CN113556825A

CN113556825A - Polling method based on multi-attribute decision in point coordination

Info

Publication number: CN113556825A
Application number: CN202110751915.6A
Authority: CN
Inventors: 任塨晔; 柏荣胜; 黄镠; 吕卫祥
Original assignee: 724th Research Institute of CSIC
Current assignee: 724th Research Institute of CSIC
Priority date: 2021-07-02
Filing date: 2021-07-02
Publication date: 2021-10-26
Anticipated expiration: 2041-07-02
Also published as: CN113556825B

Abstract

The invention provides a polling method based on multi-attribute decision in point coordination, which determines an STA polled by an AP at a specific time in a PCF and a data packet additionally sent when polling the STA. First, the AP calculates the urgency of transmitting a packet to each STA, which is related to the remaining validity period of the packet, the priority of the packet, and the length of the packet. Secondly, the AP calculates how urgent each STA is polled, which is related to the time the STA last polled, the number of data packets the STA sends, and the average snr at which the AP receives the STA signal. The AP solves the two values using a multi-attribute decision method of dynamic weighted summation, where each attribute weight is related to the position of the current time in the PCF period. The AP determines the STA polled at the current time based on the urgency of sending the data packet to each STA and the urgency of each STA being polled, and selects and sends the data packet with the maximum urgency in the corresponding transmission queue as the data packet sent by the current polling.

Description

Polling method based on multi-attribute decision in point coordination

Technical Field

The present invention relates to a scheduling policy in an infrastructure network, and more particularly, to a polling method for an AP in a PCF.

Background

The IEEE Std 802.11-1999 edition proposes two MAC layer channel access modes in the wireless local area network: a Distributed Coordination Function (DCF) and a Point Coordination Function (PCF), a work cycle of a whole Basic Service Set (BSS) may also be divided into two periods: a DCF period and a PCF period. In the DCF period, an Access Point (AP) and a mobile STAtion (STAtion) in the BSS Access a channel in a contention manner using a Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA) protocol. In the PCF period, the AP controls the channel access of STAs in the entire BSS in a polling manner, and the non-polled STAs cannot transmit signals. The PCF operating mode avoids delays caused by channel access collisions and helps to guarantee the quality of service for a particular service.

To avoid sending too many frames at the time of polling, the IEEE Std 802.11-1999 standard utilizes an omni-directional antenna feature to combine polling, data transmission, and acknowledgement in one frame for transmission, thereby performing more polling in a limited time. The frames sent by the AP include CF-Poll (no Data), Data + CF-Poll, Data + CF-Ack + CF-Poll, and CF-Ack + CF-Poll (no Data), and the frames sent by the STA include Data, Data + CF-Ack, CF-Ack (no Data), and Null.

One important issue facing APs operating in PCF mode is the rules for polling STAs. The IEEE Std 802.11-1999 standard does not specify AP polling rules. The general approach is to poll the STAs in sequence and send the packets in sequence in the order in which they entered the queuing sequence. Although this method can ensure fairness among STAs (polling is performed every certain period), the impact of the polling sequence on the service quality is ignored. The STA often interacts with the remote server for various different types of services, which also have different requirements for quality of service. For example, the STA may simultaneously run services such as live video, web browsing, online gaming, file downloading, etc. The live video service has high requirements on transmission rate and time delay, but can tolerate a certain packet loss rate, while the online game service has high requirements on packet loss rate and time delay, but has low requirements on transmission rate. In order to ensure the service quality of each service, the AP operating in PCF mode should determine the STA polling at the current time and the data packet sent when polling the STA according to the attributes of each data packet and the service type carried by it.

When performing the polling decision, the AP needs to consider the following two problems in addition to the attribute of the packet itself. On the one hand, the polling rule of the AP should consider the characteristics of the wireless channel. The wireless channel has problems of interference, attenuation, etc., and packet interaction between the AP and the STA is not always successful. Under the condition that other conditions are consistent, the AP should poll the STA with better channel quality preferentially, so as to fully utilize the limited PCF period to transmit more information. On the other hand, the polling rule of the AP should consider the attributes of the STAs. The STA as a user terminal may also generate traffic data, which also has an important impact on the quality of service. Under other conditions, the AP should poll more STAs with more data packets to be sent, and ensure certain polling opportunities for other STAs.

Disclosure of Invention

The invention aims to solve the polling decision problem of the AP working in PCF mode in the wireless local area network, namely determining the STA polled by the AP at a specific moment in PCF time period and transmitting a data packet when polling the STA. The invention adopts a multi-attribute decision-making method of dynamic weighted summation to solve the problem, comprehensively considers the factors of the remaining validity period of the data packet, the priority of the data packet, the length of the data packet, the time from the STA to the last polling, the number of data packets to be sent by the STA, the average signal-to-noise ratio of the AP to receive the STA signal and the like, can effectively improve the service quality, and simultaneously ensures the fairness of the STAs.

In order to solve the technical problems, the invention is realized by the following technical scheme: the AP considers the following factors when determining the STA to be polled and the data packet to be sent: the method comprises the following steps of firstly determining the STA to be polled by adopting a dynamic weighted sum multi-attribute decision method, then determining the data packet sent when polling the STA, wherein the weight of each attribute is related to the position of the current time in a PCF time period, and the specific steps comprise:

step 1: AP calculates to STA_iUrgency of transmission of data packet s_i，t；

Step 2: AP computing STA_iUrgency to be polled v_i，t；

And step 3: AP compute poll STA_iIntensity w of_i，t＝max{s_i，t，v_i，t}；

And 4, step 4: AP determines polling STA at current moment_j′Wherein i ═ argmax_{i＝1，2，...，N}{w_i，t}；

And 5: the AP judges whether the AP is to be sent to the STA_i′If there is, the AP sends a Data + CF-Poll packet or a Data + CF-Ack + CF-Poll packet to the STA_i′Wherein the data part is the destination address of STA_i′And sending the data packet with the maximum urgency, otherwise, the AP sends the CF-Poll packet or the CF-Ack + CF-Poll packet to the STA_i′。

A BSS is provided with 1 AP and N (N is more than or equal to 1) STAs. The set of STAs is denoted as S, S ═ { STAi }, i ═ 1, 2. The AP maintains a transmission queue for each STA, in which data packets to be sent to each STA are stored. At time t, AP maintains STA_iThe corresponding transmission queue is marked as

For a particular STA_i，

Each packet within has corresponding attributes including remaining validity period, priority, length, etc. When the AP polls the STAi, it needs to consider these factors comprehensively and determine the data packet transmission that can improve the service quality most. The problem can be modeled as a multi-attribute decision problem. And (4) aggregating various attributes of the decision-making object into a utility value in the multi-attribute decision-making, wherein the object with the highest utility value is the target of the decision-making. Defining AP to STA_iUrgency of transmission of data packet s_i，tIs composed of

Maximum utility value of each data packet in the packet. Also, the AP records the STA_iIncluding STA_iTime since last polled, STA_iNumber of data packets to be transmitted, AP receiving STA_iThe average signal-to-noise ratio of the signal, which together determine the STA_iUrgency to be polled v_i，tThe problem can also be modeled as a multi-attribute decision problem. In summary, the polling decision of the AP is related to two factors: the urgency of the AP to transmit packets to the STAs and the urgency of the STAs to be polled. The AP considers these two factors together to finally determine the STA to be polled and the data packet to be transmitted when polling the STA.

Due to the limited PCF time period, the location of the PCF time period at the current time also affects the polling decision of the AP. When the PCF time period is about to end, the AP should give priority to how to transmit more packets in the remaining time period or to guarantee the transmission quality in the remaining time period when making a decision. Therefore, the polling decision of the AP is a dynamic multi-attribute decision problem, that is, the position of the PCF period at the current time also affects the decision result under the condition that other factors are unchanged.

At time t, the AP operates in PCF mode, and determines the STA to be polled and the data packet sent when polling the STA according to the following steps:

step 1: AP calculates to STA_iUrgency of transmission of data packet s_i，t(i＝1，2，...，N)：

(1) AP culling

And the remaining data packets with the validity period less than or equal to zero. For the

Inner data packet

Defining a residual validity period u_i，j，t＝t_{remove，i，j}-t, wherein t_{remove，i，j}Indicating the absolute time at which the packet failed, is specified by the source node when generating the packet. AP computation

The remaining validity period of each data packet in the packet sequence, and

middle removing u_i，j，tData packets less than or equal to 0 to obtain transmission queue

j＝1，2，...，J_i，t(J_i，t≥0)。

(2) For the

Each data packet in

Its following three attributes were calculated:

a) residual effective period u_i，j，t: some services require that packets be delivered to the destination node within a certain time limit. The source node specifies a validity period when generating the packet, and the service is valid when the packet is delivered to the destination node within the validity period. When the validity period has expired without the packet reaching the destination node, the intermediate node (router, AP, etc.) discards the packet. Definition u_i，j，t＝t_{remove，i，j}-t, wherein t_{remove，i，j}Indicating the absolute time at which the packet failed, is specified by the source node when generating the packet.

b) Priority p_i，j: when the upper layer protocol generates a data packet, the priority p of the data packet is determined according to a certain strategy_i，j∈[0，1]0 represents the lowest priority, and 1 represents the highest priority.

c) Length l_i，j: the packet length refers to the length of an upper layer protocol frame carried by the MAC layer, in bytes. In the IEEE 802.11 standard, this length is referred to as a MAC layer Service Data Unit (MSDU).

(3) Because the dimensions of the three attributes of the data packet and the influence on the urgency of the AP to send each data packet are different, the weighted sum multi-attribute decision needs to be made on the three attributesThe attribute is normalized to obtain normalized value

And

a) for u_i，j，tThe normalization method comprises the following steps: u. of_i，j，tThe smaller the packet size, the closer the current time is to the expiration time of the validity period of the corresponding packet, and the stronger the urgency of transmitting the packet. Meanwhile, the validity period (time from generation to expiration) of each packet is different from one packet to another, and therefore the following normalization method is adopted:

wherein

It indicates the packet validity length, t_{generate，i，j}As data packets

At the absolute time of generation at the source node.

Indicating the relative persistence time of the packet. In the case where a packet is not invalidated, the longer the relative persistence time, the higher the urgency of the AP to transmit the packet. f. of_u，n(x) Is defined as

b) For p_i，jThe normalization method comprises the following steps: the value range of the priority is [0, 1 ]]And the higher the priority, the higher the urgency of the AP to transmit the packet, so the normalized value

c) For l_i，jThe normalization method comprises the following steps: due to the fact that wireless channel loss is large and interference is prone to happen, certain packet loss rate exists in data packet transmission. To increase the transmission success rate, the node should preferentially send a shorter packet. However, the packet length difference of different services is large, and the normalization process needs to be performed by using the average packet length. Note L_tIs the average length of the data packets in all transmission queues in the AP, then

Indicating the relative length of the data packet. Calculated by the following formula

wherein

(4) Determining the weight omega of the three attributes according to the current time t_u，t、ω_p，t、ω_l，t. Calculating the weight ω_u，t、ω_p，t、ω_l，tFirst determine the boundary value of each attribute weight at the beginning of PCF time interval

And boundary value at the end of PCF period

They firstly satisfy the following requirements for weight:

at the beginning of the PCF, the AP prioritizes the remaining validity period and priority of the packets, and towards the end of the PCF, the AP prioritizes the packet length to send as many packets as possible during the remaining period. Therefore, the boundary values for the set weights satisfy the following relationship:

let the starting time of the current PCF time interval be t_startPCF period length of T_pcfThen ω is calculated according to the following formula_u，t、ω_p，t、ω_l，t：

The above formula defines the change rule of the weights of the three attributes, wherein ω is_u，t and ω_p，tDecreases linearly with increasing t, and ω_l，tThe linear increase with the increase of t reflects the change of the influence degree of different attributes of the AP in decision making over time.

(5) Based on the normalized values

And corresponding three attribute weights omega_u，t、ω_p，t、ω_l，tObtaining AP sending data packet by using weighted summation method

S degree of urgency of_i，j，tComprises the following steps:

(6) by definition, an AP towards an STA_iUrgency of transmission of data packet s_i，tIs composed of

Maximum utility value of each data packet therein, i.e.

Step 2: AP computing STA_iUrgency to be polled v_i，t(i＝1，2，...，N)：

(1) AP calculates corresponding STA_iThe following three attributes:

a) time c since last polled_i，t：c_i，t＝t-t_i, wherein t_iRepresents the absolute time that STAi was last polled, which is recorded by the AP;

b) number of data packets to be transmitted r_i，t: at STA_iThe corresponding field in the format of the frame (Data + CF-Ack, Data, CF-Ack, Null) sent to the AP records the transmission queue of the STA

(

Is a transmission queue from which failed packets have been dropped,

) Number of data packets contained in (a)Does not contain the currently transmitted data packet), the AP can record the value after receiving the packet;

c) AP receiving STA_iAverage signal-to-noise ratio q of signal_i，t: AP records last M times of receiving STA_iThe signal-to-noise ratio of the reply packet is calculated, and the average value is calculated to obtain q_i，t；

(2) Normalizing the three attributes to obtain normalized values

And

a) for c_i，tThe normalization method comprises the following steps: the longer the time is from the last time of being polled, the higher the urgency is for the corresponding STA to be polled, so that the data packets generated by the STA can be transmitted in time. Remember h_tFor the average time interval in which the STAs are polled, then

Representation of STA_iThe relative time interval to be polled is calculated using the following formula

wherein

b) For r_i，tThe normalization method comprises the following steps: STA (station)_iThe more data packets to be sent, the more STAs_iThe higher the urgency to be polled. Note R_tThe number of packets waiting to be transmitted is averaged for each STA, i.e.

Then

Representation of STA_iRelative to the number of packets waiting to be sent, the following formula is used to calculate

wherein f_r，n(x)＝f_c，n(x)；

c) For q_i，tThe normalization method comprises the following steps: STA received by AP_iThe larger the average signal-to-noise ratio of the signal, it means that the AP sends to the STA_iIs encapsulated by STA_iThe greater the probability of correct reception, and thus the AP polls the STAs_iThe higher the tendency of (c). Note q_max，t＝max_{i＝1，2，...，N}{q_i，t}，q_min，t＝min_{i＝1，2，...，N}{q_i，tIs calculated according to the following formula

(3) Determining the weight omega of the three attributes according to the current time t_c，t、ω_r，t、ω_q，t. Calculating the weight ω_c，t、ω_r，t、ω_q，tFirst determine the boundary value of each attribute weight at the beginning of PCF time interval

And boundary value at the end of PCF period

They firstly satisfy the following requirements for weight:

at the beginning of PCF, the AP prioritizes the time each STA was polled last and the number of data packets to be sent, and towards the end of PCF, the AP prioritizes the average snr of the received STA signals to send as many packets as possible during the remaining time period. Therefore, the boundary values for the set weights satisfy the following relationship:

let the starting time of the current PCF time interval be t_startPCF period length of T_pcfThen ω is calculated according to the following formula_c，t、ω_r，t、ω_q，t：

The above formula defines the change rule of the weights of the three attributes, wherein ω is_c，t and ω_r，tDecreases linearly with increasing t, and ω_q，tThe linear increase with the increase of t reflects the change of the influence degree of different attributes of the AP in decision making over time.

(4) Based on the normalized values

And corresponding three attribute weights omega_c，t、ω_r，t、ω_q，tObtaining STA by applying weighted summation method_iUrgency to be polled v_i，tComprises the following steps:

and step 3: for each STA_iAP calculates s corresponding to the AP_i，t and v_i，tWhich respectively characterize the AP-polling STA from different angles_iThe urgency of the application. Defining AP polling STAs_iIntensity w of_i，tThe following were used:

w_i，t＝max{s_i，t，v_i，t} (18)

w_i，tcomprehensively reflects AP polling STA_iThe urgency of the application.

And 4, step 4: AP determines polling STA at current moment_i′Wherein i ═ argmax_{i＝1，2，...，N}{w_i，t}。

And 5: after determining the polled STA, the AP needs to determine the data packet sent to the STA. Judgment of

Whether or not it is empty, if

If null, the AP sends a CF-Poll frame (when the last polled STA returned a frame containing no Data portion) or a CF-Ack + CF-Poll frame (when the last polled STA returned a frame containing a Data portion) to the STA_i′Otherwise, the AP sends a Data + CF-Poll frame (when the last frame returned by the polled STA contains no Data portion) or a Data + CF-Ack + CF-Poll frame (when the last frame returned by the polled STA contains a Data portion) to the STA_i′The Data part is a Data packet

wherein

The decision process of the AP is ended up to this point.

The invention has the following beneficial effects:

1. the invention can effectively reduce the service time delay and improve the service quality. The invention comprehensively considers two factors: the attribute of the data packet to be sent cached by the AP and the attribute of each STA in the BSS. The attributes of the data packets include the remaining validity period, priority and length, and the attributes of the STA include the time since the last polling, the number of data packets to be transmitted and the average signal-to-noise ratio of the STA signal received by the AP.

2. The invention dynamically adjusts the weight of each attribute according to the position of the current time in the PCF time period, and reflects the change of the influence degree of different attributes on the decision at different times in the PCF.

3. The method is simple and easy to implement, and a large amount of calculation load can not be introduced.

4. The invention is compatible with IEEE Std 802.11-2016 standard. The MAC frame header defined by the IEEE Std 802.11-2016 standard contains a QoS Control field. When the STA sends Data + CF-Ack, Data, CF-Ack and Null frames, the Data volume of the Data to be sent can be added in the field, and the AP can obtain the information.

Drawings

Fig. 1 is an AP polling decision flow diagram;

FIG. 2 is a diagram of AP calculation s_i，tA flow chart of (1);

FIG. 3 is AP calculation v_i，tIs described.

Detailed Description

The invention discloses a polling method based on multi-attribute decision in point coordination, which determines an STA polled by an AP at a specific time and a data packet additionally transmitted when the STA is polled. The polling decision of the AP is related to two factors: the urgency of the AP to transmit packets to the STAs and the urgency of the STAs to be polled. The whole polling decision problem is modeled as a multi-attribute decision problem, and the factors of the remaining validity period of a data packet, the priority of the data packet, the length of the data packet, the time from the STA to the last polling, the number of data packets to be sent by the STA, the average signal-to-noise ratio of signals of the AP receiving STA and the like are considered. And in consideration of the influence of the position of the PCF time period at the current moment on the decision result, a dynamic weighted summation method is adopted to solve the multi-attribute decision problem. The invention discloses a polling decision execution step of an AP, definition of each attribute, a normalization method, a weight determination method and the like. The invention is further illustrated by the following figures and examples. The embodiments are carried out on the premise of the technical scheme of the invention, and detailed embodiments and processes are given, but the scope of the invention is not limited to the following embodiments.

When the AP executes the polling method of the present invention, the AP, the STA, or the protocol in which they operate need to have the following functions:

1. when an application layer protocol of a source node generates a data packet, the generation time, the failure time and the priority of the data packet need to be added to the data packet;

the AP needs to record the latest polling moment of each STA;

3, the AP needs to record the average time interval of all the polled STAs;

the MAC layer protocol supports the STA to notify the AP of the number of data packets to be sent, and the AP acquires and records the information when receiving the reply frame of the STA;

and 5, the AP needs to record the signal-to-noise ratio of the latest M times of receiving reply packets of each STA, and calculates to obtain the average signal-to-noise ratio.

The AP decision process of the present invention is shown in fig. 1, which includes five steps:

step 1: AP calculates to STA_iUrgency of transmission of data packet s_i，t(i 1, 2.., N), fig. 2 gives a detailed process of the calculation;

step 2: AP computing STA_iUrgency to be polled v_i，t(i 1, 2.., N), fig. 3 gives a detailed process of the calculation;

and step 3: AP computing polling STA_iIntensity w of_i，t＝max{s_i，t，v_i，t}(i＝1，2，...，N)；

And 4, step 4: AP determines polling STA at current moment_i′Wherein i ═argmax_{i＝1，2，...，N}{w_i，t}；

And 5: judgment of

Whether or not it is empty, if

If null, the AP sends a CF-Poll frame (when the last polled STA returned a frame containing no Data portion) or a CF-Ack + CF-Poll frame (when the last polled STA returned a frame containing a Data portion) to the STA_i′Otherwise, the AP sends a Data + CF-Poll frame (when the last frame returned by the polled STA contains no Data portion) or a Data + CF-Ack + CF-Poll frame (when the last frame returned by the polled STA contains a Data portion) to the STA_i′Wherein the Data part is a Data packet

The decision process of the AP is ended up to this point.

Assume that 1 BSS contains 1 AP and 4(N ═ 4) STAs. The whole BSS works in a DCF and PCF alternative mode, the repetition period is 1s, the DCF time interval and the PCF time interval respectively occupy 500ms, namely T_dcf＝T_pcf500 ms. Assume that the current time is t_current3800ms, the AP operates in PCF mode, and the current PCF period is from t_startStarting at 3500 ms. At t_currentAt this time, the AP determines the polling STA and the data packet transmitted when polling the STA according to the flow shown in fig. 1. The AP first executes step 1, calculates the direction STA_iUrgency of transmission of data packet s_i，t(i ═ 1, 2, 3, 4). The various attributes of the packets to be sent by the AP to each STA are shown in table 1, where the failed packets have been rejected. The attributes of each normalized packet are shown in table 2.

Table 1 various attributes of data packets to be sent by an AP to each STA

Table 2 normalized attributes of each packet

The attributes of each STA recorded by the AP are shown in table 3. The normalized class attributes of each STA are shown in Table 4, where h_t＝245ms。

Table 3 AP recorded various attributes of each STA

Table 4 normalized attributes of STAs

The boundary values for the weights of the attributes are set as follows:

at t_currentAt time 3800ms, the weights of the attributes are calculated according to equations (5) to (7) and (14) to (16)

ω_u，t＝0.34，ω_p，t＝0.28，ω_l，t＝0.38，ω_c，t＝0.24，ω_r，t＝0.28，ω_q，t＝0.48 (21)

According to table 2, table 4 and formula (21), the urgency s of the AP to send the data packet to each STA is calculated_i，tUrgency v of each STA being polled_i，tAnd the strength w of AP polling each STA_i，tAs shown in table 5. According to the rule described in FIG. 1, at t_currentAt time 3800ms, the AP decides to poll STA3 and sends the 1 st packet in the STA3 transmission queue at the time of the poll.

TABLE 5 s of STAs_i，t、v_i，t and w_i，tValue of

Claims

1. A polling method based on multi-attribute decision in point coordination is characterized in that: the AP considers the following factors when determining the STA to be polled and the data packet to be sent: the method comprises the following steps of firstly determining the STA to be polled by adopting a dynamic weighted sum multi-attribute decision method, then determining the data packet transmitted when polling the STA, wherein the weight of each attribute is related to the position of the current time in a PCF time period, and the steps comprise:

step 1: AP calculates to STA_iUrgency of transmission of data packet s_i，t；

Step 2: AP computing STA_iUrgency to be polled v_i，t；

And step 3: AP computing polling STA_iIntensity w of_i，t＝max{s_i，t，v_i，t}；

And 4, step 4: AP determines polling STA at current moment_i′Wherein i ═ argmax_{i＝1，2，...，N}{w_i，t}；

2. The method according to claim 1, wherein the method comprises: the AP adopts a dynamic weighted sum multi-attribute decision method to calculate the STA_iUrgency of transmission of data packet s_i，tThe method comprises the following steps:

(1) for and STA_iCorresponding transmission queue

Each data packet in

The following three attributes were calculated:

a) residual effective period u_i，j，t：u_i，j，t＝t_{remove，i，j}-t, wherein t_{remove，i，j}Indicating an absolute time of failure of the data packet, as determined by the source node when generating the data packet;

b) priority p_i，j: calibrated by upper layer protocol, p_i，j∈[0，1]0 represents the lowest priority, 1 represents the highest priority;

c) length l_i，j: the number of bytes of the MAC layer service data unit;

(2) normalizing the three attributes to obtain normalized values

And

(3) determining the weight omega of three attributes according to the current time_u，t、ω_p，t、ω_l，t；

(4) Calculating AP sending data packet

The urgency of (2):

(5) computing AP to STA_iUrgency of sending a packet:

3. the method according to claim 1, wherein the method comprises: the AP adopts a dynamic weighted sum multi-attribute decision method to calculate the STA_iUrgency to be polled v_i，tThe method comprises the following steps:

(1) AP calculates corresponding STA_iThe following three attributes:

a) time c since last polled_i，t：c_i，t＝t-t_i, wherein t_iRepresentation of STA_iThe absolute time of last polling;

b) number of data packets to be transmitted r_i，t: the information is transmitted by STA_iProviding the AP;

c) AP receiving STA_iAverage signal-to-noise ratio q of signal_i，t: AP records STA received last M times_iThe signal-to-noise ratio of the reply packet is calculated, and the average value is calculated to obtain q_i，t；

(2) Normalizing the three attributes to obtain normalized values

And

(3) determining the weight omega of three attributes according to the current time_c，t、ω_r，t、ω_q，t；

(4) Computing STA_iUrgency to be polled v_i，t：

4. The method according to claim 2, wherein the method comprises: the remaining effective period u_i，j，tPriority p_i，jAnd length l_i，jThe normalization method comprises the following steps:

a) for u_i，j，tThe normalization method comprises the following steps: note t_{generate，i，j}As data packets

At the absolute time, t, generated by the source node_{valid，i，j}For the length of the validity period of the data packet, t_{valid，i，j}＝t_{remove，i，j}-t_{generate，i，j}Calculated according to the following formula

wherein ：

b) for p_i，jThe normalization method comprises the following steps:

c) for l_i，jThe normalization method comprises the following steps: note L_tThe average length of the data packets in all transmission queues in the AP is calculated according to the following formula

wherein ：

5. the method according to claim 2, wherein the method comprises: the remaining effective period u_i，j，tPriority p_i，jAnd length l_i，jWeight ω of (d)_u，t、ω_p，t and ω_l，tIn relation to the current time t, it includes:

a) let omega_u，t、ω_p，t、ω_l，tThe initial values at the beginning of the PCF period are

The value at the end of the PCF period is

And satisfy

b) Let the starting time of the current PCF time interval be t_startPCF period length of T_pcfω is calculated according to the following formula_u，t、ω_p，t、ω_l，t：

6. The method of claim 3, wherein the method comprises: the time c since the last polling_i，tNumber r of data packets to be transmitted_i，tAnd AP receiving STA_iAverage signal-to-noise ratio q of signal_i，tThe normalization method comprises the following steps:

a) for c_i，tThe normalization method comprises the following steps: remember h_tThe average time interval for which each STA is polled is calculated as follows

wherein ：

b) for r_i，tThe normalization method comprises the following steps: note R_tThe number of packets waiting to be transmitted for each STA on average is calculated as follows

wherein f_r，n(x)＝f_c，n(x)；

c) For q_i，tThe normalization method comprises the following steps: note q_max，t＝max_{i＝1，2，...，N}{q_i，t}，q_min，t＝min_{i＝1，2，...，N}{q_i，tIs calculated according to the following formula

7. The method of claim 3, wherein the method comprises: the time c since the last polling_i，tNumber r of data packets to be transmitted_i，tAnd AP receiving STA_iAverage signal-to-noise ratio q of signal_i，tWeight ω of (d)_c，t、ω_r，t and ω_q，tIn relation to the current time t, the calculation method comprises:

a) let omega_c，t、ω_r，t、ω_q，tThe initial values at the beginning of the PCF period are

The value at the end of the PCF period is

And satisfy

b) Let the starting time of the current PCF time interval be t_startTotal PCF period length of T_pcfω is calculated according to the following formula_c，t、ω_r，t、ω_q，t：