drawings.

Embodiment as shown in Figure 1 be embodiment be the preferential LTE equity dispatching method of a kind of handling capacity, including N number of use Family { u₁, u₂..., u_N, set u_iIt is each user, u_i∈{u₁, u₂..., u_N}；

Step 1: handling capacity is adjusted；

Step 1-1: set the signal to noise ratio of each user asThe bit error rate of business isDistribute to each user RB Resource Block quantity beUtilize formula Calculate the throughput evaluation index of each user

Step 1-2: set a width of B of band shared by each RB resource_RB, utilize formulaCalculate The handling capacity of each user

Step 2: priority update；

Step 2-1: utilize formulaCalculate handling capacity desired valueAnd utilize formulaCalculate the nondimensionalization throughput value of each user

Step 2-2: utilize formulaCalculate the desired value after nondimensionalizationUtilize formulaCalculate variances sigma_i, utilize formulaCalculate weight coefficient

Step 2-3:

Utilize formulaCalculate and update the dispatching priority ρ of each user_i；

Step 3: high throughput users is dispatched；

Step 3-1: set k user and meetThe priority of k user is { ρ₁, ρ₂..., ρ_k, calculate minimum Priority ρ_a=min (ρ_j), ρ_j∈{ρ₁, ρ₂..., ρ_k}；

Step 3-2: by { ρ₁, ρ₂..., ρ_kCarrying out descending, scheduler is carried out according to the size of each User Priority Successively scheduling.；

Step 4: low throughput users is dispatched；

Step 4-1: for remaining N-k user, its priority is { ρ respectively_k+1, ρ_k+2..., ρ_N, utilize formulaρ_q∈{ρ_k+1, ρ_k+2..., ρ_NCalculate N-k user desired value ρ_b；

Step 4-2: utilize formula ρ_N-k=min (ρ_a, ρ_b) calculate ρ_N-k, set unified for the priority of remaining N-k user It is set to ρ_N-k, scheduler is according to ρ_N-kRemaining N-k user is scheduling.

Being specifically described the present invention as a example by N=5 below, representative basis data are as shown in table 1, table 2:

Sequence number	Project	Data
			1	TTI(ms)	1
2	System bandwidth (MHz)	20
			3	RB sum (individual)	80
4	Each RB resource occupation bandwidth BRB(kHz)	180

Table 1

Table 2

The preferential LTE fairness dispatching method of handling capacity that this example describes, comprises the steps: that handling capacity is adjusted, excellent First level updates, and high throughput users is dispatched, and low throughput users is dispatched.

Step 1: handling capacity is adjusted；

Calculate user { u₁, u₂..., u₅Throughput evaluation index

According to B_RB, calculate the handling capacity of each user

$T_{u_i}=B_{RB}*{\mathrm{CEI}}_{u_i}\{3.31,2.30,1.62,2.24,3.24\}Mbos;$

Step 2: priority update；

Step 2-1: calculate handling capacity desired valueAnd the handling capacity of each user is carried out dimensionless Change processes

Step 2-2: solve the desired value after nondimensionalizationDraw variance

And solve weight coefficient

Step 2-3: update the dispatching priority of each user according to weight coefficient

Step 3: high throughput users is dispatched；

Step 3-1: meetNumber of users k=2, i.e. u₁And u₅, its priority be respectively 0.26,0.24}, calculate Minimum priority ρ_a=min (ρ_j)=0.24；

Step 3-2: by { 0.26,0.24} carries out descending, and scheduler is according to user { u₁, u₅The size of priority enters Row successively scheduling；

Step 4: low throughput users is dispatched；

Step 4-1: for remaining 3 user { u₂, u₃, u₄, its priority be respectively 0.08,0.31,0.10}, meter Calculate its desired value

Step 4-2: calculate ρ_N-k=min (0.24,0.17)=0.17, by { u₂, u₃, u₄Priority unified be set to 0.17, submit to scheduler, scheduler according to 0.17 to { u₂, u₃, u₄Be scheduling.

Emulation experiment:

The methods such as the TFFG method of the present invention and RR, PF and SJF carry out simulation comparison, and scheduling number of users is from 10 progressively Increasing to 30, the LTE system handling capacity of gained, the packet loss and the method fairness that occur in scheduling process see attached respectively Fig. 2 to 4.

As in figure 2 it is shown, the TFFG of the present invention gives the good user of channel quality scheduling power more preferably, therefore produce and compare RR Handling capacity higher with SJF algorithm, and along with participating in the increase of scheduling number of users, throughput of system is consequently increased.

As it is shown on figure 3, the packet loss of the TFFG of the present invention is between RR and SJF, its reason is that SJF is all the time with the shortest Job priority, as dispatching criterion, does not consider user's state in which environment, and therefore its packet loss is always positioned at a high position, and RR is then only It is to have taken into account each user in line with fair principle simply.

As shown in Figure 4, the TFFG of the present invention, while considering high throughput users, has taken into account the public affairs of low throughput users Levelling, therefore, its fairness is positioned among RR Yu SJF, and along with the increase of scheduling user, fairness is gradually lowered.

Should be understood that the present embodiment is merely to illustrate the present invention rather than limits the scope of the present invention.In addition, it is to be understood that After having read the content that the present invention lectures, the present invention can be made various changes or modifications by those skilled in the art, these etc. Valency form falls within the application appended claims limited range equally.

Claims (5)

1. the LTE equity dispatching method that handling capacity is preferential, is characterized in that, including N number of user { u₁, u₂..., u_N, set u_i It is each user, u_i∈{u₁, u₂..., u_N}；Comprise the steps:

(1-1) handling capacity is adjusted；

(1-2) priority update；

(1-3) high throughput users scheduling；

(1-4) low throughput users scheduling.

The LTE equity dispatching method that handling capacity the most according to claim 1 is preferential, is characterized in that, step (1-1) include as Lower step:

(2-1) set the signal to noise ratio of each user asThe bit error rate of business isDistribute to the RB money of each user Source number of blocks isUtilize formula Calculate the throughput evaluation index of each user

(2-2) a width of B of band shared by each RB resource is set_RB, utilize formulaCalculate each user Handling capacity

The LTE equity dispatching method that handling capacity the most according to claim 1 is preferential, is characterized in that, step (1-2) include as Lower step:

(3-1) formula is utilizedCalculate handling capacity desired valueAnd utilize FormulaCalculate the nondimensionalization throughput value of each user

(3-2) formula is utilized Calculate the desired value after nondimensionalizationUtilize formulaCalculate variances sigma_i, utilize formulaCalculate weight coefficient

(3-3)

Utilize formula Calculate and update the dispatching priority ρ of each user_i。

The LTE equity dispatching method that handling capacity the most according to claim 1 is preferential, is characterized in that, step (1-3) include as Lower step:

(4-1) set k user to meetThe priority of k user is { ρ₁, ρ₂..., ρ_k, calculate minimum priority ρ_a=min (ρ_j), ρ_j∈{ρ₁, ρ₂..., ρ_k}；

(4-2) by { ρ₁, ρ₂..., ρ_kCarrying out descending, scheduler carries out successively adjusting according to the size of each User Priority Degree.

5., according to the LTE equity dispatching method that the handling capacity described in claim 1 or 2 or 3 or 4 is preferential, it is characterized in that, step (1-4) comprise the steps:

(5-1) for remaining N-k user, its priority is { ρ respectively_k+1, ρ_k+2..., ρ_N, utilize formulaρ_q∈{ρ_k+1, ρ_k+2..., ρ_NCalculate N-k user desired value ρ_b；

(5-2) formula ρ is utilized_N-k=min (ρ_a, ρ_b) calculate ρ_N-k, it is set to unified for the priority of remaining N-k user ρ_N-k, scheduler is according to ρ_N-kRemaining N-k user is scheduling.