Apparatus and method for determining a coincidence of a position with a reference position

The present invention relates to devices and methods for determining position matching or matching with a reference position, and may be used, for example, specifically for positioning or navigating mobile terminals in wireless communication networks.

Various positioning techniques are available to find a person with a mobile terminal. Perhaps the best known system for positioning and / or outdoor navigation is the Satellite Assisted Positioning System (GPS). For positioning and / or navigation in building interiors and / or indoor areas, such as infrared systems, RFID (radio frequency identification) systems, or IEEE 802.11 WLAN (wireless local area network) field strength assessments. Various approaches are known. Currently, GPS systems are available as a reliable method only for outdoor areas. More recent extensions include high-sensitivity receivers and so-called A-GPS (assisted). GPS) and others show an attempt to create a technology that can be used inside the building. Here, A-GPS is a combination of a satellite-based GPS system and the reception of so-called assistance information from a mobile wireless network. However, at present, these techniques do not provide the desired average accuracy. Infrared and RFID systems are generally not available in full range and must meet special prerequisites.

For example, with the increasing popularity of wireless wireless networks based on WLAN standards, these wireless networks are serving as the basis for new positioning methods.

In general, the positioning methods that have been used so far include, for example, triangulation, neighborhood relations, and time-measured lalation of lalation using field strength evaluation. Whether these methods require the location of the fixed transmitter and / or base station to be known, or whether they are pre-trained at a typical location in the area covered by the positioning method. This is a positioning method in which either is performed.

In WLAN-based positioning systems, the so-called received signal strength (RSS) fingerprint method is often adopted as the basic method. This method is based on the assumption that the signal strength of the radio signals of some radio stations received and / or receivable at the current location uniquely characterizes the location or location. For multiple reference locations or reference locations, the current location is the current measurement (if there is a database of identification of radio stations received and / or receivable there, as well as the field strength of the corresponding radio signal. From transmitter identification and signal strength values), it can be estimated by performing a match between the currently measured measurements and the database reference values. For each reference point, this matching assesses how similar the pre-recorded and / or reference value is to the current measurement at the current location. The most similar reference point then estimates the current location.

For the reference database, the signal strength is experimentally determined by test measurements at a sufficient number of points. This creates a list of base stations (access points) with the associated received field strength and quality for each location where the test measurement was performed. In a WLAN execution example, such a reference database may include, for example, the following parameters.

<tables num="1"><img file="JP2010540916A_D0001.tif" /></tables>

That is, the table contains the following information: --Reference point identification (RID) --MAC address of receiving station --Received field strength of access point (RSSI; 46560 means -46.560dbm.) --Position in Cartesian metric coordinates (x, y, z; 24583 means 245.83m), and --Time when measurements were captured WLAN signals should theoretically be measured only at relatively low field strengths, but exhibit relatively unreliable behavior as to whether they are "measurable" or "non-measurable." The PGS (percentage) column shows, in percentages, how many times this station was seen during the period of acquisition of measurements. (That is, PGS = 90 means that the station was measured 9 times on average in 10 measurements.) PGS value is the teaching of the reference position and / or reference measurement packet for each radio transmitter. It should be judged at the time of the above and understood as an evaluation standard of its reliability. Within a measurement time frame, there are a fixed number of possible measurements from the radio transmitter, for example with a fixed sampling interval of 200 ms. The PGS value is a percentage of the radio transmitter's (RSSI) value that was really measured within the measurement time frame with respect to potentially possible measurements. The reference position is captured, for example, every 200 ms, over a longer time frame (eg, 6 to 10 seconds), ideally for a calibration period. In this regard, FIG. 5 shows an example of the waveform of the received signal of a particular radio transmitter, which radio transmitter can only be received relatively unreliably at the measurement position. Over a 10 second measurement time frame, the particular radio transmitter can only receive for about 3 seconds, producing a PGS value of about 30% for this radio transmitter.

For positioning, the currently captured measurements are compared to the database. The most similar reference value, or the integral value of the most similar reference value, is accepted as the current position. Various methods are possible for matching, and the most widely used is the least distance method in the signal space.

The RSS fingerprint method gives good results in indoor and outdoor areas. Due to the fact that the location of the fixed radio transmitter does not need to be known, the method is well suited for unknown environments with unknown infrastructure. The fingerprint approach for matching envisions a fixed, immutable infrastructure. Many known solutions also envision a limited area where the signal of each radio transmitter can be received anywhere.

<p> RSSI values of multiple radio transmitters in which the transmitter identification pre-recorded at the reference position is the same as the transmitter identification given at that position in order to determine a match or match of the current position using the reference position. Are often compared to each other in the traditional way. The smaller the difference in RSSI value between wireless transmitters with the same transmitter identification, the higher the possibility of matching the current position with the reference position. However, this technique carries the risk of misposition estimation, for example, the number of radio transmitters for which the transmitter identification pre-recorded at the reference position is the same as the transmitter identification given at that location. If it is small, then the difference in RSSI values is also judged to be small, and an erroneously estimated match can be derived as a good one.</p><p> Starting from this background, it is an object of the present invention to provide an improved concept of matching between currently measured values and pre-recorded reference values with respect to the prior art.</p>

<p> This object is achieved by a device having the characteristics of claim 1, a navigation device according to claim 19, and a method according to claim 22.</p><p> A further embodiment of the present invention is a computer program for executing the method according to the present invention.</p><p> Considered, at the current (geographical) location, the currently given and / or measured, fixed-position radio transmitter values and / or characteristics (eg, transmitter identification and signal strength values). Matching between pre-recorded reference values and / or characteristics at the (geographical) reference position is the characteristic of the radio signal currently measured at that location and the reference of the radio signal pre-recorded at the reference position. What can be achieved by a kind of filtering of values is the result of the present invention. Here, the radio signals are pre-recorded at the reference position with a first number of radio transmitters whose transmitter identification pre-recorded at the reference position is the same as the transmitter identification given at the current position. It is divided into a second number of radio transmitters, where the transmitter identification and the transmitter identification given at the current location are different. That is, is the transmitter identification given only at the current position and not pre-recorded at the reference position, or is the transmitter identification pre-recorded only at the reference position and not given at the current position? , Whichever.</p><p> The set of measurements measured at the current location, including transmitter identification (eg, MAC address) for radio transmitter identification and associated signal strength indicator (RSSI), is as follows: In the measurement packet (measurement) It shall be referred to as packet: MP). According to a preferred embodiment of the invention, a given measurement at the current position and a pre-recorded measurement at the reference position are prefiltered into three groups. On the other hand, all wireless transmitters are separated from the measurement packets that are also included in the reference measurement packets inherent in the matching. A radio transmitter received at the current position and deviating (too audible) from the reference measurement packet implies that it cannot be at the reference position. These radio transmitters received in addition to the current location cannot be used in direct matching, i.e., matching between radio transmitters that have the same transmitter identification in the current measurement packet and the reference measurement packet. In addition, radio transmitters that were not received (inaudible) at the current location may be listed in reference data or reference measurement packets at the reference position. These radio transmitters and / or their measurements that were not received at the current location also require special processing and therefore direct matching, ie radios with the same transmitter identification in the current measurement packet and the reference measurement packet. Not supplied for matching between transmitters.</p><p> The matching unit matches the current measurement at the current position with all the reference points in the reference data in question. That is, the degree of matching between the current position and all the reference points in question is determined. The amount of matched reference points and / or positions adopts, for example, the last position of the mobile device as the start position, assuming that the user has not moved from that last position beyond a fixed maximum distance since then. It can be arbitrarily restricted by. Limitations of reference points to be compared can be made dynamically by factors such as movement model, quality of current position fixing, computing power limits or memory limits.</p><p> In each comparison between the reference measurement packet and the current measurement packet, the degree of matching is determined, which determines how well the measured value received at the current position matches the pre-recorded measured value at the reference position. .. The degree of matching for that position is determined based on the given characteristics of the radio signal, and both the characteristics of the first number of radio transmitters and the characteristics of the second number of radio transmitters are matched. The degree of matching is entered separately for the characteristics of the first number of radio transmitters and the characteristics of the second number of radio transmitters, which are taken into account in determining the degree. In the first number of radio transmitters, the pre-recorded transmitter identification at that location matches the transmitter identification given at the current location. In the second number of radio transmitters, the transmitter identification is given only at the current position and not pre-recorded at the reference position, or is pre-recorded only at the reference position and given at the current position. Either not.</p><p> According to a preferred embodiment, the characteristics of the first number of radio transmitters are weighted more strongly than the characteristics of the second number of radio transmitters. This is because the number of transmitters such that the transmitter identification pre-recorded at the reference position matches the transmitter identification given at that position was pre-recorded at the reference position in the calculation of the degree of matching. A second number of radio transmitters that differ from the transmitter identification and the transmitter identification given at that location, i.e., the transmitter identification was given only at that location and was not pre-recorded at the reference position, or Means that it is considered more strongly than the number of radio transmitters pre-recorded only at the reference position and not given at the current position. The weights of the properties of the first and second numbers complement, for example, the other complementarily.</p><p> More preferred embodiments and further embodiments are the subject of the dependent claims. Radio transmitters whose transmitter identification includes measurements of the radio transmitter that were received only at the current position and were not pre-recorded at the reference position, and that were pre-recorded only at the reference position and not received at the current position. By including it, the accuracy of determining the degree of matching between the current position and the reference position can be significantly increased. This can result in significant improvements in positioning and / or navigation results.</p><p> Preferred embodiments of the present invention will be described in more detail below with reference to the accompanying drawings.</p>

<figref num="1">It is a flowchart for demonstrating the method of determining the agreement between the present position and the reference position according to the embodiment of this invention.</figref><figref num="2">It is a figure which shows an example of a series of measurement packets.</figref><figref num="3">It is the schematic of the apparatus for determining the agreement between the present position and the reference position according to the Embodiment of this invention.</figref><figref num="4">It is a block diagram of the means for determining the degree of position matching according to the embodiment of this invention.</figref><figref num="5">It is a figure which shows the typical waveform of the wireless transmitter which can be received with low reliability.</figref>

For subsequent description, the same or similar functional elements include the same reference number in different embodiments, and thus the description of these functional elements is interchangeable with each other in the various embodiments illustrated below. It should be noted that there is.

Next, the concept of the invention for determining the matching or matching between the current geographical position and the geographical reference position will be described based on FIGS. 1 to 4.

To obtain the reference position, the user records the radio fingerprint as part of the database, for example, during the teaching phase for later positioning. In practice, teaching can be done, for example, using a PDA or smartphone. The geographic map around the target is filed, for example, as a bitmap. Upon teaching, the user marks the current position on the map and begins capturing the measurements. In practice, higher density teaching points have been found to significantly increase the amount of computation during positioning, but with little significant improvement in results.

However, collecting fingerprints manually is only feasible within a limited area. In urban areas and in the center of urban areas, so-called calibration boxes can be adopted for this purpose. They have a high precision GPS unit (so-called differential GPS) combined with a high precision inertial sensor (eg, an accelerometer and an electronic compass). These calibration boxes can collect fingerprints without manual intervention as they travel through cities and urban areas. Positioning via differential GPS continues for up to 20 minutes for position determination even if the GPS system fails, but high-precision inertial sensor technology enables average accuracy of tens of centimeters even in urban environments. So, for example, a covered road does not cause any problems with the calibration box.

FIG. 1 is a flowchart for illustrating a method of determining a match between the current position of the mobile terminal and the reference position.

The method of recognizing the match, schematically illustrated in FIG. 1, comprises a first step S1 of recognizing and / or also providing the characteristics of the radio signal of the fixedly placed radio transmitter at the current position, determining and / or The characteristics of the radio signal provided include transmitter identification that identifies the radio transmitter. Therefore, this means that, in particular, the transmitter identification of the radio transmitter is recognized in step S1. Further, in a preferred embodiment of the present invention, the received electric field strength, received power spectrum, signal-to-noise power ratio (SNR: S / N ratio), incident angle, propagation time, polarization or phase location (phase) of the radio signal. The electromagnetic characteristics of the radio signal, such as location), are recognized.

In the second step S2, the radio signal and / or the radio transmitter associated with the radio signal has the transmitter identification pre-recorded at the reference position the same as the transmitter identification given at the current position. Number of 1 N_eqA second number N that differs from the wireless transmitter in the transmitter identification pre-recorded at the reference position and the transmitter identification given at the current position._neqSeparated and / or filtered into wireless transmitters. That is, is the transmitter identification given only at the current position and not pre-recorded at the reference position, or is the transmitter identification pre-recorded only at the reference position and not given at the current position? , Whichever. According to the embodiment, the second step S2 further comprises a second number N._neqNumber of radio transmitters that were not received at the current position, that is, radio transmitters with pre-recorded characteristics at the reference position but no characteristics given at the current position._nhIncludes sub-steps for selecting. Number of radio transmitters not received at the current location N_nhThe larger the value, the less likely the current position corresponds to the reference position. In addition, step S2 includes an additional substep, in which there is no pre-recognized electromagnetic characteristic at the reference position, but there is an electromagnetic characteristic given at the current position, added at the current position. Number of wireless transmitters received in N_htmBut the second number N_neqIs selected from. Number of additional radio transmitters received at current location N_htmThe larger the value, the less likely the current position corresponds to the reference position. Therefore, the second number N_neqRadio transmitters are the number of radio transmitters that were not received at the current location N_nhAnd the number of additional radio transmitters received at the current location N_htmFrom N_neq= (N_nh+ N_htm).

Based on the characteristics of the radio signal given from step S1, the degree of matching and / or the distance value acc for the current position is determined in the third step S3 and the first number N_eqWireless transmitter characteristics and 2's complement N_neqBoth of the characteristics of the wireless transmitter of are taken into account in determining the degree of matching, and the first number N_eqWireless transmitter characteristics and 2's complement N_neqFor the characteristics of the wireless transmitter, the degree of matching is input separately. According to the embodiment, the first number N_eqThe characteristic of the wireless transmitter is the second number N_neqIt is weighted more strongly than the characteristics of the wireless transmitter of, and will be dealt with in detail below.

According to a mobile device in practice of the present invention, step S1 recognizing and / or providing the characteristics of a wireless signal is a mobile terminal and / or such as a WLAN-enabled PDA, a Bluetooth-enabled PDA, or a mobile phone. Performed by the client. To this end, the client comprises means for certifying and / or providing the radio signal characterization of the fixedly placed radio transmitter, which characteristics generally identifies the fixedly placed radio transmitter. And it is characterized by electromagnetic signal characteristics such as received electric field strength, received spectrum or received signal-to-noise ratio. Identification and / or identification of features of a fixedly located radio transmitter can be, for example, a MAC (media access control) address, base station identification, or cell identification.

The characteristics of the radio signal of the fixedly arranged radio transmitter are combined with the so-called measurement packet MP (i). This fact is illustrated in FIG.

Figure 2 illustrates three temporally continuous measurement packets MP (1), MP (2), MP (3), i.e. i = 1,2,3 from a WLAN network. MP (i) has multiple MAC addresses 22 and associated RSSI values, RSSI_kIncluding (i), index k indicates the kth radio transmitter. This means that the MAC addresses of the fixedly placed radio transmitters and their RSSI values received by the client are combined into the measurement packet MP (i) at each time interval.

FIG. 3 shows a device 30 for determining a match between the current position and the reference position according to the embodiment of the present invention. Here, the fixedly placed radio transmitter and / or base station can be received at the current location.

To this end, the device 30 comprises means 32 for providing the radio signal characteristics of the radio transmitter fixedly located at the current position, and the radio signal characteristics provided are the radio transmitter. Includes transmitter identification such as identifying MAC address. To this end, the means 32 may be coupled with a receiving antenna 33 for receiving the characteristics of the radio signal, such as the electromagnetic characteristics. In fact, the measurement packet MP (i) described above can be meant by the characteristics of the radio signal. The certified and / or provided measurement packet MP (i) is the first number N of the radio signal and / or the radio transmitter associated with them._eqWireless transmitter and 2's complement N_neqIs given to means 34 for separation with the radio transmitter. Where the first number N_eqRadio transmitters include a radio transmitter whose transmitter identification pre-recorded at the considered reference position matches the transmitter identification determined at the current position. 2's complement N_neq= (N_nh+ N_htm) Radio transmitters either the transmitter identification is given only at that position and not pre-recorded at the reference position, or is pre-recorded only at the reference position and not given at that position. Includes wireless transmitter. First number N_eqWireless transmitter and 2's complement N_neqIn order to determine the radio transmitter of, the means for separation 34 may be combined with a database 35 in which pre-recorded characteristics of radio signals from multiple reference positions, ie reference measurement packets, are stored. This means that the database 35 contains, for example, various pre-recorded measurement packets, each associated with a reference position. These pre-recorded measurement packets should then be referred to as the reference measurement packet RP. Therefore, means 34 separates the radio signal into two groups. The first group 36 is the first number N_eqIncludes the characteristics of the radio signal of the radio transmitter, while the second groups 37,38 include the second number N_neqIncludes the characteristics of the radio signal of the radio transmitter of. As already mentioned above, the second groups 37,38 are a group of radio signal characteristics of the radio transmitter not received at the current position and additional radio signals from the radio transmitter received at that position. It can be further subdivided into groups of characteristics 38.

The device 30 further includes means 39 for determining the degree of matching with respect to the current position, which can be combined with both means 34 for separation and database 35. The means 39 is formed to determine the degree of matching based on the given characteristics 36,37,38 of the radio signal, the first number N._eqRadio signal characteristics of wireless transmitters 36 and 2's complement N_neqThe radio signal characteristics 37,38 of the radio transmitter of the first number Neq are taken into account in determining the degree of matching and / or the distance value acc, and the radio signal characteristics 36 and the second number N of the radio transmitter of the first number Neq._neqThe radio signal characteristics 37,38 of the radio transmitter of the above input the degree of matching separately, that is, are weighted separately.

Means 32 provides signals from different base stations and / or radio transmitters with different signal strengths at any location and / or at any location, along with associated transmitter identification. For WLAN networks, such electronic fingerprints include a unique MAC address for each WLAN device and / or WLAN radio transmitter, and a list of received signal strengths associated with them, thereby characterizing the current location. .. It doesn't matter where the wireless transmitter is.

Positioning essentially consists of two steps. The first is the matching of the currently measured measurement packets with the fingerprints and / or reference measurement packets in database 35, and the second is the selection of suitable location candidates on the one hand and the other as well. Weighting and coupling of candidate positions to estimated positions.

In the matching phase, the deviation between the currently measured measurement packet MP (i) and the reference measurement packet RP in database 35 is determined. In fact, means 34 and 39 of the device 30 according to the present invention provide this purpose.

In order to find a possible candidate position for the current position from the various reference positions stored in the matching phase, means 39 uses the currently measured measurement packet and the pre-recorded reference measurement packet at the reference position. Determine the degree of matching between and. A schematic block schematic of means 39 for determining the degree of matching and / or distance value acc is shown in FIG.

As already mentioned above, the means 39 has a first number N on the input side._eqThe electromagnetic characteristics of the wireless transmitter 36, for example RSSI value, are supplied. In addition, a second number N, such as RSSI value_neqRadio transmitter electromagnetic characteristics 37,38 appear at the input of means 39. Where the first number N_eqRadio transmitter characteristics 36 include both signal characteristics measured at the current position and signal characteristics pre-recorded at the reference position.

According to the embodiment, in block 41, the signal characteristics pre-recorded at the reference position and the first number N_eqA difference is formed from the given electromagnetic characteristics at the current position of the radio transmitter. For example, a difference in RSSI value of the radio transmitter is formed in which the transmitter identification pre-recorded at the reference position matches the transmitter identification given at the current position. These RSSI value deviations ΔRSSI₁From ΔRSSI_NeqIs given to the addition block 42, which is N_eqRSSI value deviation ΔRRSI_n(n = 1, .... N_eq) Is added and the total value is ΣΔRSSI_nIs calculated. N_eqIs the first number N of radio transmitters in both the measurement packet and the reference packet_eqIs. The function ΔRSSI calculates the distance between two signal strength values. The Euclidean distance of the measurement, in decibels (db), can be selected, for example, as a distance function. In this connection, distance therefore does not mean spatial distance, but mathematical deviation. Following the addition by block 42, the total value ΣRSSI_nUses the weighting factor EQW, ie EQW · ΣRSSI_nWeighted as. Here, EQW defines a weight between 0 and 1, the distance of the measurements and / or the signal strength value ΣΔRSSI._nShows how strongly important the distance is in terms of being too audible or almost inaudible compared to the radio transmitter at the current location.

If the calculation of the degree of matching is stopped at this point, a reference position that does not actually match well may be selected as a candidate rather than a reference position that is more suitable for the current position. As an example of this, N for the first reference position compared to the current position_neqIf = 1 is obtained, that is, it is assumed that only one radio transmitter identification is matched between the reference measurement packet and the current measurement packet. If the corresponding RSSI values of the matching measurement packets happen to be separated by, for example, 2.5 dB, then ΣΔRSSI₁/ N_neq= 2.5 dB is obtained. In addition, N for the second reference position compared to the current position_neqIf we get a result of = 3, that is, we assume that the three radio transmitter identifications match between the reference measurement packet and the current measurement packet. If the corresponding RSSI values are, for example, 2dB, 3dB and 4dB apart, then the overall ΣΔRSSI_n/ N_eq= 3 dB is obtained. As a result, the second reference point is evaluated to be worse than the first reference point, which causes an estimation error. Embodiments of the present invention can prevent and / or at least reduce such estimation errors.

Reference number 37 is N that was not received at the current location_nhRadio signals from a number of radio transmitters, i.e. radio transmitters that have pre-recorded characteristics at the reference position but no characteristics given at that position, i.e. radio transmitters that cannot be received at the current position. Characterize the characteristics of. At block 43, the malus function and / or the malus value M_{nh, m}() (m = 1, ..., N_nh) Can be defined for each of the unreceived radio transmitters. This is the Marus value M_{nh, m}() (m = 1, ..., N_nh) Means that it can be defined for each station that is within the reference value but not within the current measurement. This may depend, for example, on how reliably the corresponding station that was not received could have been received at the reference position in the past. The result is a previously good reception capability of the station that was not received, i.e. a high RSSI value, for example, a high malus value. Therefore, according to the embodiment, the malus value M_{nh, m}() (m = 1, ..., N_nh) Is a group of stations that have not been received at the current position can directly proportional to the quasi RSSI value. In addition, the Malus function M_{nh, m}() (m = 1, ..., N_nh) Can be combined with the PGS value of the corresponding radio transmitter that was not received. A small PGS value in the reference database is, for example, the corresponding Marus value M._{nh, m}() (m = 1, ..., N_nhOnly small values of) can be derived. This is, for example, the following formula, M_{nh, m}It can be calculated by (PGS) = fixed malus + dynamic malus. Here, dynamic malus = fixed malus * PGS / 100. Therefore, according to the embodiment, the function M for the malus value for the radio transmitter that was not received._{nh, m}() (m = 1, ..., N_nh) Depends on the properties related to the received field strength and, like the model, the properties pre-recorded at a reference point, such as the environment or the quality of the measurements. Mnh malus values M for radio transmitters not received at the current position_{nh, m}() (m = 1, ..., N_nh) Is handed over to block 44 and the M of the radio transmitter that was not received_nhThe first sum of the Malus values ΣM_{nh, m}Determine ().

Additional N received at the current location_htmThe characteristics of the radio signal from the radio transmitters are given by reference number 38. What is meant by this is a radio transmitter that does not have the electromagnetic characteristics pre-recorded at the reference position, but does have the electromagnetic characteristics given at the current position. In block 45, the Malus function M_{htm, r}() (r = 1, ..., N_htm) And / or the Marus value may be associated with each of the additional radio transmitters received at the current location. This was not found in the reference position, but for radio transmitters included within the current readings, the Marus value M_{htm, r}() (r = 1, ..., N_htm) Can be defined. Here, the Marus value M_{htm, r}() (r = 1, ..., N_htmThe function for) depends on the current RSSI measurements of the radio transmitter, such as the environment, the quality of the measurements, the age of the reference data, as well as the model. According to one embodiment of the invention, this means that the means 39 for determining is a malus value M._{htm, r}() (r = 1, ..., N_htm) Is formed to be associated with an additional received radio transmitter at the current location, meaning that it depends on characteristics related to the received field strength of the radio signal, such as the RSSI value. Therefore, according to the embodiment, the malus value M_{htm, r}() (r = 1, ..., N_htm) Is directly proportional to the reference RSSI value of the station additionally received at the current position. In addition, the Malus function M_{htm, r}() (r = 1, ..., N_htm) Can be combined with the PGS value of the corresponding radio transmitter received additionally. For example, a smaller PGS value in the reference database is the corresponding Malus function M._{htm, r}() (r = 1, ..., N_htm) Can be derived.

Additional received wireless transmitter N_htmThe Marus values are passed to the addition block 46 and N_htmAdd up the Marus values and add the second total value ΣM_{htm, r}().

According to the embodiment, the first total value ΣM of the malus value of the wireless transmitter that was not received._{nh, m}(), And the second total value of the radio transmitter received additionally ΣM_{htm, r}() Is added and weighted by the weighting factor (1-EQW). That is, (1-EQW) and (ΣM)_{htm, r}() + ΣM_{nh, m}()).

Finally, according to the embodiment, a weighted sum of the differences between the pre-recorded electromagnetic characteristics at the reference position and the electromagnetic characteristics given at that position of the first number Neq radio transmitter. Value EQW ΣΔRSSI_nAnd the weighted total value of the Marus value (1-EQW) · (ΣM_{htm, r}() + ΣM_{nh, m}()) Is (N_eq+ M_nh+ N_htm) Normalizes to give the distance value acc between the current position and the considered reference position. The distance value acc is, for example,

<maths num="1"><img file="JP2010540916A_D0002.tif" /></maths>

It is calculated according to. When the distance value acc is determined according to Eq. (1), the greater the coincidence between the current position and the considered reference position, the smaller the distance value acc. This is because the larger the match, the more the sum of the differences ΣRSSI_nIs smaller, and the total value of the malus value ΣM_{nh, m}(), ΣN_{htm, r}It means that () becomes smaller. According to the embodiment, the distance value acc corresponds to the degree of matching.

According to other embodiments, the degree of matching can be the reciprocal of the distance value acc, or can be calculated by (1-acc) if acc never exceeds 1. .. This means that the smaller the distance acc, the greater the degree of matching. Of course, other law of calculation is possible as well, the first number N_eqWireless transmitter characteristics and 2's complement N_neq= (N_nh+ N_htmFor the characteristics of the wireless transmitter in), the degree of matching is input separately.

In embodiments of the invention, therefore, each additional station received or not received increases the distance acc. The processing of different stations in the fingerprints and current measurements affects accuracy. That is, stations not found in the memorized reference fingerprint, but appearing in the current measurement, give a strong indicator of the fact that this fingerprint is not suitable.

In a large area, the distance calculation for all memorized reference fingerprints according to Eq. (1) can take a lot of time. Therefore, preselection of the reference fingerprint is advantageous. The final calculated position of the terminal can already give an indication of the current position. The surrounding digital map can further limit the number of candidates. However, such area restrictions also pose a danger. If the estimated position is misplaced, the positioning can no longer be restored and will get stuck in its "worst" position. That is, it is also always a matter of checking the absolute quality of candidate positions. For example, if the absolute quality of the best candidate is too poor due to a very weak RSSI value, matching should be restarted without region limitation. At the end of the matching phase, multiple possible locations and / or candidate locations are obtained from the estimates that could have been determined for the current location.

The best position candidate determined in the matching phase is adopted for the so-called position calculation phase in the navigation device including the device 30 according to the present invention. To this end, a navigation device according to the present invention further includes means for outputting an estimate of the current position based on the degree of matching and / or distance acc communicated by device 30.

A means for outputting an estimate for that position calculates the position and / or position estimate for the mobile device from candidate positions that are not higher or lower than the default limits for the degree of matching and / or distance acc. .. Here, individual positions are not considered very much, but rather are arranged in the context of overall movement. The result represents an estimate of the current position of the terminal.

A simple implementation is, for example, the calculation of the weighted average of candidate positions. The value depends on the degree of matching of the candidate positions, for example, the reciprocal of the degree of matching, where the weights of the candidate positions are formed in the averaging. This so-called k-weighted nearest neighbor method gives really reasonable results. The average positioning error is a few meters.

However, estimates of the current position can also be determined by more complex methods, such as probability calculations such as the Bayesian method and Markov chains. Here, the basic idea is to compensate for errors in individual position estimation by subsequently considering and optimizing the entire path. The most probable route calculations may further include additional data such as surrounding maps. Motion estimation filters such as the Kalman filter can also be employed. Such filters estimate the direction of motion, velocity and acceleration based on the final position and establish future predictions. The matching between the prediction and the estimated position can recognize and correct incredible movements and jumps.

In the matching and location calculation phase, the surrounding digital map can provide important additional information. A digital map of the surroundings, where you can see the possible paths, helps to recognize the impossible movements and correct the impossible movements to the possible paths. Thereby, the positioning accuracy can be significantly increased. Digital maps consist of possible paths (active maps) that include only inaccessible areas (negative maps), or represent structural states (real maps) such as floors, walls, or doors. It can be either. In the field of vehicle navigation, it is customary to adopt a positive map. These include important metadata as well as roads and routes (eg, one-way road driving directions and speed limits). Positive maps are meaningful here as accessible roads and routes make up only a small portion of the land.

However, in buildings and halls, or on factory premises, routes are only restricted by minor obstacles. Here, the negative map or the real map is obvious. Positive maps for vehicle navigation are commercially available in several standard formats (such as GDF, SIF or ArcView), but for maps inside and around buildings, standards have been developed so far. It has not been. The traditional WLAN positioning method provides the user with assistance as a dedicated tool, leaving a map of the surroundings in the proper format. Basic information can often be obtained from bitmaps or CAD diagrams. Therefore, the development of 3D maps suitable for town or city and national navigation, including building details as well as streets and public buildings, is an important step towards standardizing map formats. is there.

In summary, it is pointed out that the concept of the present invention is also realized in software, depending on the situation. An example example is a digital storage medium with an electronically readable control signal that can work with a programmable computer system and / or microcontroller such that the corresponding method is executable, specifically a floppy disk, CD or It can be a DVD. Thus, in general, the invention is a computer program having program code stored on a machine-readable carrier for performing the methods of the invention when executed on a computer and / or a microcontroller. It is also a product. Thus, in other words, the invention can be realized as a computer program having program code to perform the method when executed on a computer and / or a microcontroller.