CN106921530A - The superframe method of testing of procedure-oriented automatic industrial wireless network - Google Patents

Abstract

The present invention relates to the radio network technique towards commercial Application, the superframe method of testing of procedure-oriented automatic industrial wireless network is specifically claimed.Can the inventive method is mainly used in test equipment under test (routing device or field apparatus) correctly configure superframe, and correctly carry out data transmit-receive according to superframe information.The inventive method takes into full account interference of the factors such as environmental disturbances and clock drift to testing, using the time offset of Least Square Method superframe, on the premise of it need not change equipment under test function and communication pattern, the error of superframe configuration and superframe to equipment under test is tested, and has reached high efficiency, high accuracy and the effect being easily achieved.

Description

The superframe method of testing of procedure-oriented automatic industrial wireless network

Technical field

The present invention relates to the industry wireless network technology of procedure-oriented automation, specifically a kind of procedure-oriented is automatic The superframe method of testing of chemical industry wireless network.

Background technology

The industry wireless network technology of procedure-oriented automation has the advantages that low cost, low energy consumption and easily uses.With The popularization and application of the industry wireless network of procedure-oriented automation, to be realized with a low cost " ubiquitous perception " and " overall process optimization control The industry measurement and control system of system " turns into hot spot technology, and with vast potential for future development, represents industrial automation system technology One of developing direction, will play a significant role, in oil day improving production efficiency, simplifying the aspects such as industrial processes Highly energy-consuming, the high pollution industries such as the exploitation of right gas, petrochemical industry, metallurgy, sewage disposal have wide practical use.Industry wireless network WIA-PA (Wireless network for Industrial Automation-Process Automation) is the field An emerging technology for world level.

In order to ensure the real-time and reliability of data transfer, WIA-PA industry wireless networks are used based on beacon IEEE STD 802.15.4:2006 superframe structures.With the expansion of WIA-PA industry wireless network application scales, WIA-PA sets Can standby superframe will turn into a key factor of restriction network performance by rational distribution and correct execution.To tested The implementation status and time migration error of equipment superframe are tested, thus it is ensured that the real-time and reliable communication of equipment under test.Superframe Test is into a test link of equal importance with apparatus function test.

For WIA-PA testing protocol consistencies specification (the WIA-PA Conformance Test of industrial process automation Specification for use in Industrial Process Automation) it is directed to WIA-PA standards and is surveyed The reference frame of examination, is industry wireless network equipment, system and engineer applied meeting in WIA-PA industrial wireless communication agreements Degree provides the test specification of standard, is that the equipment between wireless industrial product certification and the different agreement realization of multiple producers is mutual Connection is provided and ensured.But WIA-PA testing protocol consistencies specification only provides testing process and content, the test of correlation is not given Technology.Research currently for the uniformity test technology of WIA-PA agreements is at the early-stage, especially for the test skill of superframe Art has no research.There is problems with existing WIA-PA test systems and method of testing：1) only considered holding for equipment under test Row flow, does not carry out overall execution time test from the angle of superframe to system；2) not to influence superframe each influence because Element is tested, and whether the skew of non-testing time is in superframe tolerance interval；3) some method of testings need modification tested The packet sending and receiving requirement of equipment and partial function, reduce the efficiency of test.

The content of the invention

It is super to test equipment under test execution superframe not to be accounted for for existing WIA-PA protocol conformance test methods The influence of frame time, causes there is larger error to the test of equipment, also just cannot accurately test out equipment under test superframe Implementation status, the present invention proposes a kind of superframe method of testing of procedure-oriented automatic industrial wireless network.

The technical scheme that is used to achieve the above object of the present invention is：A kind of procedure-oriented automatic industrial wireless network Superframe method of testing, comprise the following steps：

Test system and equipment under test are set up into the structure of the hub-and-spoke configuration of single-hop or star-like and netted combination；

Make equipment under test and test system synchronization using time synchronized command frame, it is tested using the setting of resource allocation command frame The superframe information of equipment so that equipment under test can be received and dispatched continuously；

Test system periodically sends request of data message in each sending time slots, and waits ACK and data to respond Message；

The response time stamp for sending time and data response message of statistics request message, using least square method, According to the reception time of regression equation calculation data response message, and the conspicuousness of regression equation is detected, if returned significantly, Continue next step；If returning not notable, superframe mistake；

Super frame tim e deviant is calculated using the timestamp for receiving time and data response message of data response message, and Check the super frame tim e deviant whether be data response message reception the time unbiased esti-mator；

If it is, thinking unbiased esti-mator effectively, super frame tim e skew is illustrated in the range of acceptable error, superframe is accurate Really, otherwise it is assumed that unbiased esti-mator is invalid, superframe mistake.

The equipment under test includes routing device and field apparatus；

Wherein, field apparatus is arranged on industry spot, for being responsible for connection or control production process；Routing device is responsible for existing The management of field device and message are forwarded.

The test system periodically sends request of data message in each sending time slots, and waits ACK and data Response message, if test system receives data response message, then it is assumed that be once effective data access process；If no Receive response message, then it is assumed that this data access process inefficient, be not analyzed.

The a plurality of link that can continuously receive and dispatch is configured in the superframe of the equipment under test.

The regression equation is equation of linear regression：

Wherein,It is the least-squares estimation of y, y is the reception time of data response message, and x is request of data message The transmission time,

Wherein, (x_i,y_i) it is sample point,It is sample data x₁,x₂...x_nAverage value,It is sample data y₁, y₂...y_nAverage value, n is the number of times that event occurs, the i.e. number of sample point.

The conspicuousness of the inspection regression equation, specially：

As F ＞ F_aWhen (1.n-2), return significantly, otherwise, then it is assumed that return not notable；Wherein, α is quantile, by looking into F distribution tables understand F_a(1, n-2), n is the number of times that event occurs, the i.e. number of sample point, Q_eIt is residual sum of squares (RSS), U is flat to return Fang He, and

Wherein, It is regression straight line slope, (x_i,y_i) it is sample point,It is sample data x₁, x₂...x_nAverage value,It is sample data y₁,y₂...y_nAverage value,It is the least-squares estimation of y.

The super frame tim e deviant is the difference of the timestamp for receiving time and data response message of data response message.

The inspection super frame tim e deviant whether be data response message the reception time unbiased esti-mator, that is, examine The validity of unbiased esti-mator is surveyed, specially：

If formula (10) is set up, unbiased esti-mator is effective；Otherwise, unbiased esti-mator is invalid；It is super frame tim e deviant, θ is that the response message calculated by formula (4) receives the time, i.e.,

Wherein,It is the least-squares estimation of b,It is the expectation of x and y；

It is the difference of the reception time of transmission time and ACK that message transmitting time delay is data message.

The present invention has advantages below and beneficial effect：

1. the present invention proposes a kind of superframe method of testing of procedure-oriented automatic industrial wireless network, takes into full account The actual test demand of the industry wireless network of Kernel-based methods automation, need not change equipment under test function and receive and dispatch special On the basis of data frame, high efficiency, high-precision superframe method of testing are realized；

2. the regression equation that the present invention is made up of construction request of data message and data response message, and check recurrence The conspicuousness of equation, judges whether equipment under test can respond in the range of superframe regulation to data request data, enters And detect whether the superframe of equipment under test correctly runs；

3. the present invention calculates the reception time for sending time and ACK of request of data message, draws the transmission of data message Time delay, by detecting that super frame tim e offsets the validity relative to transmission delay, tests out the drift condition of equipment under test superframe, The interference for test due to factors such as environment and hardware handles can be effectively prevented from；

4. the perfect method of testing of WIA-PA superframes of the present invention, specify that the method for testing and flow of superframe so that WIA-PA uniformity tests are more complete, simplify testing process, are easy to tester to perform test job, and test in practice Feasibility is demonstrate,proved.

The present invention proposes a kind of superframe method of testing of procedure-oriented automatic industrial wireless network, tested for testing Whether the superframe implementation status and super frame tim e offset error of WIA-PA equipment meet requirement, have taken into full account that procedure-oriented is automatic The actual test demand of the industry wireless network of change, realizes convenient, high efficiency, the test of high-precision superframe.

Towards WIA-PA Wireless Communication Equipment and tdma communication mechanism, by sending periodic data message, it is proposed that A kind of method that judgement equipment under test correctly performs superframe, including the ruuning situation and super frame tim e drift condition of test superframe have Effect distinguishes time migration caused by environment and hardware handles, is favorably improved testing efficiency.

Brief description of the drawings

Fig. 1 is superframe test system schematic diagram in the present invention；

Fig. 2 is super frame tim e test schematic diagram in the present invention；

Fig. 3 is superframe method of testing flow chart in the present invention；

Fig. 4 is the structure design schematic diagram of network layer data bag in the present invention；

Fig. 5 is the schematic diagram of packet sending and receiving time in the present invention.

Specific embodiment

Below in conjunction with the accompanying drawings and embodiment the present invention is described in further detail.

As shown in Figure 1 to use WIA-PA network stelliform connection topology configuration in the present invention, wherein equipment under test is route Equipment or field apparatus；

Using resource allocation command frame, time and the superframe information of equipment under test are set so that equipment under test can be carried out Continuous transmitting-receiving, test system configurations into the superframe structure matched with equipment under test, as shown in Figure 2；

Test system sends request of data message when each sending time slots starts, and wait test system ACK and Data response message, if receiving correct ACK and data response message, then it is assumed that be once effective data communication process；

Testing process of the invention is illustrated in figure 3, according to the form of the data message of WIA-PA Protocol network layers, every When individual sending time slots start, a data request command message is periodically sent, continuously transmit 20 data messages.Record has Imitate the transmission time Lacal_Data_Time of data message_i, the reception time Global_ACK_Time of ACK_iAnd response datagram The reception time Global_Data_Time of text_i, and the time stamp data Global_Data_ for receiving response data message Timestamp_i, the form of timestamp is as shown in Figure 4 in data message；

Transmission time x of request of data message and the reception time y of data response message should meet unary linear relation, with The increase of the time of transmission, the reception time should be presented the trend of linear increase, that is, meet formula (1)：

Wherein,It is the least-squares estimation of a, b and y；

Using the method for least square method, regression equations of the y on x can be calculated；

Using 20 independent tests to equipment under test, sample is can obtain：

(Lacal_Data_Time₁,Global_Data_Time₁),

(Lacal_Data_Time₂,Global_Data_Time₂),

...,

(Lacal_Data_Time₂₀,Global_Data_Time₂₀),

The least-squares estimation of parameter a, b can be calculated, respectively as shown in formula (2) and (3)：

Wherein,WithThe least-squares estimation of b and a is represented,WithRepresent the expectation of sample；

The regression side of reception time of the transmission time of data request command message on data response command message Shown in journey, such as formula (4)：

Judge that the equation of linear regression set up is meaningful, otherwise it is assumed that the transmission time or the time of receiving of system are incorrect；

Two kinds of situation maps of packet sending and receiving time in the present invention are illustrated in figure 5, wherein there are indivedual points to be scattered in 5 (a) Outside error range, but most points belong to superframe normal operation on regression equation；A little all fall in error in 5 (b) In the range of, but it is basic not on regression equation, belong to super frame tim e skew；The straight line obtained using formula (4) should be with Sample fitting obtains best straight line, and the regression equation set up for utility formula (4) is meaningful, it is necessary to x, Linear relationship between y carries out significance test；

(5) are assumed in checking, if H₀Set up, illustrate to return not significantly, superframe is not properly executed, and otherwise illustrates back Return significantly, i.e., superframe correctly runs；

H₀:B=0, H₁:b≠0 (5)

Each sample point (x_i,y_i) may be by formula (4) and obtain a corresponding pointsFormula (6) can be utilized Residual sum of squares (RSS) is calculated, regression sum of square is calculated using formula (7)；

Wherein, It is regression straight line slope, Q_eIt is residual sum of squares (RSS), U is regression sum of square；

Can be pushed away using the definition of the theorem and F substeps of regression equation significance test, H₀During establishment, formula (8) into It is vertical；

When α=0.05, look into F distribution tables and understand F_0.05(1,20)=4.35；

As F ＞ F_aNegate H when (1.n-2)₀, that is, return significantly, otherwise, then it is assumed that return not notable.

The transmission delay of message can be determined using formula (9)：

Time_Delay_i=Local_Data_Time_i-Global_ACK_Time_i (9)

It is inclined super frame tim e can be calculated using the time for receiving data response message and the timestamp of data response message Shifting value Time_Offset_i.In theory, super frame tim e deviant should be in response to the unbiased esti-mator that message receives the time；

According to the error requirements of WIA-PA testing standards, if the unbiased estimator of super frame tim e deviant possesses effectively Property, then it is assumed that superframe is accurate, otherwise, then it is assumed that equipment under test superframe mistake；

Wherein, θ is that the response message calculated by formula (4) receives the time,It is super frame tim e deviant,It is message Transmission delay；

Compare the conclusion that obtains of mean square error using formula (10) although it cannot be guaranteed thatThanUnbiasedness more preferably, it is but right Can more accurately, more reflect in this test processDegrees of offset.

Claims (8)

1. a kind of superframe method of testing of procedure-oriented automatic industrial wireless network, it is characterised in that comprise the following steps：

Test system and equipment under test are set up into the structure of the hub-and-spoke configuration of single-hop or star-like and netted combination；

Make equipment under test and test system synchronization using time synchronized command frame, equipment under test is set using resource allocation command frame Superframe information so that equipment under test can be received and dispatched continuously；

Test system periodically sends request of data message in each sending time slots, and waits ACK and data response message；

The response time for the sending time and data response message stamp of statistics request message, using least square method, according to The reception time of regression equation calculation data response message, and the conspicuousness of regression equation is detected, if returned significantly, continue Next step；If returning not notable, superframe mistake；

Super frame tim e deviant is calculated using the timestamp for receiving time and data response message of data response message, and is checked The super frame tim e deviant whether be data response message reception the time unbiased esti-mator；

If it is, thinking unbiased esti-mator effectively, super frame tim e skew is illustrated in the range of acceptable error, superframe is accurate, no Then think that unbiased esti-mator is invalid, superframe mistake.

2. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The equipment under test includes routing device and field apparatus；

Wherein, field apparatus is arranged on industry spot, for being responsible for connection or control production process；Routing device is responsible for scene and is set Standby management and message are forwarded.

3. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The test system periodically sends request of data message in each sending time slots, and waits ACK and data response message, If test system receives data response message, then it is assumed that be once effective data access process；If not receiving response Message, then it is assumed that this data access process inefficient, is not analyzed.

4. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The a plurality of link that can continuously receive and dispatch is configured in the superframe of the equipment under test.

5. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The regression equation is equation of linear regression：

$\hat{y}=\hat{a}+\hat{b}x---\left(1\right)$

Wherein,It is the least-squares estimation of y, y is the reception time of data response message, when x is the transmission of request of data message Between,

$\hat{b}=\frac{n\underset{i=1}{\overset{n}{\Σ}}{x}_i{y}_i-\left(\underset{i=1}{\overset{n}{\Σ}}{x}_i\right)\left(\underset{i=1}{\overset{n}{\Σ}}{y}_i\right)}{n\underset{i=1}{\overset{n}{\Σ}}{x}_i^2-{\left(\underset{i=1}{\overset{n}{\Σ}}{x}_i\right)}^2}=\frac{\underset{i=1}{\overset{n}{\Σ}}(x_i-\stackrel{\‾}{x})(y_i-\stackrel{\‾}{y})}{\underset{i=1}{\overset{n}{\Σ}}{(x_i-\stackrel{\‾}{x})}^2}---\left(2\right)$

$\hat{a}=\frac{1}{n}\underset{i=1}{\overset{n}{\Σ}}{y}_i-\frac{\hat{b}}{n}\underset{i=1}{\overset{n}{\Σ}}{x}_i=\stackrel{\‾}{y}-\hat{b}\stackrel{\‾}{x}---\left(3\right)$

Wherein, (x_i,y_i) it is sample point,It is sample data x₁,x₂...x_nAverage value,It is sample data y₁,y₂...y_n's Average value, n is the number of times that event occurs, the i.e. number of sample point.

6. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The conspicuousness of the inspection regression equation, specially：

$F=\frac{U}{Q_e/(n-2)}~F(1,n-2)---\left(8\right)$

As F ＞ F_aWhen (1.n-2), return significantly, otherwise, then it is assumed that return not notable；Wherein, α is quantile, is distributed by looking into F Table understands F_a(1, n-2), n is the number of times that event occurs, the i.e. number of sample point, Q_eIt is residual sum of squares (RSS), U is regression sum of square, And

$Q_e=\underset{i=1}{\overset{n}{\Σ}}{(y_i-\hat{y})}^2---\left(6\right)$

$U=\underset{i=1}{\overset{n}{\Σ}}{({\hat{y}}_i-\stackrel{\‾}{y})}^2=\underset{i=1}{\overset{n}{\Σ}}{(\hat{a}+\hat{b}{x}_i-\stackrel{\‾}{y})}^2=\underset{i=1}{\overset{n}{\Σ}}{(\stackrel{\‾}{y}-\hat{b}\stackrel{\‾}{x}+\hat{b}{x}_i-\stackrel{\‾}{y})}^2={\hat{b}}^2\underset{i=1}{\overset{n}{\Σ}}{(x_i-\stackrel{\‾}{x})}^2---\left(7\right)$

Wherein, It is regression straight line slope, (x_i,y_i) it is sample point,It is sample data x₁,x₂...x_n Average value,It is sample data y₁,y₂...y_nAverage value,It is the least-squares estimation of y.

7. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The super frame tim e deviant is the difference of the timestamp for receiving time and data response message of data response message.

8. the superframe method of testing of procedure-oriented automatic industrial wireless network according to claim 1, it is characterised in that The inspection super frame tim e deviant whether be data response message the reception time unbiased esti-mator, that is, detect that unbiased is estimated The validity of meter, specially：

$E{({\widehat{\mathrm{\&theta;}}}_1-\mathrm{\&theta;})}^2<E{({\widehat{\mathrm{\&theta;}}}_2-\mathrm{\&theta;})}^2---\left(10\right)$

If formula (10) is set up, unbiased esti-mator is effective；Otherwise, unbiased esti-mator is invalid；It is super frame tim e deviant, θ is The response message calculated by formula (4) receives the time, i.e.,

$\hat{y}=(\stackrel{\&OverBar;}{y}-\hat{b}\stackrel{\&OverBar;}{x})+\frac{\underset{i=1}{\overset{n}{\&Sigma;}}(x_i-\stackrel{\&OverBar;}{x})(y_i-\stackrel{\&OverBar;}{y})}{\underset{i=1}{\overset{n}{\&Sigma;}}{(x_i-\stackrel{\&OverBar;}{x})}^2}x---\left(4\right)$

Wherein,It is the least-squares estimation of b,It is the expectation of x and y；

It is the difference of the reception time of transmission time and ACK that message transmitting time delay is data message.