JP6965772B2 - Signal receiver, parameter estimation method and program - Google Patents

Description

The present invention relates to a signal receiver, a parameter estimation method and a program.

In the technical field of communication and measurement, extracting desired information from a signal is an indispensable technique. Therefore, there are various methods for extracting a desired signal from a signal that fluctuates independently of the desired signal.

For example, noise removal technology suppresses noise such as Gaussian noise whose intensity distribution and correlation distribution are known. Here, the noise removal technique includes a filter technique. A filter that minimizes the intensity of noise for a desired specific signal is called a matched filter. Filters are classified into analog filters and digital filters. The analog filter is realized as an analog circuit and operates in the frequency domain of the signal. A digital filter is realized as an operation, that is, an algorithm, performed on a discrete sample of a signal.

Further, in the noise canceling technique, a noise component is discriminated from a transmission signal and a reception signal, and a signal waveform that cancels the noise component is added in advance to the transmission waveform to bring the reception signal closer to the original transmission signal.

In addition, statistical methods may be used to remove noise after the fact for analysis.

In addition, equalization technology is known in communication. The equalization technology corrects signal distortion due to reflection and Doppler shift in the communication path. Especially for dealing with reflections, calculations are performed to estimate the channel parameters that describe the channel delay characteristics. In simple equalization, a particular known signal waveform is inserted into the signal to estimate the channel parameters, and the parameters that best reproduce the distortion of the signal waveform are calculated.

Alternatively, an inverse system may be calculated that cancels out the distortion of the signal waveform.

There is also a technique called judgment feedback type equalization, which is different from the simple equalization technique. In the determination feedback type equalization technology, after the data symbol is hard-determined using the estimation result of the channel parameter, the channel parameter is updated using the signal waveform in the symbol section, and the channel parameter is used as the subsequent data symbol. It is used to estimate and improve the judgment accuracy of past symbols.

Further, in the method called blind equalization, the existence of a known signal waveform is not premised, and the entire received signal is subject to equalization. In blind equalization, for example, since each symbol in the distorted signal waveform is considered to be approaching the Gaussian distribution by being randomly superimposed, the channel parameter so as to move the shape of the signal distribution away from the Gaussian distribution. The channel parameters are estimated by calculations such as optimizing.

Further, a signal source separation technique is known in audio signal processing. In this technique, signals emitted from a plurality of signal sources are received in a superposed state, and then separated into signals arriving from each signal source. In particular, blind separation technology minimizes the correlation of signals between receivers, for example, from signals received by multiple different receivers when the signal of the signal source is unknown, and is independent and in the form of a signal distribution. Each signal and its components are estimated by calculation such as separating into multiple signals that deviate from the Gaussian distribution.

Patent Document 1 discloses a technique for estimating the delay of multipath discretely by MMSE (Minimum Mean Square Error). Patent Document 2 discloses a technique for regressing a phase as a curve up to a quadratic. Further, in this document, the non-linearity of a signal is treated in the same manner as noise. Patent Document 3 discloses a technique for approximately linearly modeling inter-subcarrier interference in OFDM (Orthogonal Frequency Division Multiplexing) as a synthesis between adjacent carriers.

Special Table 2006-502625 Special Table 2009-538054 JP-A-2016-092454

In addition, each disclosure of the above prior art documents shall be incorporated into this document by citation. The following analysis was made by the present inventors.

As described above, there are various techniques for extracting desired information from a signal.

If noise removal technology is used, noise can be removed from a signal in which noise is randomly added, and signal parameters can be measured precisely. However, it is not possible to eliminate systematically changing signals. This is because it is difficult to distinguish between the signal components necessary for estimating the signal parameters and the unnecessary signal components in the systemically changing signals.

It is also conceivable to use noise canceling technology to remove systematically changing signals. However, since the systematically changing signal itself also depends on the signal parameters, it is not possible to simply separate the part that depends on the signal parameters from the part that does not, and even when using noise canceling technology, the above The problem is difficult to solve.

Any signal source can be separated by using the signal source separation technology. However, in this technique, since a plurality of receiving devices are usually used and the parameters themselves are not estimated and estimation accuracy is not required, high-precision estimation cannot be expected. It is also conceivable to identify a signal location that is not affected by unnecessary signals and measure the signal parameters from that location, but such a method lacks most of the information about the signal parameters of the original signal and is high. Accurate estimation is not possible.

By using the equalization technology, it is possible to estimate a kind of signal parameter called a data symbol and a communication path parameter from a signal systematically distorted in the communication path. However, the equalization technique deals with a model in which the same transmitted signal is superimposed with different time delays. In addition, the equalization technology specializes in estimating data symbols and channel parameters, and its main purpose is to demodulate the data symbols. In other words, in the equalization technology, the estimation accuracy of the parameter itself does not matter, and high-precision estimation cannot be expected.

In addition, parameter estimation problems typically involve calculations that minimize or maximize the objective function (eg, root mean square error). At that time, if iterative calculation is required to search for a solution, the following problems may occur.

First, adjustments may be required for reliable convergence of solutions. Second, the calculation time may be long. Thirdly, it may be difficult to implement the algorithm based on the above calculation as hardware such as FPGA (Field Programmable Gate Array). Fourth, there are cases where an exact solution cannot be obtained.

If the model of the signal parameter to be estimated is linear, a deterministic matrix operation can be performed, so that the above problem can be solved. However, since the change in the time direction is a non-linear change, the above problem becomes more serious in the measurement of the signal parameter caused by the change in the time direction.

An object of the present invention is to provide a signal receiving device, a parameter estimation method, and a program that contribute to estimating signal parameters with high accuracy and high speed from a signal that receives systematic non-linear fluctuation.

According to the first viewpoint of the present invention or the disclosure, an ideal reception signal when it is assumed that a receiving unit that receives a signal including a signal parameter to be measured and the signal parameter are fixed to a specific value. The generated unit and the signal parameter are used as the reference signal coefficient, and the received signal received by the receiving unit is modeled as a superposition of the reference signal. The receiving unit includes an estimating unit that estimates the solution of the minimum average squared error regarding the received signal received as the signal parameter, and the estimating unit calculates the minimum average squared error in a plurality of phase regions of the signal. A signal receiver is provided that runs over the data symbol interval.

According to the second viewpoint of the present invention or the disclosure, in a signal receiving device including a receiving unit that receives a signal including a signal parameter to be measured, it is ideal when it is assumed that the signal parameter is fixed to a specific value. The step of generating a typical received signal as a reference signal, the signal parameter as a coefficient of the reference signal, the received signal received by the receiving unit is modeled as a superposition of the reference signal, and the modeled received signal And the step of estimating the solution of the minimum average squared error regarding the received signal received by the receiving unit as the signal parameter, and the step of estimating the signal parameter includes the calculation of the minimum average squared error of the signal phase. A method of estimating parameters is provided that is performed over multiple data symbol intervals in the region.

According to the third viewpoint of the present invention or the disclosure, the signal parameter is fixed to a specific value in a computer mounted on a signal receiving device including a receiving unit that receives a signal including a signal parameter to be measured. The process of generating an ideal received signal in the assumed case as a reference signal, the signal parameter as a coefficient of the reference signal, and the received signal received by the receiving unit are modeled as a superposition of the reference signal, and the model The process of estimating the signal parameter by executing the process of estimating the solution of the minimized average squared error regarding the converted received signal and the received signal received by the receiving unit as the signal parameter is the process of estimating the signal parameter. A program is provided that performs the calculation of the above over multiple data symbol intervals in the phase region of the signal.
Note that this program can be recorded on a computer-readable storage medium. The storage medium may be a non-transient such as a semiconductor memory, a hard disk, a magnetic recording medium, or an optical recording medium. The present invention can also be embodied as a computer program product.

According to the viewpoints of the present invention and the disclosure, a signal receiving device, a parameter estimation method, and a program that contribute to estimating signal parameters with high accuracy and high speed from a signal that receives systematic nonlinear fluctuations are provided. ..

It is a figure for demonstrating the outline of one Embodiment. It is a figure for demonstrating the signal parameter of the present disclosure. It is a figure for demonstrating the signal parameter of the present disclosure. It is a figure for demonstrating the estimation about the parameter of time fluctuation. It is a figure which shows an example of the internal structure of the signal receiving apparatus which concerns on 1st Embodiment. It is a figure which shows an example of the internal structure (processing structure) of a parameter estimation part. It is a figure which shows an example of the hardware configuration of the signal receiving apparatus which concerns on 1st Embodiment. It is a flowchart which shows an example of the operation of the signal receiving apparatus which concerns on 1st Embodiment. It is a figure for demonstrating the basis function which concerns on 1st Embodiment. It is a figure for demonstrating the frequency minute shift rate which concerns on 1st Embodiment. It is a figure for demonstrating the effect which concerns on 1st Embodiment. It is a figure for demonstrating the effect which concerns on 1st Embodiment. It is a figure which shows an example of the internal structure of the signal receiving apparatus which concerns on 2nd Embodiment.

First, an outline of one embodiment will be described. It should be noted that the drawing reference reference numerals added to this outline are added to each element for convenience as an example for assisting understanding, and the description of this outline is not intended to limit anything. Further, the connecting line between the blocks in each figure includes both bidirectional and unidirectional. The one-way arrow schematically shows the flow of the main signal (data), and does not exclude interactivity. Further, in the circuit diagram, block diagram, internal configuration diagram, connection diagram, and the like shown in the disclosure of the present application, although not explicitly stated, an input port and an output port exist at the input end and the output end of each connection line, respectively. The same applies to the input / output interface.

The signal receiving device 100 according to one embodiment includes a receiving unit 101, a generating unit 102, and an estimating unit 103 (see FIG. 1). The receiving unit 101 receives a signal including a signal parameter to be measured. The generation unit 102 generates an ideal received signal as a reference signal assuming that the signal parameters are fixed to a specific value. The estimation unit 103 models the signal parameter as a reference signal coefficient and the reception signal received by the reception unit 101 as a superposition of the reference signals, and the minimum average of the modeled reception signal and the reception signal received by the reception unit 101. The solution of the square error is estimated as a signal parameter. Further, the estimation unit 103 executes the calculation of the minimum mean square error over a plurality of data symbol sections in the phase region of the signal.

The signal receiving device 100 according to one embodiment generates a reference signal as a virtual signal with fixed signal parameters. Further, the signal receiving device 100 models the received signal actually received by the receiving unit 101 as a linear superposition model of the reference signal having the signal parameter as a coefficient, and MMSE of the actual received signal and the modeled received signal. Estimate (measure) signal parameters by estimation. That is, although the details will be described later, the received signal is modeled as in the equation (1) described later, and the solution of the modeled received signal and the minimum mean square error of the received signal is in the equation (4). Obtained like this. Since the solution of equation (4) can be obtained by solving simultaneous equations, signal parameters can be estimated with high accuracy and high speed from a signal (received signal) that receives systematic nonlinear fluctuations.

The signal receiving device 100 can estimate linear signal parameters in the phase region with high accuracy and high speed based on the minimum mean square error (MMSE; Minimum Mean Square Error) even when the signal is subject to non-linear fluctuations such as modulation. .. Further, the types of signal parameters may be plural, and for example, the signal parameters may have different values depending on the data symbol. As an example of the signal parameter to be measured, the initial phase, frequency, time increase rate of frequency and the like can be considered, but other components of any temporal function may be used.

Further, as will be described in detail later, by modeling the received signal, not only the signal parameters but also the time fluctuation parameters due to the fluctuation of the reference time of the signal can be estimated with high accuracy (precision) by the same method. Can be done. More specifically, a communication device usually has a time reference inside it in the form of a combination of an oscillator and a counter. The fluctuation of the time reference appears as a non-linear fluctuation of the signal. The cause of the apparent fluctuation of signal parameters may be caused by the above-mentioned time fluctuation. However, when the fluctuation of the signal parameter is caused by the time fluctuation, the fluctuation of the signal parameter itself becomes non-linear. As a result, it becomes difficult to estimate the signal parameters at high speed. On the other hand, when the time fluctuation is small, the signal parameters can be roughly estimated at high speed by the method disclosed in the present application.

Further, as described above, the signal receiving device 100 can estimate an arbitrary signal parameter linear in the phase region from a signal subject to systematic nonlinear fluctuation with high accuracy and high speed. This fact means that it is possible to measure signal parameters by intercepting the transmitted signal without affecting the sender of the signal to be measured. This eliminates the need for the signal sender to require a dedicated signal waveform transmission for measuring signal parameters. In particular, even when the signal parameters deviate due to fluctuations in the time direction of the signal, it can be said that it is a further advanced purpose to measure the signal parameters with high accuracy and high speed. This is because when the signal parameter deviates from the original value, the deviation may be caused by a clock error of the transmitting device or the like.

Specific embodiments will be described in more detail below with reference to the drawings. In each embodiment, the same components are designated by the same reference numerals, and the description thereof will be omitted.

[First Embodiment]
The first embodiment will be described in more detail with reference to the drawings.

In the first embodiment, the same number of reference signals (hereinafter, also referred to as basis functions) as the signal parameters to be measured are generated. The reference signal is a signal including systematic non-linear fluctuation of the signal, and is a signal (signal waveform, function) on the assumption that the signal parameter is fixed to a certain value.

In the first embodiment, the received signal is modeled as a superposition of a number of reference signals, each with a fixed signal parameter. By this modeling, all the non-linear fluctuations of the received signal are passed on to the non-linearity of the reference signal, and the signal parameters themselves maintain the linearity (see FIGS. 2 and 3). In addition, in FIG. 3, three parameters are signal parameters to be measured, and the reference signal corresponding to each of these three parameters is shown.

In the first embodiment, a virtual received signal when the signal parameter to be measured is fixed to a certain value is generated and used as a reference signal (basis set). Further, in the first embodiment, the signal parameter is redefined as a coefficient of the basis function. As a result, systematic unnecessary variation of the signal can be removed and signal parameters can be estimated at the same time by matrix calculation.

In the present disclosure, to be measured i (i is a positive integer, the same applies hereinafter) th signal parameters is denoted by a _i, f an ideal signal waveform when fixed with a i _{= 1} as a function of time _i Notated as (t). The _f i (t) is a basis function (reference signal). f i _(t) is already have information of the non-linear variation of the modulation or the like.

The received signal θ (t) in the phase region can be expressed (modeled) by the following equation (1) by the signal parameters, the basis function and the noise function n (t).

[Equation 1]

The signal parameter _ai is expressed as a linear one in the equation (1). However, in practice, since what is discretely sampled as a received signal is obtained, k (k is a positive integer, the same applies hereinafter) when the th sample time and t _k, equation (1) is the following Equation (2) is obtained.

[Equation 2]

Further, as shown in the following equation (3), each variable is defined as a vector and a matrix.

[Equation 3]

In equation (3), K is the number of samples and P is the number of signal parameters to be measured.

Given the above, the solution that minimizes the mean square error for the received signal modeled by the actually acquired received signal, the basis function, and its coefficient signal permeter is a matrix as shown in Eq. (4) below. It is known that it can be expressed by the product of and vectors.

[Equation 4]

In equation (4), "T" on the right shoulder represents the transpose of the matrix, and "-1" represents the inverse matrix.

In the first embodiment, according to the above equation (4), an exact solution can be obtained by an operation limited to a matrix operation without causing adjustment for solution search and instability of calculation time as in a nonlinear model. .. Further, when the above calculation is executed, the information lost due to the non-linear fluctuation caused by the modulation or the like is recovered and the noise is removed at the same time.

The above method according to the first embodiment estimates the signal parameters using the information of all the samples of the received signal, and has high measurement accuracy.

Next, consider the case where the deviation of the signal parameters is caused by the time fluctuation. In this case, if the time variation at time t is τ (t) and the original signal in the phase region is θ (t), the received signal θ'(t) can be obtained by the following equation (5).

[Equation 5]

Therefore, if the time variation τ (t) is always sufficiently small, the received signal θ'(t) can be approximated as shown in the following equation (6) except for the signal discontinuity point (discontinuity point of higher order differential coefficient).

[Equation 6]

Here, the time fluctuation (time variation tau (t)), using the parameters _{b i} and the basis function _u i (t), expressed by the equation (7) below.

[Equation 7]

Then, the following equation (8) is obtained.

[Equation 8]

と置き替えた結果に等しい。 The above equation (8) is expressed in the equation (1).

Is equal to the result of replacing with.

Therefore, it is possible to estimate the parameters _{b i} by MMSE in the same manner as equation (4) (see FIG. 4). Note that FIG. 4B shows the amount of time fluctuation at each time, and FIG. 4A shows the received signal waveform at each time.

The received signal s (t) itself can be expressed by the signal r (t) in the amplitude region and the signal θ (t) in the phase region as shown in the following equation (9).

[Equation 9]

In equation (9), j is an imaginary unit. Further, the following equation (11) can be derived by calculating the function that can be expressed by the following equation (10).

[Equation 10]

[Equation 11]

The real part of the equation (11) represents the phase, and the imaginary part represents the amplitude. That is, if the logarithm of the received signal is calculated, not only the phase but also both the phase and the amplitude can be handled in a unified manner, and the mean square error can be minimized by the following equation (12).

[Equation 12]

であり、右肩の「*」は複素共役を表す。その他の記号の定義は式（４）と同一である。 In equation (12),

And the "*" on the right shoulder represents the complex conjugate. The definitions of other symbols are the same as in Eq. (4).

In the above method, since the linearity in the phase region is used as one of the means, when the calculation is executed by the equation (12), the linearity in the amplitude region is lost instead. Therefore, the signal parameter in the amplitude region may be measured in advance separately from the above calculation, and only the signal parameter in the phase region may be calculated by the equation (4).

Further, in Eqs. (4) and (12), all the samples at each time are treated equally and the signal parameter estimation calculation is performed, but there are cases where a highly reliable sample and a low sample are generated depending on the time. In such a case, the measurement accuracy of the signal parameter can be further improved by defining the weight c (t) of the sample at each time and performing the estimation calculation by the weighted MMSE.

とすると、式（１２）は下記の式（１３）の重み付きＭＭＳＥに拡張することができる。 Specifically, a K × K matrix having each of | c (t ₁ ) | ² , | c (t ₂ ) | ² , ..., | c (t _K ) | ^{2 as diagonal components}

Then, the equation (12) can be extended to the weighted MMSE of the following equation (13).

[Equation 13]

を

に置き換えれば良い。 Further, in order to perform the weighted MMSE only for the signal parameters in the phase region, the “*” in the equation (13) is not taken into consideration.

of

You can replace it with.

According to the disclosure of the present application, not only the measurement of desired signal parameters but also, for example, a minute signal that cannot be specified without using the method disclosed in the present application can be embedded in a signal having a higher intensity.

[Configuration of signal receiver]
Subsequently, the configuration of the signal receiving device 1 according to the first embodiment will be described.

FIG. 5 is a diagram showing an example of the internal configuration of the signal receiving device 1 according to the first embodiment. In the first embodiment, it is assumed that the systematic fluctuation of the signal is caused by the modulation, and the data symbol string for the modulation is known. Further, in the first embodiment, the signal to be measured varies depending on the data and signal parameters.

Referring to FIG. 5, the signal receiving device 1 includes a signal receiving unit 10, a detection unit 11, a phase conversion unit 12, a reference signal generation unit 13, a parameter estimation unit 14, and a parameter output unit 15. Consists of.

The signal receiving unit 10 is a means for receiving a signal including a signal parameter to be measured as information. Specifically, the signal receiving unit 10 receives a signal that varies depending on signal parameters and other information, and performs sampling. The signal receiving unit 10 outputs the sampling result as I / Q data (In-Phase data / Quadrature data) to the detection unit 11 which is a subsequent digital signal processing unit (digital signal processing circuit).

The detection unit 11 detects the modulated signal based on the I / Q data received from the signal reception unit 10 and synchronizes the time. At this time, particularly in the case of synchronous detection, rough amplitude and initial phase information of the signal are obtained. The detection unit 11 outputs the I / Q data, the rough amplitude, and the initial phase information after the time synchronization is performed to the phase conversion unit 12.

The phase conversion unit 12 converts the time-synchronized I / Q data received from the detection unit 11 into phase data or a pair of phase data and weight data. The phase is the declination of each signal point when I data (In-Phase data) is assigned to the real axis of the complex plane and Q data (Quadrature data) is assigned to the imaginary axis of the complex plane. The declination usually takes a value of 0 to 2π, and there are some parts that are discontinuous with time. However, the phase conversion unit 12 adds a multiple of 2π and converts the data into continuous data with respect to time. Generate phase data.

The phase conversion unit 12 outputs a set of phase data and weight data to the parameter estimation unit 14.

The reference signal generation unit 13 is a means for generating an ideal received signal as a reference signal when it is assumed that the signal parameters are fixed to a specific value. More specifically, the reference signal generation unit 13 generates a reference signal from the data symbol string for modulation. Only the same types of reference signals as the types of signal parameters to be estimated are generated. That is, the reference signal generation unit 13 generates the same number of reference signals as the signal parameters to be measured. More specifically, the reference signal generation unit 13 samples each reference signal so as to have the same number of samples as the phase data output by the phase conversion unit 12, and outputs the phase component to the parameter estimation unit 14.

The parameter estimation unit 14 uses the signal parameter as the coefficient of the reference signal, models the received signal as a superposition of the reference signals, and solves the modeled reception signal and the minimum average square error of the received signal received by the signal reception unit 10. Is a means of estimating as a signal parameter. At that time, the parameter estimation unit 14 executes the calculation of the minimum mean square error over a plurality of data symbol sections in the phase region of the signal.

The parameter estimation unit 14 performs a matrix calculation (Equation 4) of a plurality of signal parameters from the phase data and weight data of the received signal received from the phase conversion unit 12 and the phase data of the plurality of reference signals received from the reference signal generation unit 13. , Equation 12 or Equation 13). The parameter estimation unit 14 outputs the estimation result of the signal parameter to the parameter output unit 15.

In the first embodiment, the case where the parameter estimation unit 14 calculates the weighted MMSE will be described, but the parameter estimation unit 14 may calculate the unweighted MMSE. That is, the parameter estimation unit 14 estimates the signal parameter by calculating the minimum mean square error or the weighted minimum mean square error with respect to the sample point of the received signal received by the signal reception unit 10.

The parameter output unit 15 holds or outputs a plurality of signal parameters received from the parameter estimation unit 14 as signal parameter measurement results to the outside.

FIG. 6 is a diagram showing an example of the internal configuration (processing configuration) of the parameter estimation unit 14.

The parameter estimation unit 14 includes a sample data buffer 21, a weight data buffer 22, a reference signal buffer 23, an F ^{* T} Cw buffer 24, an F ^{* T} CF buffer 25, a simultaneous equation calculation unit 26, and the like. It is configured to include a solution output unit 27.

The reference signal buffer 23 includes a buffer for each type of generated reference signal. For example, the reference signal buffer 23 includes a reference signal 1 buffer, a reference signal 2 buffer, and a reference signal 3 buffer.

Each of the sample data buffer 21, the weight data buffer 22, and the reference signal buffer 23 stores the data received from the phase conversion unit 12 and the reference signal generation unit 13 shown in FIG.

の値を計算し、求めるべき信号パラメータの個数と同数の数値を格納する。 The F ^{* T} Cw buffer 24 is represented by the above equation (13) from the values stored in the sample data buffer 21, the weight data buffer 22, and the reference signal buffer 23.

Calculate the value of and store the same number of values as the number of signal parameters to be obtained.

値を計算し、求めるべき信号パラメータの個数の二乗の数の数値を格納する。 The F ^{* T} CF buffer 25 is represented by the above equation (13) from the values stored in the weight data buffer 22 and the reference signal buffer 23.

Calculate the value and store the number of squares of the number of signal parameters to be obtained.

が取り得る値のパターンが少なければ、予め逆行列を計算し、保持しておいても良い。 The simultaneous equation calculation unit 26 calculates the value of the plurality of signal parameters a shown in the equation (13) from ^{the F * T} Cw buffer 24 and the F ^{* T CF buffer 25, and outputs the value as a solution of the simultaneous equations.} This calculation can be performed by Gauss-Jordan or the like. Alternatively, the inverse matrix may be calculated explicitly. for example,

If the pattern of values that can be taken by is small, the inverse matrix may be calculated and held in advance.

The solution output unit 27 outputs the estimated signal parameter, and delivers the signal parameter to the parameter output unit 15 of FIG.

FIG. 7 is a diagram showing an example of the hardware configuration of the signal receiving device 1 according to the first embodiment. As shown in FIG. 7, the signal receiving device 1 includes a CPU (Central Processing Unit) 31 and a memory 32. The configuration shown in FIG. 7 is not intended to limit the hardware configuration of the signal receiving device 1. The signal receiving device 1 may include hardware (not shown). Alternatively, the number of CPUs and the like included in the signal receiving device 1 is not limited to the example shown in FIG. 7, and for example, a plurality of CPUs may be included in the signal receiving device 1.

The memory 32 is a RAM (Random Access Memory), a ROM (Read Only Memory), and an auxiliary storage device (hard disk or the like).

Each of the processing modules shown in FIGS. 5 and 6 is realized, for example, by the CPU 31 executing a program stored in the memory 32. In addition, the program can be downloaded via a network or updated using a storage medium in which the program is stored. Further, the processing module may be realized by a semiconductor chip. That is, it suffices if there is a means for executing the function performed by the processing module by some hardware and / or software. In particular, the parameter estimation unit 14 may be realized by FPGA, GPU (Graphics Processing Unit) or the like in addition to the realization by the CPU 31.

[Operation of signal receiver]
In the first embodiment, a case where the center frequency and the symbol rate of the signal subjected to MSK (Minimum Shift Keying) modulation are precisely measured will be described as an example.

For MSK, the center frequency f _c, the frequency of the symbol interval data is 0 f _{c +} Δ, if the frequency of the symbol interval is a data 1 was f _c - [delta, the symbol rate is R = 4Δ .. The configuration of the signal receiving device 1 according to the first embodiment is as shown in FIG.

FIG. 8 is a flowchart showing an example of the operation of the signal receiving device 1 according to the first embodiment.

First, when the MSK signal arrives, the signal receiving unit 10 receives the MSK signal and performs sampling (step S101).

Next, the detection unit 11 synchronizes the time of the MSK signal, and outputs the I / Q data after the time synchronization to the phase conversion unit 12 (step S102).

Next, the phase conversion unit 12 converts the I / Q data of the MSK signal into phase data, and in parallel, also converts it into amplitude data (step S103). Phase converter 12, as weight data c _{(t n)} amplitude data, and outputs to the parameter estimation unit 14 with the phase data θ _{(t n).}

From the known data symbol sequence, the reference signal generation unit 13 has a reference signal (basis function) f ₀ (t), f _{1 (t) corresponding to the center frequency and the symbol rate, and a basis function f 2} (base function) corresponding to the initial phase. t) is generated (step S104).

The basis function is as shown in the following equation (14).

[Equation 14]

In equation (14), α (t) and β (t) are values that depend on the history of data in order to maintain phase continuity. For example, when the data symbol string is {0 1 0 1 0 0 1 1 0}, the basis function f ₀ (t) becomes the basis function related to the center frequency (see FIG. 9 (a)), and f ₁ (t) becomes. It is a basis function related to the symbol rate (see FIG. 9B).

Arbitrary signal parameters can be estimated by the method of defining the basis function in the reference signal generation unit 13. For example, _{by adding a basis function such that f 3} (t) = t ² , the rate of increase in frequency can also be estimated. As described above, in the first embodiment, at least three or more basis functions can be handled, and at least one or more basis functions thereof differ depending on the data value (data symbol string).

The reference signal generation unit 13 samples each of the basis functions according to the time-synchronized received signal, and outputs them as f ₀ (t _n ), f ₁ (t _n ), and f ₂ (t _n ).

The parameter estimation unit 14 is based on equations (3) and (13) from _{c (t n} ), θ (t _n ), f ₀ (t _n ), f ₁ (t _n ), and f ₂ (t _{n). The signal parameters a 0} , a ₁ , and a ₂ are estimated by MMSE (assuming w = θ) (step S105).

The signal parameter a ₀ is the measurement result of the center frequency, and 4 × a ₁ is the measurement result of the symbol rate. Since the signal parameter a ₂ is the measurement result of the initial phase and is a value obtained by the regression analysis of the received signal, when the phase of the first point of the I / Q data is simply measured or the average of the phases is simply calculated. Compared, it is more accurate.

The parameter output unit 15 outputs the estimation result (step S106). Specifically, the parameter output unit 15 outputs signal parameters a ₀ and a ₁ (and a ₂ if necessary). Signal parameters a ₀ to the parameter output unit 15 _outputs, a ₁ from f _c, measuring the center frequency and symbol rate by calculating Δ is completed.

The parameter estimation unit 14, a different clock shift the center frequency f _c and the symbol rate R, even if slightly shifted from the design value, the frequency fine shift rate and data for the data "0" and "1" The frequency minute shift rate can be estimated. In that case, the basis function in the equation (8) is, for example, the following equation (15) (see FIG. 10).

[Equation 15]

Note that FIG. 10A shows a basis function related to the information “0”, and FIG. 10B shows a basis function related to the information “1”.

と置く。このことにより、中心周波数ｆ_ｃ及びシンボルレートＲがそれぞれ異なるクロックずれにより、設計値からわずかにシフトしている場合にも、周波数微小シフト率が推定できる。具体的には、パラメータ推定部１４は、信号パラメータｂ_０、ｂ_１として、データ「０」に対する周波数微小シフト率とデータ「１」に対する周波数微小シフト率が推定できる。 Further, the received waveform is θ'(t), and the ideal received waveform or a virtual received waveform (including phase continuity) based on a rough estimation result is θ (t).

And put. Thus, the different clock shift the center frequency f _c and the symbol rate R, if you are slightly shifted from the design value also frequency micro shift ratio can be estimated. Specifically, the parameter estimation unit 14 can estimate the frequency minute shift rate for the data "0" and the frequency minute shift rate for the data "1" as the _{signal parameters b 0} and b _1.

Further, from the signal parameters b ₀ and b ₁ , a minute shift rate of the center frequency can be obtained as (b ₀ + b ₁ ) / 2, and a minute shift rate of the symbol rate can be obtained as _{2 (b 0} − b _1).

In this way, the signal receiving device 1 can estimate the signal parameters even in an environment where the signal to be measured can fluctuate due to the fluctuation of the clock related to the transmission or reception of the signal. Specifically, in the signal receiving device 1, the approximate value of the signal parameter to be measured when there is no fluctuation of the clock is known, or the approximate value of the signal parameter to be measured is estimated in advance. In some cases, the signal parameters can be estimated.

[effect]
In the first embodiment, if the frequency is simply measured in each symbol section, the frequency systematically fluctuates in the symbol section due to unknown information such as the influence of the filter and particularly the group delay, and the measurement is highly accurate. Can not be expected at all (see FIG. 11). Note that FIG. 11A shows the phase fluctuation, and FIG. 11B shows the frequency fluctuation. In addition, although frequency fluctuations are systematic within the symbol section, the fluctuations also depend on the history of past data and fluctuate randomly over a long period of time. Therefore, it is difficult to directly predict the influence of filters and the like.

However, by using the method disclosed in the present application, it is possible to perform regression analysis that incorporates systematic fluctuations of signals over a large number of symbol sections, even if the information regarding the effects of filters and the like is not completely possessed. , The signal parameters in the phase region can be measured with high accuracy. That is, the method disclosed in the present application does not detect unknown signals as performed by blind signal processing, but obtains desired signal parameters buried by systematic signal fluctuations with high accuracy and high speed regardless of whether they are known or unknown. Can be measured. In other words, by using the blind signal processing technology as the preprocessing of the disclosure of the present application, the blind signal processing technology can be combined with the disclosure of the present application.

Moreover, since the signal parameters result in the solution of simultaneous equations, high-speed and reliable processing is possible. For example, when measuring frequency as a signal parameter, even if there is a systematic fluctuation of the signal, the calculation is executed by the phase difference between two points in the region where there is no fluctuation or the average of the local frequencies. Estimated calculation in. However, even if such an estimation calculation is not executed, the estimation accuracy can be maximized by using the information of all the sample points for regression as disclosed in the present application (see FIG. 12).

By applying the method disclosed in the present application, it is possible to obtain a signal receiving device or a signal analysis device that linearly estimates signal parameters buried in fluctuations due to data. Further, according to the method disclosed in the present application, when the signal is distorted by information unrelated to the desired information, only the desired information can be extracted with high accuracy and high speed. As a result, it is possible to measure a physical quantity without affecting the signal source, measure the specifications from an arbitrary modulated signal, measure a weak signal parameter, and the like.

[Second Embodiment]
Subsequently, the second embodiment will be described in detail with reference to the drawings.

[Configuration of signal receiver]
FIG. 13 is a diagram showing an example of the internal configuration of the signal receiving device 1 according to the second embodiment. The first and second embodiments differ in that the reference signal depends on an unknown received data sequence rather than a known data sequence. Referring to FIG. 13, the signal receiving device 1 according to the second embodiment further includes a signal demodulation unit 16, a decoding / information restoration unit 17, and a memory 18. Although the memory 18 is explicitly shown in FIG. 13, the signal receiving device 1 shown in FIG. 5 also includes a memory necessary for executing various processes.

The signal receiving unit 10 has the same function as the signal receiving unit 10 described with reference to FIG. That is, the signal receiving unit 10 receives a signal that varies depending on signal parameters and other information, and performs sampling.

The detection unit 11 has the same function as the detection unit 11 described with reference to FIG. That is, the detection unit 11 outputs the I / Q data, the rough amplitude, and the initial phase information after the time synchronization is performed to the phase conversion unit 12. Further, the detection unit 11 has a function of outputting digital data such as I / Q data to the signal demodulation unit 16 in addition to the function described with reference to FIG. Further, the detection unit 11 may receive the signal parameter estimation result from the memory 18 and perform more detailed feedback analysis.

The phase conversion unit 12 has the same function as the phase conversion unit 12 described with reference to FIG. That is, the phase conversion unit 12 outputs the set of the phase data and the weight data to the parameter estimation unit 14.

The reference signal generation unit 13 receives the restoration result of the data series in which the received signal is modulated from the decoding / information restoration unit 17, and generates each reference signal according to the data series. That is, the reference signal generation unit 13 generates a reference signal using the data restored by the decoding / information restoration unit 17. The reference signal generation unit 13 outputs the phase component of the reference signal to the parameter estimation unit 14 as in the first embodiment.

The parameter estimation unit 14 outputs the estimation result of the signal parameter to the parameter output unit 15 as described with reference to FIG. Further, the parameter estimation unit 14 may output the signal parameter estimation result to the memory 18 for more detailed analysis of the received signal. As described above, when the memory 18 is used, the memory 18 feeds back the estimation result of the signal parameter by outputting the estimation result of the signal parameter to the detection unit 11.

The parameter output unit 15 holds the signal parameter measurement result or outputs the signal parameter measurement result to the outside as described with reference to FIG.

The signal demodulation unit 16 receives digital data such as I / Q data of the received signal from the detection unit 11, demodulates the received signal according to the modulation method, and outputs it to the decoding / information restoration unit 17. At this time, noise removal technology and equalization technology can be used.

The decoding / information restoration unit 17 receives the demodulation result of the received signal from the signal demodulation unit 16, performs error correction processing and the like, identifies the data series that is the source of the modulation, and outputs the data series to the reference signal generation unit 13.

[Operation of signal receiver]
Hereinafter, the points different from the first embodiment of the signal receiving device 1 according to the second embodiment will be mainly described.

The detection unit 11 synchronizes the time of the MSK signal, outputs the I / Q data after the time synchronization to the phase conversion unit 12, and also outputs the data necessary for signal demodulation to the signal demodulation unit 16. In the case of MSK, the I / Q data and the initial phase contained therein are usually sufficient.

Next, the signal demodulation unit 16 demodulates the MSK signal. Since the phase of the MSK signal depends on the data sequence itself, demodulation is performed sequentially for each symbol interval.

The signal demodulation unit 16 outputs the data series of “0” or “1”, which is the demodulation result, to the decoding / information restoration unit 17.

Next, the decoding / information restoration unit 17 decodes the data series received from the signal demodulation unit 16 when it is encoded. Further, by encoding the decoding result again if necessary, the decoding / information restoration unit 17 obtains an estimated data sequence for signal modulation, for example, {0 1 0 1 0 0 1 10}.

When the reference signal generation unit 13 receives the data series, the reference signal generation unit 13 generates a reference signal by the same method as that described in the first embodiment.

The parameter estimation unit 14 estimates the signal parameters from the reference signal and the phase region reception signal, and then stores the estimation result in the memory 18. The estimation result stored in the memory 18 is fed back by the detection unit 11 and can be used for more accurate detection and demodulation.

In the second embodiment, the frequency shift for each data symbol is measured. At that time, by the above method, for example, a low-speed and minute FSK signal can be embedded in the MSK signal, and the FSK signal can be demodulated after the MSK signal is demodulated. Alternatively, it is possible to consider that the frequency shift is caused by the Doppler shift and measure the relative speed and relative acceleration of the sender.

Further, in the flowchart used in the above description, a plurality of steps (processes) are described in order, but the execution order of the steps executed in each embodiment is not limited to the order of description. In each embodiment, the order of the illustrated steps can be changed within a range that does not hinder the contents, for example, each process is executed in parallel. In addition, the above-described embodiments can be combined as long as the contents do not conflict with each other.

Although the industrial applicability of the present invention is clear from the above description, the present invention is suitably applicable to signal analysis / specification measurement, physical quantity measurement, signal interception, communication, orbit estimation and the like. In particular, in the communication field, it is possible to utilize the detection of a minute signal for communication and embed the signal in another signal. In orbit estimation, the frequency, its time derivative, second-order time derivative, etc. are estimated with high accuracy for each data, and the part caused by the Doppler shift is separated from it and used for the relative position calculation of the signal transmitter. can do.

Some or all of the above embodiments may also be described, but not limited to:
[Appendix 1]
This is the signal receiving device according to the first viewpoint described above.
[Appendix 2]
The signal to be measured varies depending on the data and signal parameters, preferably the signal receiving device of Appendix 1.
[Appendix 3]
The generation unit generates the reference signal in the same number as the signal parameters to be measured, preferably the signal receiving device of Appendix 1 or 2.
[Appendix 4]
The estimation unit estimates the signal parameters by calculating the minimum mean square error or the weighted minimum mean square error with respect to the sample point of the received signal received by the reception unit, preferably any one of Supplementary notes 1 to 3. The signal receiving device according to.
[Appendix 5]
Further provided with a restoring unit that restores data from the received signal received by the receiving unit.
The signal receiving device according to any one of Supplementary note 1 to 4, wherein the generation unit generates the reference signal by using the data restored by the restoration unit.
[Appendix 6]
In the environment where the signal to be measured can fluctuate due to the fluctuation of the clock related to the transmission or reception of the signal, the estimation unit can obtain the approximate value of the signal parameter to be measured when the fluctuation of the clock does not exist. The signal receiving device according to any one of Supplementary note 1 to 5, which estimates the signal parameter when it is known or the approximate value of the signal parameter to be measured is estimated in advance. ..
[Appendix 7]
The type of signal parameter to be measured is at least 3 or more, and the type of signal parameter that contributes to signal fluctuation differs depending on the data value, preferably the signal according to any one of Supplementary notes 1 to 6. Receiver.
[Appendix 8]
The signal receiving device according to any one of Appendix 1 to 7, wherein the signal parameter to be measured includes a signal amplitude.
[Appendix 9]
The receiving unit receives the signal subjected to MSK (Minimum Shift Keying) modulation, and receives the signal.
The signal receiving device according to any one of Supplementary notes 1 to 8, wherein the estimation unit executes a calculation regarding a weighted minimum mean square error with the amplitude of the signal as a weight.
[Appendix 10]
This is the method of estimating the parameters according to the second viewpoint described above.
[Appendix 11]
This is the program related to the third viewpoint described above.
Note that the form of Appendix 10 and the form of Appendix 11 can be expanded to the forms of Appendix 2 to the form of Appendix 9 in the same manner as the form of Appendix 1.

Each disclosure of the above-mentioned patent documents cited shall be incorporated into this document by citation. Within the framework of the entire disclosure (including the scope of claims) of the present invention, it is possible to change or adjust the embodiments or examples based on the basic technical idea thereof. Further, various combinations or selections of various disclosure elements (including each element of each claim, each element of each embodiment or embodiment, each element of each drawing, etc.) within the framework of all disclosure of the present invention. Is possible. That is, it goes without saying that the present invention includes all disclosure including claims, and various modifications and modifications that can be made by those skilled in the art in accordance with the technical idea. In particular, with respect to the numerical range described in this document, it should be interpreted that any numerical value or small range included in the range is specifically described even if there is no other description.

1,100 Signal receiving device 10 Signal receiving unit 11 Detection unit 12 Phase conversion unit 13 Reference signal generation unit 14 Parameter estimation unit 15 Parameter output unit 16 Signal demodulation unit 17 Decoding / information restoration unit 18, 32 Memory 21 Sample data buffer 22 Weight data buffer 23 Reference signal buffer 24 F ^{* T} Cw buffer 25 F ^{* T} CF buffer 26 Simultaneous equation calculation unit 27 Solution output unit 31 CPU
101 Receiver 102 Generator 103 Estimator

Claims (10)

A receiver that receives a signal including the signal parameter to be measured, and
A generator that generates an ideal received signal as a reference signal assuming that the signal parameters are fixed to a specific value.
The signal parameter is used as the coefficient of the reference signal, the received signal received by the receiving unit is modeled as a superposition of the reference signal, and the modeled received signal and the received signal received by the receiving unit are the minimum mean squares. An estimation unit that estimates the solution of the error as the signal parameter,
With
The estimation unit is a signal receiving device that executes the calculation of the minimum mean square error over a plurality of data symbol sections in the phase region of the signal. The signal receiving device according to claim 1, wherein the signal to be measured varies depending on data and signal parameters. The signal receiving device according to claim 1 or 2, wherein the generation unit generates the same number of reference signals as the signal parameters to be measured. Any one of claims 1 to 3, wherein the estimation unit estimates the signal parameters by calculating the minimum mean square error or the weighted minimum mean square error of the received signal received by the reception unit with respect to the sample point. The signal receiving device according to. Further provided with a restoring unit that restores data from the received signal received by the receiving unit.
The signal receiving device according to any one of claims 1 to 4, wherein the generation unit generates the reference signal using the data restored by the restoration unit. In the environment where the signal to be measured can fluctuate due to the fluctuation of the clock related to the transmission or reception of the signal, the estimation unit can obtain the approximate value of the signal parameter to be measured when the fluctuation of the clock does not exist. The signal receiving device according to any one of claims 1 to 5, which estimates the signal parameters when they are known or the approximate values of the signal parameters to be measured are estimated in advance. .. The signal according to any one of claims 1 to 6, wherein the type of the signal parameter to be measured is at least 3 or more, and the type of the signal parameter contributing to the fluctuation of the signal differs depending on the data value. Receiver. The signal receiving device according to any one of claims 1 to 7, wherein the signal parameter to be measured includes a signal amplitude. In a signal receiving device including a receiving unit that receives a signal including a signal parameter to be measured,
A step of generating an ideal received signal as a reference signal assuming that the signal parameter is fixed to a specific value, and
The signal parameter is used as the coefficient of the reference signal, the received signal received by the receiving unit is modeled as a superposition of the reference signal, and the minimum average square of the modeled received signal and the received signal received by the receiving unit is used. Including the step of estimating the solution of the error as the signal parameter.
The step of estimating the signal parameter is a method of estimating the parameter, in which the calculation of the minimum mean square error is executed over a plurality of data symbol sections in the phase region of the signal. A computer mounted on a signal receiver that has a receiver that receives signals that include signal parameters to be measured.
Processing to generate an ideal received signal as a reference signal assuming that the signal parameters are fixed to a specific value, and
The signal parameter is used as the coefficient of the reference signal, the received signal received by the receiving unit is modeled as a superposition of the reference signal, and the modeled received signal and the received signal received by the receiving unit are the minimum mean squares. The process of estimating the solution of the error as the signal parameter is executed.
The process of estimating the signal parameters is a program that executes the calculation of the minimum mean square error over a plurality of data symbol sections in the phase region of the signal.

Images