JP5521179B2 - Position estimation device for electronic key system - Google Patents

Description

  The present invention relates to a position estimation device for an electronic key system that performs various controls by estimating the position of an electronic key.

  In recent years, many vehicles are equipped with an electronic key system that uses an electronic key that can transmit an ID code unique to the vehicle by wireless communication as a key of the vehicle. In this electronic key system, when the in-vehicle device mounted on the vehicle receives the ID code transmitted from the electronic key, the in-vehicle device collates the ID code of the electronic key with the ID code registered in advance in itself. If these ID codes match and ID verification is established, control of the in-vehicle device, for example, locking / unlocking the door lock of the vehicle and starting of the engine are permitted or executed (see, for example, Patent Document 1).

  This electronic key system does not require any button operation on the electronic key when transmitting the ID code from the electronic key, that is, automatically transmits the ID code from the electronic key. Specifically, in an electronic key system, a request signal requesting the electronic key to transmit an ID code is intermittently transmitted from the in-vehicle device, and when the electronic key receives the request signal, the electronic key is in response to the request. An ID code signal including a code is transmitted to the vehicle, and ID verification is performed using the ID code of the ID code signal received by the in-vehicle device.

  Further, the electronic key system includes an engine start system that can start and stop the engine only by operating an engine switch provided in the vehicle. This engine start system performs ID verification with an electronic key that has entered the vehicle, and when the ID switch is established, if the engine switch is operated with the brake pedal depressed, Starts. For this reason, if the driver carries the electronic key in a pocket of clothes or puts it in a bag or the like, the engine is started.

JP 2002-29385 A

Kenichi Koga, 7 others, “Basic study on application of ESPRIT method to near-field source estimation”, IEICE Technical Report 2009, Vol. 109, No. 218, p. 91-96 M.M. Feder and E.C. M. Feder and E. Weinstein, “Parameter estimation of superimposed signals using the EM algorithm” (USA), IEEE paper (IEEE Trans. Acoust ), Speech and Signal Proc, 1988, Vol. 36, No. 4, p. 477-489 Takahiro Hayashi and two others, “A study on DOA estimation using EM, SAGE algorithm”, IEICE Technical Report, 2003, Vol. 103, No. 22, p. 57-62 P. J. et al. Chen and J.H. F. Baume (PJChung and JFBohme), “DOA estimation using fast EM and SAGE algorithms” (USA), Signal Processing, November 2002, 82nd Volume, p. 1753-1762

  By the way, in this type of electronic key system, permission to lock and unlock the door lock is conditional on ID verification outside the vehicle, and engine start permission is conditional on ID verification inside the vehicle. Therefore, the vehicle has an in-vehicle transmitter capable of transmitting a radio signal to a predetermined communication area (external communication area) around the vehicle, and an in-vehicle transmission capable of transmitting a radio signal to a predetermined communication area (in-vehicle communication area) in the vehicle interior. Depending on whether the driver is outside or inside the vehicle, either one of these transmitters is selectively operated and a quest signal is transmitted. Then, the vehicle determines whether the electronic key exists outside the vehicle or inside the vehicle based on whether the ID code signal is returned from the electronic key in response to the request signal transmitted to which communication area, and the determination Perform actions based on results.

  In light of the fact that the electronic key system adopts such a communication type, in this type of electronic key system, the output of the transmission radio wave (radio signal) transmitted from the external transmitter and the in-vehicle transmitter is between the external communication area and the in-vehicle It must be adjusted so that it does not overlap with the communication area. Specifically, the boundary of the in-vehicle communication area composed of the area of the predetermined signal strength transmitted from the in-vehicle transmitter and the boundary of the in-vehicle communication area composed of the area of the predetermined signal intensity transmitted from the in-vehicle transmitter almost coincide. It must be. However, since the communication area setting requires delicate setting work, the in-vehicle communication area may overlap the outside-vehicle communication area in some cases. For this reason, it is a very difficult task to combine the in-vehicle communication area and the out-of-vehicle communication area.

  Therefore, it is conceivable to employ a position estimation technique using an adaptive antenna that mounts an adaptive antenna on a vehicle and accurately estimates the position of the electronic key. Various methods have been devised for the position estimation technique using this adaptive antenna. However, since the position estimation accuracy can be improved and the calculation processing can be relatively suppressed, for example, the initial position is acquired in advance and the position is based on this initial position. Therefore, it is conceivable to use a position estimation technique for estimating the position. As a position estimation method using such an initial position, a neighborhood ESPRIT method (see Non-Patent Document 1), an EM method (Expectation-Maximization) (see Non-Patent Documents 2 and 3), a SAGE method (Space-Alternating Generalized Expectation) maximization) (see Non-Patent Documents 3 and 4).

  In position estimation using these initial positions, an algorithm for estimating some initial position is separately prepared for obtaining the initial position, and the position is estimated using the initial estimated position obtained by using this as an initial position. However, if an algorithm for estimating the initial position is prepared separately, the arithmetic processing becomes complicated and a device with high processing capability is required. Further, if a simple initial estimation algorithm is used, the accuracy of the initial estimated position is low, the accuracy of the final estimated position is deteriorated, and the amount of calculation required for the position estimation process is increased. Therefore, there has been a demand for a position estimation device for an electronic key system that can reduce the calculation process of position estimation while facilitating setting of a communication area.

  The present invention has been made in view of such circumstances, and its purpose is to reduce the calculation load of position estimation using an array antenna and to simplify the communication area setting of a transmitter that allows an electronic key to take a response action. An object of the present invention is to provide an electronic key system position estimation device that can be realized.

Hereinafter, means for achieving the above-described object and its operation and effects will be described.
According to the first aspect of the present invention, an ID reply request signal is transmitted from the transmitter of the communication master, and the electronic key transmits an ID signal in response to the ID reply request signal. When receiving with an antenna, the electronic key system confirms whether or not ID collation is established by collating an ID code included in the ID signal while calculating a radio wave arrival direction of the ID signal by a position estimation algorithm using the array antenna. In the position estimation device, an initial stage required for the calculation of the position estimation algorithm is confirmed by confirming whether the electronic key has returned the ID signal in response to the ID reply request signal transmitted from which transmitter. An initial position calculating means for calculating a position; and the position estimation algorithm based on the initial position calculated by the initial position calculating means. That a key position estimating means using estimates the position of the electronic key is set to its gist.

  According to the configuration, by checking which transmitter the electronic key returns in response to the request signal transmitted from which transmitter, it is confirmed in which communication area the electronic key is located. From the confirmation result, an initial position necessary for calculation of the position estimation algorithm is estimated. Then, using the initial position determined from the estimation result, the position of the electronic key is estimated by the position estimation algorithm. Therefore, since the initial position of the position estimation algorithm is determined by looking at which transmitter the electronic key responds to and returns the ID signal, it is not necessary to calculate the initial position using a complicated calculation formula. Therefore, the calculation load for the initial position is reduced, and as a result, the calculation load for estimating the position of the electronic key can be reduced.

  The detailed position of the electronic key takes a determination format that is performed using a position estimation algorithm when an ID signal from the electronic key is received by the array antenna. For this reason, the ID reply request transmitted from the transmitter need only be a signal that allows the electronic key to take a response action when the electronic key approaches the communication master to some extent, so the communication area does not need to be set in detail, It suffices to take a rough area. For this reason, since it is not necessary to set the communication area of the transmitter strictly, it is possible to eliminate the area setting operation of the transmitter that requires a very large work load.

  According to a second aspect of the present invention, in the position estimation device of the electronic key system according to the first aspect, the initial position calculation means transmits the ID reply request signal from a plurality of transmitters in a time-sharing manner, and By checking which transmitter the ID signal received by the array antenna is a response to which transmitter, the electronic key is located in the communication area of the transmitter, and the communication area that has been determined The gist is to set one point as the initial position.

  According to this configuration, the ID reply request signal is transmitted in a time-division manner, in other words, every predetermined time only from one of the transmitters, and the reception timing of the ID signal in response to the time-divided ID reply request signal From which transmitter area is specified, any one point in this communication area is set as the initial position. Therefore, it is possible to determine the current position of the electronic key, that is, the initial position, by simple control of transmitting signals in order from a plurality of transmitters.

  According to a third aspect of the present invention, in the position estimation device for an electronic key system according to the first or second aspect, each communication area of the transmitter is overlapped with the one located next to each other. This is the gist.

  According to this configuration, since the edges of the transmitter communication area are overlapped with each other adjacent to each other, it is difficult to form a gap between the communication areas, and simplifies the setting work of the communication area. it can.

  Invention of Claim 4 is a position estimation apparatus of the electronic key system as described in any one of Claims 1-3. WHEREIN: The said transmitter is provided in each door of the vehicle as the said communication master, The said The gist of the communication area of the transmitter is that it is formed around each door of the vehicle.

  According to this configuration, a general vehicle is equipped with a transmitter at each door of the vehicle in order to realize an electronic key system. In this configuration, this existing transmitter is used to It is possible to calculate an initial position necessary for the calculation of the estimation algorithm. Therefore, it is possible to calculate the initial position using existing parts without mounting a new article.

  According to the present invention, it is possible to reduce the calculation load of position estimation using an array antenna, and it is possible to simplify the communication area setting of the transmitter that causes the electronic key to perform a response operation.

The block diagram which shows schematic structure of an electronic key system. The top view which shows the communication area of a vehicle. The figure which shows the transmission timing of a transmitter, and the reception timing of a receiver. The flowchart which shows control of a vehicle. The figure which shows the positional relationship of an array antenna and a wave source.

Hereinafter, an embodiment in which the present invention is applied to an electronic key system of a vehicle will be described with reference to FIGS.
As shown in FIG. 1, the vehicle 2 includes an electronic key that can perform vehicle operations such as locking / unlocking a door lock and starting / stopping an engine without the driver actually operating the vehicle key. The system, so-called key operation free system is installed. In the key operation free system, an electronic key 1 that can transmit an ID code unique to a key by wireless communication is used as a vehicle key, and ID collation is executed by narrow-band wireless communication between the electronic key 1 and the vehicle 2. A door lock or an engine corresponds to an in-vehicle device.

  The key operation-free system includes an entry system in which ID verification is automatically performed when the vehicle 2 approaches or leaves the vehicle 2 and door locking / unlocking is permitted, or actual vehicle key operation during engine start / stop operation There is a one-push engine start system as a function that can perform an engine start / stop operation by a simple switch operation without requiring a switch. This will be described below. The vehicle 2 includes a verification ECU (Electronic Control Unit) 21 that performs ID verification when performing narrow-area wireless communication (hereinafter referred to as smart communication) with the electronic key 1, and the vehicle 2. A main body ECU 31 that manages the state of the engine and an engine ECU 32 that performs engine ignition control and fuel injection control are provided. The verification ECU 21, the main body ECU 31, and the engine ECU 32 are connected through an in-vehicle LAN (Local Area Network) 30.

  The verification ECU 21 includes a first LF transmitter 22, a second LF transmitter 23, a third LF transmitter 24, and a fourth LF as transmitters that are embedded in each door of the vehicle 2 and can transmit signals in an LF (Low Frequency) band. A transmitter 25 and a fifth LF transmitter 26 are connected. These LF transmitters 22 to 26 can transmit the request signal Srq from their respective transmitting antennas to their surroundings. The LF transmitters 22 to 26 form communication areas A1, A2, A3, A4, and A5 having a predetermined signal strength of the request signal Srq around each door of the vehicle 2 (see FIG. 2). These communication areas A1, A2, A3, A4, and A5 are formed in a substantially circular shape, and the edges of the adjacent areas are overlapped, in other words, overlapped.

  The verification ECU 21 is connected to a UHF receiver 27 that is embedded in the indoor ceiling of the vehicle 2 and can receive a UHF (Ultra High Frequency) band signal. The UHF receiver 27 is equipped with an array antenna 27a composed of a plurality of receiving elements (antenna elements). The array antenna 27a is a kind of directional antenna in which a large number of small antennas are arranged in a plane, and the radio wave arrival direction, radio wave arrival position, and the like can be calculated. Further, the LF transmitters 22 to 26 are arranged at substantially equal intervals from the array antenna 27a.

  The verification ECU 21 is provided with a memory 21a that stores an ID code and a verification unit 21b that performs ID verification. The collation unit 21b compares the ID code registered in the memory 21a with the ID code of the electronic key 1 to perform ID collation. The verification ECU 21 transmits the UHF band request signal Srq from the LF transmitters 22 to 26 in a time-sharing manner to form communication areas A1, A2, A3, A4, A5 in which the periphery including the vehicle 2 is divided into five ( (See FIG. 3). Note that these communication areas A1, A2, A3, A4, and A5 correspond to divided areas. Further, the time division refers to an operation of sequentially transmitting the request signal Srq while taking a certain time difference.

  On the other hand, the electronic key 1 is provided with a communication control unit 11 as a control unit for performing wireless communication with the vehicle 2 according to the electronic key system. The communication control unit 11 includes a memory 11a in which an ID code is stored as a unique key code. Connected to the communication control unit 11 are an LF receiving unit 12 capable of receiving an externally transmitted UHF band signal and a UHF transmitting unit 13 capable of transmitting a UHF band signal in accordance with an instruction from the communication control unit 11. Yes. When receiving the request signal Srq from the vehicle 2, the communication control unit 11 transmits an ID code signal Sid including its own ID code from the UHF transmission unit 13.

  When the verification ECU 21 receives the ID code signal Sid returned from the electronic key 1 by the array antenna 27a, the verification ECU 21 calculates the arrival direction of the radio wave by the array antenna 27a. When the verification ECU 21 recognizes that the electronic key 1 is located outside the vehicle, the verification ECU 21 performs external verification as ID verification. When the verification outside the vehicle is confirmed, the door lock device 38 locks and unlocks the door lock via the main body ECU 31. Allow or execute Further, when the verification ECU 21 recognizes that the electronic key 1 is located in the vehicle, it performs in-vehicle verification as ID verification, and when it is confirmed that the in-vehicle verification is established, permits the power switch operation and the engine start operation by the engine switch 35. .

  For the position estimation using the array antenna 27a, for example, the near wave source ESPRIT method which is a kind of the ESPRIT method is used. By the way, the ESPRIT method is a calculation method for obtaining the direction of arrival of radio waves using a rotation invariant that holds between adjacent subarrays when the wave source is far away. In other words, when the wave source is far away, the position of the wave source can be specified by the ESPRIT method. However, with the simple ESPRIT method, the rotation invariant is not established when the wave source is in the vicinity, and the position of this near wave source cannot be estimated. There is a present situation.

  Thus, the near wave source ESPRIT method has been devised as an ESPRIT method capable of estimating the position even in the vicinity of the wave source. The position estimation algorithm of the near wave source ESPRIT first estimates the initial position of the radio wave source, and calculates a transformation matrix from the position obtained by the initial estimation. The conversion matrix is a matrix that performs conversion by multiplying the array mode vector for the initial estimated position by a constant. When this transformation matrix is used, a pseudo rotation invariant is established. The near wave source ESPRIT method estimates the wave source position using a calculation method similar to the ESPRIT method using this pseudo rotation invariant.

  The key operation free system of this example is provided with a key position estimation device 3 that realizes the key position estimation by the array antenna 27a with a simple configuration. The key position estimation device 3 of this example is a device that calculates a position by using the detection positions by the LF transmitters 22 to 26 as the initial estimated position when determining the key position by an algorithm that requires the initial position. Yes.

  Hereinafter, the key position estimation device 3 will be described in detail. The verification ECU 21 is provided with an initial position calculation unit 21c as an initial position calculation means for calculating an initial position for obtaining the position of the electronic key 1. When the initial position calculation unit 21c receives the ID code signal Sid with the array antenna 27a, the initial position calculation unit 21c confirms which reception signal at this time is a response to the request signal Srq transmitted by which LF transmitter, and the electronic key 1 communicates. It is determined which of the areas A1 to A5 is located. Then, the initial position calculation unit 21c sets one point of the communication area where the electronic key 1 is located, in this example, the center P of the communication area as the initial position of the position estimation algorithm.

  Further, the verification ECU 21 is provided with a key position estimation unit 21d as key position estimation means for estimating the position of the electronic key 1 based on the initial position calculated by the initial position calculation unit 21c. The key position estimation unit 21d uses the center P calculated by the initial position calculation unit 21c as the initial position, estimates the radio wave arrival direction and radio wave emission position of the ID code signal Sid by a position estimation algorithm (neighboring wave source ESPRIT method), The position of key 1 is calculated.

Next, the control operation of the in-vehicle device including the position estimation by the verification ECU 21 will be described with reference to FIG. As an example of the position estimation method, the neighborhood wave source ESPRIT method is given.
First, as shown in FIG. 4, the verification ECU 21 transmits the request signal Srq from each of the LF transmitters 22 to 26 to each of the communication areas A1, A2, A3, A4, and A5 in a time division manner (step S1). That is, as shown in FIG. 3, the request signals Srq are transmitted in order from the first LF transmitter 22 so that the transmission timings of the request signals Srq from the LF transmitters 22 to 26 do not overlap. In the present embodiment, the first LF transmitter 22 is followed by the second LF transmitter 23, then the third LF transmitter 24, the fourth LF transmitter 25, and the fifth LF transmitter 26, and the LF transmitters 22 to 26 in this order. Transmits a request signal Srq. The verification ECU 21 transmits a request signal Srq from each of the LF transmitters 22 to 26 so as to repeat this one set intermittently, with one cycle in which the request signal Srq is transmitted once from each of the LF transmitters 22 to 26 as one set. Let

  Subsequently, as shown in FIG. 4, the verification ECU 21 determines whether or not the ID code signal Sid is received by the UHF receiver 27 (array antenna 27a) (step S2). And collation ECU21 will complete | finish a process, if ID code signal Sid is not received (step S2: NO). On the other hand, when the verification ECU 21 receives the ID code signal Sid (step S2: YES), the verification ECU 21 proceeds to step S3.

  In step S3, initial position estimation is performed by the initial position calculation unit 21c of the verification ECU 21. At this time, the initial position calculation unit 21c estimates in which of the communication areas A1 to A5 the electronic key 1 exists from the reception timing of the received ID code signal Sid. That is, as shown in FIG. 3, the initial position calculation unit 21c estimates the initial position of the position estimation algorithm based on the reception timing of the ID code signal Sid transmitted by the electronic key 1 in response to the transmitted request signal Srq. To do. For example, when the electronic key 1 is located in the communication area A1 of the first LF transmitter 22, the initial position calculation unit 21c receives the ID from the electronic key 1 at the reception timing immediately after transmitting the request signal Srq from the first LF transmitter 22. Since the code signal Sid is received, it is estimated that the electronic key 1 exists in the communication area A1 of the first LF transmitter 22.

Then, the initial position calculation unit 21c assumes that the electronic key 1 exists at the center P1 as one point of the communication area A1 where the electronic key 1 is estimated to exist, and sets the XY coordinates (x ₀ , y ₀ ) of the center P1. The initial position for obtaining the position of the electronic key 1 is set (step S4). The center P1 of the communication area is set as the initial position because the position where the electronic key 1 exists is determined as the center P1 of the communication area A1 regardless of the position of the electronic key 1 in the communication area A1. This is because the probability that the distance is closest is increased. Further, since the center P1 of the communication area A1 is on the door, there is a high possibility that the electronic key 1 entering the communication area A1 is located nearby. For this reason, the center of the communication area where the electronic key 1 exists is set as the initial position for obtaining the position of the electronic key 1.

  Subsequently, the key position estimation unit 21d estimates the position of the electronic key 1 using the initial position calculated by the initial position calculation unit 21c (step S5). That is, the key position estimation unit 21d estimates the position of the electronic key 1 using the near wave source ESPRIT method. Therefore, position estimation will be described in the order of formulation, ESPRIT method, and application of ESPRIT method to the near wave source.

<Formulation>
Let L be the number of wave sources. FIG. 5 shows a narrowband wave source of the l-th wave (l = 1, 2,..., L) located at (x, y) = (x ₁ , y ₁ ) = β ₁ and elements arranged on the x-axis. It is the figure which showed the positional relationship with the K element equal interval linear array of the space | interval d. When an L-wave incoming wave is incident on the array antenna, the received signal xk of the _k -th (k = 1, 2,..., K) antenna element can be expressed as follows.

ここに、Ｆ_ｌはｌ番目の波源の複素振幅であり、ｄ_ｋ（β_ｌ）はｌ番目の波源とｋ番目のアレー素子との距離、η_ｋ（β_ｌ）はｌ番目の波源の位置β_ｌからアレー素子位置までの伝搬による位相遅れである。また、ｗ_ｋはｋ番目のアレー素子における熱雑音、λ_ｌはｌ番目の波源の波長を表す。式（１）を用い、入力ベクトルＸを次のように定義する。

Here, F _l is the complex amplitude of the l-th wave source, d _k (β _l ) is the distance between the l-th wave source and the k-th array element, and η _k (β _l ) is the position of the l-th wave source. is the phase delay due to the propagation of from β _l to the array element position. W _k represents thermal noise in the k-th array element, and λ _l represents the wavelength of the l-th wave source. Using equation (1), the input vector X is defined as follows.

このとき、振幅ベクトルＦ、アレーのモード行列Ａ、熱雑音ベクトルＷを次のようにそれぞれ定義する。

At this time, the amplitude vector F, the array mode matrix A, and the thermal noise vector W are respectively defined as follows.

なお、^Ｔは転置を表す。このとき、アレーアンテナへの入力ベクトルは次のように表すことができる。

^T represents transposition. At this time, the input vector to the array antenna can be expressed as follows.

このようにして定義された入力ベクトルＸの相関行列Ｒ_ｘｘは、次式のように表すことができる。

The correlation matrix R _xx of the input vector X defined in this way can be expressed as follows:

ここで、Ｈは複素共役転置を、Ｅ［・］は期待値を表す。各アンテナの熱雑音は互いに無相関であり、入射波とも無相関であるので、スナップショット数が無限である場合、熱雑音電力をσ^２とすると、Ｗ＝σ^２Ｉ（Ｉ：単位行列）となる。

Here, H represents a complex conjugate transpose, and E [•] represents an expected value. Since the thermal noise of each antenna is uncorrelated with each other and uncorrelated with the incident wave, if the number of snapshots is infinite and the thermal noise power is σ ² , W = σ ² I (I: unit matrix) It becomes.

<ESPRIT method>
The ESPRIT method is established on the assumption that the wave source is far away and the incoming wave can be regarded as a plane wave. Consider a case where the l-th (l = 1, 2,..., L) wave source that is sufficiently far away comes from an angle θ _l (positive in the counterclockwise direction) with the y-axis. Then, when the wave source is in the vicinity, the component a _k (β _l ) of the mode vector a (β _l ) expressed by the equation (8) is changed as the following equation.

ここで、第１，２，…，Ｋ−１素子からなるサブアレーを第１サブアレー、第２，３，…，Ｋ素子からなるサブアレーを第２サブアレーと呼ぶこととする。このとき、Ｊ_１，Ｊ_２を次のように定義する。

Here, the subarray consisting of the first, second,..., K-1 elements is called the first subarray, and the subarray consisting of the second, third,. At this time, J ₁ and J ₂ are defined as follows.

第１サブアレーのモード行列はＪ_１Ａ、第２サブアレーのモード行列はＪ_２Ａと表すことができる。すると、これらサブアレーのモード行列の間には次の関係式が成り立つ。

The mode matrix of the first subarray can be expressed as J ₁ A, and the mode matrix of the second subarray can be expressed as J ₂ A. Then, the following relational expression holds between the mode matrices of these subarrays.

ここに、diag［・］は対角行列を表す。式（１７）は回転不変式（ｒｏｔａｔｉｏｎａｌ ｉｎｖａｒｉａｎｃｅ）と呼ばれ、ＥＳＰＲＩＴ法の基本原理を表す式である。これは第１サブアレーの第ｌ波のモードベクトルＪ_１ａ（β_ｌ）をexp｛ｊφ_ｌ｝倍すると、第２サブアレーのモードベクトルＪ_２ａ（β_ｌ）となることを意味し、サブアレー同士が同形かつ、波源が遠方であるからこそ成り立つものである。

Here, diag [•] represents a diagonal matrix. Expression (17) is called a rotation invariance and is an expression representing the basic principle of the ESPRIT method. This means that when the mode vector J ₁ a (β _l ) of the l-th wave of the first sub-array is multiplied by exp {jφ _l }, the mode vector J ₂ a (β _l ) of the second sub-array is obtained. Is the same shape and the wave source is far away.

Now, the eigenvalue problem for the correlation matrix R _xx obtained by the equation (11) is solved.

上記を解いて得られるＫ個の固有値を、値の大きいものから順にλ_１，λ_２，…，λ_Ｋとする。言い換えれば、式（２１）のΛの対角成分λ_ｋが固有値である。このうち、大きいものからＬ個の固有値に対応する固有ベクトルｖ_１，ｖ_２，…，ｖ_Ｌの張るＬ次元部分空間は、行列Ａを構成するＬ個の列ベクトルの張る部分空間に一致する。したがって、以下を満たすＬ次の正則な行列Ｔが唯一存在する。

Let _K eigenvalues obtained by solving the above be λ ₁ , λ ₂ ,..., Λ _{K in} descending order. In other words, the diagonal component λ _k of Λ in equation (21) is an eigenvalue. Among these, the L-dimensional subspace spanned by the eigenvectors v ₁ , v ₂ ,..., _VL corresponding to the L eigenvalues from the largest matches the subspace spanned by the L column vectors constituting the matrix A. Therefore, there is only one L-order regular matrix T that satisfies the following.

したがって、式（２３）をＡ＝Ｅ_ＳＴ^−１と変形し、式（１７）へ代入すると次式が導かれる。ここで、式（２３）におけるＥ_ｓが信号部分空間に属する固有ベクトルである。

Therefore, when the equation (23) is transformed to A = E _S T ⁻¹ and substituted into the equation (17), the following equation is derived. Here, the eigenvector belonging to E _s is the signal subspace in equation (23).

ここで、Ｅ_Ｘ，Ｅ_Ｙ，Ψはそれぞれ以下のように定義した。

Here, E _X , E _Y , and Ψ were defined as follows.

Ｅ_Ｘ，Ｅ_Ｙはそれぞれ式（２６），（２７）より計算可能であるため、式（２５）よりΨを算出可能である。よって、式（２８）より、Ψを固有値展開することでＴ^−１，Φを求めることができ、Φが求まれば式（１８），（１９）より到来角度θ_ｌを算出することができる。

Since E _X and E _Y can be calculated from Expressions (26) and (27), respectively, Ψ can be calculated from Expression (25). Therefore, T ⁻¹ and Φ can be obtained by developing eigenvalues of Ψ from Expression (28), and when Φ is obtained, the arrival angle θ _l can be calculated from Expressions (18) and (19). .

<Application of ESPRIT method to nearby wave source>
In the above, it was confirmed that the ESPRIT method can be applied when the wave source is far away. However, when the wave source is in the vicinity, the received signal is in an irregularly spaced array in which the phase difference of the received signals between adjacent elements is not constant in spite of the equally spaced array. Therefore, a pair of subarrays having the same shape that satisfies the rotation invariant (17) that is the principle expression of the ESPRIT method cannot be extracted.

  Therefore, in this method, a transformation matrix B that satisfies the following equation is used.

式（２９）は、「位置β_ｌ＝(ｘ_ｌ，ｙ_ｌ)を表すモードベクトルａ（β_ｌ）を、変換行列Ｂにより変換すると、ａ（β_ｌ）の（ｘ_ｌ＋ｊｙ_ｌ）倍となる」ことを意味する。そこで、このような変換行列Ｂが得られれば，以下の方法により波源位置を推定することが可能である。

Equation (29), "the position _{β l = (x l, y} l) are presented mode vector a a (beta _l), when converted by the transformation matrix _{B, (x l + jy l} ) of a (beta _l) times and Means. Therefore, if such a transformation matrix B is obtained, the wave source position can be estimated by the following method.

  First, Expression (29) is expressed as follows using the mode matrix A.

ここに、変換行列Ｂはモード行列ＡをＡΩという行列に変換することがわかる。この式（３０）を回転不変式と呼ぶ。また、ω（β_ｌ）が波源位置β_ｌと１対１対応する。次に、式（２３）をＡ＝Ｅ_ＳＴ^−１と変形し、式（３０）へ代入すると，次式が導かれる。

Here, it can be seen that the conversion matrix B converts the mode matrix A into a matrix of AΩ. This equation (30) is called a rotation invariant equation. Further, ω (β _l ) has a one-to-one correspondence with the wave source position β _l . Next, when equation (23) is transformed to A = E _S T ⁻¹ and substituted into equation (30), the following equation is derived.

ここで、Ｅ_Ｘ，Ｅ_Ｙ，Ψはそれぞれ以下のように定義した。

Here, E _X , E _Y , and Ψ were defined as follows.

以上より、式（３３）は式（２５）と等しく、ＥＳＰＲＩＴ法が適用可能であることがわかる。また、式（３３）より得られるΨは、Ωが対角行列であることから、式（３６）のようにΩの対角成分を固有値、Ｔ^−１の列ベクトルを固有ベクトルとする行列である。したがって、Ψを固有値展開することによりΩが得られ、更に式（３１），（３２）を用いることで波源位置β_ｌを求めることができる。すなわち、式（３０）を満たす変換行列Ｂを得ることができれば、近傍波源に対してＥＳＰＲＩＴ法を適用することができ、波源位置が推定可能となる。

From the above, it can be seen that Expression (33) is equal to Expression (25), and that the ESPRIT method can be applied. Further, Ψ obtained from the equation (33) is a matrix having a diagonal component of Ω as an eigenvalue and a column vector of T ⁻¹ as an eigenvector, as Ω is a diagonal matrix, as in equation (36). . Therefore, Ω is obtained by developing eigenvalues of Ψ, and the wave source position β ₁ can be obtained by using the equations (31) and (32). That is, if the transformation matrix B satisfying Expression (30) can be obtained, the ESPRIT method can be applied to the nearby wave source, and the wave source position can be estimated.

  Since the transformation in Equation (29) is a nonlinear transformation, there is no matrix B that satisfies Equation (29) for all positions. Therefore, in this method, the transformation matrix B is calculated using the following equation.

ここで、^leftは左逆行列を表す。すなわち、Ａ^leftＡ＝Ｉである。モード行列ＡはＫ×Ｌ行列であり、素子数Ｋ＞波源数Ｌより縦長の行列である。また、Ａの列ベクトル、すなわち各波源の位置β_ｌに対するモードベクトルａ（β_ｌ）はそれぞれ一次独立であるため、左逆行列Ａ^leftが存在する。したがって、式（３７）により変換行列Ｂを算出することが可能である。モード行列Ａは波源位置β_ｌより算出されるものであり、推定前の段階では未知の行列だということである。したがって、まず初期推定を行い、おおよその波源位置からモード行列を算出し、これを用いて変換行列の計算を行うこととなる。

Here, ^left represents a ^left inverse matrix. That is, A ^left A = I. The mode matrix A is a K × L matrix, and is a vertically long matrix with the number of elements K> the number of wave sources L. In addition, since the column vector of A, that is, the mode vector a (β _l ) for each wave source position β _l is linearly independent, there is a left inverse matrix A ^left . Therefore, it is possible to calculate the transformation matrix B by the equation (37). The mode matrix A is calculated from the wave source position β ₁ , and is an unknown matrix before the estimation. Therefore, initial estimation is performed first, a mode matrix is calculated from the approximate wave source position, and a transformation matrix is calculated using this.

In this embodiment, the transformation matrix B is calculated with the center P1 of the communication area A1 obtained previously as the initial position (x ₀ , y ₀ ). The conversion matrix B is a matrix that performs conversion by multiplying the mode vector A with respect to the center P1 as the initial position by a constant.

Then, the position of the electronic key 1 is estimated using the near wave source ESPRIT method (step S6).
Next, the collation unit 21b of the collation ECU 21 compares the ID code of the ID code signal Sid with the ID code of the memory 21a to perform ID collation (step S7). And the collation part 21b will complete | finish a process, if ID collation is not materialized (step S7: NO).

  On the other hand, if ID collation is materialized (step S7: YES), collation part 21b will control in-vehicle equipment based on the position of electronic key 1 (step S8). That is, if the position of the electronic key 1 is outside the vehicle, the verification unit 21b determines that the verification outside the vehicle is established, and permits or executes door lock locking / unlocking by the door lock device 38 via the main body ECU 31. Exit. On the other hand, if the position of the electronic key 1 is in the vehicle, the collating unit 21b determines that the in-vehicle collation is established, permits the switching of the power state by the engine switch 35, and ends the process.

  In the case of this example, when the vehicle 2 receives the ID code signal Sid from the electronic key 1, which LF transmitter 22 to 26 among the plurality of LF transmitters 22 to 26 is the response to the request signal Srq. By looking, it is confirmed in which communication area A1 to A5 the electronic key 1 is located. Then, the center of the communication area where the electronic key 1 is located, that is, the transmitter arrangement position, is used as the initial position, and the radio wave arrival direction of the ID code signal Sid, that is, the position of the electronic key 1 is determined by the near wave source ESPRIT method. For this reason, since it is not necessary to estimate using the initial position estimation algorithm in order to estimate the initial position, the initial position can be determined by a simple process. Therefore, when calculating the key position by an algorithm that requires position estimation, it is possible to reduce the calculation load.

As described above, according to the embodiment described above, the following effects can be obtained.
(1) By checking which LF transmitters 22-26 the electronic key 1 has returned the ID code signal Sid in response to the request signal Srq transmitted from which LF transmitter 22-26, It confirms whether it is located in the communication areas A1 to A5, and uses one point of the communication area where the electronic key 1 is located as the initial position of the position estimation algorithm. Therefore, since it is not necessary to calculate the initial position required for the calculation of the position estimation algorithm using complicated calculation, the calculation load of the initial position can be reduced, and the calculation load required for the key position estimation can also be reduced. . In addition, since the LF transmitters 22 to 26 only need to be able to transmit a signal that activates the electronic key 1 when the electronic key 1 approaches the vehicle 2 to some extent, there is no problem even if the communication area is roughly set. Therefore, it is not necessary to strictly set the communication areas of the LF transmitters 22 to 26, so that the communication area setting work of the LF transmitters 22 to 26 can be simplified.

  (2) In order to reduce the calculation load of the initial position, for example, it is assumed that a simple calculation formula is used. However, in this case, there is a concern that the calculation accuracy of the initial position deteriorates. On the other hand, in this example, since the initial position of the position estimation algorithm is calculated using the communication result of the mutual communication that the electronic key 1 and the vehicle 2 actually perform, the calculation result of the position estimation is relatively high. I can say that. Therefore, it is possible to achieve both reduction of calculation load and ensuring of position accuracy.

  (3) By transmitting the request signal Srq to each communication area in a time-sharing manner, the transmission timing is shifted, and the electronic key 1 exists from the reception timing of the ID code signal Sid of the electronic key 1 in response to one of the request signals Srq. Identify the communication area to be used. Therefore, it is used in the position estimation algorithm to specify in which communication area A1 to A5 the electronic key 1 is roughly located by simple control of transmitting signals sequentially from the plurality of LF transmitters 22 to 26, that is, to the position estimation algorithm. The initial position to be determined can be determined.

  (4) Since the edge portions in the vicinity of the communication areas A1, A2, A3, A4, and A5 overlap, it is difficult to form a gap between the communication areas, and the setting work of the communication area can be simplified.

  (5) As in the conventional electronic key system, the request signals Srq are transmitted from the LF transmitters 22 to 26 provided at the doors of the vehicle 2 to form the communication areas A1, A2, A3, A4, and A5. Therefore, if the UHF receiver 27 having the array antenna 27a is mounted while using an existing transmitter generally mounted on a vehicle, the initial position can be easily calculated using the existing components. it can.

  (6) When the electronic key 1 is present in any one of the communication areas A1, A2, A3, A4, A5, the centers P1, P2, P3, P4 of these communication areas A1, A2, A3, A4, A5 P5 is an initial position for obtaining the position of the electronic key 1. For this reason, even if the electronic key 1 is located in any of the communication areas A1, A2, A3, A4, and A5, it is possible to set the initial position at the closest position with the highest probability. Accuracy can be increased.

In addition, the said embodiment can be implemented with the following forms which changed this suitably.
In the above embodiment, the ID collation by the collation unit 21b is performed after the position calculation of the electronic key 1 in steps S3 to S6. However, the ID collation may be performed after reception of the ID code signal Sid and before step S3. You may perform simultaneously with the position calculation of the electronic key 1 of S3-S6. That is, it is only necessary to be able to control the in-vehicle device based on both the position calculation of the electronic key 1 and the ID verification establishment.

  In the above embodiment, the near wave source ESPRIT method is adopted as a position estimation method using initial estimation. However, the position estimation method is not limited to the near wave source ESPRIT method, and is adopted as a position estimation method using other initial estimation such as the EM method and the SAGE method. May be.

  In the above embodiment, the centers P1, P2, P3, P4, and P5 of the communication areas A1, A2, A3, A4, and A5 are set as the initial positions when the initial position is estimated, but the communication areas A1, A2, A3, and Another point may be set as the initial position as long as it is within A4 and A5.

  -Moreover, you may make it change an initial position according to the state of the vehicle 2. FIG. For example, as indicated by a broken-line circle in FIG. 2, when the vehicle 2 is in the locked state, the center X1, X2, X3, X4, X5 outside the communication area A1, A2, A3, A4, A5 is set to the initial position. In addition, when the vehicle 2 is after being unlocked and boarded, the center Y1, Y2, Y3, Y4, Y5 inside the communication area A1, A2, A3, A4, A5 is set as the initial position. According to such a configuration, a point closer to the position where the electronic key 1 exists can be set as an initial position for obtaining the position of the electronic key 1 by selecting the point according to the situation.

  In the above embodiment, the LF transmitters 22 to 26 are provided at each door to form the communication areas A1, A2, A3, A4, A5. However, the LF transmitters 22 to 26 are provided at the front side, in other words, at the doors of the driver seat and the passenger seat. The communication area may be formed by only the LF transmitters 22 and 26 provided. In this way, the number of LF transmitters can be reduced, and the cost can be reduced.

In addition, the position of the LF transmitter is not limited to the door, and can be set to an arbitrary position of the vehicle 2 as long as the communication area can cover a wide area around the vehicle 2.
In the above configuration, as shown in FIG. 1, the electronic key 1 is provided with the lock switch 18 and the unlock switch 19, and by operating the lock switch 18 and the unlock switch 19, the lock signal S1 or The unlocking signal Sul may be transmitted from the UHF transmitter 13 and received by the UHF receiver 27 of the vehicle 2 so that the door lock of the vehicle 2 can be unlocked or locked.

  In the above embodiment, the frequency of the radio wave used in the electronic key system is not necessarily limited to LF or UHF, and other frequencies can be used. Further, the frequency when the radio wave is transmitted from the vehicle 2 to the electronic key 1 and the frequency when the radio wave is returned from the electronic key 1 to the vehicle 2 are not necessarily different from each other, and may be the same frequency.

  -In the said embodiment, although applied to the position estimation of the electronic key 1 of an electronic key system, it is not necessarily applied only to the vehicle 2, If the communication object which uses an electronic key is a communication object, the employer will be especially It is not limited.

-In the said embodiment, although applied to the electromagnetic wave, you may apply to a sound wave.
Next, a technical idea that can be grasped from the above embodiment will be described together with its effects.
(A) In the position estimation device for an electronic key system according to any one of claims 1 to 4, the initial position calculation means uses the initial position as the center of the communication area. System position estimation device.

  According to this configuration, since the center of the communication area is set as the initial position, even if there is an electronic key at any position in the communication area, it is the point closest to the position where the electronic key exists, and the position of the electronic key is The accuracy of the initial position for obtaining can be improved.

  (B) In the position estimation device for an electronic key system according to any one of claims 1 to 4 and (A), the initial position calculation means determines the initial position when the vehicle as the communication master is in a locked state. Position estimation of the electronic key system, wherein the position is the center outside the vehicle in the communication area and the initial position is the center inside the vehicle in the communication area when the vehicle is unlocked and boarded apparatus.

  According to this configuration, when the vehicle is in a locked state, in other words, when the electronic key approaches the vehicle from the outside of the vehicle, the center of the outside of the communication area is set as the initial position, and when the vehicle is unlocked from this locked state and put in, in other words, When the electronic key is present in the vehicle, the center of the communication area outside the vehicle is set as the initial position. For this reason, the point close to the position where the electronic key exists can be set as the initial position for obtaining the position of the electronic key by selecting the point according to the situation.

  (C) In the position estimation device for an electronic key system according to any one of claims 1 to 4, (a), and (b), the position estimation algorithm uses the initial position calculated by the initial position calculation means. Based on this conversion matrix, a pseudo rotation invariant is calculated based on this conversion matrix, and the arrival of the ID signal is calculated from the rotation invariant. A position estimation apparatus for an electronic key system, wherein the algorithm is an algorithm for estimating a direction, that is, a position of the electronic key.

  DESCRIPTION OF SYMBOLS 1 ... Electronic key, 2 ... Vehicle, 11 ... Communication control part, 11a ... Memory, 12 ... LF receiving part, 13 ... UHF transmission part, 21 ... Collation ECU, 21a ... Memory, 21b ... Collation part, 21c ... Initial position calculation , 21d... Key position estimation unit, 22... 1 LF transmitter, 23... 2 LF transmitter, 24... 3 LF transmitter, 25 ... 4 LF transmitter, 26 ... 5 LF transmitter, 27. 27a ... array antenna, 30 ... in-vehicle LAN, 31 ... main body ECU, 32 ... engine ECU, 35 ... engine switch, 37 ... courtesy switch, 38 ... door lock device, A1, A2, A3, A4, A5 ... communication area, P1, P2, P3, P4, P5 ... center, Sid ... ID code signal, Srq ... request signal.

Claims (4)

When an ID reply request signal is transmitted from the transmitter of the communication master, the electronic key transmits an ID signal in response to the ID reply request signal, and when the communication master receives the ID signal by the array antenna, the array antenna is In the position estimation device of the electronic key system for checking whether or not the ID verification is established by checking the ID code included in the ID signal while calculating the radio wave arrival direction of the ID signal by the used position estimation algorithm,
An initial stage for calculating an initial position required for the calculation of the position estimation algorithm by confirming which ID signal is returned by the electronic key in response to the ID reply request signal transmitted from which transmitter Position calculating means;
Position estimation of an electronic key system, comprising: key position estimation means for estimating the position of the electronic key using the position estimation algorithm based on the initial position calculated by the initial position calculation means apparatus. The position estimation device for an electronic key system according to claim 1,
The initial position calculating means transmits the ID reply request signal from a plurality of transmitters in a time-sharing manner, and confirms to which transmitter the ID signal received by the array antenna is the response. The position estimation device of the electronic key system according to claim 1, wherein the electronic key system is located in a communication area of the transmitter, and one point of the calculated communication area is set as the initial position. . The position estimation device for an electronic key system according to claim 1 or 2,
Each communication area of the transmitter is overlapped with adjacent ones by an edge portion. A position estimation device for an electronic key system. In the position estimation apparatus of the electronic key system as described in any one of Claims 1-3,
The transmitter is provided at each door of the vehicle as the communication master, and a communication area of the transmitter is formed around each door of the vehicle.

Images