JP5305324B2 - Distance measuring method, distance measuring receiving station apparatus and position measuring system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable a high-resolution delay time measurement in an indoor multipath environment. <P>SOLUTION: A multicarrier signal in which a subcarrier frequency is synchronized is received by a receiving station R. Frequency domain constellation data 26 as phase and amplitude data of constellation data 18 is obtained in each subcarrier. When a signal received from a transmitting station T is represented as a combined delay signal of a plurality of delay signals having different amplitudes A<SB>i</SB>, delay times T<SB>i</SB>and phases &Theta;<SB>i</SB>corresponding to propagation paths, a relative distance between the transmitting station T and the receiving station R is obtained from the delay times T<SB>i</SB>for reducing a residual error between the frequency domain constellation data 26 and the delay signal. <P>COPYRIGHT: (C)2009,JPO&amp;INPIT

Description

  The present invention relates to distance and position measurement using radio. In particular, the present invention relates to a highly accurate measurement method, measurement apparatus, and measurement system that are suitable for indoor use in a multipath environment and have a narrow signal band.

In recent years, position recognition of mobile devices outdoors is facilitated by GPS or the like, and many applied technologies (positioning applications) have been proposed accordingly. Many applications can be conceived if the same position recognition is performed indoors, but radio waves from GPS do not reach indoors. There are also the following problems indoors. That is,
(1) There are many radio wave propagation obstacles such as walls and furniture in the vicinity of the mobile device whose location you want to know, and the level of the reflected radio waves from these propagation obstacles is high, and there is a mutual delay time difference. Since the difference is small, it is difficult to separate the direct wave and the reflected wave, and the accuracy of position detection deteriorates.
(2) Nevertheless, indoor scales often require higher measurement accuracy (eg, within 1 m, within 3 ns of delay time difference) than outdoors.
From the above, there is no measurement method that has been decisive in the past, and therefore, many positioning applications are not used indoors as outdoors.

  In wireless three-dimensional position measurement, signals from a plurality of transmitting stations whose positions are known are received, a specific signal in the received signal is detected, and a delay time from each transmitting station is measured. In general, the distance to each transmitting station is calculated and the position of the receiving station is determined. In the method using three transmitting stations (TOA method), it is necessary to obtain the absolute value of the propagation delay time of each transmitting station and receiving station. In the method using four transmitting stations (TDOA method), each transmitting station and the receiving station are received. It is only necessary to obtain a delay time (delay time difference) relative to the station.

  Whether the TOA method or the TDOA method is used, the delay time (difference) is obtained by precisely determining the timing of the incoming signal at the receiver. If there are multiple paths, the decision is ambiguous and leads to measurement errors. In order to apply indoors, it is necessary to decompose a multipath signal having an arrival time difference of about several nanoseconds and determine a delay time of only a direct wave included therein. Conventionally, this is done by multiplying a template signal in a transmission signal by the same template signal while changing the delay time, and calculating a correlation between them in a time domain correlation method (for example, Patent Document 1). ) And a method of obtaining a delay time spectrum by performing inverse Fourier transform on a received signal (for example, Patent Document 2). Since it is known that the Fourier transform of the correlation coefficient results in a power spectrum in the frequency domain, the former can be reduced to the latter, and mathematically both have equivalent performance as methods based on the Fourier transform. It can be seen as having. In these methods, the modulation band of the transmission signal must be wide to achieve high resolution. For example, when trying to realize a resolution of 1 m (delay time difference: 3 nsec), the observation bandwidth W of the frequency response of the communication channel becomes W = (1/2) × (1 / (3 nsec)) = 167 MHz, which is widely used indoors. A bandwidth that is an order of magnitude larger than the 16 MHz bandwidth of IEEE 802.11g, which is a standard for wireless LANs, is required.

The above limit is known as the uncertainty principle of time / frequency resolution in Fourier transform, and corresponds to the fact that it is necessary to lower the frequency resolution (take a wider observation frequency band) in order to increase the time resolution. When separating a direct wave from a multipath wave with high temporal resolution, the correlation method or inverse Fourier transform method has mathematically equivalent characteristics, so it cannot be removed from its uncertainty principle. It can be said that the technique is insufficient to achieve the accuracy required for measurement in a signal format such as IEEE802.11g.
Japanese Patent Laid-Open No. 2002-14152 (paragraphs 0008 to 0015, FIG. 2) JP2007-300284A (paragraph 0006)

  An object of the present invention is to realize a delay time measuring method, an apparatus, and a system capable of obtaining high measurement accuracy even in an indoor multipath environment by a completely new algorithm in order to overcome the above-described conventional limitations. To do.

  In order to achieve the above object, the present invention uses a multicarrier signal in which a plurality of carriers are synchronized as a transmission signal, and the constellation data of each subcarrier (the modulation signal point expressed on the IQ plane) is a propagation path delay. The relative distance is measured while removing the multipath interference based on the fact that it receives the phase rotation corresponding to the time difference and receives characteristic deformation when there is multipath interference.

That is, the distance measuring method of the present invention is a multicarrier in which the phase relationship of a plurality of subcarriers including time reference information S (see FIG. 2) transmitted from the transmitting station T is synchronized as shown in FIG. A signal is received by the receiving station R, each subcarrier signal constellation data 18 of the multicarrier signal at the time of reception of the time reference information S is obtained, and the phase and amplitude of the constellation data 18 are set for each subcarrier. Normalizing with the identification signal 19 after demodulating the carrier to obtain the frequency domain constellation data 26 from which the modulation component has been removed, setting the receiving station R to the position to be measured, and receiving the multicarrier signal from the transmitting station T Obtaining the frequency domain constellation data 26 at the time of reception of the time reference information S, and the transmitting station T The received signal et, depending on the propagation path, each amplitude A _i, the delay time T _{i (uppercase} Greek letter "tau", hereinafter the same), _(distinguish each propagation path i) phase theta _i is different When the delay signal is expressed as a combined delay signal, the remaining frequency domain constellation data 26 of the received signal and the delay signal calculated from the amplitude A _i , delay time Τ _i , and phase Θ _{i of the} delay signal A fitting step of evaluating the difference with a predetermined likelihood function and determining each of the amplitude A _i , the delay Τ _i , and the phase Θ _{i in} which the evaluated residual is equal to or less than a predetermined value (FIG. 9). S1 to S11) and a value of the smallest delay among the determined plurality of delays _i , and the relative distance between the transmitting station and the receiving station from the value of the smallest delay _i Distance determination process ( Equipped with a 9).

  Here, the time reference information S can be applied at the moment when the phase of each subcarrier is aligned with a predetermined value, and a specific signal may be inserted into the transmission signal for that purpose, or multicarrier communication These preamble signals usually contain the above-mentioned equivalents for determining the demodulation synchronization point, so that signal may be used. "Subcarrier phase relationship is synchronized" means that one or more time points where the phase of each subcarrier takes a predetermined value relationship during the signal duration used for measurement are set at a specific location. It means that. “Normalization with demodulated identification signal” means that a demodulated signal including amplitude error and phase error is identified and reproduced by a predetermined threshold, and for amplitude, the demodulated signal is divided by the amplitude of the identified and reproduced signal, and the phase value Is to remove the modulation component by subtracting the phase of the identified and reproduced signal from the demodulated phase. “Frequency domain constellation” refers to a plot of the amplitude and phase of each subcarrier on the IQ plane, details of which will be described later. The “propagation path” refers to a path path through which a direct wave from the transmitting station T reaches the receiving station R, and includes a direct wave path and a multipath wave path. The “fitting process” is to set parameters of one curve so that one curve matches the other one curve within an allowable range. The “smallest delay value” refers to the delay time of the delay signal having the shortest delay time in the delay spectrum obtained by performing the above-mentioned fitting considering a plurality of delay signals. Usually, this delay time corresponds to the delay time of the direct wave from the transmitting station T.

  As described above, this measurement method does not detect a time difference by Fourier transform, but directly obtains a wave from a feature amount of a figure drawn on the IQ plane of the subcarrier, and is bound by the uncertainty principle associated with the Fourier transform processing. High accuracy measurement is possible.

If the multicarrier signal is an OFDM modulated signal as shown in FIG.
When an OFDM signal is used, the algorithm of this measurement method can be executed only by information obtained by the standard demodulation process of the OFDM signal. Further, there is an advantage that special hardware such as a high-speed A / D converter is unnecessary.

  In the fitting step of the present invention, for example, as shown in FIGS. 7 and 9, each delay signal when the combined delay signal is a combination of a plurality of delay signals propagating through the propagation paths is converted into an IQ plane. A direct wave fitting step (S1 to S4) that evaluates a residual obtained by subtracting a delay signal having the smallest delay time from the delay signal from the measured frequency domain constellation expressed by the upper arc using a predetermined likelihood function. ) (FIG. 9), and if it is determined that the residual value is not sufficient, a multipath wave fitting step is performed again including a delay signal corresponding to a large propagation delay time. (S6 to S8), and the delay time difference selection step repeats the direct wave fitting step (S1 to S4) and the multipath wave fitting step (S6 to S8). Of the delay Τ which is more determined, comprising the step of selecting the smallest Τ as the reception delay time difference.

  With this configuration, the direct wave can be separated from the multipath wave, and further, the direct wave can be separated first by fitting sequentially from the one with the smallest delay time で. Can be omitted, and high-speed processing is possible.

  In the fitting step, as the likelihood function, (1) a method of evaluating by the length of the residual arc, and (2) evaluation by a standard deviation of the distribution of the residual point sequence corresponding to each subcarrier. (3) A method of evaluating with a standard deviation of curvature of the residual point sequence corresponding to each subcarrier can be used.

  When such a likelihood function is used, the calculation can be performed only with the simple geometric property of the curve on the IQ plane, so that a calculation with good line-of-sight and high-speed processing can be performed. In addition, the likelihood function is an example in (1) to (3) above, and any other function may be used as long as it is a function that can evaluate the match between the other two curves.

The distance measurement receiving station according to the present invention, for example, as shown in FIG. 3, transmits a multicarrier signal transmitted from the transmitting station T and including time reference information (see FIG. 2) and synchronized in subcarrier phase relationship. Means for obtaining constellation data 18 (and FIG. 4 (b)) of each subcarrier signal of the multicarrier signal received at the receiving station and at the reception timing of the time reference information S; Means for obtaining frequency domain constellation data 26 from which modulation components have been removed by normalizing the phase and amplitude of the modulation data 18 with the identification signal 19 after demodulation of each subcarrier, and receiving the signal at a reference distance from the transmitting station T. A channel equalization circuit 34 for presetting the frequency constellation data to be gathered at one point when the station is placed; The received signal, in response to the propagation path, each amplitude A _i, the delay time T _i, when the phase theta _i is expressed as combined delay signals different delay signals, the received signal frequency domain constellation data 26 And the delay signal calculated from the amplitude A _i , the delay time Τ _i , and the phase Θ _{i of the} delayed signal are evaluated by a predetermined likelihood function, and the evaluated residual is determined in advance. Fitting means 27 for determining the amplitude A, the delay Τ _i , and the phase Θ _i that are less than or equal to a value, and the smallest delay value among the determined plurality of delay Τ _i is selected, Distance determining means (FIG. 9) for obtaining the relative distance between the transmitting station T and the receiving station R from the value of the small delay value _i .

  Here, “time reference information S”, “subcarrier phase relationship is synchronized”, “normalized by demodulated identification signal”, “propagation path”, “fitting”, and “smallest delay value” Is synonymous with the term described in the description of the distance measuring method.

  Furthermore, the position measuring system according to the present invention is, for example, as shown in FIG. 14 (a), using signals from four or more transmitting stations (T1 to T2) whose installation positions are known for distance measurement according to the present invention. It is received by the receiving station apparatus R, and a position measuring system that determines the position of the distance measuring receiving station R from the relative distance measuring data of each transmission signal can be realized.

  In addition, as shown in FIG. 14B, for example, the position measurement system according to the present invention is configured to receive at least four distance measurement reception stations (in the present invention) whose signals are transmitted from a single transmission station T. R1 to R4), and a position measurement system that determines the position for determining the position of the distance measurement transmission station T from the difference in reception delay time of each received signal can be realized.

  With this configuration, a position measurement system capable of high-precision and high-speed processing can be realized with a simple configuration.

Embodiments of the present invention will be described below with reference to the drawings. In the drawings, the same or corresponding parts are denoted by the same reference numerals, and redundant description is omitted. Moreover, what added an alphabetic character after a number like "31a" and "31b" is meant to include what is partly different from the contents of 31, but has the same configuration. When there is no need to distinguish between 31a and 31b, they are collectively referred to as “31”. Those with “dash” such as “T ′” mean that the structure of T is mostly the same, but includes a part that is different. Also, subscript numbers such as “4 ₁ ” and “4 ₂ ” are used to distinguish and indicate one when there are a plurality of identical systems, such as parallel signals. When the whole is collectively referred to, “4” is displayed.

[principle]
Hereinafter, the principle of delay time measurement according to the present invention will be described before the description of the embodiments of the present application. The present invention measures the delay time by utilizing the fact that the phase transition of each subcarrier of the multicarrier signal changes depending on the propagation delay time, and can be widely used in a multicarrier communication system capable of demodulating the amplitude and phase characteristics. Is. Since the knowledge of the multicarrier signal is necessary for understanding the invention of the present application, the case where it is applied to an OFDM signal will be described as an example here as a typical one.

  In OFDM, transmission data is divided into a plurality of subchannels (carriers) and transmitted in parallel. The subcarriers are synchronized in phase (subcarriers are generated from a single reference frequency generator, and the same symbol is modulated). In this case, the timing of one of the zero crossing points of these signals coincides with each other, and the respective frequency intervals (angular frequency: Δω) are set at equal intervals at the symbol rate interval from the orthogonal condition. In OFDM, it is obtained by performing inverse Fourier transform (generally using fast inverse Fourier transform from calculation efficiency) while generating a complex frequency group corresponding to the subcarrier frequency, and converting it to a signal in the time domain. On the receiving side, the received OFDM signal is Fourier transformed to obtain a frequency domain signal. At this time, the complex frequency generator used in the Fourier transform on the reception side must be synchronized with the frequency generator on the transmission side and have the same phase at the corresponding transmission / reception timing point. The signal in each frequency domain obtained by demodulation is the same as the parallel signal transmitted on the subcarrier. When there is a propagation delay between the transmitter and the receiver, the complex sine wave between the transmitter and the receiver is synchronized by this delay time.

However, in an actual system, the synchronization does not need to be completely matched, and it is sufficient if the demodulation symbol points can be distinguished from adjacent ones because an identification / reproduction process is performed. In general, the allowable difference is deviated from the timing point estimated by the receiver. In such a case, the signal after Fourier transformation undergoes phase rotation corresponding to this shift, but the rotational phase amount θ is an amount proportional to the subcarrier frequency for each subcarrier. If the subcarriers are all modulated with the same symbol and are taken at equal frequency intervals as in OFDM, the amount of phase rotation will be equal at each subcarrier. That is, θ = nΔωτ ₀ (n: natural number) where τ _{0 is} the deviation of the timing point and nΔω is the frequency difference from the reference subcarrier. When this is plotted on the complex plane (hereinafter referred to as “IQ plane”) for each subcarrier, an arc having a constant opening degree around the origin O is formed. Such a subcarrier plotted on the IQ plane is hereinafter referred to as a “frequency domain constellation”. When each subcarrier is modulated with a different symbol, the amplitude phase of the virtually unmodulated subcarrier is constructed from the demodulated signal, that is, the frequency is normalized by using the demodulated identification signal An area constellation can be obtained.

FIG. 13 is a diagram for explaining the principle and procedure of the measurement method of the present invention. As shown in FIG. 13A, consider a case where the multicarrier signal of the transmitting station T (position: P0) is received at two points (P1, P2) having different distances. In the drawing, (b) to (f) show the above-described frequency domain constellation. The transmission frequency domain constellation (b) of the transmitting station T is gathered at one point. If the amplitude and phase disturbance given by the transmission / reception high-frequency system or antenna can be ignored, the phase difference between the subcarriers will be obtained if the signal is received at a point where the delay time is τ ₀ (P1 point), τ ₀ + τ ₁ (P2 point) θ is Δωτ _{0 at} position P1 and Δω (τ ₀ + τ ₁ ) at position P2, and the arc opening at position P2 is larger than the arc opening at position P2 by an amount corresponding to the delay time τ ₁ . The opening degree of the arc corresponds to the delay time τ ₀ + τ ₁ .

However, in reality, the transmission / reception high-frequency system and the antenna also have frequency characteristics of amplitude phase, and disturb the frequency domain constellation. Therefore, in an environment with few multipaths such as outdoors, the position P1 is separated from the transmitting station T and the receiving station R by a reference value (for example, 1 m), and the frequency domain constellation is measured. This is generally a figure scattered on the IQ plane, but it is recorded, and in the following measurement, by compensating for the dispersion, the channel of the receiving station is gathered so that the frequency domain constellation in the reference state is gathered at one point. The equalization means can be adjusted. This situation is shown in (e). Since this is compensated for including the delay τ ₀ at the reference distance, when the receiving station is placed at the point (P2) to be measured, the result becomes (f), and the opening degree of the frequency domain constellation is the reference value. This corresponds to the difference in distance from the position (τ _{1 in} delay time). Thereby, the relative distance of the receiving station R from the transmitting station T can be measured. This compensation value needs to be obtained for all the transceivers used prior to measurement.

  In other words, “channel equalization” means that the amplitude and phase of each subcarrier whose amplitude and phase are known is observed in advance, and the amplitude and phase disturbance given by the transmitter high-frequency system, transmission path, and receiver high-frequency system appearing there. To record and to compensate for each measurement. This is an operation for calibration of the measurement system. If this operation is performed with the transmission / reception antennas separated by a specified distance, the transmission path parameter determined there cancels the transmission delay of this distance, so that the distance measurement will determine this channel equalization parameter. The antenna distance at that time is set to zero, and the relative distance from there is obtained.

  Further, consider the case where there are multipath waves. The reflected wave of the multipath wave has a longer delay time than the direct wave, and the reception level is generally lower than that of the direct wave. Therefore, the frequency domain constellation has a larger arc opening and a smaller radius than the direct wave. In a multipath environment, that is, when there are a large number of reflected waves, the frequency domain constellation is not a simple arc, but a figure that is a combination of a large number of arcs with different opening degrees and radii.

Next, the delay time of the target direct wave is determined from the obtained frequency domain constellation. As described above, the frequency domain constellation is a combination of a plurality of circular arcs having openings and radii corresponding to various multipath waves. In order to determine whether or not the delay signal Ae- ^{j (ωΘ + Θ)} is included in this, as an initial condition, an appropriate A, Τ, and Θ are set, and an angular frequency ω corresponding to the subcarrier as a delay signal is set. Calculate and perform frequency domain constellation and curve fitting to obtain A, Τ, and Θ with the highest likelihood. As a result, the obtained soot becomes the target delay time. Since it is considered that there are a plurality of delayed signals in a multipath environment, A, Τ, and Θ are further extracted by applying the above steps to the residue. The delay time の of each delay signal obtained in this way corresponds to the delay time of each path.

  When obtaining the constellation, if the transmission signal is modulated, the points on the IQ plane are a plurality of points corresponding to the modulation method signal points (for example, 4 for QPSK, 64QAM). 64). In order to set this as one specific signal point, phase-shift / level adjustment according to the demodulated signal is performed on the signal after Fourier transform to remove the modulation component. (Ie, normalization with demodulated identification signal)

[First Embodiment]
[Constitution]
FIG. 1A shows an example of the configuration of a transmission station T used as a transmission signal source in the position measurement method of the present invention. The configuration of the transmitting station itself is not an element constituting claims 1 to 7 of the present application. In Claims 8 and 9, the transmitting station is an element constituting the system.
The measuring method and measuring apparatus transmission signal 2 of the invention of the present application are converted into a plurality of parallel signals by the serial / parallel conversion circuit 3 and input to each modulation circuit 4. The subcarrier signal 11 to each modulation circuit 4 is generated in the subcarrier generation circuit 10 based on the reference frequency generation circuit. For example, if the frequency of the reference frequency generation circuit is f, fl _n = n × f is there. Therefore, signals having different frequencies f are input to the modulation circuit 4.
FIG. 1B illustrates the subcarrier signal 11 ₁ having the lowest frequency, the second lowest subcarrier signal 11 ₂ , and the third lowest subcarrier signal 11 ₃ among the signals of the subcarrier signal 11. The case where one of the zeros of the amplitude coincides with the same temporal position t ₀ is illustrated. Thus, when the subcarrier signal 11 is generated from one reference frequency generation circuit, there is a certain correlation between the subcarrier signals 11 (hereinafter, such a related state is “synchronized”. Called).

  In this way, a modulated signal 8 is obtained. The modulated signal 8 is converted to an analog signal by the D / A conversion circuit 5, further converted to a predetermined frequency by the transmission RF circuit 6, and then transmitted by the transmission antenna 7.

  FIG. 2 shows a transmission signal time domain waveform (FIG. 2A) and a constellation 12 (FIG. 2B) of each subcarrier signal of the modulated signal 8. FIG. As shown in the transmission signal time domain waveform of (a), time reference information S that can be identified on the receiving side is inserted in each modulated signal 8. The time reference information S is inserted at the same timing point for each subcarrier. Further, if the time reference information S is extracted independently from a fixed-length symbol sequence constituting a communication packet, it can be considered that this is periodically inserted at regular intervals. The time reference information S does not need to be inserted specially. For example, an existing preamble signal may be used as the time reference information S. If the modulation waveform of each multicarrier is known, signal-specific properties such as amplitude change at that timing may be used. The constellation 12 in (b) shows the case of QPSK, and ideally there is a signal point at each vertex of a square centered on the origin.

  FIG. 3 shows the configuration of the receiving station R used in the position measuring method according to the first embodiment of the present invention, including the transmitting station T. A signal transmitted from the transmission antenna 7 is received by the reception antenna 22, converted to a predetermined frequency by the reception RF circuit 13, and then converted to a digital signal by the A / D conversion circuit 14. Further, each subcarrier signal 24 is demodulated using the multiplication circuit 15. Demodulation is performed by multiplying the subcarrier signal 24 by the reception local oscillation wave 31 generated by the subcarrier generation circuit 21 with reference to the reference frequency generation circuit 20 and performing frequency filtering. Thus, the necessary amplitude and phase are adjusted and output as subcarrier constellation data 18. The reference frequency generation circuit 20 is controlled to synchronize with the reference frequency generation circuit 9 on the transmission side. Such a synchronization circuit is well known and will not be described in detail here.

In this way, constellation data in each subcarrier transmitted from the transmission side is obtained as subcarrier constellation data 18. The time reference information extraction circuit 23 extracts time reference information S inserted by the transmitting station T from the signal of the subcarrier constellation data 18. The signal of the subcarrier constellation data 18 at the timing when this signal is obtained is input to one input terminal of the amplitude / phase compensation circuit 17. The identification circuit 16 identifies the subcarrier constellation data 18 and outputs the normal demodulated amplitude a ₀ and phase. For example, in the case of QPSK, the identification circuit 16 obtains a ₀ as a predetermined amplitude value from the subcarrier constellation data 18. As the phase value, any one of 45 °, 135 °, −45 °, and −135 ° is output as the identification circuit output 19.

In the amplitude / phase compensation circuit 17, the amplitude value of the subcarrier constellation data 18 is shifted to 1 / a ₀ and the phase value is shifted by the output phase value of the discrimination circuit output 19. In this way, as the output of the amplitude / phase compensation circuit 17, only a point corresponding to a specific signal point among the signal points of the constellation is output, and this corresponds to the case where an unmodulated signal is transmitted from the transmitting station T. Output is obtained.

  The signals of the plurality of amplitude / phase compensation circuit outputs 26 of each subcarrier are collected in the frequency domain constellation generation circuit 25 to generate frequency domain constellation data in which signals for each subcarrier are displayed on the IQ plane. This signal data is further processed by the delay estimation circuit 27 to estimate the target delay time. The operations of the frequency domain constellation generation circuit 25 and the delay estimation circuit 27 will be described later.

[Operation]
Based on the configuration described above, the operation of the invention according to the present embodiment will be described with reference to FIGS.
4A shows a time domain waveform of each subcarrier at the output of the multiplication circuit 15 of FIG. 3, and FIG. 4B shows subcarrier constellation data 18 corresponding to FIG. 4A. . As shown in (a), it is assumed that the time reference information S on the transmission side is detected at time U1. The waveforms of the reference frequency generation circuit 20 are shown in (i) and (ii) shown below (a). If the timing of the zero point of the waveform of the transmission side reference frequency generator 9 coincides with the timing of the time reference information S on the transmission side, the reception side reference frequency at the time reference point U1 also on the reception side as shown in (i). When the output of the generation circuit 20 is at the zero point, the transmission-side reference frequency generation circuit 9 and the reception-side reference frequency generation circuit 20 are synchronized, and the transmission-side constellation 12 and the reception-side subcarrier constellation data 18 are Match. The subcarrier constellation data 18 at that time is indicated by a black circle in FIG.

By the way, in an actual system, the synchronization does not need to be completely coincident, and it is sufficient if the symbol points for demodulation can be distinguished from adjacent ones. In general, the timing point U1 estimated by the receiving station is within a tolerance. This situation is shown in (ii) of FIG. In the figure, it is shifted by time τ _k . In such a case, each subcarrier signal undergoes phase rotation corresponding to this deviation. This is indicated by white circles in FIG. As shown in the figure, the subcarrier constellation data 18 is rotated by α _i (i is the number of the subcarrier) corresponding to the size of the time shift τ _k .

On the other hand, try changing your view. When the timing of the reference frequency generating circuit 20 is held at a constant value, a propagation delay time between the transmitting station T and the receiving station R is added, and even when the time relationship with the reference frequency generating circuit 20 is different, the delay time The subcarrier constellation data rotates corresponding to the size. Since the rotation phase amount α _i is proportional to the carrier frequency, if the frequency interval between the subcarriers is constant, the rotation phase amount is equal. When these are plotted on the IQ plane for each subcarrier, the reception level of each subcarrier is substantially the same, so a part of a circular arc having a constant opening around the origin O is drawn. This is called “frequency domain constellation”.

FIG. 5 schematically shows such a situation. A point of a specific signal of the constellation due to a difference in distance (delay time τ1, τ2) from the transmission station T (for example, transmission of subcarriers by QPSK). Only the data corresponding to [1, 1] etc.) is plotted against each subcarrier. As shown in the figure, at the transmitting station T point, τ = 0, and each point of the frequency domain constellation is overlapped. As the receiving station R moves away from the transmitting station T, the plot points are separated, and the angle between the plots relative to the origin increases. This angle corresponds to the delay times τ ₁ and τ ₂ .

FIG. 6 shows a configuration of an amplitude / phase compensation circuit 17 that converts constellation data for the obtained modulated signal into constellation data corresponding to no modulation. FIG. 6A shows the processing portions (23, 16, 17) of FIG. 3 as they are. FIG. 6B exemplifies the subcarrier constellation data 18 for the subcarrier number i and the constellation data for the identification circuit output 19. The black circle is a signal point in the subcarrier constellation data 18, and the double white circle is It is a result after performing the identification process of the data of the subcarrier constellation data 18. As shown in the figure, the black circle receives the phase rotation α _i caused by the delay time difference. The white circle is at the apex of a square centered on the origin O, and the distance from the origin O is _a0i . Based on the signal from the time reference information extraction circuit 23, the amplitude / phase compensation circuit 17 performs the normalized amplitude a _ni and the normalized phase on the signal of the subcarrier constellation data 18 at the predetermined timing point U1 by the following calculation. α _ni is obtained. That is, the amplitude of the subcarrier constellation data 18 is a _i , the phase angle is β _i , the amplitude of the discrimination circuit output 19 is a _oi , and the phase angle is β _oi ,

a _ni = a _i / a _oi (1)
α _ni = β _i -β _oi (2)

Is calculated. The result of the calculation is shown in FIG. As shown in the figure, the modulation component is removed, and the phase angle α _ni corresponding to the delay time is obtained. The above calculation is performed for each subcarrier.

  When the measurement signal constitutes a communication packet composed of a sequence of a plurality of symbols, the frequency domain constellation can be obtained independently by the number obtained by dividing the packet length by the symbol length. If the frequency domain constellation is affected by random communication noise, improvement in measurement accuracy can be expected by reducing the influence by averaging over the entire packet length. That is, in FIG. 4, not only U1, but U2,. . . . The frequency domain constellation is also obtained for Un, and instead of simply taking the average for each symbol, by taking the weighted average with the modulation power for each symbol at Uk (that is, by the square of the amplitude assigned to the symbol) The most effective averaging process can be performed (by averaging with weighting).

The frequency domain constellation generation circuit 25 temporarily stores output data of the amplitude / phase compensation circuit 17 for each subcarrier in a memory or the like.
FIG. 7A displays these data on the IQ plane, which is a curve. When the actual measured values are plotted as shown in the figure, it becomes a distorted arc, and the distortion corresponds to the multipath component. Individual direct waves and multipath waves are all arcs having different radii (corresponding to intensity), opening (corresponding to delay time) and starting positions (points indicated by double circles in FIG. 7), for example, If the received signal consists of a direct wave and a finite number of multipath waves, this distorted arc must be decomposed into a sum of a finite number of arcs. FIG. 7 (b) is an example in which the curve of FIG. 7 (a) is decomposed, and an arc (A) having a large radius and a small opening, and an arc (B) having a small radius and a large opening. Can be disassembled. The number i in the figure corresponds to the subcarrier number.

When the IQ plane is regarded as a complex plane, the arcs (b) and (b) are expressed in terms of two sets of parameters A, Τ, and Θ ((b)) by the polar coordinate representation of the signal Z = Ae− ^{j (ωΤ + Θ).} Can be represented by A ₂ , Τ ₂ , Θ ₂ ) for A ₁ , Τ ₁ , Θ ₁ , (b). Here, A is the intensity of the received wave, ω is the subcarrier angular frequency, Τ is the delay time, and Θ is the phase of the subcarrier that is the starting point. Therefore, if a composite wave that matches the actually measured frequency domain constellation shown in FIG. 8A is obtained, the received signal given by each arc becomes a delay signal corresponding to each delay time. In FIG. 7B, (A) is a direct wave, and (B) is a multipath wave. In addition, if (b) and (b) are signals from different transmitting stations T, the difference in distance between the transmitting station and the receiving station can be found from the difference in the delay time 各 々. In this specification, obtaining a synthesized wave that matches the actually measured frequency domain constellation is referred to as “fitting”.

  Next, a fitting method (a method for obtaining the parameters A, Τ, and Θ) will be described with reference to FIGS. 8 and 9. For ω, the center subcarrier frequency ω = 0, and the subcarrier frequencies are −nΔω, (−n + 1) Δω,. . -Δω, 0, Δω,. . . , (N−1) Δω, nΔω (where Δω is a frequency between subcarriers). This is convenient for the initial setting of Θ described below.

FIG. 9 is a flowchart showing a fitting procedure. This will be described with reference to FIG.
(1) First, based on the delay time い to be verified, an arc having a certain opening is created, and its amplitude A and phase Θ are set (S1). Here, Τ is an expected delay time difference, and an expected value is selected for convenience from the approximate distance from the transmitting station T, the size of the room, and the like. If the amplitude A is about 0.8 to 1 times the amplitude value of the measured subcarrier signal, and the phase Θ is set to a value that matches the position of the subcarrier at the center, the convergence is expediently from experience.
(2) Next, a value of Ae− ^{j (ωΤ + Θ)} is calculated from the set parameters (FIG. 8B), and subtracted from the measured frequency domain constellation data (FIG. 8A) (S2 ). If the parameters A, Τ, and Θ are optimum values, the residual is collected at the origin O, but usually remains as a curve (residual curve) on the IQ plane (FIG. 8C).
(3) The residual is evaluated with a predetermined likelihood function, and it is evaluated whether the residual is sufficiently small (S3). For example, it is preferable to use a likelihood function for evaluating the length of the arc, the standard deviation of the distribution density of the point sequence, and the standard deviation of the radius of curvature using the geometric characteristics of the residual curve. Specifically, it is shown in FIG.
If Z _k is a complex value at each point in the frequency domain constellation, the length of the arc is Σ | Z _k + 1−Z _k |. If what is observed is a simple arc, then drawing a correctly estimated arc will shrink it to the origin, ie zero. The standard deviation of the distribution density of the point sequence is (1 / N) Σ | Z _k + 1−Z _k | ² − ((1 / N) Σ | Z _k + 1−Z _k |) ² where the number of subcarriers is N. Yes, if it is a simple arc, it will be zero. The standard deviation of the radius of curvature is a change in the radius of curvature of a curve _obtained by interpolating the _Zk series. If the arc is a simple arc, the radius of curvature is constant and the change is zero. If they do not go to zero, it indicates the presence of different signals due to multipath. If the estimated arc used for subtraction is different from the correct signal element, it means that another arc has been added to the frequency domain constellation by the error, and the above index is considered to deteriorate. Conversely, if the signal estimation is correct, the decrease in these indices is expected to decrease, and therefore these are employed as the likelihood function.
(4) A convergence criterion is set in advance, and it is determined whether the convergence has been performed according to the criterion (S4). It is determined whether the repetitive processing for convergence is within a predetermined number of times. If it is determined that the convergence has not occurred within the predetermined number of times, A, Θ, and Τ are changed (S6), and the process returns to S1 and S2 to S4 are repeated. . If it is determined that it has converged, the value of _ｉi is output (S7), and the fitting process is terminated.

(5) If it is determined that the convergence is not sufficient even if the steps S2 to S6 are repeated a predetermined number of times or more, a delayed wave having a separate parameter is set (S8).
(6) The fitting process similar to S1-S6 is implemented with respect to the residual curve calculated | required by S2 (S9).
(7) It is determined whether the result of S9 has converged (S10). The determination content is the same as in S4. If it is determined that the value has sufficiently converged, the obtained value of _i is output (S11), and the fitting process is finished. The same process is performed including waves (S12).

  In this procedure, it is preferable to set the smallest value in the conceivable range as the set value of the delay time 、 and sequentially set the larger value. In this way, the direct wave having the smallest delay time Τ can be found at an early stage, and the subsequent processing can be stopped.

[Second Embodiment]
FIG. 11 shows a second embodiment of a transmitting station T ′ and a receiving station R ′ applied to the present invention. In this embodiment, an OFDM modulation scheme is used as a multicarrier scheme. As is well known, in the OFDM modulation system, the carrier phases of the subcarriers are synchronized, and the frequency intervals (Δω in angular frequency) are set at equal intervals at the symbol rate.

  FIG. 11A is a configuration of a general OFDM modulation scheme transmitting station T, and is different in that the modulation circuit 4 of FIG. 1 is replaced with a signal point arrangement circuit 28 and an inverse FFT circuit 29. Others are basically the same. The signal point arrangement circuit 28 is a circuit that converts the signal converted into the parallel signal string by the serial / parallel conversion circuit 3 into a complex signal (constellation) corresponding to the signal arrangement of each subcarrier. In addition to the OFDM modulation method, it is common to realize corresponding modulation by signal point arrangement, and the signal point arrangement circuit is not unique to OFDM. The inverse FFT circuit 29 converts a plurality of constellation signals into a time domain signal, whereby an OFDM modulated wave is obtained collectively.

FIG. 11 (b) shows the configuration of the receiving station R ′ of the present embodiment. The multiplication circuit 15 in FIG. 3 is replaced with an FFT circuit 30. Further, the frequency domain constellation generation circuit 33 of this figure has the function of the amplitude / phase compensation circuit 17 of FIG. 3 in addition to the function of the frequency domain constellation generation circuit 25 of FIG. The received OFDM signal is converted into a frequency domain signal by the FFT circuit 30 to obtain constellation data 32 of each subcarrier. The constellation data 32 and the output signal of the identification circuit 16 are input to the frequency domain constellation generation circuit 33 to obtain frequency domain constellation data. In this configuration, the constellation data is obtained in a batch by the processing in the FFT circuit 30 and the constellation data 32 of each subcarrier is obtained with a simple configuration as compared with the configuration of FIG.
Since the operation of the present embodiment is basically the same as that described in the first embodiment, the description thereof is omitted.

[Experimental result]
Next, experimental results according to the present invention will be described.
The experiment was (1) outdoor (ground where there is no reflecting object such as buildings within 100), (2) 15 meters square, indoors with relatively little unevenness, (3) 25 meters in size On all sides, we went inside the small auditorium where many desks and chairs were arranged in steps. As the transmitting station T, a wireless LAN station for IEEE 802.11g (central frequency 2.422 GHz, number of subcarriers 53) was used as it was.

  FIG. 12 shows the measurement error results in the environmental conditions (1) to (3). Here, the range of the distance to be measured is 1 to 9 m, and the number of measurements at each distance is 10. The estimation of the delay time was performed in the case of evaluating by the arc length and the case of evaluating by the standard deviation of the curvature distribution. As shown in the figure, a measurement error within 1 m was obtained with both methods. FIG. 12 shows, for comparison, the results measured by the above-described conventional correlation method under the same environment. As shown in the figure, according to the present invention, the measurement error can be significantly reduced compared to the conventional method even in an environment with many multipaths such as indoors.

It is an example of a structure of the transmitting station used for this measurement. It is the figure which showed the transmission waveform and constellation for every subcarrier. It is the figure which showed the structure of the receiving station which concerns on the 1st Embodiment of this invention. It is the figure which showed the time waveform and constellation data of each received subcarrier. It is the figure which showed the constellation when the position of a receiving station changed. It is the figure which showed the structure and operation | movement of an amplitude / phase compensation circuit. It is a figure which shows the relationship between the example of the measured value of a frequency domain constellation, and a delay signal. It is the figure which showed the mode of the fitting process. It is the flowchart which showed the fitting process. It is a figure explaining a likelihood function. It is the figure which showed the structure of the receiving station and transmitting station which concern on the 1st Embodiment of this invention to which OFDM is applied. It is a figure which shows the experimental result by 2nd Embodiment. It is a figure explaining the principle and procedure of the measuring method of this invention. It is a figure which shows the structure of the position measuring system which concerns on this invention.

Explanation of symbols

2 Transmission Signal 3 Serial / Parallel Conversion Circuit 4 Modulation Circuit 5 D / A Conversion Circuit 6 Transmission RF Circuit 7 Transmitting Antenna 8 Modulation Signal 9 Reference Frequency Generation Circuit 10 Subcarrier Generation Circuit 11 Subcarrier Signal 12 Constellation 13 Reception RF Circuit 14 A / D conversion circuit 15 Multiplication circuit 16 Identification circuit 17 Amplitude / phase compensation circuit 18 Subcarrier constellation information 19 Identification circuit output 20 Reference frequency generation circuit 21 Subcarrier generation circuit 22 Receiving antenna 23 Time reference information extraction circuit 24 Subcarrier signal 25 Frequency Domain Constellation Generation Circuit 26 Amplitude / Phase Compensation Circuit Output 27 Delay Estimation Circuit 28 Signal Point Placement Circuit 29 Inverse FFT Circuit 30 FFT Circuit 31 Receiving Local Oscillation Wave 32 Frequency Domain Signal 33 Frequency Domain Constellation Generation Time 34 channel equalization circuit R, R 'receiving station S time reference information T, T' transmission station

Claims (9)

Each subcarrier signal of the multicarrier signal at the reception timing of the time reference information is received by the receiving station, which receives the multicarrier signal transmitted from the transmission station and including the phase reference information and in which the phase relationships of a plurality of subcarriers are synchronized. And obtaining the frequency domain constellation data from which the modulation component is removed by normalizing the phase and amplitude of the constellation data with the identification signal after demodulation of each subcarrier for each subcarrier. ;
Setting the receiving station to a position to be measured, receiving the multi-carrier signal from the transmitting station, and obtaining the frequency domain constellation data at the reception timing of the time reference information;
When the received signal from the transmitting station is represented as a combined delayed signal of a plurality of delayed signals having different amplitudes A _i , delay times Τ _i , and phases Θ _i according to their propagation paths, the frequency domain of the received signal The residual between the constellation data and the delayed signal calculated from the amplitude A _i , the delay time Τ _i , and the phase Θ _{i of the} delayed signal is evaluated using a predetermined likelihood function, and the evaluated residual is A fitting step of determining each of the amplitude A _i , the delay Τ _i , and the phase Θ _i that are equal to or less than a predetermined value;
A distance determining step of selecting a smallest delay value from the plurality of determined delay values _i and obtaining a relative distance between the transmitting station and the receiving station from the smallest delay value;
Distance measurement method. The multicarrier signal is an OFDM modulated signal;
The distance measuring method according to claim 1. In the fitting step, each delay signal when the combined delay signal is a combination of a plurality of delay signals propagating through each propagation path is represented by an arc on an IQ plane, and from the frequency domain constellation, the delay A direct wave fitting step of evaluating a residual obtained by subtracting a delayed signal having the smallest delay time among signals using a predetermined likelihood function, and if it is determined that the value of the residual is not sufficient, Including a multi-path wave fitting step for performing the step again, including a delay signal corresponding to a large propagation delay time,
The delay time difference selection step includes a step of selecting the smallest wrinkle among the delay wrinkles obtained by repeating the fitting step as the reception delay time difference.
The distance measuring method according to claim 1 or 2. The likelihood function is evaluated by a length of the arc of the residual,
The distance measuring method according to claim 3. The likelihood function is evaluated by a variance value of the residual point sequence corresponding to each subcarrier.
The distance measuring method according to claim 3. The likelihood function is evaluated by a curvature of the residual point sequence corresponding to each subcarrier.
The distance measuring method according to claim 3. A multi-carrier signal including time reference information and having a synchronized sub-carrier phase relationship transmitted from the transmitting station is received by the receiving station, and the constellation of each sub-carrier signal of the multi-carrier signal at the time of reception of the time reference information is received. Means for obtaining data on
Means for normalizing the phase and amplitude of the constellation data for each subcarrier with an identification signal after demodulation of each subcarrier to obtain frequency domain constellation data from which modulation components have been removed;
Channel equalization means for presetting the frequency constellation data to be collected at one point when the receiving station is placed at a reference distance from the transmitting station;
When the received signal from the transmitting station is represented as a combined delayed signal of a plurality of delayed signals having different amplitudes A _i , delay times Τ _i , and phases Θ _i according to their propagation paths, the frequency domain of the received signal The residual between the constellation data and the delayed signal calculated from the amplitude A _i , the delay time Τ _i , and the phase Θ _{i of the} delayed signal is evaluated using a predetermined likelihood function, and the evaluated residual is Fitting means for determining the amplitude A _i , the delay Τ _i , and the phase Θ _i that are equal to or less than a predetermined value;
Distance determining means for selecting the smallest delay value from the plurality of determined delay values _i and determining the relative distance between the transmitting station and the receiving station from the smallest delay value;
Receiving station device for distance measurement. A position measurement system that receives signals from four or more transmission stations whose installation positions are known at one distance measurement reception station, and determines the position of the distance measurement reception station from the reception delay time difference of each transmission signal. There,
The distance measurement receiving station is the distance measurement receiving station device according to claim 7,
Position measurement system. The position measurement system receives a signal of one transmission station at four or more distance measurement reception stations whose installation positions are known, and determines the position of the distance measurement transmission station from a difference in reception delay time of each reception signal. And
The distance measurement receiving station is the distance measurement receiving station device according to claim 7,
Position measurement system.

Images