CN104765016B - Radio frequency identification and location method based on intelligent control over power - Google Patents

Abstract

The invention belongs to the technical field of mobile communication, and relates to a radio frequency identification and location method based on intelligent control over power. According to the method, the power of each reader is set into two transmitting modes of coarse-grained radiation and fine-grained radiation, a power energy level mapping function is set up, mapping from the coarse-grained radiation mode to the fine-grained radiation mode within the minimum effective energy level range is achieved, a fine-grained location radiation power sequence is generated to intelligently control the power of the readers, identity cognition sequences of labels to be located in the coarse-grained radiation mode and the fine-grained radiation mode are compared to detect whether the labels to be located are transferred out of a location environment or enter a dead zone, and neighbor reference label cognition sequences of the coarse-grained radiation mode and the fine-grained radiation mode are compared to detect whether path loss characteristics change or not. By improving the power transmitting mode of the existing radio frequency readers, the location efficiency and the location real-time performance can be effectively improved.

Description

A kind of frequency recognition positiming method controlled based on power intelligent

Technical field

The invention belongs to RF wireless communication technology field, is related to a kind of RF identification positioning controlled based on power intelligent Method.

Background technology

By noncontact, non line of sight, certification without the need for advantages such as manual intervention, support multi-tag identification operations, based on radio frequency The indoor passive location technology of identification (Radio Frequency Identification, RFID) technology is extensively and thing The fields such as stream, industry manufacture, medical treatment and amusement and recreation.

At present, the indoor passive alignment system based on RFID is referred to mainly based on LANDMARC algorithms by introducing Label motion capture environmental information substitutes the off-line data of reader, and neighbour is chosen with reference to mark using typical field intensity Euclidean distance Sign, introduce the positional information that typical residual weighting algorithm obtains label to be positioned.Compared to other location algorithms, LANDMARC is calculated Method not only reduces system cost, improves the adaptive capacity to environment of system, and precision is good.Even so, LANDMARC exists Some defects are still suffered from positioning real-time characteristic, is primarily due to：

(1) arithmetic accuracy is limited by reference label layout density.Reference label laying is more intensive, and positioning precision is higher, but Excessive reference label is laid and not only increases cost, can also cause collision and interference between extra label.

(2) arithmetic accuracy is limited by the maximum transmission power energy level of reader.Typical reader is sent out by gradually reducing Penetrate collection of letters intensity of the mode of power levels to obtain each label and indicate that transmission power energy level is bigger, the read-write scope of reader Bigger, the label of reading is more.Therefore, maximum transmission power energy level is lifted, judgement of the reader to signal strength signal intensity can be strengthened Ability and parsing degree, but single emission power levels poll can be caused time-consuming longer, affect location efficiency.

To sum up, the power emission mode of reader is studied and optimizes, for the real-time spy for improving RF identification alignment system Property, there are the important research of tool and practice significance in pole, but correlative study is still in the starting stage.

The content of the invention

The problem that the present invention need to be solved is to provide a kind of frequency recognition positiming method controlled based on power intelligent.

Based on the method, existing RF identification alignment system reduces system power dissipation on the premise of positioning precision is ensured, picks Except unnecessary reference label, suppress the read-write collision between multi-tag, effectively lift location efficiency.

1st, a kind of frequency recognition positiming method controlled based on power intelligent, is comprised the following steps：

Step 1：Setting reader radiates two kinds of power emission modes using coarseness radiation and fine granularity.

Step 2：Consider the path loss difference characteristic of different indoor positioning environment, the layout density of reference label and fixed The requirement of position precision and real-time, sets the power grade set of coarseness and fine granularity radiation.

Step 3：Multiple test points are set indoors in localizing environment, power levels are set up using nonlinear fitting mode and is reflected Penetrate function.

Step 4：Arrange localizing environment in each reader be coarseness Radiation work mode, using transmission power energy level from The little mode successively decreased step by step is arrived greatly quickly to be positioned, and generates tag identity coarseness to be positioned according to DSMC cognitive Sequence.

Step 5：For each label to be positioned, the minimum effectively energy level scope of its coarseness is set up.

Step 6：For each label to be positioned, the neighbour in the range of the minimum effectively energy level of its coarseness is recorded with reference to mark Sign, and generate neighbour's reference label coarseness cognition sequence.

Step 7：According to power levels mapping function, the minimum effectively energy level model of coarseness of each label to be positioned is mapped out Enclose the minimum effectively energy level scope of corresponding fine granularity.

Step 8：Union operation is carried out to whole elements of the minimum effectively energy level scope of fine granularity of all labels to be positioned, And by arranging from big to small, generate the fine granularity positioning radiant power sequence of reader.

Step 9：Each reader in localizing environment is set to into fine granularity Radiation work mode, according to fine granularity spoke is positioned Power sequence is penetrated, fine granularity is carried out in the way of reader transmission power is successively decreased step by step from big to small and is accurately positioned, it is special according to covering Caro method generates tag identity fine granularity cognition sequence to be positioned.

Step 10：For each label to be positioned, the neighbour in the range of the minimum effectively energy level of its fine granularity is recorded with reference to mark Sign, and generate neighbour's reference label fine granularity cognition sequence.

Step 11：Loop cycle scan mechanism is set and performs fine granularity positioning, according to Monte Carlo mode, update each and treat Neighbour's reference label fine granularity cognition sequence of positioning label.

Step 12：Using LANDMARC algorithms, the position of label to be positioned can be obtained by formula (1)：

Wherein, k is the quantity of neighbour's reference label,The estimated location of i-th label to be positioned is represented,The physical location of j-th neighbour's reference label of i-th label to be positioned is represented,Represent i-th mark to be positioned The bit-weight of the j-th neighbour's reference label signed,For i-th label undetermined and the field intensity of j-th neighbour's reference label Distance, can be obtained by formula (2)

Step 13：Comparison tag identity cognition coarse grain degree series to be positioned and tag identity to be positioned cognition fine granularity sequence Row, if the element number of particulate degree series is less than the element number of coarse grain degree series, now return to step 4, carry out coarseness fast Speed positioning.

Step 14：Neighbour's reference label fine granularity cognition sequence of each label of comparison and neighbour's reference label coarseness are cognitive Sequence, if S when the two difference is more than certain predetermined threshold value_R, now return to step 3, reset power levels mapping function M { }.

Step 15：Label newly-increased to be positioned in for localizing environment, arranges timing cycle scan mechanism, makes each reader Coarseness radiation mode is performed, tag identity cognition coarse grain degree series to be positioned are updated.

Step 16：Judge whether positioning result reaches set required precision, if meeting set constraints, close and read Device, terminates this position fixing process.

Description of the drawings：

Fig. 1 is FB(flow block) of the present invention；

Fig. 2 is indoor environment schematic diagram of the present invention.

Specific embodiment：

The purport of the present invention is to propose a kind of frequency recognition positiming method controlled based on power intelligent, and the method is adopted LANDMARC location algorithms, by controlling reader Radiation work mode, effectively control reader maximum transmission power energy level Reading to label, improves the real-time characteristic of location efficiency and alignment system.

Below in conjunction with the accompanying drawings 1 embodiment of the present invention is described further in detail.

The present invention relates to LANDMARC algorithms, by positioning scene cloth localizing environment as shown in Figure 2 is set to.By with lower section Formula completes implementation process.

Consider path loss difference characteristic, the layout density of reference label and the positioning precision of different indoor positioning environment With the requirement of real-time, setting reader is radiated using coarseness and fine granularity radiates two kinds of power emission modes, sets coarse grain The power grade collection of degree radiation is combined intoThe power grade collection of fine granularity radiation is combined intoα and β represent respectively the number of coarseness and each grade of fine granularity.

Multiple test points are set indoors in localizing environment, power levels mapping function is set up using nonlinear fitting mode M { }, for arbitrary power grade of coarseness radiationMap out corresponding fine granularity radiant power energy levelEach reader in localizing environment is set to into coarseness Radiation work mode, using transmission power energy level The mode successively decreased step by step from big to small is quickly positioned, and according to Monte Carlo mode, generates tag identity coarseness to be positioned Cognitive sequenceN represents positioning sequence label number,It is 96 EPC of i-th label to be positioned Information.

For arbitrary label to be positioned, the minimum effectively energy level scope of its coarseness is set up, such as i-th label to be positioned The minimum effectively energy level scope of coarseness isI.e. when the transmission power energy level of each reader is When, i-th label to be positioned can be detected, and the transmission power energy level for working as each reader will beWhen, can not detect To i-th label to be positioned.For arbitrary label to be positioned, the neighbour's reference in the range of the minimum effectively energy level of its coarseness is recorded Label, and neighbour's reference label coarseness cognition sequence is generated, neighbour's reference label coarseness of such as i-th label to be positioned is recognized Know that sequence isP is the number of neighbour's reference label coarseness cognition sequence.

According to power levels mapping function, the minimum effectively energy level scope correspondence of coarseness of each label to be positioned is mapped out The minimum effectively energy level scope of fine granularity, the minimum effectively energy level scope of fine granularity of such as i-th label to be positioned isWhole elements of the minimum effectively energy level scope of fine granularity of all labels to be positioned are entered Row union operation, and by arranging from big to small, generate the fine granularity positioning radiant power sequence of reader.

Each reader in localizing environment is set to into fine granularity Radiation work mode, according to fine granularity radiant power is positioned Sequence, by reader transmission power arrive greatly it is little successively decrease step by step in the way of carry out fine granularity and be accurately positioned, according to Monte Carlo mode Generate tag identity fine granularity cognition sequence to be positionedN is positioning label number.

For arbitrary label to be positioned, the neighbour's reference label in the range of the minimum effectively energy level of its fine granularity is recorded, and it is raw Into neighbour's reference label fine granularity cognition sequence, neighbour's reference label fine granularity cognition sequence of such as i-th label to be positioned isQ represents neighbour's reference label fine granularity cognition sequence number.Arrange loop cycle scan mechanism to hold Row fine granularity is positioned, and according to Monte Carlo mode, updates neighbour's reference label fine granularity cognition sequence of each label to be positioned.

Using LANDMARC algorithms, the positional information of label to be positioned is obtained.Comparison tag identity cognition coarse grain to be positioned Degree series and tag identity to be positioned cognition particulate degree series, if the element number of particulate degree series is less than the unit of coarse grain degree series Plain number, then show that there is certain label to be positioned is transferred out of localizing environment or enters into fine granularity power levels blind area, this When return 3, carry out coarseness and quickly position.Neighbour's reference label fine granularity cognition sequence of each label of comparison and neighbour are with reference to mark Coarseness cognition sequence is signed, if S when the two difference is more than certain predetermined threshold value_R, then show that the path loss of indoor positioning environment is special Property change, such as artificial disturbance strengthens, Obstacle Position changes, humiture changes, now return to step 3, weight If power levels mapping function M { }.

In position fixing process, for localizing environment in label newly-increased to be positioned, timing cycle scan mechanism is set, make each Reader performs coarseness radiation mode, updates tag identity cognition coarse grain degree series to be positioned.Judge whether positioning result reaches If meeting set constraints to set required precision, reader is closed, terminate this position fixing process.

Claims (2)

1. a kind of frequency recognition positiming method controlled based on power intelligent, is comprised the following steps：

Step 1：Setting reader radiates two kinds of power emission modes using coarseness radiation and fine granularity；

Step 2：Consider path loss difference characteristic, the layout density of reference label and the positioning accurate of different indoor positioning environment The requirement of degree and real-time, sets the power grade set of coarseness and fine granularity radiation；

Step 3：Multiple test points are set indoors in localizing environment, power levels mapping letter is set up using nonlinear fitting mode Number；

Step 4：Arrange localizing environment in each reader be coarseness Radiation work mode, using transmission power energy level from greatly to The little mode successively decreased step by step is quickly positioned, and according to DSMC tag identity coarseness cognition sequence to be positioned is generated Row；

Step 5：For each label to be positioned, the minimum effectively energy level scope of its coarseness is set up；

Step 6：For each label to be positioned, the neighbour's reference label in the range of the minimum effectively energy level of its coarseness is recorded, and Generate neighbour's reference label coarseness cognition sequence；

Step 7：According to power levels mapping function, the minimum effectively energy level scope pair of coarseness of each label to be positioned is mapped out The minimum effectively energy level scope of fine granularity answered；

Step 8：Union operation is carried out to whole elements of the minimum effectively energy level scope of fine granularity of all labels to be positioned, and is pressed Arrange from big to small, generate the fine granularity positioning radiant power sequence of reader；

Step 9：Each reader in localizing environment is set to into fine granularity Radiation work mode, according to fine granularity positioning radiation work( Rate sequence, carries out fine granularity and is accurately positioned in the way of reader transmission power is successively decreased step by step from big to small, according to Monte Carlo Method generates tag identity fine granularity cognition sequence to be positioned；

Step 10：For each label to be positioned, the neighbour's reference label in the range of the minimum effectively energy level of its fine granularity is recorded, and Generate neighbour's reference label fine granularity cognition sequence；

Step 11：Loop cycle scan mechanism is set and performs fine granularity positioning, according to Monte Carlo mode, update each to be positioned Neighbour's reference label fine granularity cognition sequence of label；

Step 12：Using LANDMARC algorithms, the positional information of label to be positioned is obtained；

Step 13：Comparison tag identity cognition coarse grain degree series to be positioned and tag identity to be positioned cognition particulate degree series, if Less than the element number of coarse grain degree series, now return to step 4, carry out coarseness quickly fixed to the element number of particulate degree series Position；

Step 14：Neighbour's reference label fine granularity cognition sequence and neighbour's reference label coarseness cognition sequence of each label of comparison Row, if S when the two difference is more than certain predetermined threshold value_R, now return to step 3, reset power levels mapping function；

Step 15：Label newly-increased to be positioned in for localizing environment, arranges timing cycle scan mechanism, performs each reader Coarseness radiation mode, updates tag identity cognition coarse grain degree series to be positioned；

Step 16：Judge whether positioning result reaches set required precision, if meeting set constraints, close reader, Terminate this position fixing process.

2. it is according to claim 1 it is a kind of based on power intelligent control frequency recognition positiming method, it is characterised in that：Step The LANDMARC algorithms adopted in rapid 12, the position of label to be positioned can be obtained by formula (1)：

$(x_{LT}^i,y_{LT}^i)=\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{w}_j^i(x_{{\mathrm{RT}}_j}^i,y_{{\mathrm{RT}}_j}^i)---\left(1\right)$

Wherein, k is the quantity of neighbour's reference label,The estimated location of i-th label to be positioned is represented, The physical location of j-th neighbour's reference label of i-th label to be positioned is represented,Represent j-th of i-th label to be positioned The bit-weight of neighbour's reference label,For i-th label undetermined and the field intensity distance of j-th neighbour's reference label, can be by Formula (2) is obtained：

$w_j^i=\frac{{\left({\mathrm{EP}}_{LR}^{ij}\right)}^{-2}}{\underset{j=1}{\overset{k}{\mathrm{\Σ}}}{\left({\mathrm{EP}}_{LR}^{ij}\right)}^{-2}}---\left(2\right).$