JP2011229090A - Preamble signal generation device, preamble signal generation method, preamble signal generation program and recording medium with preamble signal recorded - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a preamble signal generation device, a preamble signal generation method, a preamble signal generation program and a recording medium with preamble signal recorded obtaining the preamble signal which is approximated to an optimal value by searching for a short amount of time.SOLUTION: The preamble signal generation device 1 generates the preamble signal by orthogonal frequency division multiplex modulation. The preamble signal generation device 1 includes: amplitude linear search means 41 for searching for an amplitude value by a search using a linear search method; amplitude PSO means 42 for searching by a particle group optimizing method with the search amplitude value that is searched by the amplitude linear search means 41 as a starting point; phase linear search means 51 for searching for a phase value by the linear search method based on the amplitude value searched by the amplitude PSO means 42; and amplitude PSO means 52 for searching by the particle group optimization method with the search phase value searched by the phase linear search means 51 as a starting point.

Description

  The present invention relates to a technique for performing packet communication using orthogonal frequency-division multiplexing (hereinafter referred to as OFDM) modulation.

  Frame synchronization (timing synchronization) and carrier frequency synchronization in a receiving apparatus of a wireless communication system are performed using a training signal (in particular, the first packet is called a preamble signal). The training signal has different synchronization performance depending on the series. An optimal preamble signal can achieve high autocorrelation and low peak-to-average power ratio.

  An example of a method capable of generating such an optimal preamble signal is described in Patent Document 1. Since the method for generating a preamble signal described in Patent Document 1 is a method for obtaining a theoretical value using an FZC sequence, an ideal preamble signal can be obtained.

US Patent Application Publication No. 2008/0260063

  However, in the method described in Patent Document 1, a band called a guard band for preventing interference with an adjacent band in the OFDM signal cannot be set, so that an ideal preamble signal can be obtained. The signal is transmitted using the entire band allocated to. Therefore, when a null subcarrier cannot be set, out-of-band radiation increases, and there is a possibility of interfering with other communication signals.

Therefore, in order to obtain an optimal preamble signal, a guard band can be arbitrarily set in the allocated band, and a search is performed by an algorithm for generating a preamble signal, the result is determined, and the best A method for determining the preamble signal is desirable.
As an algorithm, for example, a linear search method (hereinafter sometimes referred to as a linear search method) is known. In this linear search method, an optimum preamble signal can be searched. However, if the accuracy is to be increased, the amount of calculation increases exponentially, so that there is a disadvantage that it takes too much search time.

  As another algorithm method, a particle swarm optimization method (hereinafter referred to as a PSO method) is known. This PSO method can search for a sub-optimal preamble signal even if an ideal result cannot be obtained, and can reduce the amount of calculation so that the search can be performed in a short time. is there. However, the PSO method may not obtain the expected result depending on the setting of the starting point.

  An object of the present invention is to provide a preamble signal generation device, a preamble signal generation method, a preamble signal generation program, and a recording medium on which a preamble signal is recorded, which can obtain a preamble signal approximate to an optimum value by a short-time search. And

  A preamble signal generation apparatus according to the present invention is a preamble signal generation apparatus that generates a preamble signal by orthogonal frequency division multiplexing modulation, and uses a preset initial amplitude value as a starting point for each predetermined step size. An amplitude linear search means for calculating a synchronization cost function calculated based on an amplitude value of a subcarrier carrying a preamble signal and outputting an amplitude value maximizing the synchronization cost function as a search amplitude value; and the amplitude linear Starting from the search amplitude value searched by the search means, a synchronization cost function calculated based on the amplitude value of the subcarrier carrying the preamble signal determined by the particle swarm optimization method is calculated, and the synchronization cost function Is output as the amplitude value of the subcarrier carrying the best preamble signal. Characterized by comprising an amplitude PSO means.

  The preamble signal generation method of the present invention is a preamble signal generation method for generating a preamble signal by orthogonal frequency division multiplexing modulation, and an amplitude linear search means for searching by a linear search method uses a preset initial amplitude value. Calculating a synchronization cost function calculated based on an amplitude value of a subcarrier carrying a preamble signal for each predetermined step size as a starting point, and outputting an amplitude value that maximizes the synchronization cost function as a search amplitude value And the amplitude PSO means for searching by the particle swarm optimization method based on the amplitude value of the subcarrier carrying the preamble signal determined by the search, starting from the search amplitude value searched by the amplitude linear search means. Calculate the calculated synchronization cost function and set the amplitude value that maximizes the synchronization cost function to the best Characterized in that it comprises a step of outputting as the amplitude value of the sub-carriers that carry preamble signals.

  Furthermore, a preamble signal generation program according to the present invention is a preamble signal generation program for causing a computer to generate a preamble signal by orthogonal frequency division multiplexing modulation, wherein the computer is started from a preset initial amplitude value. The synchronization cost function calculated based on the amplitude value of the subcarrier carrying the preamble signal for each predetermined step size is calculated by the linear search method, and the amplitude value that maximizes the synchronization cost function is used as the search amplitude value. Amplitude linear search means to output, and a search amplitude value searched by the amplitude linear search means as a starting point, calculated based on the amplitude value of the subcarrier carrying the preamble signal determined by the particle swarm optimization method Calculate the synchronization cost function and set the amplitude value that maximizes the synchronization cost function. Characterized in that to function as an amplitude PSO means for outputting as an amplitude value of the sub-carriers carrying the best preamble signal.

  According to the present invention, first, the amplitude linear search means searches for an amplitude value at which autocorrelation is good by a linear search method, and then the amplitude PSO means starts with the search amplitude value by the amplitude linear search means as a starting point. The optimal amplitude value is determined by the group optimization method. By doing so, the starting point in the particle swarm optimization method with a relatively short required time is set in advance as a good amplitude value even if the accuracy is reduced by roughening the step size by the linear search method in order to shorten the search time. Therefore, it is possible to obtain a preamble signal with good autocorrelation in a shorter time than when the entire search is performed by the linear search method.

  In the preamble signal generation apparatus of the present invention, further, a predetermined step size is obtained by a linear search method using the best amplitude value output from the amplitude PSO means as a starting point with a preset initial phase value. A phase linear search means for calculating a PAPR cost function calculated based on a phase value of a subcarrier carrying a preamble signal for each time, and outputting a phase value that minimizes the PAPR cost function as a search phase value; Using the best amplitude value output from the PSO means, the phase of the subcarrier carrying the preamble signal determined by the particle swarm optimization method, starting from the search phase value searched by the phase linear search means A PAPR cost function calculated based on the value is calculated, and the phase value that minimizes the PAPR cost function is determined as the best value. Provided with a phase PSO means for outputting a phase value of sub-carriers for carrying amble signal is desirable.

  Further, in the preamble signal generation method of the present invention, the phase linear search means for searching by the linear search method uses the best amplitude value output from the amplitude PSO means to set a preset initial phase value. As a starting point, a PAPR cost function calculated based on the phase value of the subcarrier carrying the preamble signal is calculated for each predetermined step size, and a phase value that minimizes the PAPR cost function is output as a search phase value. And a phase PSO means for searching by the particle swarm optimization method, using the best amplitude value output from the amplitude PSO means as a starting point for the search phase value searched by the phase linear search means Calculate the PAPR cost function calculated based on the phase value of the subcarrier carrying the preamble signal determined by , Contain and outputting a phase value that minimizes the PAPR cost function as a phase value of subcarriers carrying the best preamble signal is desirable.

  Further, in the preamble signal program of the present invention, the computer further uses the best amplitude value output from the amplitude PSO means, using a preset initial phase value as a starting point, by a linear search method, Phase linear search means for calculating a PAPR cost function calculated based on a phase value of a subcarrier carrying a preamble signal for each predetermined step size and outputting a phase value that minimizes the PAPR cost function as a search phase value The subcarrier carrying the preamble signal determined by the particle swarm optimization method using the best amplitude value output from the amplitude PSO means as the starting point and the search phase value searched by the phase linear search means And calculate a PAPR cost function calculated based on the phase value and minimize the PAPR cost function That the function as a phase PSO means for outputting a phase value of subcarriers phase values to convey the best preamble signal is desirable.

  When the amplitude value is determined by the amplitude linear search means and the amplitude PSO means, the phase value is determined by the phase linear search means and the phase PSO means in the same manner as when the amplitude width is determined. The autocorrelation depends only on the amplitude value, but since the peak-to-average power ratio (hereinafter referred to as PAPR) depends on both the amplitude value and the phase value, only the amplitude value is calculated. Instead, by determining the phase value by the phase linear search means and the phase PSO means, it is possible to generate a preamble signal that can reduce PAPR.

  If a recording medium storing the preamble signal generated by the preamble signal generation apparatus of the present invention is mounted on a wireless communication apparatus, the preamble signal is read out from the recording medium and transmitted as a training signal, and communication with good autocorrelation is achieved. Is possible.

  According to the present invention, it is possible to obtain a preamble signal approximate to an optimum value with good autocorrelation and capable of reducing PAPR by a short-time search.

It is a block diagram which shows the preamble signal generation apparatus which concerns on embodiment of this invention. 3 is a flowchart for explaining an operation in which an amplitude linear search unit of the preamble signal generation device shown in FIG. 1 searches for an amplitude value of a subcarrier signal by a linear search method. It is a table | surface for demonstrating the step size at the time of an amplitude linear search means searching by the linear search method. 6 is a flowchart for explaining an operation of the amplitude PSO unit of the preamble signal generation device shown in FIG. 1 searching for an amplitude value of a subcarrier signal by the PSO method. 3 is a flowchart for explaining an operation in which the phase linear search unit of the preamble signal generation device shown in FIG. 1 searches for a phase value of a subcarrier signal by a linear search method. It is a table | surface for demonstrating the step size at the time of a phase linear search means searching by the linear search method. 3 is a flowchart for explaining an operation in which the phase PSO unit of the preamble signal generation device shown in FIG. 1 searches for a phase value of a subcarrier signal by the PSO method. It is a figure which shows the state which the radio | wireless communication apparatus which concerns on embodiment of this invention is communicating with another radio | wireless communication apparatus. It is a figure which shows the structure of the radio | wireless transmitter shown in FIG. It is a figure which shows an example of a structure of the OFDM signal generation means shown in FIG. It is a figure which shows the packet format of the communication packet which the radio | wireless transmitter shown in FIG. 8 transmits. It is a table | surface for demonstrating the case where a null subcarrier is set when an amplitude linear search means searches by the linear search method. It is a table | surface for demonstrating the case where a null subcarrier is set when a phase linear search means searches with a linear search method.

A preamble signal generation apparatus according to an embodiment of the present invention will be described with reference to the drawings.
A preamble signal generation device 1 shown in FIG. 1 has a function of generating a preamble signal when communicating by the OFDM modulation method. The preamble signal generation device 1 functions by causing a computer to operate a preamble signal generation program. In FIG. 1, external interface means for connecting display means and other peripheral devices are not shown.

  The preamble signal generation device 1 includes an input unit 2, a storage unit 3, an amplitude value calculation unit 4, and a phase value search unit 5. The input means 2 can be a keyboard, a mouse, or the like, and is used when an initial value is set or a search is started or ended. The storage unit 3 can store various settings such as initial values instead of the input unit 2.

The amplitude value calculation means 4 searches for an amplitude value with good autocorrelation. The amplitude value calculating means 4 includes an amplitude linear search means 41 and an amplitude PSO means 42.
The amplitude linear search means 41 searches for an amplitude value that provides a good autocorrelation by the search using the linear search method. The amplitude linear search unit 41 includes an IDFT calculation unit 411, an autocorrelation calculation unit 412, a cost calculation unit 413, and a determination processing unit 414.

  The IDFT calculation means 411 performs a discrete inverse Fourier transform on the input preamble signal being searched for every predetermined step size. The autocorrelation calculating means 412 calculates autocorrelation based on the preamble signal subjected to discrete inverse Fourier transform. The cost calculation means 413 calculates a synchronous cost function. The determination processing unit 414 repeats until the search for all the step sizes is completed while determining whether or not the synchronization cost function calculated by the cost calculation unit 413 has the maximum amplitude.

  The amplitude PSO means 42 searches for an amplitude value with good autocorrelation by the PSO method using the amplitude value searched by the amplitude linear search means 41 as a starting point. The amplitude PSO unit 42 includes an IDFT calculation unit 421, an autocorrelation calculation unit 422, a cost calculation unit 423, and a determination processing unit 424.

  The IDFT calculation means 421 performs a discrete inverse Fourier transform on the preamble signal being searched input from the amplitude linear search means 41. The autocorrelation calculating means 422 calculates autocorrelation based on the preamble signal subjected to discrete inverse Fourier transform. The cost calculation means 423 calculates a synchronous cost function. The determination processing unit 424 repeats until the search is completed while determining whether the synchronization cost function calculated by the cost calculation unit 413 is the maximum amplitude value based on the personal best and the global best, and the best at the end of the search. Is output as a preamble signal of the amplitude value.

The phase value search means 5 searches for a phase value that can reduce PAPR. The phase value search unit 5 includes a phase linear search unit 51 and a phase PSO unit 52.
The phase linear search means 51 searches for a phase value with a good autocorrelation based on the search by the linear search method based on the good amplitude value output from the amplitude value search means 4. The phase linear search unit 51 includes an IDFT calculation unit 511, a cost calculation unit 512, and a determination processing unit 513.

  The IDFT calculation means 511 performs discrete inverse Fourier transform on the input preamble signal under search for each predetermined step size. The cost calculation means 512 calculates a PAPR cost function based on the preamble signal subjected to discrete inverse Fourier transform. The determination processing unit 513 repeats until the search for all the step sizes is completed while determining whether or not the PAPR cost function calculated by the cost calculation unit 512 is the minimum phase value.

  The phase PSO means 52 searches for a phase value with good autocorrelation by a search by the PSO method using the phase value searched by the phase linear search means 51 as a starting point. The phase PSO unit 52 includes an IDFT calculation unit 521, a cost calculation unit 522, and a determination processing unit 523.

  The IDFT calculation unit 521 performs a discrete inverse Fourier transform on the preamble signal being searched input from the phase linear search unit 51. The cost calculation means 522 calculates a PAPR cost function based on the preamble signal subjected to discrete inverse Fourier transform. The determination processing unit 523 repeats until the search is completed while determining whether the PAPR cost function calculated by the cost calculation unit 522 is the phase value that minimizes the personal best and the global best.

The preamble signal generation method according to the present embodiment will be described by describing the operation of the preamble signal generation apparatus according to the present embodiment configured as described above with reference to the drawings.
As shown in FIG. 2, first, initial values are set. This initial value is input by the input means 2 or stored in advance in the storage means 3 (step S100).

The initial value is i = 1 for the number of repetitions of the search, and the amplitude value of the subcarrier signal is the matrix H = [H ₀ H ₁ ... H _N−2 H _N−1 ] = [1 / T 1 / T. 1 / T] (initial amplitude value). Here, N is the number of subcarriers, and T is the step size.
When this initial value is input from the input means 2 or the storage means 3, the amplitude value search means 4 searches for the amplitude value. The search for the amplitude value by the amplitude value search means 4 is first performed by the linear search method by the amplitude linear search means 41. However, the above is an example in the case where all subcarriers are used. When null subcarriers are set, H _K (0 ≦ k <N) at the position where null subcarriers are set is null, that is, 0. Fix it.

When the initial value is input, the discrete inverse Fourier transform is performed based on the equation (1) by the IDFT calculation means 411 of the amplitude linear search means 41 in order to obtain a modulation signal in OFDM (step S110).

但し、Ｘ_nは、式（３）に示される範囲である。

Next, the autocorrelation calculation means 412 calculates autocorrelation based on equation (2) (step S120).

However, _Xn is the range shown by Formula (3).

Then, based on the autocorrelation [rho _tau calculated, the calculation of the synchronization cost function J _ACF is performed based on equation (4) the cost calculating unit 413 (step S130). Here, max ₁ (ρ _τ ) in Equation (4) means that ρ _τ is maximum in the range of −N ≦ τ <N. Moreover, max ₂ (ρ _τ) means that the maximum and becomes [rho _tau second in a range of -N ≦ τ <N.

The determination processing means 414 compares the calculated current synchronization cost function J _ACF with the maximum synchronization cost value J ^MAX _ACF indicating the maximum among the calculated synchronization cost functions J _ACF , and determines the synchronization cost. It is determined whether or not the function J _ACF is larger than the synchronization cost maximum value J ^MAX _ACF (step S140).

If the synchronization cost function _JACF is larger than the synchronization cost maximum value ^JMAX _ACF , the determination processing means 414 updates the synchronization cost maximum value by substituting the synchronization cost function _JACF for the synchronization cost maximum value ^JMAX _ACF , The matrix H that is the amplitude value at this time is substituted into the matrix H ^MAX _ACF that is the amplitude value of the subcarrier signal that maximizes the synchronization cost function J _ACF to update the amplitude value (step S150).
If the initial value of the synchronization cost maximum value J ^MAX _ACF is "0", initially calculated synchronization cost function J _ACF is which is always a positive number, the synchronization cost function than the synchronization cost maximum value J ^MAX _ACF J _ACF Is bigger. Therefore, the synchronization cost function J _ACF is substituted into the maximum synchronization cost value J ^MAX _ACF , and the matrix H = [1 / T 1 / T... 1 / T 1 / T] is substituted into the matrix H ^MAX _AFC .

If the synchronization cost function J _ACF is equal to or smaller than the maximum synchronization cost value J ^MAX _{ACF in} step S140, step S150 is not executed. Therefore, the maximum synchronization cost function J _ACF is assigned to the maximum synchronization cost value J ^MAX _ACF , and the matrix H having the maximum amplitude value is assigned to the matrix H ^MAX _ACF. Will be.

Whether or not the end condition i = ^TN is satisfied in order to determine the end of the search for all patterns according to the step size T shown in the table of FIG. 3 for the matrix H that is the amplitude value of the subcarrier signal. Is determined (step S160). The search pattern shown in the table of FIG. 3 is an example in the case of using all subcarriers. When a null subcarrier is set, the subcarrier is null, that is, the amplitude value is fixed to 0. The amplitude values of subcarriers other than the null subcarrier are all patterns according to the step size T. In that case, the termination condition of step S160 is i = ^TL . However, L is the number obtained by subtracting the number of null subcarriers from the number of subcarriers N.
If not completed, the determination processing unit 414 updates the amplitude value of the subcarrier signal to the next step size T and notifies the IDFT calculation unit 411 to perform discrete inverse Fourier transform with the new amplitude value (step S170). ).

  Thus, the subcarrier signal of the best amplitude value (search amplitude value) when the step size T is obtained by the search by the linear search method by completing all search patterns in the amplitude linear search means 41. Can be obtained.

Next, the amplitude PSO means 42 searches by the PSO method. As an initial setting, a matrix H ^MAX _ACF as the amplitude value of the subcarrier signal extracted by the amplitude linear search means 41 is set in the matrix H _Gbest as the global amplitude value which is the starting point in the PSO method, and the synchronization cost is set. The synchronization cost maximum value J ^MAX _ACF is set to the synchronization global maximum value J _Gbest of the function and is input to the IDFT calculation means 421. Further, as an initial setting, the matrix H ^m = [H ^m ₀ H ^m ₁ ... H ^m _N−2 H ^m _N−1 ] (1 ≦ m ≦ M) is randomly selected as amplitude values of M subcarrier signals. It is generated (step S200). However, when setting the null subcarrier, H ^m _k (0 ≦ k <N) at the position where the null subcarrier is set is fixed to null, that is, 0.

When the initial value is input, a discrete inverse Fourier transform is performed based on the equation (5) by the IDFT calculation means 421 of the amplitude PSO means 42 in order to obtain a modulation signal in OFDM (step S210).

但し、Ｘ_nは、式（７）に示される範囲である。

Next, the autocorrelation calculation means 422 calculates autocorrelation based on equation (6) (step S220).

However, _Xn is the range shown by Formula (7).

Next, based on the autocorrelation calculated by the equation (6), the cost calculation means 413 calculates M synchronization cost functions based on the equation (8) (step S230).

The determination processing unit 424 is configured to calculate the current M number of synchronization cost functions J ^m _ACF and the maximum synchronization personal value J ^m _Pbest (personal best) indicating the maximum among the calculated synchronization cost functions J ^m _ACF. ) To determine whether or not the synchronization cost function J ^m _ACF is larger than the synchronization personal maximum value J ^m _Pbest (step S240).

If the synchronization cost function J ^m _ACF is larger than the synchronization personal maximum value J ^m _Pbest , the determination processing unit 424 updates the synchronization personal maximum value by substituting the synchronization cost function J ^m _ACF for the synchronization personal maximum value J ^m _Pbest. together, the matrix H ^m which is randomly generated, by substituting the matrix H ^m _Pbest showing the amplitude value of the sub-carrier signals to maximize the synchronization personal maximum value J ^m _Pbest, updates the amplitude value (step S250).

If the initial value of the synchronous personal maximum value J ^m _Pbest is "0", initially calculated synchronization cost function J ^m _ACF is constantly a positive number, the synchronization personal maximum value J ^m _Pbest from the synchronization cost function J ^m _ACF is larger. Therefore, the synchronization cost function J ^m _ACF is substituted into the synchronization personal maximum value J ^m _Pbest , and the randomly generated matrix H ^m = [H ^m ₀ H ^m ₁ ... H ^m _N−2 H ^m _N−1 ] is a matrix. _Assigned to H ^m _Pbest .

If the synchronization cost function J ^m _ACF is equal to or smaller than the synchronization personal maximum value J ^m _{Pbest in} step S240, step S250 is not executed. Thus, eventually, will be synchronized personal maximum value J ^m _Pbest synchronization cost function J ^m _ACF becomes maximum in is substituted, the matrix H ^m as a maximum amplitude value is substituted into the matrix H ^m _Pbest Will be.

Next, the determination processing means 424 determines the maximum synchronization personal value J ^Z _Pbest and the synchronization global maximum value J _Gbest (global best) among the M synchronization personal maximum values J ^m _Pbest (1 ≦ m ≦ M). Compare Here, the synchronous personal maximum value J ^Z _Pbest is obtained from equation (9) (step S260).

Synchronization personal maximum value J ^Z _Pbest synchronized result of the comparison between the global maximum value J _gbest, if the relationship of the sync personal maximum value J ^Z _Pbest> synchronous global maximum value J _gbest, determination processing unit 424, synchronous personal maximum value _Substituting J ^Z _Pbest into the synchronous global maximum value J _Gbest to update the synchronization cost function, and substituting the matrix H ^Z _best as the amplitude value at this time into the matrix H _Gbest to update the amplitude value (step) S270).
As a result of step S260, if the relationship of synchronized personal maximum value J ^Z _Pbest ≦ synchronized global maximum value J _Gbest is satisfied, step S270 is not executed.

ここで、ｒ１，ｒ２は一様分布なランダムな値である。また、ｃ１，ｃ２はそれぞれ０．５である。 The determination processing unit 424 compares the synchronous global maximum value J _Gbest with a predetermined threshold value (J _DESIRED ), and determines J _Gbest > J _DESIRED that is an end condition (step S280).
If J _Gbest ≦ J _{DESIRED as} a result of step S280, the determination processing unit 424 updates the matrix H ^m that is the amplitude value of the subcarrier signal based on the equation (10), and is discrete with the new amplitude value. The IDFT calculation means 421 is notified to perform the inverse Fourier transform (step S290).

Here, r1 and r2 are uniformly distributed random values. Moreover, c1 and c2 are each 0.5.

If the termination condition is satisfied in step S280, normalization is performed by the determination processing unit 424 based on the equation (11), and a matrix G indicating an amplitude value with good autocorrelation is output (step S295). .

  In this way, the amplitude linear search means 41 searches for an amplitude value that provides a good autocorrelation in advance by the linear search method, and then the amplitude PSO means 42 determines a starting point for searching by the PSO method. Even if the step size T by the straight line search method is roughened to reduce the search time and the accuracy is lowered, the starting point in the PSO method having a relatively short time can be searched in advance from a good amplitude value. A preamble signal with good autocorrelation can be obtained in a shorter time than when the entire search is performed by the linear search method.

  Since the amplitude value is determined by the amplitude linear search means 41 and the amplitude PSO means 42, the phase value is then determined by a search by the linear search method by the phase linear search means 51 and the PSO method by the phase PSO means 52. Thus, a preamble signal with good autocorrelation and low PAPR is obtained.

First, as an initial setting, a matrix G that is an amplitude value output from the amplitude value search means 4 is set as an initial value. Also, the number of search iterations is i = 1, and the phase value of the subcarrier signal is the matrix A = [a ₀ a ₁ ... A _N−2 a _N−1 ] = [1 / P 1 / P. Let P] be the initial value. Here, N is the number of subcarriers, and P is the step size. This initial value can be input from the input means 2 or the storage means 3 (step S300). However, the above is an example in which all subcarriers are used. When null subcarriers are set, a _k (0 ≦ k <N) at the position where null subcarriers are set is fixed to an arbitrary phase. To do.

When the initial value is input, the discrete inverse Fourier transform is performed based on the equation (12) by the IDFT calculation means 511 of the phase linear search means 51 in order to obtain the modulation signal in OFDM (step S310).

ここで、式（１３）の右辺であるｍａｘ₁（｜Ｘ_n｜²）は、０≦ｎ＜Ｎの範囲で最大となる｜Ｘ_n｜²を意味する。 Next, the PAPR cost function is calculated by the cost calculation means 512 based on the equation (13) (step S320).

Here, max ₁ (| X _n | ² ), which is the right side of Expression (13), means | X _n | ² that is maximum in the range of 0 ≦ n <N.

The determination processing unit 513 compares the calculated PAPR cost function J _PAPR with the PAPR cost minimum value J ^MIN _PAPR indicating that it is the smallest among the calculated PAPR cost functions J _{PAPR. It} is determined whether _{PAPR is} smaller than PAPR cost minimum value ^MIN _PAPR (step S330).

If the PAPR cost function J _PAPR is smaller than the PAPR cost minimum value ^MIN _PAPR , the determination processing unit 523 updates the PAPR cost minimum value by substituting the PAPR cost function J _PAPR into the PAPR cost minimum value J ^MIN _PAPR. The matrix A that is the current phase value is substituted into the matrix A ^MIN _PAPR that is the phase value of the subcarrier signal that minimizes the PAPR cost function J _PAPR to update the phase value (step S340).
When the initial PAPR cost minimum value J ^MIN _{PAPR is set} to a sufficiently large value (for example, 100 or more), the PAPR cost function J _PAPR calculated first is smaller than the PAPR cost minimum value J ^MIN _PAPR . Accordingly, the PAPR cost function J _PAPR is substituted into the PAPR cost minimum value J ^MIN _PAPR , and the matrix A = [1 / P 1 / P... 1 / P 1 / P] is substituted into the matrix A ^MIN _PAPR .
If the PAPR cost function J _PAPR is greater than or equal to the PAPR cost minimum value J ^MIN _{PAPR in} step S330, step S340 is not executed. Therefore, finally, the minimum PAPR cost function J _PAPR is assigned to the minimum PAPR cost value J ^MIN _PAPR , and the matrix A that is the minimum phase value is assigned to the matrix A ^MIN _PAPR. Will be.

Whether or not the end condition i = ^PN is satisfied in order to determine the end of the search for all patterns according to the step size P shown in the table of FIG. 6 for the matrix A that is the phase value of the subcarrier signal. Is determined (step S350). The search pattern shown in the table of FIG. 6 is an example in the case of using all subcarriers. When a null subcarrier is set, the subcarrier is fixed to an arbitrary phase, so that it is not a null subcarrier. The subcarrier phase values are all patterns according to the step size P. In that case, the termination condition in step S160 becomes i = P ^L. However, L is the number obtained by subtracting the number of null subcarriers from the number of subcarriers N.
If not completed, the determination processing unit 513 updates the phase value of the subcarrier signal to the next step size P and notifies the IDFT calculation unit 511 to perform discrete inverse Fourier transform with the new phase value (step S360). ).

  Thus, the subcarrier signal of the best phase value (search phase value) in the case where the step size P is obtained by the search by the linear search method by completing all search patterns in the phase linear search means 51. Can be obtained.

Next, the phase PSO means 52 searches by the PSO method. As an initial setting, a matrix A ^MIN _PAPR as a phase value of a subcarrier signal extracted by the phase linear search means 51 is set in a matrix A _Gbest as a global phase value that is a starting point in the PSO method, and a PAPR cost is set. The PAPR cost minimum value J ^MIN _PAPR is set in the PAPR global minimum value J _Gbest of the function and is input to the IDFT calculation means 521. Further, as the initial setting, further matrix _{^{A m = [a m 0 a}} m 1 ... a m N-2 a m N-1] (1 ≦ m ≦ M) is random as the phase value of the M subcarrier signals It is generated (step S400). However, when setting the null subcarrier, a _k (0 ≦ k <N) at the position where the null subcarrier is set is fixed to an arbitrary phase.

When the initial value is input, the discrete inverse Fourier transform is performed based on the equation (14) by the IDFT calculation means 521 of the phase PSO means 52 in order to obtain a modulation signal in OFDM (step S410).

Next, M PAPR cost functions are calculated based on the equation (15) by the cost calculating means 522 (step S430).

The determination processing means 523 includes the calculated current M PAPR cost functions J ^m _PAPR and the personal minimum value J ^m _Pbest (personal best) indicating the smallest among the calculated PAPR cost functions J ^m _PAPR. To determine whether the PAPR cost function J ^m _PAPR is smaller than the PAPR personal minimum value J ^m _Pbest (step S430).

If the PAPR cost function J ^m _PAPR is smaller than the PAPR personal minimum value J ^m _Pbest , the determination processing means 523 updates the PAPR personal minimum value by substituting the PAPR cost function J ^m _PAPR for the PAPR personal cost function J ^m _Pbest. together, they are substituted into the matrix a ^m _Pbest showing the phase value of the subcarrier signals to minimize the PAPR personal minimum J ^m _Pbest the matrix a ^m that is randomly generated, and updates the phase value (step S440).

_If the initial value of the PAPR personal minimum value J ^m _Pbest is a sufficiently large value (for example, 100 or more), the PAPR cost function J ^m _PAPR calculated first is smaller than the PAPR personal minimum value J ^m _Pbest . Therefore, PAPR cost function J ^m _PAPR is assigned to PAPR personal minimum J ^m _Pbest, randomly generated matrix _{^{A m = [a m 0 a}} m 1 ... a m N-2 a m N-1] is a matrix _Assigned to A ^m _Pbest .
If the PAPR cost function J ^m _PAPR is greater than or equal to the PAPR personal minimum value J ^m _Pbest in step S430, step S440 is not executed. Thus, finally, the PAPR personal minimum J ^m _Pbest will be PAPR cost function J ^m _PAPR having the smallest is substituted, the matrix A ^m _Pbest a minimum phase value for the matrix A ^m assignment Will be.

Next, the determination processing means 523 determines the PAPR personal minimum value J ^Z _Pbest and PAPR global minimum value J _Gbest (global best) that are the smallest among the M PAPR personal minimum values J ^m _Pbest (1 ≦ m ≦ M). Compare Here, the PAPR personal minimum value J ^Z _Pbest is obtained from equation (9) (step S450).

PAPR personal minimum J ^Z _Pbest and PAPR global minimum J _gbest and a result of comparison, if the relationship between the PAPR personal minimum J ^Z _Pbest <PAPR global minimum J _gbest, determination processing unit 523, PAPR personal minimum by substituting J ^Z _Pbest the PAPR global minimum J _gbest, updates the PAPR global minimum, by substituting the matrix a ^Z _Pbest a phase value at that time into the matrix a _gbest, updates the phase value ( Step S460).
If the result of step S450 is PAPR personal minimum value J ^Z _Pbest ≧ PAPR global minimum value J _Gbest , step S460 is not executed.

ここで、ｒ１，ｒ２は一様分布なランダムな値である。また、ｃ１，ｃ２はそれぞれ０．５である。 The determination processing unit 523 compares the PAPR global minimum value J _Gbest with a predetermined threshold value (J _DESIRED ), and determines J _Gbest <J _DESIRED as an end condition (step S470).
Result of the step S470, if J _Gbest ≧ J _DESIRED, determination processing unit 523, a matrix A ^m is the phase value of the subcarrier signal, and updated based on equation (16), discrete in the new phase value The IDFT calculation means 521 is notified to perform the inverse Fourier transform (step S480).

Here, r1 and r2 are uniformly distributed random values. Moreover, c1 and c2 are each 0.5.

If the end condition is satisfied in step S470, the determination processing unit 523 ends the process. As a result, the determination processing means 523 outputs the A _Gbest preamble signal whose amplitude value is finally obtained in step S460 with the matrix G obtained in step S295 (see equation (11)). .

  Thus, autocorrelation depends only on the amplitude value, but PAPR depends on both the amplitude value and the phase value. Therefore, the phase linear search means 51 is based on the amplitude value searched by the amplitude value search means 4. And the phase PSO means 52 can determine the phase value. Therefore, it is possible to obtain a preamble signal approximate to an optimum value with good autocorrelation and capable of reducing PAPR by a short search.

  In addition, since the preamble signal generation device 1 can arbitrarily set and search for null subcarriers, out-of-band radiation can be generated by using the preamble signal generated by the preamble signal generation device 1 as a training signal in wireless communication. It can be suppressed and interference with other communication signals can be prevented.

The preamble signal obtained by the preamble signal generation device 1 can be stored in a recording medium. The recording medium only needs to be non-volatile and readable, and electrical, optical, and magnetic writing and reading are not limited.
For example, it is possible to rewrite a flexible disk, magneto-optical disk, CD (-ROM, -R, -R / W), DVD (-ROM, -R, -R / W, -RAM, etc.) that can be read from a computer. DAT, 8mm tape, memory card, hard disk, etc. When incorporated in a wireless communication device, there are ROM, flash, EPROM, E ² PROM, and the like. In addition, a preamble signal can be incorporated together with a control circuit incorporated in an LSI or the like.

Here, the radio communication apparatus according to the present embodiment incorporating a recording medium in which a preamble signal is stored will be described with reference to FIGS.
As shown in FIG. 8, a radio communication apparatus a according to the present embodiment includes a radio transmission apparatus 10 and a radio reception apparatus 10x, and a plurality of antennas by another radio communication apparatus b and MIMO-OFDM modulation. Packet communication is performed via In the present embodiment, the wireless communication device a and another wireless communication device b communicate with each other by four antennas A11 to A13 and A21 to A24. FIG. 8 shows an example in which the number of antennas on the transmission side, the number of antennas on the reception side, the number of spatial streams, and the number of FEC encoders used are four.

The wireless communication device “a” has a transmission bandwidth of 80 MHz and a maximum transmission rate of 1 Gbps or more using a 5 GHz band while maintaining backward compatibility with the IEEE 802.11a standard and the IEEE 802.11n standard. is there.
Since the present invention relates to the preamble signal and the packet format of the communication packet transmitted by the wireless transmission device 1, the description of the wireless reception device 10x is omitted.

  Next, the configuration of the wireless transmission device 10 will be described with reference to FIG. As shown in FIG. 9, the wireless transmission device 10 is formed by a control unit 10a and an OFDM signal generation unit 10b. The control unit 10a has a function of controlling the OFDM signal generation unit 10b to generate communication packets in a predetermined order, or to generate transmission data and output it to the OFDM signal generation unit 10b.

  The OFDM signal generation unit 10b has a function of transmitting a preamble signal and transmission data from the control unit 10a by OFDM modulation. As shown in FIG. 10, the OFDM signal generation means 10b includes a preamble pilot memory section 11, a scrambler section 12, an encoder parser section 13, an FEC (Forward Error). Correction) Encoder unit (FEC encoder) 14, Spatial stream parser unit 15, Interleaver unit 16, Mapper unit 17, STBC (Space-Time Block Code) encoder unit (STBC encoder) 18.

  Further, the OFDM signal generation means 10b includes a multiplexer unit (Mux) 19, a first CS (Cyclic Shift) inserter unit (CS inserter) 20, a spatial mapper unit (Spatial mapper) 21, and a PTS unit (Partial Transmit Sequence) 22. Inverse Fast Fourier Transform (IFFT) 23, second CS inserter section (CS inserter) 24, GI (Guard interval) inserter section (GI inserter) 25, and windowing section (Windowing) 26 And a transmission filtering unit (TX filter) 27.

  The preamble pilot memory unit 11 is a memory in which a preamble signal transmitted prior to transmission information (Data) and a pilot signal inserted into the transmission information are stored. The preamble pilot memory unit 11 outputs a preamble signal instructed by the control means 10a. The ROM storing the preamble signal obtained by the preamble signal generation device 1 is built in the wireless transmission device 10 as the preamble pilot memory unit 11.

The scrambler unit 12 randomizes the bits “0” and “1” using a generator polynomial of G (X) = X ⁷ + X ⁴ +1 for the transmission bit sequence from the control means, ”And“ 1 ”should not appear continuously.
The encoder parser unit 13 divides the scrambled transmission information into four FEC encoder units 14 and distributes them to the FEC encoder unit 14.
The FEC encoder unit 14 performs forward error correction encoding to enable error correction. The FEC encoder unit 14 performs encoding by BCC (Binary Convolutional Codes) calculated using a generator polynomial of rate 1/2 G ₀ = 133 ₈ and G ₁ = 171 ₈ .

The spatial stream parser unit 15 distributes transmission information to a spatial stream used in MIMO-OFDM modulation by decomposing the encoded data from the FEC encoder unit 14 into four blocks.
The interleaver unit 16 rearranges the arrangement of the transmission information distributed by the spatial stream parser unit 15 according to a predetermined rule.
The mapper unit 17 transmits BPSK (Binary Phase Shift Keying) modulation, QPSK (Quadrature Phase Shift Keying) modulation, and QAM (Quadrature Amplitude Modulation) modulation in order to transmit transmission information that is interleaved bit information as a radio signal. For example, bit information is mapped to symbols.

The STBC encoder unit 18 uses the spatio-temporal block code (STBC) to change the arrangement points from the spatial stream (Spatial Stream: SS) to the spatio-temporal stream (Space-Time Stream: STS) for the mapped transmission information (symbol). ), It is possible to obtain a greater diversity gain.
The multiplexer unit 19 selects transmission information from the STBC encoder unit 18 and a preamble signal or pilot signal from the preamble pilot memory unit 11.

The first CS inserter unit 20 adds a cyclic shift to the transmission information (symbol). By adding a cyclic shift to the transmission information, unintended beam forming of the radio signal is prevented.
The spatial mapper unit 21 maps the transmission information (symbol) to which the cyclic shift is added according to the spatial mapping matrix.
The PTS unit 22 performs different phase rotation for each of the first to fourth bands to reduce PAPR.

The inverse fast Fourier transformer 23 performs OFDM modulation by performing inverse Fourier transform (IFFT) on the signal from the PTS unit 22.
Similar to the first CS inserter unit 20, the second CS inserter unit 24 adds a cyclic shift to the transmission information (symbol). The second CS inserter unit 24 is used exclusively with the first CS inserter unit 20.
The GI inserter unit 25 adds a guard interval to the OFDM-modulated symbol output from the second CS inserter unit 24.
Since the out-of-band spectrum increases when the signal periodicity changes greatly between adjacent symbols, the windowing unit 26 increases the attenuation of the spectrum by performing smoothing so that the change between adjacent symbols is small. And suppress the out-of-band spectrum. The transmission filtering unit 27 performs band limitation on the transmission signal from the windowing unit 26.

  Here, the communication packet transmitted by the wireless transmission device 10 will be described with reference to FIG. In this specification, the packet format of a communication packet transmitted by the wireless transmission device 1 according to the present embodiment is referred to as a “VHT (Very High Throughput) Mixed Mode” packet format, and is hereinafter simply abbreviated as a VTH packet format. .

  As shown in FIG. 11, the VHT packet format includes preamble formats of L-STF (Legacy-Short Training Field: 8 μS), L-LTF (Legacy-Long Training Field: 8 μS), and L-SIG (Legacy-Signal Filed: 4 μS), HT (High Throughput) −SIG (8 μS), and examples of packet formats following HT-SIG include VHT-SIG (8 μS), VHT-STF (4 μS), VHT-LTF 1 to 4 (4 μS). , Data (4 μS / 3.6 μS).

  L-STF, L-LTF, and L-SIG correspond to IEEE802.11a STF, LTF, and SIG, and also correspond to IEEE802.11n L-STF, L-LTF, and L-SIG. HT-SIG corresponds to HT-SIG of IEEE 802.11n. Preamble signals from L-STF to VHT-LTF 1 to 4 in these VHT packet formats are stored in the preamble pilot memory unit 11 as subcarrier information.

  In particular, by using the preamble signal generated by the preamble signal generation device 1 as the preamble signal of VHT-STF and VHT-LTF1 to 4, the wireless transmission device 10 is excellent in autocorrelation and capable of low PAPR. be able to.

Example 1
The number of subcarrier signals is N = 16, of which the number of null subcarriers is 4. H _k (k = 0, k = 1, k = 14, k = 15) at the position where the null subcarrier is set is fixed to null, that is, 0. After searching with the step size in the amplitude linear search means 41 as T = 5, the search in the amplitude PSO means 42 was performed. As a comparative example, the search was performed by changing the step size T from 1 to 9 only by the amplitude linear search means 41.
The initial value is i = 1 for the number of repetitions of the search, and the amplitude value of the subcarrier signal is the matrix H = [H ₀ H ₁ H ₂ H ₃ ... H ₁₂ H ₁₃ H ₁₄ H ₁₅ ] = [0 0 1 / T 1 / T ... 1 / T 1 / T 0 0] (initial amplitude value). However, T is a step size.
When this initial value is input from the input means 2 or the storage means 3, the amplitude value search means 4 searches for the amplitude value. The search for the amplitude value by the amplitude value search means 4 is first performed by the linear search method by the amplitude linear search means 41. The search pattern at this time is the search pattern according to FIG. 12 in which H _k (k = 0, k = 1, k = 14, k = 15) at the position where the null subcarrier is set is fixed to null, that is, 0. Become. The search termination condition is i = ^TL . However, L is the number obtained by subtracting the number of null subcarriers 4 from the number of subcarriers N = 16, that is, 12.

  First, Table 1 shows the results of the maximum synchronization cost function and search time when the step size T is changed from 1 to 9 only by the amplitude linear search means 41.

  From this table (1), it can be seen that, for example, in the search at the step size T = 7 by the linear search method, the maximum synchronization cost function is 30.8 and the search time is 23.8 hours.

  Therefore, after the amplitude linear search means 41 searches for the step size as T = 5, the amplitude PSO means 42 searches by the PSO method. The target value was set to 30.8, which is equivalent to T = 7 in the linear search method alone. As a result, when the search time is 23.8 hours only with the linear search method, the search is performed by the PSO method with the amplitude PSO means 42 using the result of the amplitude linear search means 41 as a starting point, so that the step size is increased by the linear search method The search could be completed with a search time of 25 minutes and a search time of 27.3 seconds at T = 5.

(Example 2)
Next, the number of subcarrier signals is N = 16, of which the number of null subcarriers is 4, and _ak (k = 0, k = 1, k = 14, k = 15) where null subcarriers are set Is fixed to an arbitrary phase, this time 0. As the amplitude of the subcarrier, the amplitude value G obtained as a result of searching with the amplitude PSO means 42 after searching with the step size in the amplitude linear search means 41 of Example 1 as T = 5 was used. After searching with the step size in the phase linear search means 51 as P = 5, the search in the phase PSO means 52 was performed. As a comparative example, the search was performed by changing the step size P from 1 to 9 only by the phase linear search means 51.
The initial value is i = 1 for the number of repetitions of the search, and the phase value of the subcarrier signal is the matrix A = [a ₀ a a ₂ a ₃ ... A ₁₂ a ₁₃ a ₁₄ a ₁₅ ] = [0 0 1 / P 1 / P... 1 / P 1 / P 0 0] (initial phase value). However, P is a step size.
When this initial value is input from the input means 2 or the storage means 3, the phase value search means 5 searches for the phase value. The search for the phase value by the phase value search means 5 is first performed by the linear search method by the phase linear search means 51. The search pattern at this time is a search pattern according to FIG. 13 in which a _k (k = 0, k = 1, k = 14, k = 15) at the position where the null subcarrier is set is fixed to 0. In addition, the termination condition of the search will be i = P ^L. However, L is the number obtained by subtracting the number of null subcarriers 4 from the number of subcarriers N = 16, that is, 12.
First, Table (2) shows the minimum PAPR cost function and the search time result when the step size P is changed from 1 to 9 only by the phase linear search means 51.

  From this table (1), it can be seen that, for example, in the search at the step size P = 9 by the linear search method, the minimum PAPR cost function is 0.754 dB and the search time is 5.5 days.

  Therefore, after the phase linear search means 51 searches for the step size as P = 5, the phase PSO means 52 searches by the PSO method. The target value was set to 0.75 dB equivalent to P = 9 in the linear search method alone. As a result, it takes 5.5 days for the search time with the linear search method alone, but by performing the search with the PSO method in the phase PSO means 52 using the result of the phase linear search means 51 as a starting point, the step size is increased with the linear search method The search was completed at a search time of 6.9 minutes at P = 5 and a search time of 114 seconds.

  From Example 1 and Example 2, it can be seen that searching by the PSO method after searching by the linear search method can be performed in a shorter time than by searching only by the linear search method.

  The present invention can be widely applied not only to a preamble signal in IEEE standard wireless communication but also to a preamble signal used in OFDM modulated wireless communication.

DESCRIPTION OF SYMBOLS 1 Preamble signal generator 2 Input means 3 Storage means 4 Amplitude value search means 41 Amplitude linear search means 411 IDFT calculation means 412 Autocorrelation calculation means 413 Cost calculation means 414 Judgment processing means 42 Amplitude PSO means 421 IDFT calculation means 422 Autocorrelation calculation Means 423 Cost calculation means 424 Determination processing means 5 Phase value search means 51 Phase linear search means 511 IDFT calculation means 512 Cost calculation means 513 Determination processing means 52 Phase PSO means 521 IDFT calculation means 522 Cost calculation means 523 Determination processing means a, b Wireless communication device 10 Wireless transmission device 10x Wireless reception device 10a Control unit 10b OFDM signal generation unit 11 Preamble pilot memory unit 12 Scrambler unit 13 Encoder parser unit 14 FEC Coder unit 15 Spatial stream parser unit 16 Interleaver unit 17 Mapper unit 18 STBC encoder unit 19 Multiplexer unit 20 First CS inserter unit 21 Spatial mapper unit 22 PTS unit 23 Inverse fast Fourier transformer 24 Second CS inserter unit 25 GI inserter unit 26 Window 27 Transmitting filtering unit A11-A14, A21-A24 Antenna

Claims (7)

A preamble signal generation device that generates a preamble signal by orthogonal frequency division multiplexing modulation,
Using a preset initial amplitude value as a starting point, a synchronization cost function calculated based on the amplitude value of a subcarrier carrying a preamble signal for each predetermined step size is calculated by a linear search method. An amplitude linear search means for outputting an amplitude value that maximizes as a search amplitude value;
Using the search amplitude value searched by the amplitude linear search means as a starting point, calculate a synchronization cost function calculated based on the amplitude value of the subcarrier carrying the preamble signal determined by the particle swarm optimization method, An amplitude PSO means for outputting an amplitude value that maximizes the synchronization cost function as an amplitude value of a subcarrier carrying the best preamble signal. Using the best amplitude value output from the amplitude PSO means, using the preset initial phase value as a starting point, the phase value of the subcarrier carrying the preamble signal for each predetermined step size is obtained by the linear search method. Phase linear search means for calculating a PAPR cost function calculated based on the phase and outputting a phase value that minimizes the PAPR cost function as a search phase value;
Using the best amplitude value output from the amplitude PSO means, starting from the search phase value searched by the phase linear search means, the subcarriers carrying the preamble signal determined by the particle swarm optimization method And a phase PSO means for calculating a PAPR cost function calculated based on the phase value and outputting a phase value that minimizes the PAPR cost function as a phase value of a subcarrier carrying the best preamble signal. The preamble signal generation device according to Item 1. A preamble signal generation method for generating a preamble signal by orthogonal frequency division multiplexing modulation,
Amplitude linear search means for searching by a linear search method calculates a synchronization cost function calculated based on the amplitude value of a subcarrier carrying a preamble signal at a predetermined step size starting from a preset initial amplitude value. Outputting an amplitude value that maximizes the synchronization cost function as a search amplitude value;
The amplitude PSO means for searching by the particle swarm optimization method is calculated based on the amplitude value of the subcarrier carrying the preamble signal determined by the search, starting from the search amplitude value searched by the amplitude linear search means. Calculating a synchronization cost function, and outputting an amplitude value that maximizes the synchronization cost function as an amplitude value of a subcarrier carrying the best preamble signal. The phase linear search means for searching by a straight line search method uses the best amplitude value output from the amplitude PSO means to carry a preamble signal for each predetermined step size starting from a preset initial phase value. Calculating a PAPR cost function calculated based on the phase value of the subcarrier to output, and outputting a phase value that minimizes the PAPR cost function as a search phase value;
The phase PSO means to be searched by the particle swarm optimization method is determined by searching using the best amplitude value output from the amplitude PSO means and using the search phase value searched by the phase linear search means as a starting point. PAPR cost function calculated based on the phase value of the subcarrier carrying the preamble signal to be calculated, and outputting the phase value that minimizes the PAPR cost function as the phase value of the subcarrier carrying the best preamble signal 4. A preamble signal generation method according to claim 3, further comprising the step of: A preamble signal generation program for causing a computer to generate a preamble signal by orthogonal frequency division multiplexing modulation,
The computer,
Using a preset initial amplitude value as a starting point, a synchronization cost function calculated based on the amplitude value of a subcarrier carrying a preamble signal for each predetermined step size is calculated by a linear search method. Amplitude linear search means for outputting an amplitude value that maximizes as a search amplitude value,
Using the search amplitude value searched by the amplitude linear search means as a starting point, calculate a synchronization cost function calculated based on the amplitude value of the subcarrier carrying the preamble signal determined by the particle swarm optimization method, A preamble signal generation program that functions as an amplitude PSO unit that outputs an amplitude value that maximizes a synchronization cost function as an amplitude value of a subcarrier carrying the best preamble signal. The computer,
Using the best amplitude value output from the amplitude PSO means, using the preset initial phase value as a starting point, the phase value of the subcarrier carrying the preamble signal for each predetermined step size is obtained by the linear search method. A phase linear search means for calculating a PAPR cost function calculated based on the phase and outputting a phase value that minimizes the PAPR cost function as a search phase value;
Using the best amplitude value output from the amplitude PSO means, starting from the search phase value searched by the phase linear search means, the subcarriers carrying the preamble signal determined by the particle swarm optimization method A PAPR cost function calculated based on the phase value is calculated, and a phase value that minimizes the PAPR cost function is output as a phase value of a subcarrier carrying the best preamble signal. 5. A preamble signal generation program according to 5. 3. A recording medium on which a preamble signal generated by the preamble signal generation apparatus according to claim 1 is stored.

Images