CN102625316B - Random back-off distributed coordination scheduling method - Google Patents

Abstract

The present invention provides a random backoff distributed coordinated scheduling technology, including: (1) independently scheduling users in each sector through the principle of proportional fair scheduling; (2) according to the type of scheduled users: users in the sector or The sector edge user selects the cooperative sector; (3) each sector broadcasts its own scheduling information to the cooperative cell after the independent scheduling is completed and the cooperative sector is determined; (4) according to the scheduling information of its own sector and For the scheduling information of other cells, the scheduling conflict between cells is handled through the interference coordination mechanism of random backoff, and the scheduling coordination between cells is performed. Through this random backoff coordinated scheduling technology, the distributed processing of coordinated scheduling is realized, which can be realized in the existing network architecture, and the performance of cell edge users is greatly improved.

Description

A Distributed Coordinated Scheduling Method with Random Backoff

technical field

The invention belongs to the technical field of wireless communication, in particular to a distributed coordination scheduling technology of random backoff.

Background technique

In order to make full use of limited spectrum resources, a full frequency reuse scheme is adopted. However, when the network adopts full frequency reuse, the system is in a state of limited interference, especially for cell edge users. In order to eliminate the interference of users and improve the spectrum efficiency of the cell and the spectrum efficiency of the users at the edge of the cell, a coordinated scheduling technique is proposed. That is, multiple cells work together to eliminate interference to other cells. Since the coordinated scheduling technology requires joint processing and joint optimization between cells, a large amount of information interaction between cells is required, and a central node is required to handle coordination and other issues, which is difficult to achieve under the existing network architecture.

Contents of the invention

The purpose of the present invention is to overcome the problem that the existing coordinated scheduling technology needs global optimization, in order to realize the coordinated scheduling technology under the existing network architecture, overcome the large amount of information interaction between the cells brought about by the coordinated scheduling technology, and the whole network For the optimization problem, a distributed coordinated scheduling method based on random backoff is proposed, which improves the spectral efficiency of cell edge users while ensuring the average spectral efficiency of the cell.

To achieve the above object, the present invention adopts the following technical solutions:

(1) Each site uses the principle of proportional fair scheduling to independently schedule users in each sector of the cell, and each resource block in the same sector can only be allocated to one user;

(2) Divide independently scheduled users into sector center users and sector edge users, and use coordinated scheduling technology to select the cooperative sector required for coordinated scheduling for sector edge users;

(3) After the cooperative sector is determined, the cell broadcasts the shared cooperation information to the cell where the coordinated sector is located;

(4) After each station receives the cooperation information shared by the neighboring cells, if the neighboring cells have a cooperation request for the sector on the resource block, then determine whether the scheduled user on the resource block is a sector-edge user, if yes Users at the edge of the sector will back off with random probability, and users at the center of the sector will back off from the sector.

The concrete method of described step (1) is as follows:

For any time t, any cell n, the resource block i of the cell at this time is allocated to user k ^* , k ^* is determined by the following formula:

${\mathrm{<span\; class="notranslate">\; k</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; arg</span>}\mathrm{<span\; class="notranslate">\; max</span>}}\frac{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; k\; n</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}$

, _rkn (t) represents the maximum transmission rate of user k on cell n, T _k represents the total historical throughput of user k, after resource block allocation at each moment, the historical throughput is calculated according to T _k =T _k +r _kn (t) update, repeat the above process until all sector resources are allocated, use I _s (k, i) to represent the resource allocation indicator factor of sector s, and I _s (k, i)=1 to represent the sector s i resource blocks are allocated to user k, and I _s (k, i)=0 means that the i th resource block of sector s is not allocated to user k.

The concrete method of described step (2) is as follows:

(2-1) Classification rules for user categories

When user k satisfies , it is determined that user k is a sector edge user, when user k satisfies When , it is determined that user k is the sector center user, Indicates the average received signal-to-interference-noise ratio of user k, and γ _thr is the decision threshold of edge users, which can be determined by the proportion of sector edge users;

(2-2) Selection of collaboration type

Classify the user k ^* selected in step (1) according to the division rules in (2-1). If user k ^* is a sector center user, a non-cooperative transmission scheme is adopted, and only the local station transmits data for it. If user k ^* is a sector edge user, the coordinated scheduling transmission scheme is adopted;

(2-3) Selection of cooperative sectors in coordinated scheduling

If user k ^* adopts the coordinated scheduling transmission scheme, then user k ^* selects the two sectors with the strongest interference signals in all cells as the cooperative sectors in coordinated scheduling

The concrete method of described step (3) is as follows:

make Indicates that the cooperative sector in cell b includes the sector set of users, if Indicates that site b needs a sector on the i-th resource block cooperation, if Indicates that site b does not need sectors on the i-th resource block Collaboration, Sector The site is site b to site The way to share collaborative information is to send to site

The concrete method of described step (4) is as follows:

i. Initialization:

make i=1, then enter step ii, Indicates the site All received against sector shared collaboration information, Indicates all required sectors Collaborative collection of sites;

ii. Judging whether there is an adjacent site on the resource block i for the sector There is a collaboration request:

like Indicates that there is a need for collaboration, go to step iii, if Indicates that there is no need for collaboration, go to step v;

iii. Judging sectors Whether user k ^* scheduled on resource block i is a sector edge user:

If it is a sector edge user, go to step iv; if it is a sector center user, make user k ^* slave sector Back off block i on the resources to reduce interference, enter step v;

iv. Cell edge users use The probability backoff of :

Generate a (0, 1) uniformly distributed random number tempVal, if order user k ^* slave sector Back off on the resource block i of , go to step v; if Go directly to step v; P _thr represents the probability that the user backs off from the corresponding resource block when there is one collaboration demand;

v. Make i=i+1, if i≤N, then return to step ii to coordinate the next resource block;

If i>N, end sector Scheduling coordination, N represents the total number of resource blocks.

The present invention has the following beneficial effects: firstly, the present invention independently schedules each cell, selects the scheduled user, and determines its cooperative cell by the type of the scheduled user. Then broadcast the scheduling information and request cooperation information, and finally deal with the conflict of scheduling information between cells through the back-off mechanism. Avoiding the introduction of central nodes can be realized under the existing network architecture.

Specific implementation method

The present invention will be further described below in conjunction with embodiment.

The core idea of the present invention is: due to full frequency multiplexing, users at the edge of the cell are severely interfered and have poor performance, so a coordinated scheduling technique is proposed. However, the realization of coordinated scheduling technology requires global optimization and central node scheduling, which is difficult to achieve under the existing framework. In order to overcome the above problems, we propose a distributed coordinated scheduling technology with random backoff. Each cell independently schedules users in the cell, and then broadcasts its own scheduling information and coordination requests to surrounding cells. According to its own scheduling information and the scheduling information of other cells, if a scheduling conflict occurs, the cell will use an adaptive random backoff technology to complete the conflict coordination. Avoid using a central node so that each cell can complete its scheduling independently.

Example

User MIMO cell, the cell includes B cells (one cell corresponds to one site), each cell is divided into 3 sectors, each sector distributes the same users, the coordinated scheduling technology is used between the cells to eliminate the difference between the cells interfering. Include the following steps:

(1) Independent scheduling within the sector;

(2) Selection of the cooperative sector;

(3) Scheduling information sharing among cooperative cells;

(4) Scheduling and coordination between cells;

Specific steps are as follows:

(1) Independent scheduling within the sector, the specific steps are as follows:

Each site independently schedules the users in the cell, and the scheduling scheme is proportional fair scheduling. The scheme is as follows.

Intra-cell proportional fair scheduling strategy: For any time t and any cell n, the resource block i of the cell at this time is allocated to user k ^* , and k ^* is determined by the following formula:

${\mathrm{<span\; class="notranslate">\; k</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; arg</span>}\mathrm{<span\; class="notranslate">\; max</span>}}\frac{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; k\; n</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}$

, r _kn (t) represents the maximum transmission rate of user k in cell n, and T _k represents the total historical throughput of user k. After the allocation of resource blocks at each moment, the historical throughput needs to be updated, that is, T _k =T _k + r _kn (t). In order to ensure the orthogonality in the cell, each resource block in the same sector can only be allocated to one user. Repeat the above process until all resources (three sectors) are allocated. May wish to use I _s (k, i) to represent the resource allocation indicator factor of sector s, if I _s (k, i) = 1, it means that the i-th resource block of sector s is allocated to user k, otherwise I _s (k , i)=0 means not assigned to the user.

(2) The selection of the cooperation sector, the specific steps are as follows:

Step 1: Classify the users k ^* selected by proportional fair scheduling in the cell

In the present invention, there are two types of users: intra-sector users and sector-edge users. The user mainly determines its type according to the average received SINR of the pilot signal. Suppose the average received SINR of user k is When user k satisfies , it is determined as a cell edge user, otherwise it is determined as an intra-sector user. Where γ _thr is the edge user decision threshold.

Step 2: Different types of users adopt different transmission schemes

If the user is an intra-sector user, a non-cooperative transmission scheme is adopted, and only the serving node transmits data for it. If user k is a sector-edge user, the user selects the two sectors with the strongest received signals in all cells to join its coordination set.

(3) Scheduling information sharing between cooperative cells, the specific steps are as follows:

After the scheduling is completed, each station will send the resource allocation plan of the corresponding cooperative user in the cell to the station where the cooperative sector selected by the local user is located. use Indicates that the cooperation set in cell b contains sectors The set of all users of , let Obviously, if It means that station b needs a sector on the i-th resource block cooperation, otherwise Indicates that there is no collaboration requirement. Sector The site is Then site b and site The way to share scheduling information is to send to site

(4) Scheduling and coordination between cells, the specific steps are as follows:

After each site receives the scheduling information shared by neighboring cells, it coordinates the scheduling of its own cell based on it, so as to realize cooperation. by site sector As an example, after receiving all Shared scheduling information for in Indicates all required sectors After coordinating sites), the scheduling scheme in this sector is coordinated according to the random backoff method, and the specific steps are as follows:

i. Initialization: command Go to step ii.

ii. Judging whether there is an adjacent site on the resource block n sector There is a collaboration request: if If there is a need for collaboration, go to step iii; otherwise, go to step v.

iii. Judging sectors Whether user k ^* scheduled on resource block n is a cell-edge user: if it is a cell-edge user, go to step iv; otherwise, user k ^* retreats from the resource to reduce interference, that is, Go to step v.

iv. Cell edge users use Probabilistic backoff: Generate a (0, 1) uniformly distributed random number tempVal, if Then user k ^* backs off from the resource, that is, Go to step v; otherwise, go directly to step v.

v. Let n=n+1, if n≤N, go back to step ii to coordinate the next resource block; otherwise, end the scheduling coordination of this sector.

In the above algorithm, we adopt a random backoff mechanism for cell edge users, where _Pthr represents the probability of users backing off from the corresponding resource block when there is one coordination demand (its value can be determined empirically). According to the above algorithm, if there are multiple nodes requesting cooperation on this resource block, the probability that users in the cooperative sector will back off from this resource block increases, which is reasonable, because the transmission on this resource block will give multiple peers Users in adjacent cells cause serious interference, so the gain brought by backoff is more obvious. The purpose of adopting the random backoff mechanism for the cell edge users is to prevent both cooperating parties from backing off at the same time, resulting in waste of resources.

Claims (4)

1. A distributed coordinated scheduling method of random backoff, characterized in that, comprising the following steps: (1) Each site uses the principle of proportional fair scheduling to independently schedule users in each sector of the cell, and each resource block in the same sector can only be allocated to one user; (2) Divide independently scheduled users into sector center users and sector edge users, and use coordinated scheduling technology to select the cooperative sector required for coordinated scheduling for sector edge users; (3) After the cooperative sector is determined, the cell broadcasts the shared cooperation information to the cell where the coordinated sector is located; (4) After each station receives the cooperation information shared by the neighboring cells, if the neighboring cells have a cooperation request for the sector on the resource block, then determine whether the scheduled user on the resource block is a sector-edge user, if yes Users at the edge of the sector will back off with random probability, and users at the center of the sector will back off from the sector; The concrete method of described step (4) is as follows: i. Initialization: make i=1, then enter step ii, Indicates the site All received against sector shared collaboration information, Indicates all required sectors Collaborative collection of sites; ii. Judging whether there is an adjacent site on the resource block i for the sector There is a collaboration request: if Indicates that there is a need for collaboration, go to step iii, if Indicates that there is no need for collaboration, go to step v; iii. Judging sectors Whether user k ^* scheduled on resource block i is a sector edge user: If it is a sector edge user, go to step iv; if it is a sector center user, make user k ^* slave sector Back off on resource block i to reduce interference, enter step v; iv. Cell edge users use The probability of backoff: Generate a (0,1) uniformly distributed random number tempVal, if order user k ^* slave sector Back off on the resource block i of , go to step v; if Go directly to step v; P _thr represents the probability that the user backs off from the corresponding resource block when there is one coordination demand; v. Let i=i+1, if i≤N, return to step ii to coordinate the next resource block; if i>N, end the sector Scheduling coordination, N represents the total number of resource blocks. 2. according to the described distributed coordinated scheduling method of a kind of random backoff of claim 1, it is characterized in that: the concrete method of described step (1) is as follows: For any time t, any cell n, the resource block i of the cell at this time is allocated to user k ^* , k ^* is determined by the following formula: ${\mathrm{<span\; class="notranslate">\; k</span>}}^{<span\; class="notranslate">\; *</span>}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; k</span>}}{\mathrm{<span\; class="notranslate">\; arg</span>}\mathrm{<span\; class="notranslate">\; max</span>}}\frac{{\mathrm{<span\; class="notranslate">\; r</span>}}_{\mathrm{<span\; class="notranslate">\; k\; n</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; t</span>}<span\; class="notranslate">\; )</span>}{{\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; k</span>}}}$ , r _kn (t) represents the maximum transmission rate of user k on cell n, and T _k represents the total historical throughput of user k. After the allocation of resource blocks at each moment, the historical total throughput is calculated according to T _k =T _k +r _kn (t) is updated, repeat the above process until all sector resources are allocated, use I _s (k,i) to represent the resource allocation indicator factor of sector s, and I _s (k,i)=1 to represent the resource allocation indicator factor of sector s The i-th resource block is allocated to user k, and I _s (k,i)=0 indicates that the i-th resource block of sector s is not allocated to user k. 3. according to the described distributed coordinated scheduling method of a kind of random backoff of claim 1, it is characterized in that: the concrete method of described step (2) is as follows: (2-1) Classification rules for user categories When user k satisfies , it is determined that user k is a sector edge user, when user k satisfies When , it is determined that user k is the sector center user, Indicates the average received signal-to-interference-noise ratio of user k, and γ _thr is the decision threshold of edge users; (2-2) Selection of collaboration type Classify the user k ^* selected in step (1) according to the division rules in (2-1). If user k ^* is a sector center user, a non-cooperative transmission scheme is adopted, and only the local station transmits data for it. If user k ^* is a sector edge user, the coordinated scheduling transmission scheme is adopted; (2-3) Selection of cooperative sectors in coordinated scheduling If user k ^* adopts the coordinated scheduling transmission scheme, then user k ^* selects the two sectors with the strongest interference signals in all cells as the cooperative sectors in coordinated scheduling 4. according to the described distributed coordinated scheduling method of a kind of random backoff of claim 1, it is characterized in that: the concrete method of described step (3) is as follows: make Indicates that the cooperative sector in site b contains the sector set of users, if Indicates that site b needs a sector on the i-th resource block cooperation, if Indicates that site b does not need sectors on the i-th resource block Collaboration, Sector The site is site b to site The way to share collaborative information is to send to site