System and method for locating a target using a transceiver array

• US 9,285,456 B2
• Filed: 04/19/2010
• Issued: 03/15/2016
• Est. Priority Date: 04/17/2009
• Status: Expired due to Fees

• Current Assignee: Universite de Technologie de Troyes
• Original Assignee: Universite de Technologie de Troyes
• Inventors: Snoussi, Hichem, Honeine, Paul, Richard, Cdric
• Primary Examiner(s): Phan, Dao
• Application Number:
  US13/264,719
• Publication Number:
  US 20120154219A1
• Time in Patent Office: 2,157 Days
• Field of Search: 342/451, 342/458, 342463-465, 455/456.1
• US Class Current: 1/1
• CPC Class Codes: G01S 5/02521 using a radio-map
• CPC Class Codes: G01S 5/0289 of multiple transceivers, e.g. in ad hoc networks
• CPC Class Codes: G01S 5/0294 Trajectory determination or predictive filtering, e.g. target tracking or Kalman filtering

Abstract

The invention relates to a system and a method for locating at least one target (X) using an array of transceivers or sensors (S), in which at least a portion has a known geographic location, each comprising data processing means (S1) implementing at least one algorithm (AA, AS, AR, AF) for locating a target, means (S2) for transmitting/receiving a signal that decreases with the distance, the sensor array (S) covering at least one geographic area or area (Z), characterized in that they implement for each instant (t) an exchange of data or similarity data (DS) between the sensors (S) and a leading sensor (SL), and a distribution determination (DP) of the probability of the location of the target (X) using at least one regression algorithm (AR) on the basis of the similarity data (DS).

Claims

1. A method for locating and/or tracking at least one target with a network of transmitter/receivers, called sensors, of which at least part have a known geographic location, each comprising data processing means implementing at least one algorithm for locating and/or tracking target(s), means for transmitting/receiving a signal that decreases with distance and communication means, the network of sensors covering at least one geographic area, called the region, the method comprises, for each instant (t), an iteration of the following steps:

selection of a sensor, called the leader, for each target, using at least one leader selection algorithm, for the implementation of the subsequent steps,

exchange of data, called similarity data, between the leader sensor and at least part of the sensors in the network, these data being representative of power measurements of the signal between the sensors and the target and between the sensors themselves,

determination of a probability distribution of the position of the target in the region by the leader sensor, using at least one probabilistic matrix regression algorithm, determining, on the basis of the similarity data, a probability distribution of the scalar products between the positions of the sensors and the positions of the target, for estimating said probability distribution of the position of the target.

2. A method according to claim 1, the method further comprises, when the target(s) is (or are) a passive target or targets with respect to the signals of the sensors, a step of detection of at least one target present in the region, by at least one of the sensors at an initial instant, and further comprises, at each instant (t) an iteration of a step of selection of a set of n sensors, using at least one activation algorithm, prior to the step of selection of the leader sensor, the latter being then selected from within this set using the selection algorithm.

3. A method according to claim 1, wherein when the target(s) is (or are) a mobile sensor or sensors, this (or these) mobile sensor or sensors is (or are) selected using the selection algorithm, as leader sensor(s) and select(s), using at least one activation algorithm, a set of n neighboring sensors for estimating the position of the target within the region.

4. A method according to claim 1, the method further comprises, at each instant (t), an iteration of a step of tracking of the target by the leader sensor, by updating the position of the target using at least one variational filtering algorithm based on a variational filter providing an estimate of the position of the target.

5. A method according to claim 4, wherein the variational filter used during the updating step relies on the use of a confidence function of the position of the target based on the probability distribution of the position of the target determined during the determination step, and on the use of a model, called a transition model, based in particular on a temporal correlation of an assumed trajectory of the target from one instant to another.

6. A method according to claim 4, wherein, when there exist several targets and the targets are passive, the leader sensor of each of the targets employs, during the updating step, a step of probabilistic association of the data representative of the position of the target and of its trajectory, emanating from the variational filter used by this leader sensor, for determining to which target a set of data belongs.

7. A method according to claim 4, wherein the tracking step comprises, at each instant (t), an iteration of a step of predicting the position(s) of the target(s) at the following instant.

8. A method according to claim 7, wherein the step of predicting the position(s) of the target(s) at the following instant is accompanied, when the target(s) is (are) a passive target or targets with respect to the signal of the sensors, of a reiteration of the step of selecting a set of n sensors, using at least one activation algorithm and of the step of selecting a leader sensor using the selection algorithm.

9. A method according to claim 7, wherein the step of predicting the position(s) of the target(s) at the following instant is followed, when the target(s) is (are) a passive target or targets with respect to the signal of the sensors, by a step of sending, by the leader sensor at the instant (t) to the sensor leader of the following instant, of data representative of sufficient statistics for updating the variational filter at the following instant and a reiteration of the method.

10. A method according to claim 7, wherein the step of predicting the position(s) of the target(s) at the following instant is followed, when the target(s) is (are) a mobile sensor or sensors, by a step of erasing the data processed by the processing means of the leader sensor, with the exception of the data representative of sufficient statistics for updating the variational filter of that leader sensor in the following instant, followed by a reiteration of the method.

11. A method according to claim 1, wherein the similarity data are considered to be Euclidean scalar products between the sensors and the target and between the sensors themselves, in a reproducing Kernel Hilbert space.

12. A method according to claim 1, wherein the probability distribution of the position of the target within the region is a Gaussian.

13. A method according to claim 9, wherein the data representative of sufficient statistics are representative of a mean and a covariance of the estimated position of the target.

14. A system for locating and/or tracking at least one target with a network of transmitter/receivers, called sensors, of which at least part have a known geographic location, each comprising data processing means, means for transmitting/receiving a signal that decreases with distance and communication means, the network of sensors covering at least one geographic area, called the region, wherein the data processing means implement at least one algorithm for locating and/or tracking target(s) by the implementation of the method according to claim 1.

15. A system according to claim 14, wherein the means for transmitting/receiving a signal that decreases with distance and communication means are the same means, allowing both measurements and communication for the implementation of said method.