CN109342998B - Bionic-based position fingerprint indoor positioning system and method - Google Patents

Bionic-based position fingerprint indoor positioning system and method Download PDF

Info

Publication number
CN109342998B
CN109342998B CN201811488576.1A CN201811488576A CN109342998B CN 109342998 B CN109342998 B CN 109342998B CN 201811488576 A CN201811488576 A CN 201811488576A CN 109342998 B CN109342998 B CN 109342998B
Authority
CN
China
Prior art keywords
pulse
signal
acceleration sensor
signals
stepping
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811488576.1A
Other languages
Chinese (zh)
Other versions
CN109342998A (en
Inventor
侯涛
权美静
刘富
王柯
韩志武
游子跃
赵宇峰
王跃桥
宋阳
姜守坤
刘云
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN201811488576.1A priority Critical patent/CN109342998B/en
Publication of CN109342998A publication Critical patent/CN109342998A/en
Application granted granted Critical
Publication of CN109342998B publication Critical patent/CN109342998B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02—Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/06—Position of source determined by co-ordinating a plurality of position lines defined by path-difference measurements
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
    • H04W4/02—Services making use of location information
    • H04W4/025—Services making use of location information using location based information parameters
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W84/00—Network topologies
    • H04W84/18—Self-organising networks, e.g. ad-hoc networks or sensor networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

The invention provides a bionic-based position fingerprint indoor positioning system and a bionic-based position fingerprint indoor positioning method, wherein a stepping signal acquired from an acceleration sensor module is transmitted to an upper computer; and the upper computer receives the stepping signals, constructs pulse fingerprint characteristic vectors corresponding to the stepping signals, matches the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizes user position positioning by using a WKNN algorithm. The bionic scorpion vibration source positioning mechanism takes the pulse number as the fingerprint characteristic, and the fingerprint characteristic is obtained by processing the position information through the bionic scorpion nervous system and is an index which can represent the signal characteristic with finer granularity than RSS, so that the information of the target position can be described more clearly, and the more accurate user position positioning is realized.

Description

Bionic-based position fingerprint indoor positioning system and method
Technical Field
The invention relates to the technical field of indoor positioning, in particular to a bionic position fingerprint indoor positioning system and method.
Background
The indoor positioning technology is a key problem Based on LBS (Location Based Service). The fast and accurate indoor positioning technology can improve the life quality and the work efficiency of people, for example, in a nursing home and a nursing center, the positions of the elderly residents, the disabled and the dementia patients need to be monitored in real time; due to the shielding of buildings, signals of a GNSS (Global Navigation Satellite System) cannot be received indoors, and it becomes difficult to accurately acquire user position information indoors, so that the method has important practical significance for research on indoor positioning technology.
Currently, the indoor positioning method is mainly based on AOA (Angle of Arrival), TOA (Time of Arrival), TDOA (Time Difference of Arrival), and the like. However, these methods have strong dependence on the quality of the signal, and if the signal is slightly attenuated or unstable, the positioning accuracy is reduced, and in a complex indoor environment, the calculation is difficult. Location technology based on LF (Location Fingerprint) is based on RSS (Received Signal Strength), it is relatively easy to obtain information characteristics of Location fingerprints, and it is not necessary to know the Location of AP (Access Point) in advance, and it has been widely used in recent years. Since RSS is the superposition of indoor multipath signals and is easily affected by indoor multipath effects, RSS values measured even in the same indoor environment will change greatly, and thus the performance of the RSS-based location fingerprint positioning method is not stable, resulting in low positioning accuracy.
Therefore, the prior art is subject to further improvement.
Disclosure of Invention
In view of the defects in the prior art, the invention aims to provide a position fingerprint indoor positioning system and method based on bionics, which solve the problems that the indoor position positioning method in the prior art is superposition of multipath signals, and positioning performance is unstable and positioning accuracy is low due to the fact that the indoor position positioning method is easily interfered by indoor multipath effects.
The first embodiment disclosed by the invention is a bionic-based position fingerprint indoor positioning system, which comprises the following components:
the system comprises a plurality of acceleration sensor modules, and an Ethernet data transmission module, a microprocessor and an upper computer which are connected with the acceleration sensors;
the acceleration sensor module is laid out according to the position of a receptor at the tail end of the leg of the scorpion and is used for collecting a stepping signal sent by a user;
the Ethernet data transmission module is used for transmitting the stepping signals collected in the acceleration sensor module to an upper computer;
the microprocessor is used for controlling the Ethernet data transmission module to transmit the stepping signal to the upper computer;
and the upper computer is used for receiving the stepping signals, constructing pulse fingerprint characteristic vectors corresponding to the stepping signals, matching the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizing user position positioning by using a WKNN algorithm.
Optionally, the number of the acceleration sensor modules is 8, the acceleration sensor modules are distributed on a circumference with a radius of a preset length, and the direction of linear advancement of the acceleration sensor modules is set to 0 °, so that the angles of the 8 acceleration sensor modules arranged on the circumference are ± 18 °, ± 54 °, 90 °, and ± 140 °.
Optionally, the system further includes: the analog-to-digital conversion module and the constant current adaptive regulator module;
the analog-to-digital conversion module is used for converting an analog signal of a step signal received from the acceleration sensor into a corresponding digital signal;
the constant-current adapter module is arranged between the acceleration sensor module and the analog-to-digital conversion module and is used for receiving a stepping signal transmitted by the acceleration sensor and adjusting the stepping signal;
the acceleration sensor module, the constant current adaptive regulator module and the analog-to-digital conversion module are electrically connected.
Optionally, the host computer still includes: a pulse fingerprint feature database construction module;
the pulse fingerprint feature database construction module comprises: the device comprises a reference signal acquisition unit, a neuron modeling unit, a neural network construction unit, a signal conversion unit and a pulse fingerprint characteristic vector construction unit;
the reference signal acquisition unit is used for acquiring step signals sent by users at a plurality of reference points arranged in an indoor positioning area;
the neuron modeling unit is used for simulating neurons of scorpions by utilizing an LIF neuron model;
the neural network construction unit is used for establishing a pulse neural network by utilizing the n/1 neuron configuration of the scorpion; extracting n-1, 2,3,5, 7;
the signal conversion unit is used for converting the step signals acquired at each receptor position in the pulse neural network into corresponding pulse signals;
the pulse fingerprint feature vector construction unit is used for constructing pulse fingerprint feature vectors of pulse signals on the positions of the receptors, and constructing a pulse fingerprint feature database by using the pulse fingerprint feature vectors and the relative positions of the pulse fingerprint feature vectors.
A second embodiment provided by the present invention is a method for indoor positioning by the location fingerprint indoor positioning system, wherein the method comprises:
acquiring a stepping signal sent by a user by using an acceleration sensor module; the acceleration sensors are distributed according to the positions of receptors at the tail ends of the legs of the scorpions;
converting an analog signal of a step signal received from the acceleration sensor into a corresponding digital signal;
transmitting the stepping signals collected in the acceleration sensor module to an upper computer;
and the upper computer receives the stepping signals, constructs pulse fingerprint characteristic vectors corresponding to the stepping signals, matches the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizes user position positioning by using a WKNN algorithm.
Optionally, the method further includes: constructing a pulse fingerprint feature database; the step of constructing the pulse fingerprint feature database comprises the following steps:
collecting step signals sent by a user on a plurality of reference points arranged in an indoor positioning area;
simulating the neurons of the scorpions by using an LIF neuron model;
establishing a pulse neural network by utilizing the n/1 neuron configuration of the scorpion; extracting n-1, 2,3,5, 7;
converting the step signal acquired at each receptor position in the impulse neural network into a corresponding impulse signal;
and constructing pulse fingerprint feature vectors of the pulse signals at the positions of the receptors, and constructing a pulse fingerprint feature database by using the pulse fingerprint feature vectors and the relative positions of the pulse fingerprint feature vectors.
Optionally, the method includes: the number of the acceleration sensor modules is 8, the acceleration sensor modules are distributed on a circumference with a radius of a preset length, the linear advancing direction of the acceleration sensor modules is set to be 0 degree, and the angles of the 8 acceleration sensor modules on the circumference are +/-18 degrees, +/-54 degrees, +/-90 degrees and +/-140 degrees respectively.
Optionally, the method further includes:
and the upper computer performs waveform display and data storage on the received stepping signals.
Alternatively, each LIF neuron model is described as follows:
dU(t)＝(-U(t)+RI(t))*dt/τm
wherein I (t) represents the excitation current of the input neuron, R represents the resistance, U (t) is the membrane potential, τmIs a time constant.
The system and the method have the advantages that the system and the method for positioning the position fingerprint indoors based on bionics are provided, and step signals collected in the acceleration sensor module are transmitted to an upper computer; and the upper computer receives the stepping signals, constructs pulse fingerprint characteristic vectors corresponding to the stepping signals, matches the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizes user position positioning by using a WKNN algorithm. The bionic scorpion vibration source positioning mechanism takes the pulse number as the fingerprint characteristic, and the fingerprint characteristic is obtained by processing the position information through the bionic scorpion nervous system and is an index which can represent the signal characteristic with finer granularity than RSS, so that the information of the target position can be described more clearly, and the more accurate user position positioning is realized.
Drawings
FIG. 1 is a schematic structure diagram of a bionic-based indoor location and fingerprint positioning system provided by the invention;
FIG. 2 is a schematic diagram of the BCSS location of scorpions;
FIG. 3 is a schematic structural diagram of the system provided by the invention in a specific application;
FIG. 4 is a schematic diagram of the structure of a neural network in the system provided by the present invention;
FIG. 5 is a schematic diagram of a sensor layout in an embodiment of use of the system of the present invention;
FIG. 6 is a schematic diagram of step signals received by various sensors at different times compared to their corresponding amplitudes in an example of use of the system of the present invention;
FIG. 7 is a schematic graph comparing the landing signal received by each sensor to its corresponding membrane voltage and pulse rate at different times for an embodiment of the system of the present invention;
FIG. 8 is a flow chart illustrating the steps of the indoor positioning method of the present invention;
fig. 9 is a cumulative distribution function of positioning error based on n/1(n ═ 1,2,3,5, 7) configuration and RSS based location fingerprint positioning method.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Example 1
A first embodiment disclosed in the present invention is a bionics-based position fingerprint indoor positioning system, as shown in fig. 1, including:
the system comprises a plurality of acceleration sensor modules 110, and an Ethernet data transmission module 120, a microprocessor 130 and an upper computer 140 which are connected with the acceleration sensors 110;
the acceleration sensor module 110 is laid out according to the positions of receptors at the tail ends of the legs of the scorpions and is used for collecting step signals sent by users;
the ethernet data transmission module 120 is configured to transmit the stepping signal collected by the acceleration sensor module to the upper computer 140;
the microprocessor 130 is configured to control the ethernet data transmission module to transmit the stepping signal to the upper computer 140;
the upper computer 140 is configured to receive the step signal, construct a pulse fingerprint feature vector corresponding to the step signal, match the constructed pulse fingerprint feature vector with a fingerprint feature vector stored in a pre-established pulse fingerprint feature database, and implement user position location by using a WKNN algorithm.
Specifically, the system disclosed by the invention utilizes bionics and a scorpion vibration source positioning mechanism to carry out indoor user position fingerprint positioning, and different pulse numbers are obtained in different directions due to the fact that the scorpions instruct neurons to carry out pulse coding on vibration signals under the combined action of excitation and inhibition. The pulse signals are used as the position fingerprint characteristics, so that more accurate positioning can be realized, and therefore in a specific embodiment, the acceleration sensor is used for simulating a receptor at the tail end of the leg of the scorpion; the acceleration sensor is correspondingly arranged according to the structure of a receptor at the tail end of the leg of the scorpion. As shown in fig. 2, the leg of the scorpion is disposed with 8 susceptors, and preferably, the number of the acceleration sensor modules is set to 8, and the acceleration sensors are distributed on a circumference having a radius of a preset length with reference to the arrangement of the susceptors, and the direction in which the acceleration sensors linearly advance is set to 0 °, so that the angles of the arrangement of the 8 acceleration sensor modules on the circumference are ± 18 °, ± 54 °, ± 90 °, and ± 140 °, respectively.
The acceleration sensor modules 110 are distributed indoors and used for collecting stepping signals sent by indoor users and transmitting the stepping signals to the upper computer through the Ethernet data transmission module, the upper computer extracts pulse fingerprint features of the stepping signals, matches the extracted pulse fingerprint features with fingerprint feature vectors stored in a pulse fingerprint feature database, and finally realizes the positioning of the user positions by using a WKNN algorithm.
Specifically, the system further includes: the analog-to-digital conversion module and the constant current adaptive regulator module;
the analog-to-digital conversion module is configured to convert an analog signal of the step signal received from the acceleration sensor 110 into a corresponding digital signal;
the constant-current adapter module is arranged between the acceleration sensor module and the analog-to-digital conversion module and is used for receiving a stepping signal transmitted by the acceleration sensor and adjusting the stepping signal;
the acceleration sensor module, the constant current adaptive regulator module and the analog-to-digital conversion module are electrically connected.
The microprocessor controls the analog-to-digital conversion module to carry out operation of converting analog signals into digital signals on the stepping signals collected in the acceleration sensor, and controls the Ethernet data transmission module to transmit the signals to the upper computer.
Specifically, in order to realize accurately positioning the user position according to the obtained step signal, the upper computer further comprises: a pulse fingerprint feature database construction module;
the pulse fingerprint feature database construction module comprises: the device comprises a reference signal acquisition unit, a neuron modeling unit, a neural network construction unit, a signal conversion unit and a pulse fingerprint characteristic vector construction unit;
the reference signal acquisition unit is used for acquiring step signals sent by users at a plurality of reference points arranged in an indoor positioning area;
the neuron modeling unit is used for simulating neurons of scorpions by utilizing an LIF neuron model;
the neural network construction unit is used for establishing a pulse neural network by utilizing the n/1 neuron configuration of the scorpion; extracting n-1, 2,3,5, 7;
the signal conversion unit is used for converting the stepping signals acquired at the position of each receptor BCSS in the pulse neural network into corresponding pulse signals;
the pulse fingerprint feature vector construction unit is used for constructing pulse fingerprint feature vectors of pulse signals at the positions of the receptors BCSS, and constructing a pulse fingerprint feature database by using the pulse fingerprint feature vectors and the relative positions of the pulse fingerprint feature vectors.
The pulse fingerprint characteristic database is established in advance, and the pulse signals corresponding to the stepping signals collected in the acceleration sensor are identified and the position of a user is positioned according to the pulse fingerprint characteristic vectors of the indoor reference points stored in the pulse fingerprint characteristic database.
The system disclosed in the present invention is described in more detail in the following embodiments with reference to fig. 3, and the system includes the following components:
1. the acquisition platform is built: the system comprises 8 acceleration sensor modules, a KD5200 constant-current adjuster, an STM32F103 microprocessor, an AD7606 analog-to-digital conversion module, a W5500 Ethernet data transmission module and a controller, wherein the controller is arranged on an upper computer: a LabVIEW data acquisition and processing system and a Matlab data positioning system; the system is arranged on the indoor ground by using 8 acceleration sensor modules in a manner of imitating the distribution of BCSS (Basitarsus composite slits sendilla, midtarsal composite suture receptors) at the tail ends of the legs of scorpions, namely 8 acceleration sensor modules are distributed on a circle with the radius of 2m and the angles are +/-18 degrees, +/-54 degrees, +/-90 degrees and +/-140 degrees respectively; an STM32F103 microprocessor is used for controlling an AD7606 analog-to-digital conversion module to acquire multi-channel vibration signal data; the W5500 Ethernet data transmission module sends data to a PC upper computer in real time; the LabVIEW data acquisition and processing system performs waveform display and data storage on the received data; the Matlab data positioning system realizes positioning by utilizing a position fingerprint indoor positioning method imitating a scorpion vibration source positioning mechanism.
2. Establishing an offline pulse fingerprint feature database:
the method comprises the following steps: setting m reference points in an indoor positioning area, and acquiring a user stepping signal on each reference point to obtain an original signal of each reference point as follows:
S1(t)、S2(t)、S3(t)、S4(t)、S5(t)、S6(t)、S7(t)、S8(t)；
step two: after receiving the vibration signal, the BCSS on each leg of the scorpion firstly activates an instruction neuron on the leg to excite the instruction neuron; then the instruction neuron activates a mesogenic inhibitory neuron to inhibit the instruction neuron in the opposite direction; and finally, under the combined action of excitation and inhibition, instructing the neurons to perform pulse coding on the vibration signals, and finally deciding the response direction according to the pulse coding by the brain of the scorpion. The neurons of the scorpions were simulated using LIF (Leaky Integrated-and-Fire) neuron models, one LIF neuron model for each of the 8 sensor orientations. The mathematical model of the neuron model is described as follows:
dU(t)＝(-U(t)+RI(t))*dt/τm(1)
wherein I (t) represents the excitation current of the input neuron, R represents the resistance, U (t) is the membrane potential, τmIs a time constant.
Order: du (t) ═ U (t + Δ t) -U (t), according to formula (1):
U(t+Δt)＝(-U(t)+RI(t))*dt/τm+U(t) (2)
step three: as shown in fig. 4, a spiking neural network was established using the 3/1 neuron configuration of the scorpion, in which the command neuron activated the mesogenic suppressor neuron and suppressed its opposite direction by 3 command neurons in the 3/1 neuron configuration, producing 1 excitatory input and 3-fold suppressive inputs. Eight BCSSs are coded clockwise from the right front leg of the scorpionNumber e1、e2、e3、e4、e5、e6、e7、e8If the target is in the e-th position3Generating a vibration signal near a leg, wherein the command neuron of the BCSS of the leg receives the e-th excitation signal first and reverses the excitation signal first6、e7And e8The command neuron of the strip-leg BCSS produces an inhibitory effect, likewise, item e7After receiving the excitation signal input, the command neuron of the strip-leg BCSS also sends the excitation signal to the e-th neuron2、e3And e4The command neurons of the legged BCSS produce inhibitory effects.
E thkExcitation input of (k ═ 1,2, … …, 8) neurons
Figure BDA0001895140280000091
Comprises the following steps:
Figure BDA0001895140280000092
where k is the number corresponding to the BCSS that generated the excitation input,
Figure BDA0001895140280000093
the excitation generated by the neuron is inputted with a corresponding signal.
E thk( e k1,2, … …, 8) neuron-corresponding inhibitory inputs
Figure BDA0001895140280000094
Comprises the following steps:
Figure BDA0001895140280000095
wherein
Figure BDA0001895140280000096
To generate the number of the corresponding BCSS for the inhibit input,
Figure BDA0001895140280000097
for signals corresponding to inhibitory inputs generated by neuronsW is the weight of the inhibitory input, and w is taken to be 0.01 in combination with the characteristics of the impulse neuron and the application model.
Obtaining the e-th product according to the formulas (3) and (4)kThe excitation current of the input neuron of each neuron is:
Figure BDA0001895140280000098
substituting the formula (5) into the formula (2) to obtain the membrane potential of the neuron:
U(t+Δt)＝(-U(t)+RIm(t))*dt/τm+U(t) (6)
when the membrane potential U (t + Deltat) reaches the threshold membrane potential U of the neuronthWhen it is, i.e. U (t + Δ t) ≧ UthTime, neuron ekWill emit a pulse and then the membrane potential will drop to the resting potential U of the neuronrest；
Step four: the original signal of each reference point is converted into different pulse numbers in 8 directions through the third step, and the pulse number corresponding to the ith reference point is Ni1,Ni2，…，Ni8. Constructing a pulse fingerprint feature vector N of a vibration signal at an ith reference pointiComprises the following steps:
Ni＝[Ni1,…，Nir，…，Ni8，Mi](7)
Figure BDA0001895140280000101
wherein N isirNumber of pulses, M, converted for the signal acquired by the r receiver at the i reference pointiThe total number of pulses converted from the signals collected by 8 receivers at the ith reference point, wherein r is the number of the receivers, and r is 1,2, …, 8;
step five: referring to fig. 5, the positions (x, y) of all the reference points are stored in the pulse fingerprint feature database together with the corresponding pulse fingerprint feature vector N, so as to form a pulse fingerprint feature database D:
Figure BDA0001895140280000102
wherein (x)i,yi) And (3) the position coordinate of the ith reference point, m is the number of the reference points, and i is 1,2, … and m.
3. And (3) in an online positioning stage:
the method comprises the following steps: collecting the step signal of the positioning point, and extracting the pulse fingerprint characteristic vector of the step signal according to the step four;
step two: and matching the pulse fingerprint feature vector with a reference point position fingerprint stored in a pulse fingerprint feature database in the step five of the off-line library building stage, and realizing position estimation by using a WKNN algorithm.
Example 2
A second embodiment provided by the present invention is a method for indoor positioning by the location fingerprint indoor positioning system, as shown in fig. 8, including:
step S81, acquiring a stepping signal sent by a user by using an acceleration sensor module; the acceleration sensor is laid out according to the position of a receptor BCSS at the tail end of the leg of the scorpion;
step S82, converting the analog signal of the step signal received from the acceleration sensor into a corresponding digital signal;
step S83, transmitting the stepping signals collected in the acceleration sensor module to an upper computer;
and step S84, the upper computer receives the stepping signal, constructs a pulse fingerprint characteristic vector corresponding to the stepping signal, matches the constructed pulse fingerprint characteristic vector with a fingerprint characteristic vector stored in a pre-established pulse fingerprint characteristic database, and realizes user position positioning by using a WKNN algorithm.
Optionally, the method further includes: constructing a pulse fingerprint feature database; the step of constructing the pulse fingerprint feature database comprises the following steps:
collecting step signals sent by a user on a plurality of reference points arranged in an indoor positioning area;
simulating the neurons of the scorpions by using an LIF neuron model;
establishing a pulse neural network by utilizing the n/1 neuron configuration of the scorpion; extracting n-1, 2,3,5, 7;
converting the step signal acquired at each receptor position in the impulse neural network into a corresponding impulse signal;
and constructing pulse fingerprint feature vectors of the pulse signals at the positions of the receptors, and constructing a pulse fingerprint feature database by using the pulse fingerprint feature vectors and the relative positions of the pulse fingerprint feature vectors.
Preferably, the number of the acceleration sensor modules is 8, the acceleration sensor modules are distributed on a circumference with a radius of a preset length, and the direction of linear advancement of the acceleration sensor modules is set to 0 °, so that the angles of the 8 acceleration sensor modules arranged on the circumference are ± 18 °, ± 54 °, 90 °, and ± 140 °, respectively.
Further, the above steps further include the steps of:
and the upper computer performs waveform display and data storage on the received stepping signals.
Alternatively, each LIF neuron model is described as follows:
dU(t)＝(-U(t)+RI(t))*dt/τm
wherein I (t) represents the excitation current of the input neuron, R represents the resistance, U (t) is the membrane potential, τmIs a time constant.
The positioning system comprises 8 acceleration sensor modules, a KD5200 constant current adaptive regulator, an STM32F103 microprocessor, an AD7606 analog-to-digital conversion module, a W5500 Ethernet data transmission module, a LabVIEW data acquisition and processing system and a Matlab data positioning system. The specific implementation steps are as follows:
the method comprises the following steps: as shown in FIG. 2, the basal tarsal compound suture receptors (BCSS) at the distal ends of the legs of the scorpions are distributed on a circle having a radius of 2.5cm, and if the straight line direction of the basal tarsal compound suture receptors is 0 °, the angles of the BCSS are + -18 °, + -54 °, + -90 °, + -140 °, respectively, and eight BCSSs are numbered as e in the clockwise direction from the right front leg1、e2、e3、e4、e5、e6、e7、e8. Will Yangzhou KekeA KD1200l piezoelectric acceleration sensor array produced by the electronic Limited company is arranged on the indoor ground according to the distribution of the BCSS of scorpions, wherein the radius of the distribution is 2 m;
step two: the sensor is connected with the input end of the KD5200 constant current adjuster through a low-noise signal line to realize the adjustment of signals;
step three: the output end of the KD5200 constant-current adjuster leads out a connecting terminal with the positive electrode and the negative electrode connected with the AD7606 module through a BNC joint;
step four: the AD7606 module is connected with an AD7606 module interface of the STM32F103 microprocessor through a flat cable to realize the analog-to-digital conversion of data;
step five: a network module interface of the STM32F103 microprocessor is connected with the W5500 Ethernet data transmission module through a flat cable;
step six: the W5500 Ethernet data transmission module is connected with a PC upper computer through a network cable, performs network data communication by using a UDP protocol, and transmits data to the PC upper computer;
step seven: utilizing a LabVIEW data acquisition and processing system of an upper computer to perform waveform display and data storage on received data;
step eight: and (4) importing the stored data into a Matlab data positioning system for data preprocessing and realizing target positioning by utilizing a position-based fingerprint algorithm.
The verification experiment of the system and the method provided by the invention specifically comprises the following steps:
the method comprises the following steps: as shown in FIG. 5, the design positioning area is a concentric circular ring shape with an inner circle radius of 2m, on which 8 acceleration sensors are arranged following the distribution of the BCSS of scorpions, such as A1、A2、A3、A4、A5、A6、A7、A8As shown. The excircle radiuses are respectively 2.5m, 3.0m, 3.5m, 4.0m, 4.5m, 5.0m, 5.5m and 6.0m, the excircle is inscribed with regular octagons, the top point of each regular octagon is selected as a positioning reference point, 64 reference points are totally arranged, and 8 acceleration sensors are arranged according to the distribution of the BCSS of the scorpions;
step two: acquiring a user stepping signal on each reference point, wherein the acquired vibration source information is a stepping signal of one user, and 120 times of signal samples are acquired at 64 positioning reference points respectively; in the on-line positioning stage, 15 positioning test points with random positioning areas are selected, and 20 signal samples are acquired respectively. As shown in fig. 6, the one-step signal collected within 0.5s is a signal sample, the sampling frequency is about 7.5KHz, and the raw signal of each reference point is obtained as follows:
S1(t)、S2(t)、S3(t)、S4(t)、S5(t)、S6(t)、S7(t)、S8(t)；
step three: and (3) performing least square detrending and wavelet dessication on the original signal to obtain:
S1'(t)、S2'(t)、S3'(t)、S4'(t)、S5'(t)、S6'(t)、S7'(t)、S8'(t)；
step four: a pulse neural network is established by utilizing 3/1 neuron configurations of scorpions, signal samples are converted into different pulse numbers in 8 directions through n/1 neuron configurations, and the pulse numbers are used as position characteristics of positioning points. As shown in fig. 7, is the discharge pulse obtained after the signal passed the 3/1 configuration.
Step five: in an off-line stage, extracting n/1(n is 1,2,3,5 and 7) configuration position characteristics and RSS (received signal strength) position characteristics based on scorpions, and respectively establishing fingerprint databases; the neural network model with 3/1 configuration is preferably used for feature extraction in the invention.
Step six: and in an online stage, matching the pulse fingerprint characteristic vector of the positioning point with the position fingerprint in the fingerprint database, and realizing position estimation by using a WKNN algorithm.
As shown in fig. 9, which is a cumulative distribution function of positioning errors of the n/1(n ═ 1,2,3,5, and 7) configuration and the RSS-based location fingerprint positioning method, it can be seen from fig. 9 that the location fingerprint indoor positioning system and method provided by the present invention can position the user location more accurately than the RSS-based positioning, and the user location positioned by using the neural network model with 3/1 configuration is the most accurate between different n/1 configurations.
The invention provides a bionic-based position fingerprint indoor positioning system and a bionic-based position fingerprint indoor positioning method, wherein a stepping signal acquired from an acceleration sensor module is transmitted to an upper computer; and the upper computer receives the stepping signals, constructs pulse fingerprint characteristic vectors corresponding to the stepping signals, matches the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizes user position positioning by using a WKNN algorithm. The bionic scorpion vibration source positioning mechanism takes the pulse number as the fingerprint characteristic, and the fingerprint characteristic is obtained by processing the position information through the bionic scorpion nervous system and is an index which can represent the signal characteristic with finer granularity than RSS, so that the information of the target position can be described more clearly, and the more accurate user position positioning is realized.
It should be understood that equivalents and modifications of the technical solution and inventive concept thereof may occur to those skilled in the art, and all such modifications and alterations should fall within the scope of the appended claims.

Claims (8)

1. A bionics-based position fingerprint indoor positioning system, comprising: the system comprises a plurality of acceleration sensor modules, and an Ethernet data transmission module, a microprocessor and an upper computer which are connected with the acceleration sensors;
the acceleration sensor module is laid out according to the position of a receptor at the tail end of the leg of the scorpion and is used for collecting a stepping signal sent by a user;
the Ethernet data transmission module is used for transmitting the stepping signals collected in the acceleration sensor module to an upper computer;
the microprocessor is used for controlling the Ethernet data transmission module to transmit the stepping signal to the upper computer;
the upper computer is used for receiving the stepping signals, constructing pulse fingerprint characteristic vectors corresponding to the stepping signals, matching the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizing user position positioning by using a WKNN algorithm;
the host computer still includes: a pulse fingerprint feature database construction module;
the pulse fingerprint feature database construction module comprises: the device comprises a reference signal acquisition unit, a neuron modeling unit, a neural network construction unit, a signal conversion unit and a pulse fingerprint characteristic vector construction unit;
the reference signal acquisition unit is used for acquiring step signals sent by users at a plurality of reference points arranged in an indoor positioning area;
the neuron modeling unit is used for simulating neurons of scorpions by utilizing an LIF neuron model; there is a LIF neuron model in each sensor direction;
the neural network construction unit is used for establishing a pulse neural network by utilizing the n/1 neuron configuration of the scorpion; wherein n is 1,2,3,5, 7;
the signal conversion unit is used for converting the step signals acquired at each receptor position in the pulse neural network into corresponding pulse signals;
the pulse fingerprint characteristic vector construction unit is used for constructing pulse fingerprint characteristic vectors of pulse signals at the positions of the receptors and constructing a pulse fingerprint characteristic database by using the pulse fingerprint characteristic vectors and the relative positions of the pulse fingerprint characteristic vectors;
the step signal of each reference point is converted into different pulse signals in 8 directions, and the pulse signal corresponding to the ith reference point is Ni1,Ni2，…，Ni8；
Constructing a pulse fingerprint feature vector N of a vibration signal at an ith reference pointiComprises the following steps:
Ni＝[Ni1,…，Nir，…，Ni8，Mi]
Figure FDA0002424369840000021
wherein N isirPulse signal, M, converted from step signal acquired by the r-th receiver at the i-th reference pointiThe stepping signals collected by 8 receivers at the ith reference point are converted into the sum of pulses, r is the number of the receivers, and r is 1,2, … and 8.
2. The bionics-based positional fingerprint indoor positioning system according to claim 1, wherein the number of the acceleration sensor modules is 8, which are distributed on a circumference having a radius of a preset length, and the direction in which the straight line advances is set to 0 °, so that the angles of the 8 acceleration sensor modules arranged on the circumference are ± 18 °, 54 °, 90 °, and 140 °, respectively.
3. The bionics-based location-fingerprinting indoor positioning system of claim 2, further comprising: the analog-to-digital conversion module and the constant current adaptive regulator module;
the analog-to-digital conversion module is used for converting an analog signal of a step signal received from the acceleration sensor into a corresponding digital signal;
the constant-current adapter module is arranged between the acceleration sensor module and the analog-to-digital conversion module and is used for receiving a stepping signal transmitted by the acceleration sensor and adjusting the stepping signal;
the acceleration sensor module, the constant current adaptive regulator module and the analog-to-digital conversion module are electrically connected.
4. A method for indoor positioning using the location fingerprint indoor positioning system of claim 1, comprising:
acquiring a stepping signal sent by a user by using an acceleration sensor module; the acceleration sensor is laid out according to the position of a receptor BCSS at the tail end of the leg of the scorpion;
converting an analog signal of a step signal received from the acceleration sensor into a corresponding digital signal;
transmitting the stepping signals collected in the acceleration sensor module to an upper computer;
and the upper computer receives the stepping signals, constructs pulse fingerprint characteristic vectors corresponding to the stepping signals, matches the constructed pulse fingerprint characteristic vectors with fingerprint characteristic vectors stored in a pre-established pulse fingerprint characteristic database, and realizes user position positioning by using a WKNN algorithm.
5. The method of indoor positioning of claim 4, further comprising: constructing a pulse fingerprint feature database; the step of constructing the pulse fingerprint feature database comprises the following steps:
collecting step signals sent by a user on a plurality of reference points arranged in an indoor positioning area;
simulating the neurons of the scorpions by using an LIF neuron model;
establishing a pulse neural network by utilizing the n/1 neuron configuration of the scorpion; extracting n-1, 2,3,5, 7;
converting step signals acquired at each receptor BCSS position in the impulse neural network into corresponding impulse signals;
and constructing pulse fingerprint feature vectors of the pulse signals at the positions of each receptor BCSS, and constructing a pulse fingerprint feature database by using the pulse fingerprint feature vectors and the relative positions of the pulse fingerprint feature vectors.
6. The method of indoor positioning of claim 5, comprising: the number of the acceleration sensor modules is 8, the acceleration sensor modules are distributed on a circumference with a radius of a preset length, the linear advancing direction of the acceleration sensor modules is set to be 0 degree, and the angles of the 8 acceleration sensor modules on the circumference are +/-18 degrees, +/-54 degrees, +/-90 degrees and +/-140 degrees respectively.
7. The method of indoor positioning of claim 4, further comprising:
and the upper computer performs waveform display and data storage on the received stepping signals.
8. The method for indoor localization according to claim 4, wherein each LIF neuron model is described as follows:
dU(t)＝(-U(t)+RI(t))*dt/τm
wherein I (t) represents the excitation current of the input neuron, R represents the resistance, U (t) is the membrane potential, τmIs a time constant.
CN201811488576.1A 2018-12-06 2018-12-06 Bionic-based position fingerprint indoor positioning system and method Active CN109342998B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811488576.1A CN109342998B (en) 2018-12-06 2018-12-06 Bionic-based position fingerprint indoor positioning system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811488576.1A CN109342998B (en) 2018-12-06 2018-12-06 Bionic-based position fingerprint indoor positioning system and method

Publications (2)

Publication Number Publication Date
CN109342998A CN109342998A (en) 2019-02-15
CN109342998B true CN109342998B (en) 2020-06-09

Family

ID=65297325

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811488576.1A Active CN109342998B (en) 2018-12-06 2018-12-06 Bionic-based position fingerprint indoor positioning system and method

Country Status (1)

Country Link
CN (1) CN109342998B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110245707B (en) * 2019-06-17 2022-11-11 吉林大学 Human body walking posture vibration information identification method and system based on scorpion positioning
CN110097171B (en) * 2019-06-19 2020-04-21 吉林大学 Activity track positioning method and system based on scorpion micro-vibration positioning mechanism
CN112858717B (en) * 2021-02-20 2022-04-19 吉林大学 Bionic airflow sensor and airflow detection device
CN113483885B (en) * 2021-07-06 2022-05-31 吉林大学 Composite pulse vibration source positioning method based on scorpion hair seam coupling positioning mechanism

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101612704B1 (en) * 2009-10-30 2016-04-18 삼성전자 주식회사 Apparatus and Method To Track Position For Multiple Sound Source
CN104457755B (en) * 2014-10-31 2017-06-09 清华大学 A kind of location acquiring method
CN104865555B (en) * 2015-05-19 2017-12-08 河北工业大学 A kind of indoor sound localization method based on sound position fingerprint
CN105282758A (en) * 2015-09-06 2016-01-27 华南理工大学 Self-adaptive dynamic construction method of WIFI indoor positioning system fingerprint database
CN106412839B (en) * 2016-09-12 2019-09-10 南京邮电大学 Based on secondary partition and the matched indoor positioning of fingerprint gradient and tracking
CN108737979B (en) * 2018-05-11 2020-05-22 华南理工大学 Indoor positioning method

Also Published As

Publication number Publication date
CN109342998A (en) 2019-02-15

Similar Documents

Publication Publication Date Title
CN109342998B (en) Bionic-based position fingerprint indoor positioning system and method
CN108989976B (en) Fingerprint positioning method and system in intelligent classroom
CN104865555B (en) A kind of indoor sound localization method based on sound position fingerprint
CN203092551U (en) Domestic service robot based on Kinect and FPGA (Field-programmable Gate Array)
CN101957442A (en) Sound source positioning device
Sun et al. Indoor geomagnetic positioning using the enhanced genetic algorithm-based extreme learning machine
CN107018522A (en) A kind of localization method of the unmanned aerial vehicle base station based on Multi-information acquisition
CN104394588B (en) Indoor orientation method based on Wi Fi fingerprints and Multidimensional Scaling
CN110245707B (en) Human body walking posture vibration information identification method and system based on scorpion positioning
CN110082716B (en) Indoor positioning system and positioning method for complex environment of hospital
CN112998668A (en) Millimeter wave-based non-contact far-field multi-human-body respiration heart rate monitoring method
CN110401978A (en) Indoor orientation method based on neural network and particle filter multi-source fusion
CN106526529A (en) Sparse representation-based direction-of-arrival estimation method in mismatched condition of steering vectors
JP2020126578A (en) Motion capture device and motion capture method for multipoint reception array based on non-propagating electromagnetic field
Sinitsyn et al. Determination of aircraft current location on the basis of its acoustic noise
Rodriguez-Labra et al. Development of a wireless robotic arm control system using piezoelectric sensors and neural networks
Liu et al. Ultrasound-based 3-D gesture recognition: Signal optimization, trajectory, and feature classification
Mantoro et al. Extreme learning machine for user location prediction in mobile environment
Kim Neural network mechanism for the orientation behavior of sand scorpions towards prey
CN113483885B (en) Composite pulse vibration source positioning method based on scorpion hair seam coupling positioning mechanism
Peremans et al. Biomimetic platform to study perception in bats
Rajai et al. Binaural sonar system for simultaneous sensing of distance and direction of extended barriers
Harper et al. Classification of plant species from CTFM ultrasonic range data using a neural network
CN113347705A (en) Ultra-wideband AOA indoor positioning method based on ICA and CS
CN111239682A (en) Electromagnetic emission source positioning system and method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant