A Passive Positioning System and Method for Coal Mine Underground with Multi-Frequency Passive Tags

Abstract

The present invention discloses a multi-frequency passive tag-based underground coal mine passive positioning system and method, belonging to the field of underground coal mine monitoring. It includes underground positioning terminals, passive positioning micro base stations, passive positioning passive tags, and an above-ground positioning server; the underground passive positioning micro base stations can be installed in roadways, fully mechanized coal mining faces, etc., and the positioning terminals can be carried by personnel or underground equipment, and the position of the positioning terminal is the position of underground personnel or equipment; the above-ground positioning server processes the collected passive tag numbers, RSSI, frequencies, and phase information, and performs calculations based on the known positions of the passive tags and the processed information, and can obtain the position where the positioning terminal is located. Using the frequencies, phases, and RSSI information of the passive tags as monitoring means to determine the position of the positioning terminal, the equipment is simple, easy to deploy, the cost of the system is low, and the positioning accuracy is high.

Description

Technical Field

[0001] The present invention relates to a passive positioning system and method for coal mines, and particularly to a passive positioning system and method for coal mines with multi-frequency passive tags suitable for use in coal mines, belonging to the field of coal mine underground monitoring. Background Art

[0002] The personnel positioning system in coal mines is an important part of the coal mine safety production management system and plays an important role in coal mine production scheduling management, safety management, and accident emergency rescue. The personnel positioning system in coal mines mainly conducts surveillance management on the positioned personnel underground, real-time monitors the dynamic distribution of the positions of underground personnel, and provides data support for underground production operations and attendance management. When an accident occurs, effective evacuation and rescue can be carried out based on the real-time positions of underground personnel, which helps to further improve the success rate of rescue. However, in the current underground positioning system, most of its positioning base stations are powered by electricity. In the event of a disaster, the coal mine system will actively cut off the underground power transmission, resulting in the inability of the existing positioning system to be used in post-disaster scenarios. Therefore, a passive underground positioning system that can be used in multiple scenarios is needed. Summary of the Invention

[0003] Aiming at the deficiencies of the prior art, a passive positioning system and method for coal mines with multi-frequency passive tags are provided. Passive tags are used to construct positioning passive positioning micro base stations, and information such as the frequency, phase, and signal strength of the passive tags is used as the basis for estimating the position of personnel. It is particularly suitable for scenarios where the environment is not ideal, position information needs to be obtained temporarily under special circumstances, and the power supply is cut off after a disaster.

[0004] To achieve the above technical objectives, a passive positioning system for coal mines with multi-frequency passive tags of the present invention includes a positioning server arranged on the ground, and a wireless access point, a positioning terminal, and a passive positioning micro base station arranged underground;

[0005] Among them, the passive positioning micro base station includes three passive positioning passive tags. The three passive positioning passive tags are installed on a metal backplane that is convenient for on-site installation. The passive positioning passive tags are passive RFID passive tags. The placement interval between the three passive RFID passive tags is less than half of the communication frequency wavelength, so as to reduce the calculation error caused by phase ambiguity when calculating the distance. The three passive RFID passive tags in the passive positioning micro base station are equivalent to three base stations, and precise positioning can be achieved by only setting one passive positioning micro base station. Each passive positioning passive tag is preset with independent identification information;

[0006] The passive positioning micro base stations are installed at intervals in the roadways and fully mechanized coal mining faces of the detection area as required. Each passive positioning micro base station is composed of three passive RFID tags. The personnel to be positioned underground carry a positioning terminal with them. The positioning terminal carried by the personnel to be positioned collects the CLUSTER, ID, EPC, RSSI information, as well as the frequency and phase information of the passive tags on the passive positioning micro base stations. Then, the information fed back by the positioning terminal is received through the wireless access points arranged in the roadways and fully mechanized coal mining faces of the detection area, and is transmitted to the positioning server through the underground wireless transmission network. Among them, the information of the passive tags on the passive positioning micro base stations is pre-recorded in the database; the positioning server processes the information such as the passive tag number, RSSI, frequency, phase, etc., and then obtains the position of the positioning terminal.

[0007] A passive positioning method for coal mines using multi-frequency passive tags, the steps are as follows:

[0008] S101, construct passive positioning micro base stations in the underground working scenario equipped with wireless access points and base stations. With a step interval of 10 cm for the positioning terminal, successively measure the signal strength values of all passive tags on the passive positioning micro base stations when the positioning terminal is 10 cm, 20 cm, up to 5 m away from the passive positioning micro base stations. Use the ElasticNet regression algorithm to establish a model of the relationship between the signal strength value and the distance:

[0009] S102, relying on the model of the relationship between the signal strength value and the distance, arrange multiple passive positioning micro base stations in the positioning scenario according to the actual positioning accuracy requirements. The coordinate information of the passive positioning micro base stations is stored in the positioning server. The positioning terminal collects the CLUSTER, ID, EPC, RSSI, frequency and phase information stored in the passive tags inside the passive positioning micro base stations. The positioning terminal emits signals of different frequencies as it moves with the personnel to be positioned in the scenario where the passive positioning micro base stations are arranged. The signals activate the passive positioning tags on the passive positioning micro base stations around the personnel to be positioned, and obtain the CLUSTER, ID, EPC, RSSI, frequency and corresponding phase information of each passive positioning tag;

[0010] S103, after the positioning terminal receives all the passive tag information returned by the passive positioning micro base stations, transmit the information to the positioning server and process it to enhance the signal;

[0011] S104, calculate the distance information fed back by the three passive positioning tags in the passive positioning micro base station with the strongest received signal, and use the mean algorithm to obtain the filtered data;

[0012] S105, subtracting the distances obtained in S104 in pairs, and then combining the coordinate information of the passive positioning passive tag on the passive positioning micro base station, and using the basic definition of the hyperbola equation to establish a positioning terminal coordinate estimation equation;

[0013] S106, using an optimization algorithm to find the coordinates (x, y) of the positioning terminal.

[0014] Furthermore, the model building process of the relationship between signal strength value and distance is as follows:

[0015] Assume that there are three passive tags in the passive positioning micro base station. The positioning terminal is measured at a step interval of 10 cm. The positioning terminal is measured at 10 cm, 20 cm, and 5 meters from the passive positioning micro base station. The signal strength values ​​of all passive tags on the passive positioning micro base station are summarized as follows:

[0016]

[0017] The subscript (1, 1) indicates the signal strength value of passive tag 1 when the positioning terminal is 10 cm away from the passive positioning micro base station, and the rest of the values ​​are similar.

[0018] 1) Establish a regression model as shown in formula (1),

[0019]

[0020] Where X is the signal strength value matrix, is the coefficient;

[0021] 2) Construct loss function, construct loss function As shown in formula (2):

[0022]

[0023] Where n is the number of measurements, α and β are parameters, is the distance vector;

[0024] 3) Solve for the optimal α, β, According to formula (3), the optimal α, β, Complete model building.

[0025]

[0026] When the optimal α,β, After that, all the parameters of the model are known, and the corresponding distance can be obtained by simply inputting the information signal strength value vector.

[0027] Further, the form of the information returned by the passive positioning passive tags on the passive positioning micro base station composed of three passive tags is as shown in formula (4):

[0028]

[0029] In the formula: S 1 represents the information set returned after the passive positioning micro base station receives the positioning terminal excitation signal for the first time, and T 11 indicates that in the case of multiple measurements, the passive tag 1 in the passive positioning micro base station first returns the information to the receiving end, and T 21 indicates that the passive tag 2 first returns the information to the receiving end, and T 31 indicates that the passive tag 3 first returns the information to the receiving end; since the electromagnetic wave frequencies sent by the positioning terminal are different each time, the phases (pha) and frequencies (fre) received from the passive tags on the passive positioning micro base station are also different each time; where id represents the label of the passive positioning micro base station, and rssi 11 , pha 11 , fre 11 respectively represent the signal strength value, frequency, and corresponding phase information returned by the passive tag with the EPC number epc 1 in the passive positioning micro base station with the id number after receiving the excitation signal of the positioning terminal for the first time. The information returned by the passive tag 2 and the passive tag 3 is similar.

[0030] Further, the specific steps for the positioning server to process the first passive positioning micro base station arranged in the scene are as follows:

[0031] Suppose n groups of data S = [S 1 , S 2 , S 3 , … S n are collected. The positioning server uses Gaussian filtering to remove outliers from the collected data. Since the communication frequency is different for each positioning, the frequencies and phases returned by the passive tags in the passive positioning micro base station are also different;

[0032] First, use various frequency and phase information to calculate the distances D1 = [d 1 , d 2 , d 3 from the positioning terminal to the three passive tags in the passive positioning micro base station. The specific calculation process for the first passive tag in the first passive positioning micro base station is as follows:

[0033] Since the value range of the positioning terminal in the positioning terminal is 0 to 2π, when the distance is far, the distances that are integer multiples of 2π will generate the same phase. Therefore, it is impossible to directly use the phase information to solve for the distance. It is necessary to establish a congruence equation using the phases of multiple frequency bands and then use the Chinese Remainder Theorem to solve the moduli of the flight times used to measure different values:

[0034] Suppose the positioning terminal emits excitation signals of n frequencies in total when obtaining n groups of measurement data, which are represented by {f 1 , f 2 , f 3 ... f n}. The phases returned by the first passive tag after receiving the excitation signal are represented by {ph 1 , ph 2 , … ph n}. Since the speeds of electromagnetic waves of different frequencies are the same, all being the speed of light c, therefore, at the same distance, the flight times τ of electromagnetic waves of different frequencies are the same. Thus, formula (5) is established

[0035]

[0036] According to the Chinese Remainder Theorem, it is obtained that:

[0037]

[0038] where

[0039] d 1 = c * τ / 2

[0040] d 1 represents the distance from the positioning terminal to the first tag in the micro base station obtained using the flight time. Similarly, the distance d 2 from the positioning terminal to the second tag in the micro base station obtained using the flight time is calculated, as well as the distance d 3 to the third tag. Then, the distance d 4 from the positioning terminal to the micro base station is obtained through the model of the relationship between the signal strength value and the distance for the processed signal strength value.

[0041] S104, using the mean algorithm, perform mean processing on d 4 respectively with d 1 , d 2 , d 3 to obtain the filtered distances dp 1 , dp 2 , dp 3 ;

[0042] dp 1 = k * d 4 + (1 - k)d1

[0043] dp 2 = k * d 4 + (1 - k)d 2

[0044] dp 3 = k * d 4 + (1 - k)d 3 。

[0045] Furthermore, the filtered distance dp 1 , dp 2 , dp 3 is subtracted pairwise,

[0046] Delt1 = dp 1 - dp 2

[0047] Delt2 = dp 2 - dp 3

[0048] Then, combined with the coordinate information of the passive tags on the passive positioning micro base station, a positioning terminal coordinate estimation equation is established using the basic definition of the hyperbola equation:

[0049] Since the passive positioning micro base station is pre-deployed, the coordinates of the three passive tags on the passive positioning micro base station are known, assumed to be (x1, y1), (x2, y2), (x3, y3), and the coordinates of the positioning terminal are represented by (x, y). The following equation is established

[0050]

[0051] Use the optimization algorithm to find a set of (x, y) values that minimize the equation f(x, y). At this time, (x, y) is the coordinate of the positioning terminal.

[0052] Beneficial effects: The present invention uses passive positioning passive tags to construct a passive positioning micro base station, establishes an underground passive positioning system for positioning underground personnel. The passive positioning micro base station does not require power supply, has a simple structure, low deployment cost of the system, and high positioning accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a schematic structural diagram of the underground passive positioning system of the passive tags of the present invention;

[0054] Figure 2 is a flow chart of the underground passive positioning method of the passive tags of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0055] The present invention will be further illustrated below in conjunction with the accompanying drawings and specific embodiments.

[0056] As shown in the Figure 1 accompanying drawings,

[0057] A coal mine underground passive positioning system and method for a multi - frequency passive tag of the present invention include an underground positioning passive positioning micro - base station, a positioning terminal carried on a person or an object for collecting positioning information, and an above - ground positioning server; the positioning information is a set of information returned from a passive tag on the passive positioning micro - base station to the positioning terminal after the passive tag is activated by an excitation signal sent by the positioning terminal, and this information is uploaded by the positioning terminal to the above - ground positioning server through the underground wireless network; the above - ground positioning server processes the received positioning information to obtain the coordinates of the positioning terminal.

[0058] The coal mine underground passive positioning system with multi - frequency passive tags includes a positioning server arranged above ground and a wireless access point, a positioning terminal, and a passive positioning micro - base station arranged underground;

[0059] Among them, the passive positioning micro - base station includes three passive positioning passive tags, and the three passive positioning passive tags are installed on a metal backplane that is convenient for on - site installation. The passive positioning passive tag is a passive RFID passive tag, and the placement interval between the three passive RFID passive tags is less than half of the communication frequency wavelength, so as to reduce the calculation error caused by phase ambiguity when calculating the distance. The three passive RFID passive tags in the passive positioning micro - base station are equivalent to three base stations, and only one passive positioning micro - base station needs to be set to achieve precise positioning. Each passive positioning passive tag is preset with independent identification information;

[0060] The passive positioning micro - base stations are installed at intervals as needed in the roadways and fully - mechanized mining faces in the detection area. Each passive positioning micro - base station is composed of three passive RFID passive tags. The underground personnel to be positioned carry a positioning terminal with them. The positioning terminal carried by the personnel to be positioned collects the CLUSTER, ID, EPC, RSSI information, as well as frequency and phase information of the passive tags on the passive positioning micro - base station, and then receives the information fed back by the positioning terminal through the wireless access points arranged in the roadways and fully - mechanized mining faces in the detection area, and transmits it to the above - mentioned positioning server through the underground wireless transmission network by using an explosion - proof network switch. Among them, the information of the passive tags on the passive positioning micro - base station is pre - recorded in the database; the positioning server processes the information such as the passive tag number, RSSI, frequency, phase, etc., and then obtains the position of the positioning terminal.

[0061] As Figure 2 shown, a coal mine underground passive positioning method for a multi - frequency passive tag of the present invention includes an offline training part and an online prediction part. The specific steps are as follows:

[0062] S101, build a passive positioning micro base station in an underground working scene where wireless access points and base stations are installed, and measure the signal strength values ​​of all passive tags on the passive positioning micro base station when the positioning terminal is 10 cm, 20 cm, and 5 meters away from the passive positioning micro base station at a step interval of 10 cm. Use the ElasticNet regression algorithm to establish a model of the relationship between signal strength value and distance:

[0063] The model building process of the relationship between signal strength value and distance is as follows:

[0064] Assume that there are three passive tags in the passive positioning micro base station. The positioning terminal is measured at a step interval of 10 cm. The positioning terminal is measured at 10 cm, 20 cm, and 5 meters from the passive positioning micro base station. The signal strength values ​​of all passive tags on the passive positioning micro base station are summarized as follows:

[0065]

[0066] The subscript (1, 1) indicates the signal strength value of passive tag 1 when the positioning terminal is 10 cm away from the passive positioning micro base station, and the rest of the values ​​are similar.

[0067] 1) Establish a regression model as shown in formula (1),

[0068]

[0069] Where X is the signal strength value matrix, is the coefficient;

[0070] 2) Construct loss function, construct loss function As shown in formula (2):

[0071]

[0072] Where n is the number of measurements, α and β are parameters, is the distance vector;

[0073] 3) Solve for the optimal α, β, According to formula (3), the optimal α, β, Complete model building.

[0074]

[0075] When the optimal α,β, After that, all the parameters of the model are known, and the corresponding distance can be obtained by simply inputting the information signal strength value vector.

[0076] S102. According to the actual positioning accuracy requirements, based on the model of the relationship between signal strength value and distance, multiple passive positioning micro base stations are arranged in the positioning scenario. The coordinate information of the passive positioning micro base stations is stored in the positioning server. The positioning terminal collects the CLUSTER, ID, EPC, RSSI, frequency, and phase information stored in the passive tags within the passive positioning micro base stations. As the positioning terminal moves with the person to be located in the scenario where the passive positioning micro base stations are arranged, it emits signals of different frequencies. The signals activate the passive positioning passive tags on the passive positioning micro base stations around the person to be located, and the CLUSTER, ID, EPC, RSSI, frequency, and corresponding phase information of each passive positioning passive tag are obtained.

[0077] The form of the information returned by the passive positioning passive tags on the passive positioning micro base station composed of three passive tags is shown in formula (4):

[0078]

[0079] In the formula: S 1 represents the information set returned after the passive positioning micro base station receives the excitation signal from the positioning terminal for the first time. T 11 indicates that in the case of multiple measurements, passive tag 1 in the passive positioning micro base station first returns the information to the receiving end. T 21 indicates that passive tag 2 first returns the information to the receiving end. T 31 indicates that passive tag 3 first returns the information to the receiving end. Since the electromagnetic wave frequencies sent by the positioning terminal are different each time, the phase (pha) and frequency (fre) received from the passive tags on the passive positioning micro base station are also different each time. Where id represents the label of the passive positioning micro base station, rssi 11 , pha 11 , fre 11 respectively represent the signal strength value, frequency, and corresponding phase information returned by the passive tag with the EPC number epc 1 in the passive positioning micro base station with the id number after receiving the excitation signal from the positioning terminal for the first time. The information returned by passive tag 2 and passive tag 3 is similar.

[0080] S103. After the positioning terminal receives all the passive tag information returned by the passive positioning micro base stations, it transmits the information to the positioning server and processes it to enhance the signal.

[0081] S104. Calculate the distance information fed back by the three passive positioning passive tags in the passive positioning micro base station with the strongest received signal, and use the mean algorithm to obtain the filtered data.

[0082] The specific steps for the positioning server to process the first passive positioning micro base station arranged in the scenario are as follows:

[0083] Suppose n groups of data S = [S 1 , S 2 , S 3 , … S n are collected. The positioning server uses Gaussian filtering to remove outliers from the collected data. Since the communication frequency is different for each positioning, the frequencies and phases returned by each passive tag in the passive positioning micro base station are also different;

[0084] First, use various frequency and phase information to calculate the distances D1 = [d 1 , d 2 , d 3 from the positioning terminal to the three passive tags in the passive positioning micro base station. The specific calculation process for the first passive tag in the first passive positioning micro base station is as follows:

[0085] Since the value range of the positioning terminal in the positioning terminal is 0 to 2π, when the distance is far, distances that are integer multiples of 2π will produce the same phase. Therefore, it is impossible to directly use the phase information to solve for the distance. It is necessary to establish a congruence equation using the phases of multiple frequency bands and then use the Chinese Remainder Theorem to solve the moduli of the flight times used to measure different values:

[0086] Suppose the positioning terminal emits excitation signals of n frequencies in total when obtaining n groups of measurement data, represented by {f 1 , f 2 , f 3 ... f n}. The phases returned by the first passive tag after receiving the excitation signal are represented by {ph 1 , ph 2 , … ph n}. Since the speeds of electromagnetic waves of different frequencies are the same, all being the speed of light c, therefore, at the same distance, the flight times τ of electromagnetic waves of different frequencies are the same. Thus, formula (5) is established

[0087]

[0088] According to the Chinese Remainder Theorem, we get:

[0089]

[0090] Where

[0091] We get d 1 = c * τ / 2

[0092] d 1 represents the distance from the positioning terminal to the first tag in the micro base station obtained using the flight time. Similarly, calculate the distance d from the positioning terminal to the second tag in the micro base station obtained using the flight time2 , and the distance d of the third tag 3 , and then obtain the distance d from the positioning terminal to the micro base station through the model of the relationship between the signal strength value and the distance with the processed signal strength value 4 .

[0093] Using the mean algorithm, for d 4 respectively perform mean processing with d 1 , d 2 , d 3 to obtain the filtered distance dp 1 , dp 2 , dp 3 ;

[0094] dp 1 = k * d 4 +(1 - k)d 1

[0095] dp 2 = k * d 4 +(1 - k)d 2

[0096] dp 3 = k * d 4 +(1 - k)d 3 .

[0097] S105. Subtract the distances obtained in S104 pairwise, and then combine the coordinate information of the passive positioning passive tags on the passive positioning micro base station, and use the basic definition of the hyperbola equation to establish an estimated equation for the coordinates of the positioning terminal;

[0098] Using the mean algorithm, for d 4 respectively perform mean processing with d 1 , d 2 , d 3 to obtain the filtered distance dp 1 , dp 2 , dp 3 ;

[0099] dp 1 = k * d 4 +(1 - k)d 1

[0100] dp 2 = k * d 4 +(1 - k)d 2

[0101] dp 3 = k * d 4 +(1 - k)d 3 .

[0102] The filtered distance dp is used with the following formula 1 , dp 2 , dp 3 to perform pairwise subtraction.

[0103] Delt1 = dp 1 - dp 2

[0104] Delt2 = dp 2 - dp 3

[0105] Combined with the coordinate information of the passive tags on the passive positioning micro base station, a positioning terminal coordinate estimation equation is established using the basic definition of the hyperbola equation:

[0106] S106. Use the optimization algorithm to find the coordinates (x, y) of the positioning terminal. Since the passive positioning micro base stations are pre-deployed, the coordinates of the three passive tags on the passive positioning micro base stations are known, assumed to be (x1, y1), (x2, y2), (x3, y3), and the coordinates of the positioning terminal to be located are represented by (x, y). The following equation is established

[0107]

Claims

1. A passive positioning system for coal mines using multi - frequency passive tags, characterized in that: it includes a positioning server arranged above the well and a wireless access point, a positioning terminal, and a passive positioning micro - base station arranged below the well; wherein the passive positioning micro - base station includes three passive positioning passive tags. The three passive positioning passive tags are installed on a metal backplane that is convenient for on - site installation. The passive positioning passive tags are passive RFID passive tags. The placement interval between the three passive RFID passive tags is less than half of the communication frequency wavelength, so as to reduce the calculation error caused by phase ambiguity when calculating the distance. The three passive RFID passive tags in the passive positioning micro - base station are equivalent to three base stations, and only one passive positioning micro - base station needs to be set up to achieve precise positioning. Each passive positioning passive tag is preset with independent identification information; The passive positioning micro - base stations are installed at intervals as required in the roadways and fully - mechanized mining faces in the detection area. Each passive positioning micro - base station is composed of three passive RFID passive tags. The positioned personnel underground carry a positioning terminal. The positioning terminal carried by the positioned personnel collects the CLUSTER, ID, EPC, RSSI information, as well as frequency and phase information of the passive tags on the passive positioning micro - base station, and then receives the information fed back by the positioning terminal through the wireless access points arranged in the roadways and fully - mechanized mining faces in the detection area, and transmits it to the above - mentioned positioning server through the underground wireless transmission network. Among them, the information of the passive tags on the passive positioning micro - base station is pre - recorded in the database; the positioning server processes the passive tag number, RSSI, frequency, and phase information to obtain the position of the positioning terminal.

2. A passive positioning method for coal mines using multi - frequency passive tags of the passive positioning system for coal mines using multi - frequency passive tags described in claim 1, characterized in that the steps are as follows: S101, construct a passive positioning micro - base station in the underground working scenario equipped with wireless access points and base stations. With a step interval of 10 cm for the positioning terminal, measure the signal strength values of all the passive tags on the passive positioning micro - base station when the positioning terminal is 10 cm, 20 cm, up to 5 m away from the passive positioning micro - base station in sequence, and establish a model of the relationship between the signal strength value and the distance using the ElasticNet regression algorithm: S102, relying on the model of the relationship between the signal strength value and the distance, arrange multiple passive positioning micro - base stations in the positioning scenario according to the actual positioning accuracy requirements. The coordinate information of the passive positioning micro - base stations is stored in the positioning server. The positioning terminal collects the CLUSTER, ID, EPC, RSSI, frequency, and phase information stored in the passive tags inside the passive positioning micro - base station. The positioning terminal emits signals of different frequencies as it moves with the positioned personnel in the scenario where the passive positioning micro - base stations are arranged. The signals activate the passive positioning passive tags on the passive positioning micro - base stations around the positioned personnel, and obtain the CLUSTER, ID, EPC, RSSI, frequency, and corresponding phase information of each passive positioning passive tag; S103, after receiving all passive tag information returned by the passive positioning micro base station, the positioning terminal transmits the information to the positioning server and processes it to enhance the signal; S104, calculating the distance information fed back by three passive positioning passive tags in the passive positioning micro base station with the strongest receiving signal, and using a mean algorithm to obtain filtered data; S105, subtracting the distances obtained in S104 in pairs, and then combining the coordinate information of the passive positioning passive tag on the passive positioning micro base station, and using the basic definition of the hyperbola equation to establish a positioning terminal coordinate estimation equation; S106, using an optimization algorithm to find the coordinates (x, y) of the positioning terminal.

3. According to the passive positioning method for coal mines using the multi-frequency passive tag of claim 2, Features The model building process of the relationship between signal strength value and distance is as follows: Assume that there are three passive tags in the passive positioning micro base station. The positioning terminal is set at a step interval of 10 cm. The positioning terminal is measured at a distance of 10 cm, 20 cm, and 5 meters from the passive positioning micro base station. The signal strength values ​​of all passive tags on the passive positioning micro base station are summarized as follows: The subscript (10, 1) indicates the signal strength value of passive tag 1 when the positioning terminal is 10 cm away from the passive positioning micro base station, and the rest of the values ​​are similar. 1) Establish a regression model as shown in formula (1): where X is a signal strength value matrix, is a coefficient; 2) Construct a loss function, construct a loss function As shown in formula (2): where n is the number of multiple measurements, and α, β are parameters, is the distance vector; 3) Solve for the optimal α and β, Use an optimization algorithm to find the optimal α and β according to formula (3). Complete the model establishment; When the optimal α and β are solved, all the parameters of the model are known. Just substituting the vector of information signal strength values, the corresponding distances can be obtained.

4. The method for passive positioning in coal mines using a multi-frequency passive tag according to claim 2, Features The form of the passive positioning passive tag return on the passive positioning micro base station composed of three passive tags is shown in formula (4): Where: S 1 represents the information set returned by the passive positioning micro base station after receiving the positioning terminal excitation signal for the first time, and T 11 indicates that in the case of multiple measurements, the passive tag 1 in the passive positioning micro base station first returns information to the receiving end, and T 21 indicates that the passive tag 2 first returns information to the receiving end, and T 31 indicates that the passive tag 3 first returns information to the receiving end; since the electromagnetic wave frequencies sent by the positioning terminal are different for each measurement, the phase pha and frequency fre received from the passive tags on the passive positioning micro base station are also different each time; where id represents the label of the passive positioning micro base station, and rssi 11 , pha 11 , fre 11 respectively represent the signal strength value, frequency, and corresponding phase information returned by the passive tag with the EPC number epc 1 in the passive positioning micro base station numbered id after receiving the excitation signal of the positioning terminal for the first time. The information returned by the passive tag 2 and the passive tag 3 is similar.

5. The method for passive positioning in coal mines using a multi-frequency passive tag according to claim 3, Features The specific steps for the positioning server to process the first passive positioning micro base station deployed in the scene are: Suppose n groups of data S = [S 1 , S 2 , S 3 , L, S n are collected. The positioning server uses Gaussian filtering to remove outliers from the collected data. Since the communication frequency is different for each positioning, the frequencies and phases returned by each passive tag in the passive positioning micro base station are also different; First, use multiple frequency and phase information to calculate the distances D1 = [d 1 , d 2 , d 3 from the positioning terminal to the three passive tags in the passive positioning micro base station. The specific calculation process for the first passive tag in the first passive positioning micro base station is as follows: Since the value range of the positioning terminal in the positioning terminal is 0~2π, when the distance is far, the distance of integer multiples of 2π will produce the same phase. Therefore, it is impossible to directly use the phase information to solve the distance. It is necessary to establish a congruence equation with the phases of multiple frequency bands, and then use the Chinese remainder theorem to measure the flight time modulus of different values: Suppose the positioning terminal emits excitation signals of n frequencies in total when acquiring n groups of measurement data, which are represented by {f 1 , f 2 , f 3 , …, f n}. The phases returned by the first passive tag after receiving the excitation signals are represented by {ph 1 , ph 2 , …, ph n}. Since the speeds of electromagnetic waves of different frequencies are the same, all being the speed of light c, therefore, at the same distance, the flight times τ of electromagnetic waves of different frequencies are the same. Thus, formula (5) is established According to the Chinese remainder theorem, we get: Among them Obtain d 1 = c * τ / 2 d 1 Indicates the distance from the positioning terminal to the first tag in the micro base station obtained using the time of flight. Similarly, calculate the distance d from the positioning terminal to the second tag in the micro base station obtained using the time of flight 2 , and the distance d to the third tag 3 . Then, obtain the distance d from the positioning terminal to the micro base station through the model of the relationship between the signal strength value and the distance by using the processed signal strength value 4 .

6. The method for passive positioning in coal mines using a multi-frequency passive tag according to claim 5, Features: Using the mean algorithm, d 4 is respectively averaged with d 1 , d 2 , d 3 to obtain the filtered distances dp 1 , dp 2 , dp 3 ; dp 1 = k * d 4 + (1 - k)d 1 dp 2 = k * d 4 + (1 - k)d 2 dp 3 = k * d 4 + (1 - k)d 3 .

7. The method for passive positioning in coal mines using a multi-frequency passive tag according to claim 6, Features: The filtered distance dp is subtracted pairwise using the following formula 1 , dp 2 , dp 3 to obtain the pairwise differences Delt1 = dp 1 -dp 2 Delt2 = dp 2 -dp 3 Combined with the coordinate information of the passive tag on the passive positioning micro base station, the basic definition of the hyperbolic equation is used to establish the positioning terminal coordinate estimation equation: Since the passive positioning micro base station is pre-deployed, the coordinates of the three passive tags on the passive positioning micro base station are known, assuming that they are (x1, y1), (x2, y2), and (x3, y3). (x, y) is used to represent the coordinates of the terminal to be located, and the following equation is established: Minf(x,y) Use the optimization algorithm to find a set of (x, y) values ​​that minimizes the equation f(x, y). At this time, (x, y) is the coordinate of the positioning terminal.

Images