EP1880230A1

EP1880230A1 - Position determining process

Info

Publication number: EP1880230A1
Application number: EP06724783A
Authority: EP
Inventors: Thomas Speidel; Thomas JÜLG
Original assignee: ADS TEC GmbH
Current assignee: ADS TEC Holding GmbH
Priority date: 2005-05-10
Filing date: 2006-05-10
Publication date: 2008-01-23
Also published as: WO2006119985A1; DE102005021579A1

Abstract

A process for determining a position inside and outside buildings uses at least one mobile transmitter unit, a so-called Client Node, and transmitting/receiving stations positioned at known locations, so-called Access Points, which communicate with each other by wireless communications, in particular by radio signals. The invention is characterised in that first a fixed time reference, i.e. the relative time difference between the system times of the respective Access Points is determined. In order to determine position, the Client Node emits a radio signal which is received by those Access Points in whose reception zone the Client Node is located. Each of these Access Points registers in its system time the moment in time when the radio signal reached it. The position of the Client Node is then calculated on the basis of the individual moments in time, the time reference determined between the system times, and the known locations of the Access Points. The particularity of the claimed process is that only the reception of Client Node signals by the Access Points is required to determine the position of the Client Node, and that no synchronisation is necessary between the system times of Access Points and Client Nodes.

Description

Method for determining position

Technical area

The invention relates to a method for determining the position inside and outside buildings, with at least one mobile transmitting unit, called ClientNode, whose position it is to determine, and positioned at known locations transmitting / receiving stations, so-called access points, by means of wireless communication technology , in particular by means of radio signals, communicate with each other.

State of the art

Positioning methods and systems designed for this purpose are known in their various design forms and are based on the most diverse positioning techniques, such as radar technology, satellite navigation technology, ultrasonic location or optical measurement technology, to name only a few. If it is a position determination with the proviso of an object location outside as well as inside of rooms, such as buildings or similar enclosed spaces, so separates, for example, the satellite-based GPS navigation and positioning technology as well as the error-prone by multiple reflections radar technology. Rather, the position of objects is determined by the radio signal transmission technology between preferably several transmission / Receiving units. Previously known such systems are characterized by complex and therefore expensive technology and partly by only insufficient positioning accuracy.

DE 103 32 551 A1 discloses a generic method for determining the position with at least one mobile transceiver unit, so-called ClientNode, whose position is to be determined in relation to at least three transceiver stations positioned at known locations, so-called AccessPoints , known. The AccessPoints and ClientNodes communicate with each other by means of radio signals. The calculation of the position of the ClientNode takes place in the respective ClientNode on the basis of a two-way transit time measurement of the radio signals, which are exchanged between ClientNode and the at least three AccessPoints. For this, the ClientNode has a complex communication and evaluation technology. The method described is suitable in particular for applications in which a quasi-autonomous position determination by the ClientNode itself is required or in which the ClientNode at least has to know its own position. In applications of the disclosed in DE 103 32 551 A1 method for position determination for a variety of ClientNodes, however, associated with the complex technical equipment of each ClientNodes high production costs of the ClientNodes prove to be disadvantageous.

From DE 102 52 934 A1, a method for continuous real-time tracking of the position of at least one mobile object can be removed, which can be used, for example, to detect the position of football during a football game. The system required for this purpose comprises at least four receivers positioned at known locations, each of which is connected via data lines to a central computer. As an essential requirement, the system is calibrated prior to commissioning, whereby the transit times in the network are determined between the individual receivers and between the receivers to the central computer. This is followed by synchronization of the receivers with each other. These measures require extensive technical and procedural precautions that complicate the practical use of this method.

Presentation of the invention

The invention is based on the object, a method for determining the position inside and outside buildings, with positioned at known locations transmitting / receiving stations, the so-called. Access Points, and at least one mobile transmitting unit, called ClientNode, by means of wireless communication, in particular by means of radio signals communicate with each other in such a way that the required for determining the position of the technical equipment of the ClientNodes can be reduced and a position determination with an accuracy of ± 25 cm is feasible. This means a determination of the signal propagation time with a sub-nanosecond accuracy. In particular, it is important to perform the position determination seamlessly inside and outside of buildings for a variety of ClientNodes, without the need for expensive additional technical components, such as time synchronizers. The presence of the position information in the respective ClientNode itself should not be necessary.

The solution of the problem underlying the invention is specified in claim 1. The dependent claims and the further description are advantageous features of the inventive concept.

The successful implementation of the method requires at least four transmitting / receiving stations (AccessPoints), which are each positioned at known locations, at least one mobile transmitting unit (ClientNode) located in the receiving area of four or more access points, and at least one control computer for the calculation the positions of the ClientNode connected to at least one AccessPoint over a network. Compared to these at least four AccessPoints it is necessary to determine the position of the mobile ClientNode, wherein the method for determining the position on a plurality of ClientsNode is expandable. For communication between ClientNodes and AccessPoints, in addition to radio methods, optical or acoustic communication methods may also be considered. Without restricting the inventive concept, the invention will be described below using the example of wireless communication.

For the following description of the method according to the invention, important terms are first explained in more detail:

The so-called client nodes can be understood as battery-operated mobile transmission units that emit so-called probe request signals (PR signals) in certain time cycles. Each ClientNode has a processor and has an individual system time. To identify the individual ClientNodes each ClientNode each has a unique identifier, which is sent out advantageously with each PR signal. The PR signals may contain further data in addition to the identifier of the respective sending ClientNodes.

The so-called AccessPoints can be interpreted as transmitting / receiving stations installed at known locations, which generate data records by exchanging radio signals among themselves and by receiving the PR signals from ClientNodes, storing them in a buffer and finally evaluating the data records forward to the at least one ControlComputer. Also between the AccessPoints a data exchange is possible. The AccessPoints provide all data required for the position calculation of the ClientNodes. The access points send in certain time cycles so-called broadcast beacon signals (BC signals), which are received by the other access points and answered with a respective response signal. The BC signals and the response signals may contain additional data. Each AccessPoints has a processor, has an individual system time and an individual identifier, which can be sent advantageously with the BC signal or a response signal. Within the system of access points, at least one so-called control computer is provided which receives the data records acquired by the access points for further processing. The ControlComputer is connected to at least one AccessPoint via a network structure. The so-called DistributionNetwork represents the necessary hardware components of the network structure for data exchange between AccessPoints and the ControlComputer, which in the simplest case consists of a normal wireless or wired Ethemet network. The other AccessPoints can also have such a network connection to the ControlComputer, but this is not absolutely necessary for the functioning of the system. Preferably, a wireless or wired Ethernet network structure is used as the network connection. The records on an AccessPoint that have been cached in the internal buffer are sent over the network connection in a compressed packet to the ControlComputer. If the AccessPoint has a network connection to the ControlComputer, the data is transferred directly via the network to the ControlComputer. If there is no direct network connection to the ControlComputer, the data is transmitted by radio to the next available AccessPoint. If this AccessPoint also has no network interface to the ControlComputer, the data is transferred back to the next available AccessPoint. This process is repeated until the data reaches an AccessPoint with a network interface to the ControlComputer. This access point then transfers data to the ControlComputer. In the event that a high number of ClientNodes and / or AccessPoints exists or if the load on the ControlComputer requires it, several ControlComputers can be integrated in parallel in the system.

Finally, the so-called ServiceArea forms the space between the ClientNodes and the AccessPoints, in which radio waves can be transmitted for signal transmission and data exchange. In order to reliably separate the requirements which occur when transmitting and receiving PR, BC or response signals in the case of a possible multiplicity of ClientNodes or AccessPoints, the method according to the invention advantageously utilizes the transmission and reception of signals the access points and the involved client nodes of the wireless network standard IEEE 802.11 or the transmission protocol CSMA-CA (Carrier Sense Multiple Access with Collision Avoidance) or with a random signal spread spectrum modulated signal. The client nodes can advantageously be additionally equipped for this purpose with a receiving unit which allows the evaluation of such transmission protocols. This allows the ClientNodes (CN) to use the BC and PR signals for protocol synchronization. These transmission protocols are used in various standardized communication methods, such as WLAN, Bluetooth or GSM. These transmission protocols ensure that each individual station involved in the radio communication between the individual ClientNodes and the respective AccesPoints or AccessPoints monitors the respective radio channel and transmits only when the radio channel is not busy. This prevents collisions and the associated interference and minimizes additional transmission attempts.

The radio communication method normally operates in time-division multiplex mode. To increase the capacity but you can also work in frequency division multiplex mode. By eliminating the need for time-consuming channel switching, the number of client nodes that can be serviced by the system increases. If a large number of AccessPoints are needed in a small area, then the system can alternatively be operated in combined time and frequency division multiplex mode, i. Part of the AccessPoint works in another available frequency band.

In many cases, apart from the localization service, there are also high data transfer rates between the individual ClientNodes and AccessPoints necessary. In this case, dedicated AccessPoints are used only for payload data communication. All AccessPoints used for location always work in the same frequency band. The current frequency band can be changed by the ControlComputer, within the available frequency bands. To control the switching of the frequency bands, commands are used which use the virtual system time as the time for the next switching of the frequency band to all AccessPoints. This virtual system time is then converted to the local system time for each AccessPoint.

The inventive method for determining position inside and outside of buildings, with at least four positioned at known locations transmitting / receiving stations, so-called AccessPoints, at least one mobile transmitting unit, called ClientNode, communicate by means of wireless communication technology, in particular by means of radio signals, and with At least one ControlComputer, which is connected to the AccessPoints via a network structure, consists of the following process steps:

In the first step, a fixed time reference, ie the relative time difference, between the system times of the respective access points is determined. It is assumed that the processors used in the client nodes and access points and used for the time measurement have a sufficiently stable clock frequency, so that after determining the fixed time reference of the system times until a next determination of the fixed time reference, the determined time reference as constant can be accepted. In the second method step, a PR signal is sent by the ClientNode, which is received by those AccessPoints in whose reception area the ClientNode is located. Each of these AccessPoints registers in its system time the time at which the PR signal arrives at it. In the third method step, based on the determined time reference between the system times and the known locations of the AccessPoints, the position of the ClientNodes is finally calculated. The peculiarity of the method according to the invention is based on the fact that only the receipt of ClientNode signals by the AccessPoints is necessary for the determination of the positions of the ClientNodes and that no synchronization of the system times of AccessPoints and ClientNodes is necessary.

The individual process steps on which the method according to the invention is based are described in more detail with reference to the following illustrative figures.

Brief description of the invention

The invention will now be described by way of example without limitation of the general inventive idea by means of embodiments with reference to the drawings. Show it:

Fig. 1 representation of the initial situation for determining the position of three

ClientNodes in a system of four AccessPoints positioned in known locations and communicating with each other.

FIG. 2 Timing diagram of a BC signal emitted by a first access point and of a response signal subsequently emitted by a second access point until it is received by the first access point. FIG.

Ways to carry out the invention, industrial usability

The arrangement shown in Figure 1 has four so-called. Access Points, which are each positioned at known locations. Three of the four AccessPoints are connected to the ControlComputer via a so-called DistributionNetwork and thus able to exchange data with the ControlComputer. Compared with the permanently positioned AccessPoints, three mobile ClientNodes are provided in FIG. 1, which are located within the reception area of the AccessPoints, ie the so-called ServiceArea. As will be described below, it is now the current position of the individual ClientNodes by appropriate radio communication with the surrounding AccessPoints and their evaluation to determine.

To simplify the further description below, instead of three, only a single ClientNode is assumed, which sends PR signals to the four AccessPoints for position determination purposes and whose position lies within the reception range of all four AccesPoints.

The measuring principle on which the method according to the invention is based is based on a so-called one-way propagation time measurement, in which the most important measured variables are the signal propagation times between the ClientNode and the respective AccessPoints. Taking into account the electromagnetic waves propagating at the speed of light, the distance between a ClientNode and a respective AccessPoint can be calculated in this way as follows:

= h = ^ct o (1) with

I _{0 is} the distance between AccessPoint and ClientNode, c is the speed of light and to the signal propagation time.

If one assumes the determination of position only a single such signal propagation time measurement between the ClientNode and a single AccessPoint, the current location of the ClientNode is on a theoretical spherical surface around the AccessPoint with a radius that corresponds to the distance of the radio signal traveled within the runtime. Thus, the ClientNode can basically be anywhere on this theoretical sphere surface. With an additional measurement to a second AccessPoint one obtains a cut line of the two spherical surfaces. By adding a third measurement, two intersections of the spherical surface remain, on which the ClientNode can be located. By the fourth Finally, AccessPoints results in a clearly defined intersection of four spherical surfaces, which is located above the room floor.

However, this previously described position calculation based on the one-way delay measurement between ClientNode and AccessPoints presupposes that the signal propagation times are present as absolute signal propagation times, which is possible, for example, with synchronized system times of the ClientNode and the AccessPoints. However, the method according to the invention is based on the fact that client nodes and access points each have non-synchronized system times. The time measurements at the individual AccessPoints thus only provide times that are unrelated to each other.

In a first step of the method according to the invention, therefore, the determination of a fixed time reference of the system times or, in other words, the determination of the relative deviations of the system times of the four individual Access Points takes place. The method advantageously used for this purpose is based on a known two-way transit time measurement between the individual access points. Such runtime measurements are carried out between all four access points.

FIG. 2 shows an example of the basic timing diagram of a transit time measurement between a first and a second access point (AP and AP-M). If necessary, the first AccessPoint (AP) can start the request for this two-way runtime measurement with the transmission of an RTMS command (RunTime Measurement Service) to the second AccessPoint (AP-M) before the start of the measurements (not shown). If the RTMS command is received by the second access point, it starts the procedure for recording signals, ie the data recording, if it is not already active. Once the measuring procedure has started, it can be repeated over and over in certain rhythms. The execution of the measurement starts with the transmission of the BC signal from the first access point (AP), at the same time the first access point stores the start time t _A pi of the transmitted spread-band code in its system time. This BC signal is always received by all AccessPoints in whose receiving area the sending first AccessPoint is located. FIG. 2 shows the reception of the BC signal by the second access point (AP-M) at time t _A p- _M .m in the system time of AP-M. In the second Access Point, the relative start time of the start of the signal recording, the time of arrival of the signal are stored from the first Access Point t _A p _M .m and / or a part of the reception signal in the system time of AP-M. The BC signal is used in the second access point for a correlation measurement to measure the relative time difference in the arrival of the BC signal at time t _A p- _M .m to the internal processor oscillator. This data record or the resulting measured value is stored in an internal buffer together with the identification of the sending first AccessPoint. The second access point then sends t _A p at time. _M. ₂ in frequency or time division multiplex mode also a BC signal which is received by the first access point at time t _A pm. This BC signal is used in the first access point for a correlation measurement to measure the relative time difference in the arrival of this BC signal at time t _A p- _M .m to the internal processor oscillator. This data record or the resulting measured value is stored in an internal buffer together with the identification of the sending second access point.

Then, the record created in the second AccessPoint is sent to the first AccessPoint. In this case, a preprocessing of the data set can still be carried out in the second access point in order to reduce or to compress the amount of data to be transmitted.

With the second AccessPoint record, its own data, the relative start times of the data records, and the start times of the runtime measurement request, the first AccessPoint can calculate the two-way signal propagation time as described below. The time difference for the entire procedure can be formulated as

'AP m ~' AP 1 ⁼ 'Θ 1 ⁺ ^' AP-M d ⁺ 'e 2> (2)

With

- _A P. _{1 of} the start time of the transmission in the AccessPoint,

, _AP .m the reception time of the signal from the AccessPoint-M at

AccessPoint, f _e .i of the signal propagation time from the AccessPoint to the AccessPoint-M, f _e . ₂ The signal propagation time from the AccessPoint-M to the AccessPoint

The time delay between the receipt of the signal and the transmission of the response radio signal from the second access point AP-M _1, which thereby becomes a first access point (AP) in the logic of the measurement algorithm, is

Δ'AP-Md ⁼ 'AP-M 2 ~' AP-M ΠV W)

With

JAP-M.2 of the start time of the transmission in the AccessPoint-M and fop- _M .m of the reception time of the signal from the AccessPoint at the AccessPoint-M.

Due to the fact that the clocks for the time measurement in the AccessPoint-M are not synchronized with the other AccessPoints, relative times must be expected. The time difference between the beginning of the signal recording and the return is

Δ'AP-M = 'AP-M 2 ~' AP-M 1 ^• (4)

The time difference between the beginning of the signal recording and the reception of the signal in the AccessPoint is

Δ * AP m ⁼ * AP m ~ 'AP 2 ^■ w)

The fact that both signals must pass through the same propagation distance, is

'.1 =' .2 - (6) Thus, the basic equation (2) can be converted into

² K = ^At fi P m + 'AP 2 - ^f AP 1 - ^At AP-M + ^Δ ' _A PM m ^■ (?)

All of these variables can be measured with the described algorithm and can correspond to (1)

<e = O (8)

be set. As a result, the runtime measurement for an AccessPoint-M can be calculated as follows:

O = 2 i ^At AP m + 'AP 2 - ^t AP 1 ^"Δ ' AP-M + ^Δ 'AP-M m) ^• (9)

The values of /ΆP.1 _I ^ AP.2, and A ^ AP-M can be determined by digital counters on the respective processors. The correlation function is used to determine Δ.λp-Mm and Δf _A pm.

From this, the deviation of the system times between the first access point (AP) and the second access point (AP-M) can now be determined

'AP-M sync ⁼ ' AP-M 1 ~ * AP 1 ^• (1 0)

and thus through

^ AP-M sync ⁼ 'AP-M m ~ ^Δ ' AP-M m ~ 'AP 1 ^■ \> *)

to calculate.

Such correlation and runtime measurements between the AccessPoints can be made on the basis of the following three signals:

- BC signals with data packets

- BC signals without data packets - Data packets and acknowledge signals used to confirm the correctness of the data transmission.

If data packets and acknowledge signals are used, two correlation and delay measurements, one on the outward and one on the return path of the data transmission, can be carried out. This means that all necessary measurement data is available between two AccessPoints in order to carry out the calculation procedure shown.

With the method described so far, the relative deviations of the system times of the four AccessPoints positioned at known locations were determined. This time reference is assumed to be known and constant in the following.

In the second step of the method according to the invention, one-way runtime measurements are performed between ClientNode and the AccessPoints. For this purpose, the ClientNode sends out a PR signal, which in the exemplary embodiment described here is received by four AccessPoints. At least the respective reception time of the PR signal is detected in the access points. The PR signal is used in the four AccessPoints, analogous to the method previously described for the BC signals, for a correlation measurement to measure the relative time difference in the arrival of the PR signal at the respective AP to the internal AP oscillator. The resulting data records or measured values are first stored with the identifier of the sending ClientNode in the internal buffer memory of the respective AccessPoints and then forwarded to the control computer for further evaluation.

In the third step of the method according to the invention in ControlComputer the

Location coordinates of the client node based on the known locations of the access points and the detected reception times for the PR signal by the at least four

Calculated access points. The algorithm described below is used. The algorithm starts from a coordinate origin chosen arbitrarily in space, from which a Cartesian coordinate system spans within which the following coordinate points are given. It is assumed that an access point is given by the vector s according to equation 2 as well as the client node by the vector e according to equation 3. A distance vector I ₀ between the access point and the client node can be calculated as the difference between the two vectors in the following way according to equation (4):

Due to the relative time measurements generated in the second method step, the equation (1) must be expanded such that now not the absolute signal delay time to, but the time relative to a time scale t _m is used. This time scale may initially be virtual or arbitrary, but in the technical realization it will correspond to the time scale, ie the clocking of the processor, of a so-called master access point selected from the four access points. The remaining AccessPoints are referred to below as Managed AccessPoints. The phase, ie the state of the time scales in the Managed Access Points and in the ClientNode can be different, but the clock frequency is generally the same for all subscribers. As a result, the time value when the ClientNode signal arrives at the AccessPoint corresponds to a relative time sum or time difference, depending on the sign.

with f _m measured differential time to ideal one-way signal propagation time, and fc _N relative CLIENTNODE time, ie the deviation of the time scale in CLIENTNODE of the time scale in the master access point.

The reference phase in the master access point changes only by the minimum drift of the processor oscillators. The phases of Managed AccessPoint time scales,

JAP, relative phase error of managed access point time, i. the time scale difference in the managed access point from the time scale in the master access point,

is assumed to be constant and known. In operation, these values must be measured and updated at certain intervals, which depend essentially on the quality of the oscillators used in the AccessPoints.

This widens (15)

'm =' θ + 'cN +' AP - (16)

The distance from the ClientNode to an AccessPoint (AP) can be calculated with (1) as follows:

The ball equation becomes the relationship

((j) ² + (/ * • - y _CN ) ² + ( ^Z AP ^"Z CN) ² = ^C ('m ^" ' CM ^" Α ^p ) (18) derived between the relative transit time measurements and the coordinate differences. Equation (18) contains four unknowns, the three ClientNode coordinates, XCN, / CN, ZCN and tcu, the time scale deviation in the ClientNode from the time scale in the Master AccessPoint. As already mentioned, the difference of the time scale of the managed access points from the master time scale is also measured by the system, but can be assumed to be known and constant here. To calculate the four unknowns, four equations of determination are necessary. This means that a (4,4) equation system, with four independent relative runtime measurements to four different AccessPoints, are necessary:

Vl * AP 1 ^{~ X} CN) ⁺ (./AP 1 ~ ^" / CN) ⁺ ( ^Z AP 1 ~ ^Z CN) ^{= C} ('m 1 ^~ ' CN ^~ * AP 1) '

To solve this non-linear (4,4) equation system and calculate the solution tuple, two steps are used. In the first step, this equation system is linearized, in the second step the solution is calculated using an iteration method. For linearization, for example, the method of Newton-Kantorowitsch can be used. The equation system must be a suitable approximation solution

be linearized. Since the calculation of the ClientNode position is ambiguous, for the approximate solution the distance to the intersection of the three spherical surfaces, which corresponds to the ClientNode position, will be less than the distance to the ambiguity point. This is necessary for the process to converge to the correct solution. In practice, therefore, the client nodes should only be turned on or initialized in areas where the geometry-dependent ambiguity points are far apart.

For the approximate solution (20), the approximate signal propagation times t ^] for the four AccessPoint measurements are calculated with the known AccessPoint coordinates (12). As an approximate solution for the measured signal propagation time arises

cL. =

FfeJ is the error term and stands for the difference between the actual and the approximated solution. If the error term is neglected, the result for the linearized equation system is a matrix representation

This equation can be combined with

and

to

Δl '= A'Δe'; (26)

shorten in the presentation. The components of the matrix A are calculated as partial derivatives at the linearization point, ie

[ ^X APι ^X CN) ^a , i = / = 1 ... 4; (27)

ΦCN) e ⁽⁰⁾ CJ Jel X 'APi * CN) + (y _A P, - / CN) + { ^Z AP, ^{~ Z} CH)'

To solve equation (26) for the desired Δe ', one multiplies on both sides with the inverted matrix A'^"1

Δe '= A'- ¹ Δl'. (31) This allows the initial value to be corrected and the improved ClientNode coordinates

be calculated. At this point, the iteration process begins until the condition

V (XcN -X ^) ² + (y _CN -. / ÄF + (* »-2ffi) ² ₊ {t _CH -Cf <ε (33)

is fulfilled. The error barrier ε is selected according to the desired accuracy. If the abort criterion for the iteration is not met, then the improved solution tuple (32) is used to jump into the iteration loop. Due to the fact that the coordinates and the time phase have changed, the matrix A must be recalculated and inverted. Subsequently, again Δe 'and the improved solution tuple are to be calculated.

From equation (32) it can be seen that with this method not only the coordinates of the client node but also its relative time, i. calculated the deviation of the time scale in the ClientNode from the time scale in the AccessPoint, at the time of the signal transmission on the ClientNode. This difference time or time phase can be used to synchronize events in the ClientNode. Of course, the achievable accuracy depends very much on how exactly the time of the signal transmission was determined on the ClientNode. If, on the other hand, the goal is to realize very simple ClientNodes in the present case, then the time of transmission is not measured and thus a considerable reduction in the hardware and software expenditure in the ClientNode is achieved.

Advantageously, before starting the position calculation for a ClientNode, the ControlComputer checks whether at least for this ClientNode Measurements are available for four different AccessPoints for the same measurement sequence. If so, the constellation with the best quality of the measurements and the best geometric distribution is used to compute the coordinates of the client node according to the described algorithm. It is thus also possible to detect areas that are not supplied with at least four AccessPoints automatically in the ControlComputer.

In addition, the method according to the invention can also be used to determine changes in the spatial arrangement of AccessPoints with respect to a selected Master AccessPoint. For this purpose, determined distances or determined signal propagation times between the individual access points are compared with known distances or signal propagation times. Detected deviations can create a system warning or a control signal.

With the method according to the invention, it is furthermore possible to determine areas which are not within the reception range of at least four AccessPoints by evaluating the datasets of the AccessPoints and the known positions of the AccessPoints. This allows the spatial constellation of AccessPoints to be tested in a given environment during operation and further expanded and optimized by adding additional AccessPoints. This represents a significant facilitation in the design and construction of a positioning and navigation system compared to conventional constellation calculation methods. The system is thus able to configure and optimize itself. In case of error, i. If one or more access points fail, the existing redundancy is automatically exploited and the coordinates of the ClientNodes are calculated with a suboptimal AccessPoint arrangement. This can also be done dynamically if the field strength of certain access points is attenuated by interference, so that the accuracy of the measurement can no longer be achieved.

By the described method for determining the position, the system can also be realized with greatly simplified ClientNodes. In the simplest case exist the client nodes only from a simple transmission and control electronics, which repeatedly sends a PR signal with the identifier of the ClientNode at intervals over time.

Thus, the accuracy of determining the times at which the signal leaves the transmitter and arrives at the receiver can be substantially increased by the following three methods:

1. Averaging of calculated maximas of the correlation functions over several measurement data sets.

2. Increasing the accuracy of the calculation of the maximum of the correlation curve by a local interpolation of the curve in the area of the maximum. The maximum, which is usually between two values of the correlation curve, can then be calculated by a geometric relationship from the local interpolation.

3. Calculation of the maximum of the correlation function by an interpolation calculation of the additional intermediate data values lying between two values of the correlation function corresponding to the sampled values. The newly created correlation curve contains significantly more data points compared to the original one. The interpolation curve between the original support values is resampled and filled with additional support values. By means of a second correlation with a new filter function adapted to the new output correlation curve, the resolution and accuracy of the calculation of the maximum of the correlation function can be substantially increased.

For a further improvement of the resolution and the accuracy of the relative propagation time measurement, the changes of the separately calculated maximum values of the correlation function in the in-phase and in the quadrature-phase evaluation branch can be used. This makes it possible to measure the change in the signal phase caused by the oscillators and the distance changes of the transmitter and the receiver with a very high resolution and accuracy. These Measurements can additionally be used to increase the accuracy and reliability of the position determination.

In spite of changing ambient temperatures, the reliability of the measurement data can be improved over a larger temperature range by integrating several temperature sensors in the AccessPoints with which the temperatures of the components which determine the signal propagation time can be measured. The current measurement data can then additionally be temperature-compensated with the aid of a previously determined map function.

In addition to increasing the resolution and accuracy of the basic measurements, to determine the distance, i. Running times and phases, to increase the integrity of the measured data, these can be weighted with the calculated current energy of the received signal, according to its height.

In problematic environments, additional sensors, such as wheel sensors, gyro sensors or beacon sensors, can be used to verify the integrity of the basic measurements and weighted according to their reliability factor determined using the additional sensors. As a result, any errors in the basic measured variables are detected and an improvement in the static error bandwidth is achieved.

In order to increase the integrity of the position determination in already known movement spaces of the client nodes, the positions calculated in a first step can be corrected by a three-dimensional map function. Systematic position errors are stored and parameterized in the map function and are used to determine the resulting positional accuracy, particularly in indoor or outdoor applications, i. in so-called IndoorNav systems, increase.

In principle, it is possible, in particular with the IndoorNav system, a so-called "virtual system time" from the measured data and the calculated gears from the local system times derived in the AccessPoints of quartz oscillators. The virtual system time is thereby obtained from a weighted averaging of the calculated individual courses of the respective quartz oscillators. This method gives a highly accurate virtual system time. So that the accuracy of this virtual system time can also be used within the system for control tasks, the calculated individual deviations are parameterized and returned to the AccessPoints. This allows the accuracy of the calculated virtual system time, via the ensemble of all quartz oscillators, to be returned to the individual oscillators in the AccessPoints. In this context, an ensemble time can alternatively be used instead of "virtual system time." With the real clock signal of the oscillator and the correction parameters, the local system time of the access points can be corrected with their time signal and thus for highly precise control tasks, for example of control systems. Commands for initiating certain actions can be generated to the AccessPoints as well as to the ClientNodes in advance with a precise activation time in the virtual system time.This virtual system time can then be converted to the local corrected system time of the hardware signal to trigger control signals with a temporal resolution and accuracy in the subnanosecond range.

Due to the accuracy of the basic measurements and the virtual system time, the data from several AccessPoints can be combined into so-called AccessPointClustem. From the difference of the phases of the basic measured variables can thus calculate the direction of arrival of the received signal. This incident direction information, in turn, can be used to perform plausibility checks, detect incorrect measurements, and increase the accuracy and integrity of the entire IndoorNav system.

Similar to an AccessPointCluster, it is also possible to build a ClientNodeCluster by using the data from multiple ClientNodes. This can be off the difference of the phases of the basic measured quantities, with known ClientNode coordinates, the direction of incidence of the received signals of the AccessPoints are calculated. If the ClientNode coordinates are only approximately known, then the infeed direction information can still be used to perform plausibility checks to accomplish the accuracy and integrity of the calculated positions.

If the client nodes belonging to a cluster are mounted on a carrier with known relative positions, the angle information of the incident receive waves of the access points and thus the spatial position of the carrier of the client node cluster can be calculated from the relative coordinate differences and the differences in the phases of the basic measurements.

So that the electronics for the measurement of the basic measured quantities can be kept as small and simple as possible, only the data required for the calculation of the basic measured quantities are recorded in the received signal and further processed. The data stream for the communication is received and decoded in a second reception branch. By building this second branch from highly integrated standard components for communications applications, the parameters that determine the cost of the overall module, such as physical size, power consumption, standard hardware and software components, etc., can be significantly optimized. However, in this case, the measurement data will lose their unique association with the sending originator, i. either a respective ClientNode or AccessPoint, which can only be restored by decoding the entire communication data stream. The assignment of the data can be restored in the ControlComputer using the following methods and / or a combination thereof:

1. By comparing the randomly varying time intervals of the reception and transmission events of the data from the measurement data branch with the data from the communication data branch, it is possible to search for an unambiguous association and restore it. 2. The calculation of the cross-correlation of the data of the communication branch with the data of the measurement data branch can also be used to restore the unambiguous assignment of the data and thus to determine the originator of the signals.

3. To further increase the reliability of the association, additional features present in both reception branches, such as signal strength, may be used.

In order for the position coordinates of a ClientNode to be calculated, data from several different AccessPoints are required, as repeatedly executed. The number of data records required for the evaluation of different AccessPoints or ClientNodes increases in addition if an AccessPointCluster or ClientNodeCluster evaluation is to be used. Due to different factors, such as the non-priority-based transfer of data from the AccessPoints or ClientNodes to the ControlComputer, the measurement data can not arrive at the processing and evaluation unit in the time sequence corresponding to the measurement. For this reason, the measurement data necessary for the respective evaluation must be sorted before the further processing and thereby brought back into the original temporal reference, in accordance with the current evaluation. For this purpose, ring-shaped buffer memories are used for each transmit and receive measurement electronics in the ControlComputer, in which the data are stored in this order after being received by the transmission network.

For each ClientNode, an additional ring buffer is set up, in which the identifications of the AccessPoints received for the last position calculations are stored with the currently used frequency band. This group of AccessPoints is also called a "Relevant Zone" for a ClientNode. If an evaluation is required for a particular ClientNode, an order to collect the necessary data is entered in the so-called "NavScheduler." As soon as this order is processed, the last record of this ClientNode is saved in its "Relevant Zone" ring buffer. used. The ring buffers of the various AccessPoints and ClientNodes in the system are now searched for the corresponding temporally closest records. Hereby, quality criteria can be applied, according to which the relative age of the data records, with respect to the ClientNode data record, will be evaluated. If one or more data records are too old, then the order for data collection can be repeatedly entered into the "NavScheduler" repeatedly and / or its update can be triggered by commands. This means that the Relevant Zone becomes a "Valid Zone." This means that the necessary processing steps can now be performed for data evaluation and, finally, for position and / or position calculation.

In many applications, in addition to the localization service, high data transfer rates between the individual ClientNodes and AccessPoints are necessary at the same time. In this case, dedicated AccessPoints are used only for payload data communication. All access points that are used for localization work on at least one "ServiceArea" in the same frequency band The current frequency band can be changed by the ControlComputer, within the available frequency bands.

To control the switching of the frequency bands, commands are used which use the virtual system time as the time for the next switching of the frequency band to all AccessPoints. This virtual system time is then converted to the local system time for each AccessPoint. The pattern for the change of the frequency bands can be selected such that an optimized temporal assignment of the frequency bands results, corresponding to the number of client nodes and the necessary position update rate. In order to be able to reliably detect newly added ClientNodes, the AccessPoints dedicated for communication tasks send a message to the ControlComputer during the log-on process of the ClientNodes, which creates a ring buffer for the Relevant Zone of the new ClientNode. As a result, the NavScheduler is able to specifically address the newly added ClientNode with corresponding control and data signals.

If different frequency bands are currently only used by client nodes with little or no data traffic, then the relative age of the data of the valid zone can become large and thus significantly degrade the positional accuracy that can be achieved with the data. In this case, immediately after the frequency band change, with appropriate commands and signals, the recording of measurement data for the individual ClientNodes is initiated. This procedure and the load-dependent pattern for changing the frequency bands ensure that all existing and newly added client nodes are taken into account for the localization service.

This method also has the advantage that the IndoorNav system can be docked to an existing communication infrastructure without having to change standard WLAN client nodes or significantly disrupting their communication function or affecting their function.

Claims

claims

1. A method for determining position inside and outside of buildings, with at least four positioned at known locations transmitting / receiving stations, so-called AccessPoints (AP), at least one mobile transmitting unit, called ClientNode (CN), by means of wireless communication technology, in particular by means of radio signals , communicate with each other, and with at least one arithmetic unit, a so-called control computer, which is connected via a network structure with the access points, wherein the following method steps are carried out:

- establishing a fixed time reference between individual non-time-synchronous system times of the respective access points,

Sending a PR signal (Probe Request signal) through the ClientNode,

Receiving the PR signal by the access points and detecting the respective reception time of the PR signal in the system time of the respective access point,

Calculating location coordinates of the client node on the basis of the known locations of the access points and the recorded reception times for the PR signal by the at least four access points, taking into account the fixed time references between the individual system times of the respective access points.

2. The method according to claim 1, characterized in that the establishment of the fixed time reference between the system times of the at least four access points takes place on the basis of the following method steps:

- Broadcasting a BC signal (broadcast beacon signal) from a first access point, simultaneously triggering a time measurement and storing the Transmission time at the first access point in the system time of the first access point,

Receiving the BC signal by at least one second access point, simultaneously triggering a time measurement in the system time of the second access point and storing the reception time in the second access point,

Sending a response signal as well as data from the second access point, the data comprising the time of reception of the BC signal and the time of transmission of the response signal in each case in the system time of the second access point,

Receiving the response signal by the first access point and storing the reception time in the system time of the first access point and the transmitted data,

Determining the relative deviation of the system times of the first and second access points on the basis of the known locations of the first and second access points, the transmission time of the BC signal from the first access point and the reception time of the response signal originating from the second access point in the first access point, respectively the system time of the first access point, as well as the time of reception of the BC signal and the time of transmission of the response signal from the second access point, in each case in the system time of the second access point.

3. The method according to claim 1 or 2, characterized in that the transmission and reception of signals of the AccessPoints and / or the at least one ClientNode using the network standard IEEE 802.11 or the transmission protocol CSMA-CA (Carrier Sense Multiple Access with Collision Avoidance) or a spread signal modulated with a pseudo-random signal is performed.

4. The method according to claim 3, characterized in that the ClientNode in addition to the transmitting unit has a receiving unit, are received with the at least signals for synchronization of the transmission protocol used.

5. The method according to any one of claims 2 to 4, characterized in that the exchange of the PR signals, BC signals and the response signals in the time division multiplex or frequency division multiplex operation or in a combination of both modes.

6. The method according to any one of claims 2 to 5, characterized in that each access point has a unique identifier and this is transmitted when transmitting BC signals and / or response signals.

7. The method according to any one of claims 1 to 6, characterized in that each ClientNode has a unique identifier and this is transmitted when sending PR signals.

8. The method according to any one of claims 6 to 7, characterized in that the respective signals of ClientNode and AccessPoints contain next to the identifier further data.

9. The method according to any one of claims 2 to 8, characterized in that temporally before the sending of a BC signal or a response signal from an access point, a trigger signal (RTMS command = Runtime Measurement Service) is sent, by receiving it in the other AccessPoints a data recording is activated.

10. The method according to any one of claims 1 to 9, characterized in that prior to sending a PR signal from the ClientNode a trigger signal (RTMS command = Runtime Measurement Service) which, when received, activates data recording in each of the AccessPoints.

11. The method according to any one of claims 1 to 10, characterized in that when receiving and / or transmitting a signal through an access point at least part of the signal, with the signal transmitted data, and / or further characteristic for the transmission and reception data stored in the AccessPoint and / or forwarded to other AccessPoints and / or to the ControlComputer.

12. The method according to any one of claims 9 to 11, characterized in that the response signal of an AccessPoint contains at least information about the start time of the data recording in the respective access point and the sending time of the response signal.

13. The method according to any one of claims 9 to 11, characterized in that a data unit is sent by an access point in time after sending the response signal, in the at least information about the start time of the data recording in the respective access point, the sending time of the response signal and at least part of the data stored in the AccessPoint is included.

14. The method according to any one of claims 6 to 13, characterized in that the response signal contains the spatial coordinates of the known location of the respective access point.

15. The method according to any one of claims 8 to 14, characterized in that the data of the signals are compressed before the respective signal is sent.

16. The method according to any one of claims 1 to 15, characterized in that the control computer is connected via a wireless or wired Ethernet network structure with the access points.

17. The method according to any one of claims 1 to 16, characterized in that based on the data obtained from the Access Points and forwarded to the ControlComputer data, the position of the at least one ClientNodes is calculated.

18. The method according to any one of claims 1 to 17, characterized in that the control computer is provided at the location of an access point.

19. The method according to any one of claims 2 to 19, characterized in that the determination of the relative deviation of the system times between the first and second access points using the following relationships:

^'■ AP M.sync ~ ^ AP Mm ^ AP-M ^ .m APA

¹ _AP - _M.Sync Deviation of the system _times of the first access point AP.1 and of the second access point AP-M ^l _AP - _Mm Receive time of the signal from the first access point at the second

AccessPoint, t _ΛPΛ Start time of the transmission in the first AccessPoint and

^ ¹ _AP - _{M. m} Time difference between the beginning of the signal recording and the return of the response signal in the second access point

20. The method according to any one of claims 1 to 19, characterized in that calculating the position e of the client node under

Using the following relationships: τ I ( ^X AP 1 - ^χ C _N Y + (y _AP 1 - y _CN f + ( ^Z AP 1 - ² CN J = ^C ('m 1 ^{~ 1} CN ^" ^ 1} ~ * CN ~ ^t AP ih Λl ( ^X AP 4 ~ ^X CN) ² + (y _A P 4 - y _CN f + ( ^Z AP * ~ ^Z CN Y = ^C ('"4 ^" * CN ^" ^ J

With

s. Spatial coordinates of an AccessPoint

e Space coordinates of the client node

difference time measured on the AccessPoint i

⁴ CAf Deviation of the time scale of the ClientNode to

system time

'AP i Deviation of the time scale of the ith access point at system time

21. The method according to any one of claims 1 to 20, characterized in that the method according to one of claims 1 to

20, is carried out in a timely manner.

22. The method according to any one of claims 1 to 21, characterized in that from the access points, a master access point is selected with a known position, and that a time-recurring review of changes in location of the other access points with respect to the master access point based on the following procedure :

Send a BC signal from the Master AccessPoint and the same time

Triggering a time measurement in the Master AccessPoint, Receiving the BC signal by a further access point APj and thus simultaneously triggering a time measurement in the access point APj,

Sending a response signal (SAP.I) from the access point APi and

Receiving the respective response signal (S _A pi) by the master

Access Point

Determining the propagation times of the BC signal between the master access point and the access point APi and / or the running time of the response signal SAP.I between the access point APj and the master access point and comparing with corresponding known transit times, and / or

Determine the distance (DMAP API) between Master AccessPoint and the

AccessPoints APj and compare with a corresponding prior art

Distance.

23. The method according to claim 22, characterized in that a deviation between a known duration or a known distance to the determined running time or the determined distance is stored and / or generates a system warning and / or a control signal.

24. The method according to any one of claims 1 to 23, characterized in that after transmission of a PR signal, a number n of access points is determined, which have received the PR signal.

25. The method according to claim 24, characterized in that for the case 1 <n <4, the position for the corresponding ClientNode approximately by a comparison, resulting from the data obtained at the n AccessPoints possible positions of the ClientNodes with the last unique known position of the ClientNode.

26. The method according to any one of claims 24 to 25, characterized in that the following method steps are carried out: Determining a location of a ClientNode from which a PR signal sent by the ClientNode of n AccessPoints with n <4 is received, determining a spatial arrangement of AccessPoints at which a PR signal of n> 4 sent by the ClientNode at the determined location AccessPoints is received.

27. The method according to any one of claims 1 to 26, characterized in that when using a plurality of ClientNodes at least a subset of ClientNodes, a so-called ClientNodeCluster is formed, that phase differences between received at the location of the ClientNodes received signals are derived from a signal , which has been sent by at least one access point, and that on the basis of the phase differences, the direction of arrival of the received signals is determined, which is used for another Plausibiltätsüberprüfung.

The method of claim 27, characterized in that the client nodes associated with a client node cluster are mounted on a contiguous carrier.