CN108363622B - Method for realizing passive time difference positioning algorithm based on multi-core DSP operation - Google Patents

Method for realizing passive time difference positioning algorithm based on multi-core DSP operation Download PDF

Info

Publication number
CN108363622B
CN108363622B CN201810085359.1A CN201810085359A CN108363622B CN 108363622 B CN108363622 B CN 108363622B CN 201810085359 A CN201810085359 A CN 201810085359A CN 108363622 B CN108363622 B CN 108363622B
Authority
CN
China
Prior art keywords
core
particle
particles
value
receiving station
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
CN201810085359.1A
Other languages
Chinese (zh)
Other versions
CN108363622A (en
Inventor
蒋伊琳
屈天开
郜丽鹏
赵忠凯
张昊平
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.)
Harbin Engineering University
Original Assignee
Harbin Engineering 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 Harbin Engineering University filed Critical Harbin Engineering University
Priority to CN201810085359.1A priority Critical patent/CN108363622B/en
Publication of CN108363622A publication Critical patent/CN108363622A/en
Application granted granted Critical
Publication of CN108363622B publication Critical patent/CN108363622B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/54—Interprogram communication
    • G06F9/545—Interprogram communication where tasks reside in different layers, e.g. user- and kernel-space
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00—Arrangements for program control, e.g. control units
    • G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46—Multiprogramming arrangements
    • G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
    • G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
    • G06F9/5038—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the execution order of a plurality of tasks, e.g. taking priority or time dependency constraints into consideration

Landscapes

  • Engineering & Computer Science (AREA)
  • Software Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

The invention discloses a method for realizing a passive time difference positioning algorithm based on multi-core DSP operation, belonging to the field of passive positioning. Firstly, transmitting TOA data acquired from each slave receiving station and position information of each receiving station to a master device and a multi-core DSP through a wireless communication module of each receiving station; the multi-core DSP performs pairing calculation on the received TOA to obtain TDOA; planning the particle population quantity of the particle swarm algorithm, and distributing each core operation task of the multi-core DSP; initializing an adaptive value of the overall optimal particle and determining a fitness function; replacing the non-globally optimal particle in the slave core with the spare particle in the master core; after obtaining a new population composed of the particles replaced by the main core from the core, iteratively and newly obtaining an approximate estimated position calculated by the particle swarm; and finally, positioning calculation is carried out in the main core by using the result obtained by the particle swarm algorithm as an initial value of the Taylor series algorithm to obtain a final positioning result. The diversity of particle populations can be increased, so that the algorithm is faster in convergence, and the solution of positioning is faster.

Description

Method for realizing passive time difference positioning algorithm based on multi-core DSP operation
Technical Field
The invention belongs to the field of passive positioning, and particularly relates to a method for realizing a passive time difference positioning algorithm based on multi-core DSP operation.
Background
With the mass use of electronic information equipment on modern battlefields, the struggle of electromagnetic space is increasing. How to accurately and quickly acquire the electromagnetic potential state of an enemy, and quickly, accurately and precisely knowing where a radiation source is the premise of acquiring the initiative of electromagnetic battles. Therefore, the fast and high-precision positioning of a target radiation source becomes a primary research subject, with the increasing update of processor technology and the improvement of a positioning algorithm, a single-core DSP cannot meet the real-time and accurate positioning of a target gradually, and a multi-core Digital Signal Processor (DSP) has the advantages of high performance, low power consumption, strong computing power, multi-core parallel efficiency and high-efficiency floating point processing capability, meets the requirements of a time difference positioning system on operation efficiency, accuracy, real-time performance and reliability, so that the application of the multi-core DSP is a necessary trend in the future positioning technology.
At present, a time difference positioning method based on precise iteration of Chan algorithm initial value selection and Taylor series expansion method is proposed by Yanjunfeng (nuclear electronics and detection technology, volume 33, phase 4, 4 months in 2013, and arrival time difference positioning based on the Chan algorithm and Taylor series hybrid algorithm), but under the conditions of non-line-of-sight transmission and poor channel performance, the positioning precision is reduced, the solved positioning initial value is not precise enough, and the solution of the Taylor series algorithm is influenced.
Disclosure of Invention
The invention aims to provide a method for realizing a passive time difference positioning algorithm based on multi-core DSP operation, which can quickly, accurately position in real time and feed back a positioning result to a computer for display in real time.
The purpose of the invention is realized by the following technical scheme:
a method for realizing a passive time difference positioning algorithm based on multi-core DSP operation comprises the following steps:
step (1): transmitting the TOA data acquired from each receiving station and the position information of each receiving station to the master device and the multi-core DSP through a wireless communication module of each receiving station;
step (2): the multi-core DSP performs pairing calculation on the received TOAs to obtain TDOA, positions the radiation source by utilizing a Taylor series-particle swarm cooperative algorithm according to the position information of the receiving station and the arrival time difference of the radiation source,
in the main core, the TOAs received by 4 groups of different receiving stations are paired one by one according to corresponding pairing rules to obtain different TOAs received by four different receiving stations at the same time, and the TDOA is obtained by subtracting the TOAs;
and (3): planning the particle population quantity of the particle swarm algorithm and the distribution of each core operation task of the multi-core DSP, wherein the multi-core DSP adopts a parallel execution mode when solving by the particle swarm algorithm, the primary core0 is responsible for the distribution and scheduling of the tasks, and the secondary core is responsible for the execution of the tasks;
and (4): initializing information of each particle, initializing an adaptive value of the overall optimal particle, determining a fitness function, and calculating and comparing individual and global optimal solutions of the particles in each kernel;
and (5): transmitting the global optimal solution in each core obtained in the fourth step to the main core for comparison to obtain the global optimal solution of the whole particle swarm, and replacing the non-global optimal particles in the secondary core with the standby particles in the main core;
and (6): obtaining new population formed by the particles replaced by the main core from the secondary core, updating the position and speed information of the new particle population, calculating the adaptive value of the particles, comparing the existing adaptive value of each particle with the prior historical best adaptive value to obtain the latest individual optimal value to replace the prior historical optimal value, comparing the latest individual optimal value of each particle, updating the global optimal value of the secondary core, and comparing the main core with the prior global optimal value to obtain the latest overall global optimal value;
and (7): judging whether the termination standard is met, and if the termination standard is met, stopping searching; otherwise, repeating the steps (4) to (6);
and (8): and positioning and calculating by using the result obtained by the particle swarm algorithm in the main core as an initial value of the Taylor series algorithm to obtain a final positioning result.
The step (1) is specifically as follows:
(1.1) the master device and the multi-core DSP send out a message for starting work to the slave receiving station;
(1.2) after receiving the broadcast information of the multi-core DSP from the receiving station, starting to acquire TOA from an acquisition system of the receiving station and putting the acquired TOA into FIFO;
(1.3) reading TOA in the FIFO from the DSP of the equipment after the FIFO is full;
and (1.4) transmitting the TOA read by the multi-core DSP and the position information of the multi-core DSP through a wireless communication module.
The step (4) is specifically as follows:
(4.1) initialization A0—A7Position and speed information of all particles in the system are obtained, and an adaptive value f of the whole optimal particle is initialized to be 0; the coordinates of the 3 slave receiving stations are (x)i,yi,zi) (i ═ 1,2,3) and a primary receiving station (x)0,y0,z0) In the same plane, the position of the target radiation source is (x, y, z), and the distance from the target radiation source to the slave device is R when the target radiation source is positioned with time differencei(i ═ 1,2,3), the distance of the target radiation source to the host device is R0The relation between the target radiation source and the position of the master-slave device can be obtained
Figure BDA0001562244290000021
In the formula,. DELTA.ti0Is a measure of the difference in arrival time of the target radiation source to the master and slave devices, c is the speed of light, where ni0Is subject to independent distribution, and 0 mean variance is σ2White gaussian noise of (1);
(4.2) bringing the position coordinates of the receiving station into the following
Figure BDA0001562244290000031
Let Δ R be [ Δ R ]10,ΔR20,ΔR30]T，n＝[n10,n20,n30]TThe formula (1) can be represented as follows
Figure BDA0001562244290000032
Because of Δ Ri0Obey a mean value of (R)i-R0) Variance is σ2So the likelihood function can be deduced from equation (2) as:
Figure BDA0001562244290000033
wherein the objective function is:
(x,y,z)＝arg{min[(ΔR-Ri+R0)T(ΔR-Ri+R0)]}(i＝1,2,3) (4)
thus, a fitness function for the population of particles can be derived as
Figure BDA0001562244290000034
(4.3) calculating the adaptive value f of each particle in the secondary nucleus according to the obtained fitness function fij(i 1,2.. 10, j 1,2.. 7), wherein i represents the number of particles in the slave core, j represents the number of slave cores, and f in each slave core is comparedijThe particle b with the best adaptation value from the kernel is obtainedi(i＝1,2...7)。
The step (5) is specifically as follows:
each slave core sends the best particle b through a message queueiTo the main core, which allocates a certain space for storing the particles biAnd comparing to obtain the global optimal solution g of the whole particle swarmbest(ii) a Randomly selecting 2 particles from the spare particles in the main core for each secondary core respectively, feeding back the position and speed information of the worst particle in each secondary core to replace the secondary core, and selecting a particle group A in the secondary core1—A7And after obtaining new particles, updating the new particles into a new particle population.
The step (6) is specifically as follows:
(6.1) assume that the velocity of the ith particle is vi＝[vix,viy,viz](i 1,2.. 10) in the position
si＝[six,siy,siz](i＝1,2...10)，
Each slave core updates the position and velocity information of the new particle population, updating according to equation (6),
Figure BDA0001562244290000041
in the formula: j is x, y, z; w is the inertial weight; c. C1，c2Is a learning factor or acceleration constant; r is1、r2Respectively represent random numbers between 0 and 1; p is a radical ofi,jThe best position found so far for the particle; p is a radical ofg,jThe best position found for all particles so far in each slave nucleus; si,j(t)，vi,j(t)，si,j(t+1)，vi,j(t +1) is the position and velocity of the ith particle at times t and t +1, respectively;
(6.2) determining the fitness f of the new population of particles according to equation (5)ij(t +1) (i ═ 1,2.. 10, j ═ 1,2.. 7), where i denotes the number of particles from the nucleus and j denotes the number of nuclei; comparing the current fitness value f of each particleij(t +1) and the previous history best adapted value fbijTo find the latest individual best adaptive value of the particle to replace the original fbijAnd putting the adaptive value and the position information of the particle into pbest, and then comparing the latest fb of all the particles in the pbest in each slave coreijFinding out the particle b with the best latest adaptive value in each secondary corei(t +1), wherein t +1 is the t +1 updating of the particles;
(6.3) mixing bi(t +1) transmitting to the main core and the global optimal solution g of the whole particle swarm in front of the main corebestComparing and updating gbestPosition of
Figure BDA0001562244290000042
The speed information is stored in the memory space.
The step (8) is specifically as follows:
(8.1) deriving the initial position of the target radiation source from the principal nucleus
Figure BDA0001562244290000043
The formula (1) is as follows
Figure BDA0001562244290000044
Performing Taylor series expansion to remove high-order components, and expressing the positioning equation as
e≈h-Gδ (7)
In the formula:
Figure BDA0001562244290000045
h is the difference value between the measured value of the distance from the target radiation source to the receiving station and the true value;
Figure BDA0001562244290000051
in formula (8):
Figure BDA0001562244290000052
for the initial position of the selected target
Figure BDA0001562244290000053
The distance between each receiving station and the receiving station can be solved by adopting a least square (WLS) method, wherein the position deviation is as follows:
Figure BDA0001562244290000054
in equation (9), Q is a covariance matrix of the distance difference measurement values;
(8.2) iterative computation is carried out in sequence, and when delta meets a preset threshold value epsilon, namely | delta x | + | delta y | + | delta z | < epsilon, the position of the target radiation source can be obtained.
The invention has the beneficial effects that:
the solution obtained by the particle swarm optimization is used as an initial value of the Newton iterative algorithm, so that the positioning accuracy is improved. And the efficiency of the newton iteration algorithm can be improved.
The advantages of the multi-core DSP are utilized, the particle swarm algorithm adopts a parallel execution mode, the operation efficiency is improved, the operation time is shortened, and therefore real-time positioning is achieved.
During the execution of the particle swarm algorithm, the worst particles of each slave core are replaced by the spare particles in the master core to form a new population, so that the diversity of the particle population can be increased, the algorithm can be converged more quickly, and the positioning can be solved more quickly.
Drawings
Fig. 1 is a schematic diagram of the method for implementing time difference positioning based on multi-core DSP operation.
FIG. 2 illustrates the manner in which the master and slave cores of the present invention execute in parallel.
FIG. 3 illustrates tasks performed by the master core and the slave core of the present invention.
Detailed Description
The following further describes embodiments of the present invention with reference to the accompanying drawings:
with reference to fig. 1, a method for implementing a passive time difference positioning algorithm based on multi-core DSP operation includes the following steps:
the method comprises the following steps: and transmitting the TOA data acquired from each receiving station and the position information of each receiving station to the master device and the multi-core DSP through a wireless communication module of each receiving station.
Firstly, the master device and the multi-core DSP send out a message for starting work to the slave receiving station, then after the slave receiving station receives the broadcast information of the multi-core DSP, the collection system of the slave receiving station starts to collect TOA and puts the TOA collected by the TOA into FIFO, after the FIFO is full, the DSP of the slave device reads the TOA in the FIFO, and then the TOA read by the slave device and the position information of the slave device are transmitted to the multi-core DSP through a wireless communication module.
Step two: and the multi-core DSP performs pairing calculation on the received TOAs to obtain TDOA, and positions the radiation source by utilizing a Taylor series-particle swarm cooperative algorithm according to the position information of the receiving station and the arrival time difference of the radiation source.
And pairing the TOAs received by the 4 groups of different receiving stations one by one in the main core according to corresponding pairing rules, finally obtaining different TOAs received by four different receiving stations at the same time, and subtracting to obtain the TDOA.
Step three: and planning the particle population quantity of the particle swarm algorithm and the distribution of each core operation task of the multi-core DSP.
The multi-core DSP adopts a parallel execution mode when solving by using a particle swarm algorithm, the primary core0 is responsible for the distribution and scheduling of tasks, and the secondary core is responsible for the execution of tasks.
The number of the particle groups is particularly important because the number of the particle groups affects the accuracy of results and the length of operation time, and here, a high-speed digital signal processor eight-core TMS320C6678 of TI company is adopted, the processor is an eight-core floating-point DSP, the highest reachable working frequency of each core is 1.25GHz, and the processor has strong operation processing capacity, so that 10 particles are allocated to each slave core, and 100 particles are allocated in the master core as spare particles to replace the non-global optimal particles of the slave core. I.e. the particles are divided into 8 groups A0—A7From 7 groups A of cores1—A7Are processed by respective processors, and the main core is a group A0The most alternative to spare particles is the inter-core communication in the form of message queue MessageQ.
Step four: initializing the information of each particle, initializing the adaptive value of the overall optimal particle, determining a fitness function, and calculating and comparing the individual and global optimal solutions of the particles in each kernel.
Initializing A before performing calculation0—A7Position and velocity information of all particles, and initializing the adaptation value f of the overall optimal particle to 0. Let the 4 slave receiving stations coordinate (x)i,yi,zi) (i ═ 1,2,3) and a primary receiving station (x)0,y0,z0) In the same plane, the position of the target radiation source is (x, y, z), and the distance from the target radiation source to the slave device is R when the target radiation source is positioned with time differencei(i ═ 1,2,3), the distance of the target radiation source to the host device is R0. The relation between the positions of the target radiation source and the master-slave device can be obtained
Figure BDA0001562244290000071
In the formula,. DELTA.ti0Is a measure of the difference in arrival time of the target radiation source to the master-slave device, and c is the speed of light. Wherein n isi0Is subject to independent distribution, and 0 mean variance is σ2The Gaussian white noise of (1) is obtained by substituting the position coordinates of the receiving station as follows
Figure BDA0001562244290000072
Let Δ R be [ Δ R ]10,ΔR20,ΔR30]T，n＝[n10,n20,n30]TThe formula (1) can be represented as follows
Figure BDA0001562244290000073
Because of Δ Ri0Obey a mean value of (R)i-R0) Variance is σ2So the likelihood function can be deduced from equation (2) as:
Figure BDA0001562244290000074
wherein the objective function is:
(x,y,z)＝arg{min[(ΔR-Ri+R0)T(ΔR-Ri+R0)]}(i＝1,2,3) (4)
thus, a fitness function for the population of particles can be derived as
Figure BDA0001562244290000075
Calculating the adaptive value f of each particle in the secondary nucleus according to the obtained fitness function fij(i 1,2.. 10, j 1,2.. 7), wherein i represents the number of particles in the secondary nucleus, and j represents the number of secondary nuclei. Comparing f in each slave coreijThe particle b with the best adaptation value from the kernel is obtainedi(i＝1,2...7)。
Step five: and transmitting the global optimal solution in each core obtained in the step four to the main core for comparison to obtain the global optimal solution of the whole particle swarm, and replacing the non-global optimal particles in the secondary core with the standby particles in the main core.
Each slave core sends the best particle b through a message queueiTo the master core, masterThe core allocates a certain space for storing these particles biAnd comparing to obtain the global optimal solution g of the whole particle swarmbest. Randomly selecting 2 particles from the spare particles in the main core for each secondary core respectively, feeding back the position and speed information of the worst particle in each secondary core to replace the secondary core, and selecting a particle group A in the secondary core1—A7And after obtaining new particles, updating the new particles into a new particle population.
Step six: the method comprises the steps of obtaining a new population by the secondary cores, forming the new population by the particles replaced by the primary cores, updating position and speed information of the new particle population, calculating adaptive values of the particles, comparing the existing adaptive values of the particles with the prior historical best adaptive values to obtain the latest individual optimal values to replace the prior historical optimal values, comparing the latest individual optimal values of each particle, updating the global optimal values of the secondary cores, and comparing the primary cores with the prior global optimal values to obtain the latest overall global optimal values. .
Since the search is performed in three-dimensional space, the velocity of the ith particle is assumed to be vi＝[vix,viy,viz](i 1,2.. 10) in the position si＝[six,siy,siz](i＝1,2...10)
Each slave core updates the position and velocity information of the new particle population, updating according to equation (6),
Figure BDA0001562244290000081
in the formula: j is x, y, z; w is the inertial weight; c. C1，c2Is a learning factor or acceleration constant; r is1、r2Respectively represent random numbers between 0 and 1; p is a radical ofi,jThe best position found so far for the particle; p is a radical ofg,jThe best position found so far for all particles in each slave core. si,j(t)，vi,j(t)，si,j(t+1)，vi,j(t +1) is the position and velocity of the ith particle at times t and t +1, respectively. The adaptation value f of the particles in the new population is determined according to equation (5)ij(t +1) (i ═ 1,2.. 10, j ═ 1,2.. 7), where i is represented byThe number of particles in the slave core, j is represented as the number of slave cores. Comparing the current fitness value f of each particleij(t +1) and the previous history best adapted value fbijTo find the latest individual best adaptive value of the particle to replace the original fbijAnd putting the adaptive value and the position information of the particle into pbest, and then comparing the latest fb of all the particles in the pbest in each slave coreijFinding out the particle b with the best latest adaptive value in each secondary corei(t +1), where t +1 is the t +1 th particle update. B is toi(t +1) transmitting to the main core and the global optimal solution g of the whole particle swarm in front of the main corebestComparing and updating gbestPosition of
Figure BDA0001562244290000082
The speed information is stored in the memory space.
Step seven: judging whether the termination standard is met, and if the termination standard is met, stopping searching; otherwise, repeating the steps from four to six.
Step eight: and positioning and calculating by using the result obtained by the particle swarm algorithm in the main core as an initial value of the Taylor series algorithm to obtain a final positioning result.
Initial position of target radiation source obtained from main nucleus
Figure BDA0001562244290000091
The formula (1) is as follows
Figure BDA0001562244290000092
The positioning equation can be expressed as
e≈h-Gδ (7)
In the formula:
Figure BDA0001562244290000093
wherein h is the difference between the measured value of the distance from the target radiation source to the receiving station and the true value.
Figure BDA0001562244290000094
In formula (8):
Figure BDA0001562244290000095
for the initial position of the selected target
Figure BDA0001562244290000096
The distance between each receiving station and the receiving station can be solved by adopting a least square (WLS) method, wherein the position deviation is as follows:
Figure BDA0001562244290000097
in equation (9), Q is a covariance matrix of the distance difference measurement values.
And (3) sequentially carrying out iterative calculation, and obtaining the position of the target radiation source when delta meets a preset threshold value epsilon, namely, the value of | delta x | + | delta y | + | delta z | < epsilon.
The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims (6)

1. A method for realizing a passive time difference positioning algorithm based on multi-core DSP operation is characterized by comprising the following steps:
step (1): transmitting the TOA data acquired from each receiving station and the position information of each receiving station to the master device and the multi-core DSP through a wireless communication module of each receiving station;
step (2): the multi-core DSP performs pairing calculation on the received TOAs to obtain TDOA, positions the radiation source by utilizing a Taylor series-particle swarm cooperative algorithm according to the position information of the receiving station and the arrival time difference of the radiation source,
in the main core, the TOAs received by 4 groups of different receiving stations are paired one by one according to corresponding pairing rules to obtain different TOAs received by four different receiving stations at the same time, and the TDOA is obtained by subtracting the TOAs;
and (3): planning the particle population quantity of the particle swarm algorithm and the distribution of each core operation task of the multi-core DSP, wherein the multi-core DSP adopts a parallel execution mode when solving by the particle swarm algorithm, the primary core0 is responsible for the distribution and scheduling of the tasks, and the secondary core is responsible for the execution of the tasks;
and (4): initializing information of each particle, initializing an adaptive value of the overall optimal particle, determining a fitness function, and calculating and comparing individual and global optimal solutions of the particles in each kernel;
and (5): transmitting the global optimal solution in each core obtained in the fourth step to the main core for comparison to obtain the global optimal solution of the whole particle swarm, and replacing the non-global optimal particles in the secondary core with the standby particles in the main core;
and (6): obtaining new population formed by the particles replaced by the main core from the secondary core, updating the position and speed information of the new particle population, calculating the adaptive value of the particles, comparing the existing adaptive value of each particle with the prior historical best adaptive value to obtain the latest individual optimal value to replace the prior historical optimal value, comparing the latest individual optimal value of each particle, updating the global optimal value of the secondary core, transmitting the updated global optimal value to the main core, and comparing the latest global optimal value with the prior global optimal value to obtain the latest overall global optimal value;
and (7): judging whether the termination standard is met, and if the termination standard is met, stopping searching; otherwise, repeating the steps (4) to (6);
and (8): and positioning and calculating by using the result obtained by the particle swarm algorithm in the main core as an initial value of the Taylor series algorithm to obtain a final positioning result.
2. The method for implementing the passive time difference positioning algorithm based on the multi-core DSP operation according to claim 1, wherein the step (1) is specifically as follows:
(1.1) the master device and the multi-core DSP send out a message for starting work to the slave receiving station;
(1.2) after receiving the broadcast information of the multi-core DSP from the receiving station, starting to acquire TOA from an acquisition system of the receiving station and putting the acquired TOA into FIFO;
(1.3) reading TOA in the FIFO from the DSP of the equipment after the FIFO is full;
and (1.4) transmitting the TOA read by the multi-core DSP and the position information of the multi-core DSP through a wireless communication module.
3. The method for implementing the passive time difference positioning algorithm based on the multi-core DSP operation according to claim 1, wherein the step (4) is specifically as follows:
(4.1) dividing the particles into 8 groups A0—A7Initialization A0—A7Position and speed information of all particles in the system are obtained, and an adaptive value f of the whole optimal particle is initialized to be 0; the coordinates of the 3 slave receiving stations are (x)a,ya,za) A 1,2,3, and a primary receiving station (x)0,y0,z0) In the same plane, the position of the target radiation source is (x, y, z), and the distance from the target radiation source to the slave device is R when the target radiation source is positioned with time differenceaThe distance from the target radiation source to the main device is R0The relation between the target radiation source and the position of the master-slave device can be obtained
Figure FDA0003342729680000021
In the formula,. DELTA.ta0Is a measure of the difference in arrival time of the target radiation source to the master and slave devices, c is the speed of light, where na0Is subject to independent distribution, and 0 mean variance is σ2White gaussian noise of (1);
(4.2) bringing the position coordinates of the receiving station into the following
Let Δ R be [ Δ R ]10,ΔR20,ΔR30]T，n＝[n10,n20,n30]TThe formula (1) can be represented as follows
Figure FDA0003342729680000022
Because of Δ Ra0Obey mean value of Ra-R0Variance is σ2So the likelihood function can be deduced from equation (2) as:
Figure FDA0003342729680000023
wherein the objective function is:
(x,y,z)＝arg{min[(ΔR-Ra+R0)T(ΔR-Ra+R0)]} (4)
thus, a fitness function for the population of particles can be derived as
Figure FDA0003342729680000031
(4.3) calculating the adaptive value f of each particle in the secondary nucleus according to the obtained fitness function fij(ii) a Wherein i represents the number of particles from the nucleus, i ═ 1,2., 10; j represents the number of secondary nuclei, j 1,2.., 7; comparing f in each slave coreijObtaining the particles b with the best adaptation value in each secondary nucleusj。
4. The method for implementing the passive time difference positioning algorithm based on the multi-core DSP operation as claimed in claim 3, wherein the step (5) is specifically as follows:
each slave core sends the best particle b through a message queuejTo the main core, which allocates a certain space for storing the particles bjAnd comparing to obtain the global optimal solution g of the whole particle swarmbest(ii) a Randomly selecting 2 particles from the spare particles in the main core for each secondary core respectively, feeding back the position and speed information of the worst particle in each secondary core to replace the secondary core, and selecting a particle group A in the secondary core1—A7And after obtaining new particles, updating the new particles into a new particle population.
5. The method for implementing the passive time difference positioning algorithm based on the multi-core DSP operation as claimed in claim 3, wherein the step (6) is specifically as follows:
(6.1) assume that the velocity of the ith particle is vi＝[vix,viy,viz]At a position si＝[six,siy,siz]；
Each slave core updates the position and velocity information of the new particle population, updating according to equation (6),
Figure FDA0003342729680000032
in the formula: w is the inertial weight; c. C1，c2Is a learning factor or acceleration constant; r is1、r2Respectively represent random numbers between 0 and 1; p is a radical ofi,jThe best position found so far for the particle; p is a radical ofg,jThe best position found for all particles so far in each slave nucleus; si,j(t)，vi,j(t)，si,j(t+1)，vi,j(t +1) is the position and velocity of the ith particle at times t and t +1, respectively;
(6.2) determining the fitness f of the new population of particles according to equation (5)ij(t + 1); comparing the current fitness value f of each particleij(t +1) and the previous history best adapted value fbijTo find the latest individual best adaptive value of the particle to replace the original fbijAnd putting the adaptive value and the position information of the particle into pbest, and then comparing the latest fb of all the particles in the pbest in each slave coreijFinding out the particle b with the best latest adaptive value in each secondary corej(t +1), wherein t +1 is the t +1 updating of the particles;
(6.3) mixing bj(t +1) transmitting to the main core and the global optimal solution g of the whole particle swarm in front of the main corebestComparing and updating gbestPosition of
Figure FDA0003342729680000041
Speed of rotationAnd store the information in the storage space.
6. The method for implementing the passive time difference positioning algorithm based on the multi-core DSP operation as claimed in claim 3, wherein the step (8) is specifically as follows:
(8.1) deriving the initial position of the target radiation source from the principal nucleus
Figure FDA0003342729680000042
The formula (1) is as follows
Figure FDA0003342729680000043
Performing Taylor series expansion to remove high-order components, and expressing the positioning equation as
e≈h-Gδ (7)
In the formula:
Figure FDA0003342729680000044
h is the difference value between the measured value of the distance from the target radiation source to the receiving station and the true value;
Figure FDA0003342729680000045
in formula (8):
Figure FDA0003342729680000046
for the initial position of the selected target
Figure FDA0003342729680000047
The distance between each receiving station and the receiving station can be solved by adopting a least square (WLS) method, wherein the position deviation is as follows:
Figure FDA0003342729680000048
in equation (9), Q is a covariance matrix of the distance difference measurement values;
(8.2) iterative computation is carried out in sequence, and when delta meets a preset threshold value epsilon, namely | delta x | + | delta y | + | delta z | < epsilon, the position of the target radiation source can be obtained.
CN201810085359.1A 2018-01-29 2018-01-29 Method for realizing passive time difference positioning algorithm based on multi-core DSP operation Active CN108363622B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810085359.1A CN108363622B (en) 2018-01-29 2018-01-29 Method for realizing passive time difference positioning algorithm based on multi-core DSP operation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810085359.1A CN108363622B (en) 2018-01-29 2018-01-29 Method for realizing passive time difference positioning algorithm based on multi-core DSP operation

Publications (2)

Publication Number Publication Date
CN108363622A CN108363622A (en) 2018-08-03
CN108363622B true CN108363622B (en) 2022-03-18

Family

ID=63007412

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810085359.1A Active CN108363622B (en) 2018-01-29 2018-01-29 Method for realizing passive time difference positioning algorithm based on multi-core DSP operation

Country Status (1)

Country Link
CN (1) CN108363622B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109491791B (en) * 2018-11-09 2021-11-19 华东师范大学 Master-slave enhanced operation method and device of NSGA-II (non-subsampled Gate-associated genetic algorithm-II) based on Shenwei many-core processor
CN110536410B (en) * 2018-12-13 2020-10-30 西安邮电大学 Positioning method based on RSS and TDOA measurement in non-line-of-sight environment
CN111273251B (en) * 2020-03-09 2022-07-22 上海无线电设备研究所 Multi-core DSP-based particle swarm exchange long-time accumulation implementation method
CN112924932A (en) * 2021-01-25 2021-06-08 吉林大学 Vehicle positioning method and system based on grid search Newton iterative algorithm

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101620270A (en) * 2009-07-23 2010-01-06 重庆邮电大学 Wireless location method based on cluster-fusion
WO2008156905A9 (en) * 2007-04-13 2010-01-28 Glowlink Communications Tecnology, Inc. Determining a geolocation solution of an emitter on earth using satellite signals
KR20110098117A (en) * 2010-02-26 2011-09-01 숭실대학교산학협력단 Method of wireless positioning using particle swarm optimization
CN104023390A (en) * 2014-05-14 2014-09-03 浙江工业大学 WSN node positioning method based on combination of PSO and UKF
CN107015852A (en) * 2016-06-15 2017-08-04 珠江水利委员会珠江水利科学研究院 A kind of extensive Hydropower Stations multi-core parallel concurrent Optimization Scheduling

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2008156905A9 (en) * 2007-04-13 2010-01-28 Glowlink Communications Tecnology, Inc. Determining a geolocation solution of an emitter on earth using satellite signals
CN101620270A (en) * 2009-07-23 2010-01-06 重庆邮电大学 Wireless location method based on cluster-fusion
KR20110098117A (en) * 2010-02-26 2011-09-01 숭실대학교산학협력단 Method of wireless positioning using particle swarm optimization
CN104023390A (en) * 2014-05-14 2014-09-03 浙江工业大学 WSN node positioning method based on combination of PSO and UKF
CN107015852A (en) * 2016-06-15 2017-08-04 珠江水利委员会珠江水利科学研究院 A kind of extensive Hydropower Stations multi-core parallel concurrent Optimization Scheduling

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Constrained Total Least-Squares Location Algorithm Using Time-Difference-of-Arrival Measurements;Kai Yang 等;《IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY》;20100331;第59卷(第3期);第1558-1562页 *
基于Chan算法和Taylor级数混合算法的到达时差定位;杨俊峰 等;《核电子学与探测技术》;20130430;第33卷(第4期);第480-482、526页 *
粒子群优化算法在多基站时差定位中的应用;成伟明 等;《指挥信息系统与技术》;20100630;第1卷(第3期);第66-69页 *

Also Published As

Publication number Publication date
CN108363622A (en) 2018-08-03

Similar Documents

Publication Publication Date Title
CN108363622B (en) Method for realizing passive time difference positioning algorithm based on multi-core DSP operation
CN110930495A (en) Multi-unmanned aerial vehicle cooperation-based ICP point cloud map fusion method, system, device and storage medium
US11067669B2 (en) Method and apparatus for adjusting point cloud data acquisition trajectory, and computer readable medium
CN110969649B (en) Matching evaluation method, medium, terminal and device for laser point cloud and map
CN105654483A (en) Three-dimensional point cloud full-automatic registration method
CN109903330A (en) A kind of method and apparatus handling data
CN104463826A (en) Novel point cloud parallel Softassign registering algorithm
CN110414569A (en) Cluster realizing method and device
CN103294912B (en) A kind of facing mobile apparatus is based on the cache optimization method of prediction
CN110231592A (en) Indoor orientation method, device, computer readable storage medium and terminal device
CN117554814A (en) Battery state of charge prediction method and device, electronic equipment and storage medium
CN113010837A (en) Method and device for selecting sensors of wireless sensor network under non-line-of-sight condition based on TDOA (time difference of arrival) positioning
CN115235527A (en) Sensor external parameter calibration method and device and electronic equipment
CN109670001A (en) Polygonal gird GPU parallel calculating method based on CUDA
Li et al. An improved pigeon-inspired optimization for clustering analysis problems
Dong et al. Indoor robot localization combining feature clustering with wireless sensor network
CN107360542A (en) One kind is based on wireless network indoor article precise positioning algorithm
Xingli et al. Research on key technologies of passive positioning system for personnel under intelligent modeling
Pen et al. An Improved Weighted K-Nearest Neighbor Positioning Method in Buildings for Power Distribution Room
CN117110983B (en) Signal source positioning method based on unmanned aerial vehicle spiral track
CN116701821B (en) Robot speed calculation method and device, electronic equipment and storage medium
Zhang et al. Study on evaluation of cylindricity errors with a hybrid particle swarm optimization-chaos optimization algorithm
Landívar et al. Trilateration-based indoor location using supervised learning algorithms
CN117392270B (en) Wire fitting method, system and computer equipment based on laser point cloud
US20230375365A1 (en) Collecting telemetry data for 3d map updates

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