CN106851667B - A kind of data processing method and device for air protocol data surface - Google Patents

A kind of data processing method and device for air protocol data surface Download PDF

Info

Publication number
CN106851667B
CN106851667B CN201710044159.7A CN201710044159A CN106851667B CN 106851667 B CN106851667 B CN 106851667B CN 201710044159 A CN201710044159 A CN 201710044159A CN 106851667 B CN106851667 B CN 106851667B
Authority
CN
China
Prior art keywords
user data
hardware
threads
cell
number
Prior art date
Application number
CN201710044159.7A
Other languages
Chinese (zh)
Other versions
CN106851667A (en
Inventor
黄勇
吴治鸣
Original Assignee
京信通信系统(中国)有限公司
京信通信系统(广州)有限公司
京信通信技术(广州)有限公司
天津京信通信系统有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 京信通信系统(中国)有限公司, 京信通信系统(广州)有限公司, 京信通信技术(广州)有限公司, 天津京信通信系统有限公司 filed Critical 京信通信系统(中国)有限公司
Priority to CN201710044159.7A priority Critical patent/CN106851667B/en
Publication of CN106851667A publication Critical patent/CN106851667A/en
Application granted granted Critical
Publication of CN106851667B publication Critical patent/CN106851667B/en

Links

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/30Arrangements for executing machine instructions, e.g. instruction decode
    • G06F9/38Concurrent instruction execution, e.g. pipeline, look ahead
    • G06F9/3885Concurrent instruction execution, e.g. pipeline, look ahead using a plurality of independent parallel functional units
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5011Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals
    • G06F9/5016Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resources being hardware resources other than CPUs, Servers and Terminals the resource being the memory
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W16/00Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
    • H04W16/18Network planning tools
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W28/00Network traffic or resource management
    • H04W28/02Traffic management, e.g. flow control or congestion control
    • H04W28/0205Traffic management, e.g. flow control or congestion control at the air interface
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W72/00Local resource management, e.g. wireless traffic scheduling or selection or allocation of wireless resources
    • H04W72/12Dynamic Wireless traffic scheduling ; Dynamically scheduled allocation on shared channel
    • H04W72/1205Schedule definition, set-up or creation
    • H04W72/1252Schedule definition, set-up or creation based on load

Abstract

The invention discloses a kind of data processing method and device for air protocol data surface, number of cells N that the hardware core total number Y and the base station for obtaining base station need support;The relationship of the hardware core total number Y of the number of cells N and the base station that are needed support according to the base station determines the hardware nucleus number M1 for handling the scheduling of cell MAC layer;According to the hardware core total number Y of the base station and the hardware nucleus number M1 dispatched for handling cell media access control MAC layer, the hardware nucleus number M2 for being used for processes user data is determined, wherein M1+M2≤Y.By the way that hardware core sum is carried out reasonable distribution, it is specifically divided into the hardware core of processing cell MAC layer scheduling, quantity M1;And the hardware core of processes user data, quantity M2.Using multi-core processor classification parallel processing achieve the purpose that meet LTE air protocol stack data surface height handle up, multiple cell, multi-user.

Description

A kind of data processing method and device for air protocol data surface

Technical field

The present invention relates to LTE (Long Term Evolution, long term evolution) technical fields more particularly to one kind to be directed to The data processing method and device of air protocol data surface.

Background technique

Standard technique one of of the LTE as forth generation mobile communication is that one kind is designed exclusively for the high broadband application of movement Wireless communication standard.Wireless interface can be divided into three protocol layers: physical layer L1, data link layer L2 and network layer L3, such as Fig. 1 It is shown.

The data surface L2 that LTE wirelessly communicates air protocol includes PDCP (Packet Data Convergence Protocol, Packet Data Convergence Protocol) layer, RLC (Radio Link Control, radio link control) layer and MAC (Media Access Control, media access control protocol) layer.It is PDCP layers of responsible data header compression/de-compression, complete Property protection, encryption/decryption, PDCP SN (Serial Number, sequence number) is safeguarded, is sequentially submitted, switch data is switched and is determined When abandon etc. functions;Rlc protocol layer is responsible for TM (Transparent Mode, transparent transmission mode), UM (Unacknowledged Mode, Unacknowledged Mode) and AM (Acknowledged Mode, affirmation mode) Three models data transmission, data point Section, cascade, recombination, the again functions such as segmentation and ARQ (Automatic Repeat Request, automatic repeat request);MAC protocol The responsible logic channel of layer and transmission channel mapping, logic channel multiplexing and demultiplexing, HARQ (Hybrid Automatic Repeat Request, hybrid automatic repeat-request), dynamic dispatching, semi-persistent scheduling and transformat selection etc. functions.

In the data surface flow chart of data processing of existing LTE air protocol, for downlink data transmission, PDCP is to coming from The data packet on upper layer increases PDCP header information, carries out head compression and is sent to RLC after encrypting;RLC is to from the wireless of PDCP Link control layers service data unit is segmented according to the scheduling designated length of MAC layer, is cascaded, filled processing and increase corresponding RLC head portion information, formed protocol data unit of wireless links control layer;Then MAC layer will be on Different Logic channel Data-reusing is to transmission channel, and multiplexing process is based on one group of set TB (Transport Block, transmission block) size and transmission Format is related to concatenating, fills, increases corresponding MAC header information and MAC control information formation MAC PDU (Protocol Data Unit, protocol Data Unit) operation;The TB formed after MAC is handled is transmitted to bottom, wirelessly sends out by eating dishes without rice or wine It is sent to terminal.

In the data surface flow chart of data processing of existing LTE air protocol, for transmitting uplink data, MAC layer will be upper The data that row receives carry out MAC header, demultiplex, send RLC for the data after demultiplexing, while will be after demultiplexing MAC control information is sent to MAC scheduling;RLC carries out the data received RLC head portion, to be sent to PDCP after recombination;PDCP The data received are decrypted, head compression is solved and goes the head PDCP, data packet to be sequentially then delivered to upper layer.

Under single carrier case, for LTE message transmission rate uplink up to 50Mbps, downlink is poly- in carrier wave up to 100Mbps In the case of conjunction, rate is even more according to the variable number of polymerization into multiple increase, while as intelligent terminal sharply increases, It is required that the number of users that base station is supported is improved in the order of magnitude.However, the data plane architecture base of the 2G and 3G air protocol of the prior art In the design architecture of deployment software single thread or multithreading on single or multiple core hardware processor, it is difficult to meet LTE high-speed The requirement of performance.

In conclusion the data plane architecture of existing air protocol be unable to satisfy it is high handle up, multiple cell, multi-user LTE The requirement of base station equipment data surface.

Summary of the invention

The present invention provides a kind of data processing method and device for air protocol data surface, to solve the prior art The data plane architecture of middle air protocol be unable to satisfy it is high handle up, multiple cell, multi-user asking of requiring of LTE base station device data face Topic.

The embodiment of the present invention provides a kind of data processing method for air protocol data surface, comprising:

The number of cells N that the hardware core total number Y and the base station for obtaining base station need support;

The relationship of the hardware core total number Y of the number of cells N and the base station that are needed support according to the base station, determination are used for Handle the hardware nucleus number M1 of cell MAC layer scheduling;

According to the hardware core total number Y of the base station and it is described for handle cell media access control MAC layer scheduling it is hard Part nucleus number M1 determines the hardware nucleus number M2 for being used for processes user data, wherein M1+M2≤Y.

Preferably, the relationship of the hardware core total number Y of the number of cells N and the base station that are needed support according to the base station, Determine the hardware nucleus number M1 for handling the scheduling of cell MAC layer, comprising:

A hardware core is occupied according to each cell and carries out the scheduling of cell MAC layer, and each cell occupies at a hardware core The principle for managing user data, determines whether Y is more than or equal to 2N;

In Y >=2N, determines that the hardware nucleus number M1 for handling the scheduling of cell MAC layer is N, determine for handling number of users According to hardware nucleus number M2=Y-N.

Preferably, the method also includes:

In Y < 2N, the hardware core for handling the scheduling of cell MAC layer according to one is the principle of two cell serves, is determined Hardware nucleus number M1, M1 for handling the scheduling of cell MAC layer are the integer not less than N/2;And it determines and is used for processes user data Hardware nucleus number M2=Y-M1.

Preferably, after the determination is for the hardware nucleus number M2 of processes user data, further includes:

Determine the cell scheduling sets of threads and user data sets of threads of cell in the base station;

Cell scheduling sets of threads is deployed on M1 hardware core of processing cell MAC layer scheduling, by user data thread Group is deployed on M2 hardware core of processes user data;

The deployment of the user of the user data sets of threads is accessed according to the load adjustment of the hardware core of processes user data.

Preferably, the load adjustment of the hardware core according to processes user data accesses the user data sets of threads The deployment of user, comprising:

According to the number of users of each user data sets of threads record access, by the least user data of the number of users accessed User data sets of threads of the sets of threads as new access user;

Or,

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, CPU is put down User data sets of threads of the user data sets of threads disposed on the smallest hardware core of equal load as new access user;

Or,

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, if maximum The difference of CPU average load and minimum CPU average load is more than given threshold, then by the hardware core of the maximum CPU average load The user of the user data sets of threads of upper deployment adjusts to the user data disposed on the hardware core of the minimum CPU average load Sets of threads.

The embodiment of the present invention also provides a kind of data processing equipment for air protocol data surface, comprising:

Acquiring unit: the number of cells N that hardware core total number Y and the base station for obtaining base station need support;

First determination unit: the hardware core of number of cells N and the base station for being needed support according to the base station are total The relationship of number Y, determines the hardware nucleus number M1 for handling the scheduling of cell MAC layer;

Second determination unit: for according to the hardware core total number Y of the base station and described for handling cell media interviews The hardware nucleus number M1 of MAC layer scheduling is controlled, determines the hardware nucleus number M2 for being used for processes user data, wherein M1+M2≤Y.

Preferably, the first determination unit, is specifically used for:

A hardware core is occupied according to each cell and carries out the scheduling of cell MAC layer, and each cell occupies at a hardware core The principle for managing user data, determines whether Y is more than or equal to 2N;

In Y >=2N, determine that the hardware nucleus number M1 for handling the scheduling of cell MAC layer is N;Second determination unit, which determines, to be used In the hardware nucleus number M2=Y-N of processes user data.

Preferably, first determination unit is also used to:

In Y < 2N, the hardware core for handling the scheduling of cell MAC layer according to one is the principle of two cell serves, is determined Hardware nucleus number M1, M1 for handling the scheduling of cell MAC layer are the integer not less than N/2;Second determination unit is determined for locating Manage the hardware nucleus number M2=Y-M1 of user data.

Preferably, described device further includes adjustment unit, it is used for:

Determine the cell scheduling sets of threads and user data sets of threads of cell in the base station;

Cell scheduling sets of threads is deployed on M1 hardware core of processing cell MAC layer scheduling, by user data thread Group is deployed on M2 hardware core of processes user data;

The deployment of the user of the user data sets of threads is accessed according to the load adjustment of the hardware core of processes user data.

Preferably, the adjustment unit, is also used to:

According to the number of users of each user data sets of threads record access, by the least user data of the number of users accessed User data sets of threads of the sets of threads as new access user;

Or,

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, CPU is put down User data sets of threads of the user data sets of threads disposed on the smallest hardware core of equal load as new access user;

Or,

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, if maximum The difference of CPU average load and minimum CPU average load is more than given threshold, then by the hardware core of the maximum CPU average load The user of the user data sets of threads of upper deployment adjusts to the user data disposed on the hardware core of the minimum CPU average load Sets of threads.

The embodiment of the present invention provides a kind of data processing method and device for air protocol data surface, obtains base station The number of cells N that hardware core total number Y and the base station need support;The number of cells N needed support according to the base station and institute The relationship of the hardware core total number Y of base station is stated, determines the hardware nucleus number M1 for handling the scheduling of cell MAC layer;According to the base station Hardware core total number Y and it is described for handle cell media access control MAC layer scheduling hardware nucleus number M1, determine for handling The hardware nucleus number M2 of user data, wherein M1+M2≤Y.By the way that hardware core sum is carried out reasonable distribution, it is specifically divided into processing The hardware core of cell MAC layer scheduling, quantity M1;And the hardware core of processes user data, quantity M2.It is handled using multicore Device classification parallel processing achieve the purpose that meet LTE air protocol stack data surface height handle up, multiple cell, multi-user.

Detailed description of the invention

To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment Attached drawing is briefly introduced, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this For the those of ordinary skill in field, without any creative labor, it can also be obtained according to these attached drawings His attached drawing.

Fig. 1 is the data surface that LTE provided in an embodiment of the present invention wirelessly communicates air protocol;

Fig. 2 is a kind of data surface software architecture schematic diagram of LTE air protocol provided in an embodiment of the present invention;

Fig. 3 is that a kind of data processing method process for air protocol data surface provided in an embodiment of the present invention is illustrated Figure;

Fig. 4 is a kind of data processing method flow chart for air protocol data surface provided in an embodiment of the present invention;

Fig. 5 is a kind of data processing equipment structural representation for air protocol data surface provided in an embodiment of the present invention Figure.

Specific embodiment

To make the objectives, technical solutions, and advantages of the present invention clearer, below in conjunction with attached drawing to the present invention make into It is described in detail to one step, it is clear that the described embodiments are only some of the embodiments of the present invention, rather than whole implementation Example.Based on the embodiments of the present invention, obtained by those of ordinary skill in the art without making creative efforts All other embodiment, shall fall within the protection scope of the present invention.

Meaning of embodiment of the present invention base station, that is, public mobile communication base station, is a kind of form of radio station, refers to one In fixed area of radio coverage, by carrying out the wireless of information transmitting between mobile switching center, with mobile telephone terminal Electricity transmitting-receiving letter radio station.

The relational language in the embodiment of the present invention is explained below.

PDCP_UL THD (PDCP Up Link Thread, PDCP uplink handle thread): be responsible for decryption, the compression of solution head, It removes PDCP header information, orderly submit the functions such as data packet to upper layer (network layer).

PDCP_DL THD (PDCP Down Link Thread, PDCP downlink processing thread): be responsible for encryption, head compression, Increase PDCP header information, timing abandons, submits the functions such as data packet to lower layer (RLC).

RLC_UL THD (RLC Up Link Thread, RLC uplink handle thread): be responsible for cascading, recombinate, reordering, Uplink ARQ, the functions such as RLC head portion information are gone.

RLC_DL THD (RLC Down Link Thread, RLC downlink processing thread): it is responsible for segmentation, filling, divides again Section, increases the functions such as RLC head portion information at downlink ARQ.

MAC_UL THD (MAC Up Link Thread, MAC uplink handles thread): it is responsible for the data solution of transmission channel It is multiplexed into logic channel, goes the functions such as MAC header information.

MAC_DL THD (MAC Down Link Thread, MAC downlink processing thread): it is responsible for the data of logic channel It is multiplexed into transmission channel, filling increases the functions such as MAC header information.

MAC_SCH_UL THD (MAC Schedule Up Link Thread, MAC dispatching uplink handles thread): it is responsible for The functions such as uplink MAC scheduling, uplink authorization.

MAC_SCH_DL THD (MAC Schedule Down Link Thread, MAC dispatching downlink handles thread): negative Blame the functions such as downlink MAC scheduling feature, downstream grants.

User data sets of threads includes PDCP uplink processing thread PDCP_UL THD, PDCP downlink in the embodiment of the present invention Handle thread PDCP_DL THD, RLC uplink handles thread RLC_UL THD, on RLC downlink processing thread RLC_DL THD, MAC Row processing thread MAC_UL THD, MAC downlink processing thread MAC_DL THD, cell scheduling sets of threads include uplink MAC scheduling Thread MAC_SCH_UL THD and downlink MAC scheduling thread MAC_SCH_DL THD.

It is thin-skinned that data processing method and device provided in an embodiment of the present invention for air protocol data surface is based on data Part framework is illustrated in figure 2 the data surface software architecture schematic diagram the present invention provides a kind of LTE air protocol, processor Available core sum is S1+S2, and one sets of threads of a kernel processes, S1 kernel is for handling S1 user data thread Group, S2 kernel is for handling S2 cell scheduling sets of threads.By by cell scheduling sets of threads and user data thread component From being reached using the multi-core processor reasonable distribution parallel processing two classes sets of threads and meet LTE air protocol stack data surface height and gulp down Spit, multiple cell, multi-user purpose.

The embodiment of the present invention provides a kind of data processing method for air protocol data surface.As shown in figure 3, being this hair A kind of data processing method flow diagram for air protocol data surface that bright embodiment provides, comprising:

Step 301: the number of cells N that the hardware core total number Y and base station for obtaining base station need support.

Specifically, the quantity for inquiring the processor cores of base station obtains hardware core total number Y.Inquire the cell that base station is supported Number, obtains number of cells N.

Step 302: the relationship of the hardware core total number Y of the number of cells N and base station that are needed support according to base station, determination are used for Handle the hardware nucleus number M1 of cell MAC layer scheduling;

Step 303: the hardware nucleus number M1 according to the hardware core total number Y of base station and for handling the scheduling of cell MAC layer is determined Hardware nucleus number M2 for processes user data, wherein M1+M2≤Y.

Specifically, in step 302, a hardware core is occupied according to each cell and carries out the scheduling of cell MAC layer, and is each Cell occupies the principle of a hardware core processes user data, determines whether Y is more than or equal to 2N.At this point, for each hardware core Load balancing considers that each cell scheduling sets of threads individually handles user's scheduling an of cell, a cell configuration one first A user data sets of threads carrys out processes user data.

In Y >=2N, determines that the hardware nucleus number M1 for handling the scheduling of cell MAC layer is N, determine for handling number of users According to hardware nucleus number M2=Y-N.

At this point, cell scheduling sets of threads needs to occupy N number of hardware core, remaining hardware core can be used for processes user data, Due to Y-N >=N, i.e., remaining hardware nucleus number is greater than or equal to cell number, thus remaining hardware core can completely or partially be used to dispose User data sets of threads can satisfy a cell in the case and configure at least one user data sets of threads to handle user Data.

For example, hardware core total number Y=8 of base station, the number of cells N=3 of base station, due to Y > 2N, then for handling cell The hardware nucleus number M1=3, the hardware nucleus number M2=5 for processes user data of MAC layer scheduling.I.e. cell scheduling sets of threads needs 3 hardware cores are occupied, remaining 5 hardware cores are used for processes user data.Wherein, 5 hardware cores are not limited to all or part of place User data sets of threads is managed, the number of users that can according to need support is disposed accordingly.Preferably, a cell is at least matched Set a hardware core processes user data sets of threads.

In Y < 2N, the hardware core for handling the scheduling of cell MAC layer according to one is the principle of two cell serves, is determined Hardware nucleus number M1, M1 for handling the scheduling of cell MAC layer are the integer not less than N/2;And it determines and is used for processes user data Hardware nucleus number M2=Y-M1.

At this point, a cell scheduling sets of threads handles user's scheduling of two cells, N/2 hardware core is at least needed to come portion Cell scheduling sets of threads is affixed one's name to, remaining hardware core is for disposing user data sets of threads.

Specifically, it when N/2 is not integer, rounds up.For example, cell number N=3, then N/2=1.5, this When M1 need to round up, i.e. M1=2.

It should be noted that in the case where hardware core total number Y is less, the hardware core of a processing cell MAC layer scheduling Or more than two cell serves, i.e., the use of more than two cells can also be handled by a cell scheduling sets of threads Family scheduling, herein with no restrictions.

As N≤Y-M1, i.e., remaining hardware nucleus number is greater than or equal to cell number, and remaining hardware core can be all or part of at this time For disposing user data sets of threads, it can satisfy a cell in the case and configure at least one user data sets of threads Processes user data.

For example, hardware core total number Y=7 of base station, the number of cells N=4 of base station, due to Y < 2N, then using a processing The hardware core of cell MAC layer scheduling is two cell serves, i.e. user's tune that a cell scheduling sets of threads handles two cells Degree, at this time for handling the hardware nucleus number M1=2, the hardware nucleus number M2=5 for processes user data of the scheduling of cell MAC layer. Due to N < M2, remaining 5 hardware cores are used to handle the user data of 4 cells.

For another example, hardware core total number Y=9 of base station, the number of cells N=5 of base station, due to Y < 2N, then using a processing The hardware core of cell MAC layer scheduling is two cell serves, i.e. user's tune that a cell scheduling sets of threads handles two cells Degree is at this time 3 for handling the hardware nucleus number M1 of cell MAC layer scheduling and rounding up, the hardware core for processes user data Number M2=6.Due to N < M2, remaining 6 hardware cores are used to handle the user data of 5 cells.

As N > Y-M1, i.e., remaining hardware nucleus number is less than cell number, and remaining hardware core can completely or partially be used for portion at this time User data sets of threads is affixed one's name to, a user data sets of threads handles the user data of at least one cell at this time.

For example, hardware core total number Y=5 of base station, the number of cells N=4 of base station, due to Y < 2N, then using a processing The hardware core of cell MAC layer scheduling is two cell serves, i.e. user's tune that a cell scheduling sets of threads handles two cells Degree, at this time for handling the hardware nucleus number M1=2, the hardware nucleus number M2=3 for processes user data of the scheduling of cell MAC layer. Due to N > M2, remaining 3 hardware cores are used to handle the user data of 4 cells.

It should be noted that working as N > Y-M1, and when N/2 is not integer, M1 can also be rounded downwards, not done herein Limitation.For example, hardware core total number Y=6 of base station, number of cells N=5, may also set up for handling the hard of cell MAC layer scheduling It is 2 that part nucleus number M1, which is rounded downwards, then is used for the hardware nucleus number M2=4 of processes user data.Due to N > M2, remaining 4 hardware cores For handling the user data of 5 cells.

It should be noted that the number of users that base station needs support is up to a hundred or thousands of orders of magnitude, therefore in LTE system It needs reasonable distribution user to each user data sets of threads, just can guarantee each hardware core load balancing.Therefore, after step 303, Also need the deployment of the user of the load adjustment access user data sets of threads according to the hardware core of processes user data.Based on needle It is each the embodiment of the invention also provides guaranteeing in the case where number of users is big to the data processing method of air protocol data surface Three kinds of methods of hardware core load balancing.

Method one: according to the number of users of each user data sets of threads record access, the number of users accessed is least User data sets of threads of the user data sets of threads as new access user.

Specifically, the number of users S that each user data sets of threads record has accessed, each user data sets of threads access one S adds one when a new user, and S subtracts one when discharging a user.When accessing new user every time, selected in M2 user data sets of threads Select user data thread of the least user data sets of threads of accessed user's number S of record as new access user.The method Relatively easy, the number of users of Yi Shixian, each user data sets of threads processing are essentially identical, but since each customer service is different, Data volume size is also just different, and there may be certain lacks of uniformity for the load of each hardware core.

Method two:

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, CPU is put down User data sets of threads of the user data sets of threads disposed on the smallest hardware core of equal load as new access user.

Specifically, each user data sets of threads records the CPU average treatment load of the affiliated hardware core in T seconds durations. When new user access, select the user data sets of threads disposed on the smallest hardware core of CPU average treatment load as newly connecing The user data sets of threads of access customer, the method relative manner one realize complexity, but can make each hardware core load balancing substantially.

Method three:

The CPU average treatment load of the affiliated hardware core in setting duration is recorded according to each user data sets of threads, if The difference of maximum CPU average load and minimum CPU average load is more than given threshold, then by the hardware core of maximum CPU average load The user of the user data sets of threads of upper deployment adjusts to the user data disposed on the hardware core of the minimum CPU average load Sets of threads.

User data sets of threads belonging to the above method one and method two are determining when being all new user access, however user Business datum amount size is dynamic change, although user's each hardware core load in access be it is balanced, with each user The variation of business datum amount size, hardware core load is also dynamic change.Therefore user belonging to dynamic adjustment user is needed Data thread group.

Specifically, the real-time CPU average treatment that each user data sets of threads records the affiliated hardware core in T seconds durations is born Lotus.It, will be on the hardware core of maximum CPU average load if the difference of maximum cpu load and minimum cpu load is more than given threshold H The user of the user data sets of threads of deployment adjusts to the user data line disposed on the hardware core of the minimum CPU average load Then journey group repeats dynamic adjustment process as above, so that the real-time CPU of the affiliated hardware core of each user data sets of threads is handled The difference of load is no more than given threshold.The method is complex, but can make the load real time equaliser of each hardware core.

The embodiment of the present invention provides a kind of data processing method for air protocol data surface, by the way that hardware core is total Reasonable distribution is carried out, the hardware core of processing cell MAC layer scheduling, quantity M1 are specifically divided into;And processes user data is hard Part core, quantity M2.Reached using two class sets of threads of multi-core processor reasonable distribution parallel processing and meets LTE air protocol stack number Handle up according to face height, multiple cell, multi-user purpose, while making each hardware core load balancing, improve hardware resource utilization and be System stability.

The embodiment of the present invention also provides a kind of data processing method for air protocol data surface, as shown in figure 4, For a kind of data processing method flow chart for air protocol data surface provided in an embodiment of the present invention, comprising:

Step 401: obtaining the hardware core total number Y of base station and the number of cells N of base station.

Step 402: whether true judging Y >=2N, if so, thening follow the steps 403, no person executes step 404.

Step 403: determining the hardware nucleus number M1=N for handling the scheduling of cell MAC layer, determine and be used for processes user data Hardware nucleus number M2=Y-N.

For example, hardware core total number Y=8 of base station, the number of cells N=3 of base station, due to Y > 2N, then for handling cell The hardware nucleus number M1=3, the hardware nucleus number M2=5 for processes user data of MAC layer scheduling.I.e. cell scheduling sets of threads needs 3 hardware cores are occupied, remaining 5 hardware cores are used for processes user data.

Step 404: determine that the hardware nucleus number for handling the scheduling of cell MAC layer is M1, M1 is the integer not less than N/2, Continue to execute step 405.

Step 405: judging whether N > Y-M1 is true, if so, thening follow the steps 406, no person executes step 407.

Step 406: one user data sets of threads of configuration handles the user data of at least one cell.

For example, hardware core total number Y=5 of base station, the number of cells N=4 of base station, due to Y < 2N, then using a processing The hardware core of cell MAC layer scheduling is two cell serves, i.e. user's tune that a cell scheduling sets of threads handles two cells Degree, at this time for handling the hardware nucleus number M1=2, the hardware nucleus number M2=3 for processes user data of the scheduling of cell MAC layer. Due to N > M2, remaining 3 hardware cores are used to handle the user data of 4 cells, then configure a user data sets of threads processing The user data of at least one cell

Step 407: configuring the user data that at least one user data sets of threads handles a cell.

For example, hardware core total number Y=7 of base station, the number of cells N=4 of base station, due to Y < 2N, then using a processing The hardware core of cell MAC layer scheduling is two cell serves, i.e. user's tune that a cell scheduling sets of threads handles two cells Degree, at this time for handling the hardware nucleus number M1=2, the hardware nucleus number M2=5 for processes user data of the scheduling of cell MAC layer. Due to N > M2, remaining 5 hardware cores are used to handle the user data of 4 cells, then configure at least one user data sets of threads Handle the user data of a cell.

Based on same inventive concept, the embodiment of the present invention also provides the data processing dress for air protocol data surface It sets, as shown in figure 5, be the data processing equipment structural diagram provided in an embodiment of the present invention for air protocol data surface, Include:

Acquiring unit 501: the number of cells N that hardware core total number Y and the base station for obtaining base station need support;

First determination unit 502: the hardware core of number of cells N and the base station for being needed support according to the base station The relationship of total number Y determines the hardware nucleus number M1 for handling the scheduling of cell MAC layer;

Second determination unit 503: for described visiting according to the hardware core total number Y of the base station and for handling cell media The hardware nucleus number M1 for asking control MAC layer scheduling, determines the hardware nucleus number M2 for being used for processes user data, wherein M1+M2≤Y.

Preferably, the first determination unit 502, is specifically used for:

A hardware core is occupied according to each cell and carries out the scheduling of cell MAC layer, and each cell occupies at a hardware core The principle for managing user data, determines whether Y is more than or equal to 2N;

In Y >=2N, determine that the hardware nucleus number M1 for handling the scheduling of cell MAC layer is N;Second determination unit, which determines, to be used In the hardware nucleus number M2=Y-N of processes user data.

Preferably, first determination unit 502 is also used to:

In Y < 2N, the hardware core for handling the scheduling of cell MAC layer according to one is the principle of two cell serves, is determined Hardware nucleus number M1, M1 for handling the scheduling of cell MAC layer are the integer not less than N/2;Second determination unit is determined for locating Manage the hardware nucleus number M2=Y-M1 of user data.

Preferably, described device further includes adjustment unit 504, it is used for:

Determine the cell scheduling sets of threads and user data sets of threads of cell in the base station;

Cell scheduling sets of threads is deployed on M1 hardware core of processing cell MAC layer scheduling, by user data thread Group is deployed on M2 hardware core of processes user data;

The deployment of the user of the user data sets of threads is accessed according to the load adjustment of the hardware core of processes user data.

Preferably, the adjustment unit 504, is also used to:

According to the number of users of each user data sets of threads record access, by the least user data of the number of users accessed User data sets of threads of the sets of threads as new access user;

Or,

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, CPU is put down User data sets of threads of the user data sets of threads disposed on the smallest hardware core of equal load as new access user;

Or,

The CPU average load that the affiliated hardware core in setting duration is recorded according to each user data sets of threads, if maximum The difference of CPU average load and minimum CPU average load is more than given threshold, then by the hardware core of the maximum CPU average load The user of the user data sets of threads of upper deployment adjusts to the user data disposed on the hardware core of the minimum CPU average load Sets of threads.

The embodiment of the present invention provides a kind of data processing equipment for air protocol data surface, obtains the hardware core of base station The number of cells N that total number Y and the base station need support;The number of cells N needed support according to the base station and the base station Hardware core total number Y relationship, determine for handle cell MAC layer scheduling hardware nucleus number M1;According to the hardware of the base station Core total number Y and the hardware nucleus number M1 dispatched for handling cell media access control MAC layer, determine for handling number of users According to hardware nucleus number M2, wherein M1+M2≤Y.By the way that hardware core sum is carried out reasonable distribution, it is specifically divided into processing cell MAC The hardware core of layer scheduling, quantity M1;And the hardware core of processes user data, quantity M2.It is reasonable using multi-core processor Distribution two class sets of threads of parallel processing achieve the purpose that meet LTE air protocol stack data surface height handle up, multiple cell, multi-user, Make each hardware core load balancing simultaneously, improves hardware resource utilization and system stability.

The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real The system for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.

These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates, Enable the manufacture of system, the instruction system realize in one box of one or more flows of the flowchart and/or block diagram or The function of being specified in multiple boxes.

These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one The step of function of being specified in a box or multiple boxes.

Although preferred embodiments of the present invention have been described, it is created once a person skilled in the art knows basic Property concept, then additional changes and modifications may be made to these embodiments.So it includes excellent that the following claims are intended to be interpreted as It selects embodiment and falls into all change and modification of the scope of the invention.

Obviously, various changes and modifications can be made to the invention without departing from essence of the invention by those skilled in the art Mind and range.In this way, if these modifications and changes of the present invention belongs to the range of the claims in the present invention and its equivalent technologies Within, then the present invention is also intended to include these modifications and variations.

Claims (10)

1. a kind of data processing method for air protocol data surface characterized by comprising
The number of cells N that the hardware core total number Y and the base station for obtaining base station need support;
The relationship of the hardware core total number Y of the number of cells N and the base station that are needed support according to the base station is determined for handling The hardware nucleus number M1 of cell MAC layer scheduling;
According to the hardware core total number Y of the base station and the hardware core dispatched for handling cell media access control MAC layer Number M1, determines the hardware nucleus number M2 for being used for processes user data, wherein M1+M2≤Y.
2. the method as described in claim 1, which is characterized in that the number of cells N that is needed support according to the base station and described The relationship of the hardware core total number Y of base station determines the hardware nucleus number M1 for handling the scheduling of cell MAC layer, comprising:
A hardware core is occupied according to each cell and carries out the scheduling of cell MAC layer, and each cell occupies a hardware core processing and uses The principle of user data, determines whether Y is more than or equal to 2N;
In Y >=2N, determines that the hardware nucleus number M1 for handling the scheduling of cell MAC layer is N, determine for processes user data Hardware nucleus number M2=Y-N.
3. method according to claim 2, which is characterized in that further include:
In Y < 2N, the hardware core for handling the scheduling of cell MAC layer according to one is the principle of two cell serves, and determination is used for Hardware the nucleus number M1, M1 for handling the scheduling of cell MAC layer are the integer not less than N/2;And it determines for the hard of processes user data Part nucleus number M2=Y-M1.
4. method as described in any one of claims 1 to 3, which is characterized in that the determination is hard for processes user data After part nucleus number M2, further includes:
Determine the cell scheduling sets of threads and user data sets of threads of cell in the base station;
Cell scheduling sets of threads is deployed on M1 hardware core of processing cell MAC layer scheduling, by user data sets of threads portion Administration is on M2 hardware core of processes user data;
The deployment of the user of the user data sets of threads is accessed according to the load adjustment of the hardware core of processes user data.
5. method as claimed in claim 4, which is characterized in that the load adjustment of the hardware core according to processes user data Access the deployment of the user of the user data sets of threads, comprising:
According to the number of users of each user data sets of threads record access, by the least user data thread of the number of users accessed User data sets of threads of the group as new access user;
Or,
The CPU average load of the affiliated hardware core in setting duration is recorded according to each user data sets of threads, CPU is averagely negative User data sets of threads of the user data sets of threads disposed on the smallest hardware core of lotus as new access user;
Or,
The CPU average load of the affiliated hardware core in setting duration is recorded according to each user data sets of threads, if maximum CPU is flat The difference of equal load and minimum CPU average load is more than given threshold, then will dispose on the hardware core of the maximum CPU average load The user of user data sets of threads adjust to the user data thread disposed on the hardware core of the minimum CPU average load Group.
6. a kind of data processing equipment for air protocol data surface characterized by comprising
Acquiring unit: the number of cells N that hardware core total number Y and the base station for obtaining base station need support;
First determination unit: the hardware core total number Y of number of cells N and the base station for being needed support according to the base station Relationship determines the hardware nucleus number M1 for handling the scheduling of cell MAC layer;
Second determination unit: for according to the hardware core total number Y of the base station and described for handling cell media access control The hardware nucleus number M1 of MAC layer scheduling, determines the hardware nucleus number M2 for being used for processes user data, wherein M1+M2≤Y.
7. device as claimed in claim 6, which is characterized in that the first determination unit is specifically used for:
A hardware core is occupied according to each cell and carries out the scheduling of cell MAC layer, and each cell occupies a hardware core processing and uses The principle of user data, determines whether Y is more than or equal to 2N;
In Y >=2N, determine that the hardware nucleus number M1 for handling the scheduling of cell MAC layer is N;Second determination unit is determined for locating Manage the hardware nucleus number M2=Y-N of user data.
8. device as claimed in claim 7, which is characterized in that first determination unit is also used to:
In Y < 2N, the hardware core for handling the scheduling of cell MAC layer according to one is the principle of two cell serves, and determination is used for Hardware the nucleus number M1, M1 for handling the scheduling of cell MAC layer are the integer not less than N/2;Second determination unit is determined to be used for handling The hardware nucleus number M2=Y-M1 of user data.
9. such as the described in any item devices of claim 6 to 8, which is characterized in that further include adjustment unit, be used for:
Determine the cell scheduling sets of threads and user data sets of threads of cell in the base station;
Cell scheduling sets of threads is deployed on M1 hardware core of processing cell MAC layer scheduling, by user data sets of threads portion Administration is on M2 hardware core of processes user data;
The deployment of the user of the user data sets of threads is accessed according to the load adjustment of the hardware core of processes user data.
10. device as claimed in claim 9, which is characterized in that the adjustment unit is also used to:
According to the number of users of each user data sets of threads record access, by the least user data thread of the number of users accessed User data sets of threads of the group as new access user;
Or,
The CPU average load of the affiliated hardware core in setting duration is recorded according to each user data sets of threads, CPU is averagely negative User data sets of threads of the user data sets of threads disposed on the smallest hardware core of lotus as new access user;
Or,
The CPU average load of the affiliated hardware core in setting duration is recorded according to each user data sets of threads, if maximum CPU is flat The difference of equal load and minimum CPU average load is more than given threshold, then will dispose on the hardware core of the maximum CPU average load The user of user data sets of threads adjust to the user data thread disposed on the hardware core of the minimum CPU average load Group.
CN201710044159.7A 2017-01-19 2017-01-19 A kind of data processing method and device for air protocol data surface CN106851667B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710044159.7A CN106851667B (en) 2017-01-19 2017-01-19 A kind of data processing method and device for air protocol data surface

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710044159.7A CN106851667B (en) 2017-01-19 2017-01-19 A kind of data processing method and device for air protocol data surface
PCT/CN2017/118081 WO2018133625A1 (en) 2017-01-19 2017-12-22 Data processing method and apparatus for air interface protocol data plane

Publications (2)

Publication Number Publication Date
CN106851667A CN106851667A (en) 2017-06-13
CN106851667B true CN106851667B (en) 2019-07-02

Family

ID=59119247

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710044159.7A CN106851667B (en) 2017-01-19 2017-01-19 A kind of data processing method and device for air protocol data surface

Country Status (2)

Country Link
CN (1) CN106851667B (en)
WO (1) WO2018133625A1 (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1937591A (en) * 2006-11-02 2007-03-28 杭州华为三康技术有限公司 Multi-core processor for realizing adaptive dispatching and multi-core processing method
WO2009109693A1 (en) * 2008-03-07 2009-09-11 Nokia Corporation A data processing arrangement
CN102868643A (en) * 2012-08-31 2013-01-09 苏州简约纳电子有限公司 Long-term evolution (LTE) data surface software architecture
CN103154897A (en) * 2010-10-14 2013-06-12 阿尔卡特朗讯公司 Core abstraction layer for telecommunication network applications
CN103838552A (en) * 2014-03-18 2014-06-04 北京邮电大学 System and method for processing multi-core parallel assembly line signals of 4G broadband communication system
CN105827654A (en) * 2016-05-26 2016-08-03 西安电子科技大学 Multi-core parallel protocol stack structure design method based on GMR-1 3G system

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7620753B1 (en) * 2005-03-17 2009-11-17 Apple Inc. Lockless access to a ring buffer
CN103906257B (en) * 2014-04-18 2017-09-08 北京邮电大学 LTE wide-band communication system computing resource schedulers and its dispatching method based on GPP

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1937591A (en) * 2006-11-02 2007-03-28 杭州华为三康技术有限公司 Multi-core processor for realizing adaptive dispatching and multi-core processing method
WO2009109693A1 (en) * 2008-03-07 2009-09-11 Nokia Corporation A data processing arrangement
CN103154897A (en) * 2010-10-14 2013-06-12 阿尔卡特朗讯公司 Core abstraction layer for telecommunication network applications
CN102868643A (en) * 2012-08-31 2013-01-09 苏州简约纳电子有限公司 Long-term evolution (LTE) data surface software architecture
CN103838552A (en) * 2014-03-18 2014-06-04 北京邮电大学 System and method for processing multi-core parallel assembly line signals of 4G broadband communication system
CN105827654A (en) * 2016-05-26 2016-08-03 西安电子科技大学 Multi-core parallel protocol stack structure design method based on GMR-1 3G system

Also Published As

Publication number Publication date
WO2018133625A1 (en) 2018-07-26
CN106851667A (en) 2017-06-13

Similar Documents

Publication Publication Date Title
CN103583074B (en) A kind of dispatching method and device
US8532030B2 (en) Techniques for initiating communication in a wireless network
CN104937977B (en) For improving in wireless network the method for handling capacity on time
CN104754748B (en) A kind of D2D resource allocation methods, data transmission method and device
CN105659685B (en) Method and apparatus for carrying out channel access in asynchronous unlicensed radio-frequency spectrum band deployment
TWI479865B (en) Method and apparatus for initializing, preserving and reconfiguring token buckets
TW200605688A (en) Method and system for managing radio resources in a time-slotted communicationwireless system
TW200840292A (en) Resource allocation method of subscriber of service negotiation system
CN102098737B (en) Cell priority level based cooperation scheduling method and system
CN103517276B (en) D2D communication method, user equipment and base station
WO2010133059A1 (en) Method and device for allocating control channel resources
CN101771462A (en) Method and device for allocating downlink control channel resource in multicarrier system
CN103686858B (en) Feedback method, base station and the user equipment of ascending control information
CN104185281B (en) Resource distribution and resource using method and device in a kind of D2D communications
CN100550691C (en) Method for dispatching group data transmission
CN101212782B (en) High-speed uplink packet access scheduling method and system
EP3110195A1 (en) Base station and method and device for scheduling on unlicensed spectrum in ue
CN101568125B (en) Wireless resource allocation method capable of supporting the frequency spectrum resource sharing
CN104363598B (en) A kind of DRB mapping methods and device
CN102761904B (en) Data transmission method, equipment and system
CN102137496B (en) Method and device for joint dispatching of PUSCH (physical uplink shared channel) resources and PHICH (physical hybrid automatic repeat request indicator channel) resources
CN101784103A (en) Method for scheduling users in a plurality of communities and wireless access network equipment thereof
WO2015149657A1 (en) Method and device for triggering buffer state reporting bsr
CN106605441A (en) Low-latency, low-bandwidth and low duty cycle operation in a wireless communication system
CN103609188B (en) Carrier resource allocation method, uplink information transmission, base station and user equipment

Legal Events

Date Code Title Description
PB01 Publication
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right
GR01 Patent grant