PROPAGATION CHANNEL ESTIMATION METHOD

Abstract

A propagation channel estimation method is provided. In-phase and quadrature components of a propagation channel estimation value in each pilot subcarrier are separated into amplitude and phase components. A propagation channel of a data subcarrier portion existing between pilot subcarriers is estimated by phase/amplitude separation linear interpolating in each separated amplitude and phase component. A reference parameter is produced by linear interpolating a portion between pilot subcarriers on a complex plane. When quadrants on the complex plane of the interpolated phase component estimation value and the reference parameter are different, a phase connecting process to cancel a discontinuity of the phases is executed to the phase component estimation value interpolated by the phase/amplitude separation linear interpolation. When the quadrants are not different or after the phase connecting process, a complex propagation channel estimation value of the data subcarrier portion is calculated from the phase and amplitude components of the propagation channel estimation value.

Claims

1. A propagation channel estimation method using pilot subcarriers inserted into an OFDM signal, comprising the steps of: separating an in-phase component and a quadrature component of a propagation channel estimation value in each of the pilot subcarriers into an amplitude component and a phase component and estimating a propagation channel of a data subcarrier portion existing between the pilot subcarriers by phase/amplitude separation linear interpolating in each of the separated amplitude component and phase component; producing a reference parameter for assisting the propagation channel estimation of the data subcarrier portion between the pilot subcarriers by linear interpolating a portion between the pilot subcarriers on a complex plane; discriminating whether or not quadrants on the complex plane of the phase component estimation value interpolated by the phase/amplitude separation linear interpolation and the reference parameter are different; if it is determined that the quadrants are different, executing a phase connecting process for cancelling a discontinuity of the phases to the phase component estimation value interpolated by the phase/amplitude separation linear interpolation; and if it is determined that the quadrants are not different or after the phase connecting process, calculating a complex propagation channel estimation value of the data subcarrier portion from the phase component and the amplitude component of the propagation channel estimation value of the interpolated data subcarrier portion between the pilot subcarriers.

2. A propagation channel estimation method according to claim 1, wherein the pilot subcarriers are arranged at predetermined intervals in each of a symbol domain and a subcarrier domain of the OFDM signal or are arranged at predetermined positions.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0022] FIG. 1 Schematic diagram for use in an explanation of an interpolation method.

[0023] FIG. 2 Flowchart for use in the explanation of the interpolation method.

DESCRIPTION OF EMBODIMENTS

[0024] Embodiments of the invention will be described hereinbelow. The embodiments which will be described hereinbelow are exemplary specific examples of the invention and various kinds of limitations which are technically preferred are added. However, it is assumed that the scope of the invention is not limited to those embodiments unless otherwise described to limit the invention in the following explanation.

[0025] As a construction of a receiver at the time of embodying the invention, a construction similar to that disclosed in foregoing NPL 3 is used. In order to demodulate a reception signal in the receiver, a pilot signal is inserted. For example, in DL, a pilot arrangement of IEEE 802. 16-2009[3] Model is a pilot arrangement of a structure in which a partial data subcarrier is sandwiched by four pilot signals, and in UL, it is a pilot arrangement of a tile structure in which all of the data subcarriers are sandwiched by four pilot signals. In DL, a pilot arrangement of Mode2 has such a form that a tile is constructed by 4 symbols and 4 subcarriers and pilot subcarriers are inserted to its four corners, In UL, it has such a form that pilot subcarriers are inserted to four corners of a tile of 7 symbols and 4 subcarriers. On the reception side, by obtaining a propagation channel estimation value of each of the known pilot subcarriers, a propagation channel of a data portion sandwiched by the pilot subcarriers is interpolation estimated.

[0026] Since the pilot subcarriers have been inserted in the signal like a foregoing example, a propagation channel fluctuation of the data subcarrier in which no pilot subcarrier is inserted is estimated. After that, by multiplying the reception signal every subcarrier by a reciprocal number of the estimated propagation channel fluctuation, an equalization is performed. As an actual interpolation method, an interpolation is performed every tile. Order of the interpolation may be set to either a subcarrier domain or an OFDM symbol domain and does not limit the invention.

[0027] An image of the interpolation method which is performed by the pilot subcarriers is shown in FIG. 1. In FIG. 1, an asterisk mark indicates a pilot subcarrier. A broken line indicates the linear interpolation method on a complex plane in the related art. A solid line indicates the phase/amplitude separation linear interpolation method according to the invention. Any one of a display of a form in which the propagation channel estimation value is separated into an in-phase component (Ich) and an quadrature component (Qch) and a complex number display in which a value obtained by multiplying Qch by j is added to Ich as will be mentioned hereinbelow can be used.

[0028] An assumption will be described. Now, assuming that a propagation channel fluctuation value in the jth subcarrier of the ith OFDM symbol is equal to h.sub.i,j,a transmission signal is equal to s.sub.i,j, and a noise is equal to a reception signal r.sub.i,j is shown by the following equation (1).

[Math 1]

r.sub.ij=h.sub.ijs.sub.ij+n.sub.ij(1)

[0029] Now, assuming that a number of a symbol in which a pilot subcarrier has been inserted is equal to p and a number of an inserted subcarrier in such a symbol is equal to q.sub.p, information which can be obtained by the receiver side is a reception signal r.sub.p,qp in the pilot subcarrier, a transmission signal s.sub.p,qp, and the reception signal r.sub.i,j in the data subcarrier. A propagation channel fluctuation value in the data subcarrier is estimated on the basis of the transmission/reception signals in the pilot subcarrier. After the estimation, a series of estimation, which will be described hereinafter, is performed in order to equalize to the transmission signal in the data subcarrier.

[0030] An embodiment of the interpolation method of the invention will be described with reference to a flowchart of FIG. 2.

[0031] Step ST1: A propagation channel estimation value in the pilot subcarrier is expressed by the following equation (2) from the reception signal r.sub.p,qp in the pilot subcarrier and the transmission signal s.sub.p,qp.

[00001]$\left[\mathrm{Math}..Math.2\right]$$\begin{array}{cc}{\hat{h}}_{p,q_p}=\frac{r_{p,q_p}}{s_{p,q_p}}& \left(2\right)\end{array}$

[0032] Step ST2: The propagation channel estimation value in the pilot subcarrier is separated into an estimated amplitude fluctuation value R.sub.p,qp and an estimated phase fluctuation value .sub.p,qp. They are expressed by the following equations (3)

[Math 3]

Rp,q.sub.p=|p,q.sub.p|,p,q.sub.p=arg p,q.sub.p(3)

Step ST3: By interpolating the estimated amplitude fluctuation value and the estimated phase fluctuation value in the pilot subcarrier, an amplitude fluctuation value and a phase fluctuation value in the data subcarrier are estimated. When the tile structure is l symbols and m subcarriers, it is assumed that i.sub.m=i mod l, i.sub.f=floor(i/l)l, j.sub.m=j mod m, and j.sub.f=floor(j/m)m. Now, when an attention is paid to one tile, since it is separated into four kinds of (p, q.sub.p)=(1+i.sub.f, 1+j.sub.f), (l+i.sub.f, m+j.sub.f), (l+i.sub.f), and (1+i.sub.f, m+j.sub.f), the estimated amplitude fluctuation value and the estimated phase fluctuation value are expressed by the following equations (4) and (5).

[00002]$\begin{array}{cc}.Math.\left[\mathrm{Math}..Math.4\right]& \\ {\tilde{R}}_{i,j}=\frac{1}{\left(l-1\right).Math.\left(m-1\right)}.Math.\left\{\left(l-1-i_m\right).Math.\left(m-1-j_m\right).Math.R_{1+i_f,1+j_f}+{\mathrm{mi}}_m.Math.{\mathrm{lj}}_m.Math.R_{l+i_f,m+j_f}+{\mathrm{mi}}_m\left(m-1-j_m\right).Math.R_{l+i_f,l+j_f}+\left(l-1-i_m\right).Math.{\mathrm{lj}}_m.Math.R_{1+i_f,1+j_f}\right\}& \left(4\right)\\ {\stackrel{~}{}}_{i,j}=\frac{1}{\left(l-1\right).Math.\left(m-1\right)}.Math.\left\{\left(l-1-i_m\right).Math.\left(m-1-j_m\right).Math.R_{1+i_f,1+j_f}+{\mathrm{mi}}_m.Math.{\mathrm{lj}}_m.Math.R_{l+i_f,m+j_f}+{\mathrm{mi}}_m\left(m-1-j_m\right).Math.R_{l+i_f,l+j_f}+\left(l-1-i_m\right).Math.{\mathrm{lj}}_m.Math.R_{1+i_f,1+j_f}\right\}& \left(5\right)\end{array}$

[0033] Step ST4: By the estimated phase fluctuation value in the data subcarrier, a direction vector of the phase/amplitude separation linear interpolation propagation channel estimation value in the data subcarrier is calculated.

[Math 5]

(cos {tilde over ()}.sub.i,j, sin {tilde over ()}.sub.i,j) is obtained from {tilde over ()}.sub.i,j

Step ST5: As shown in the following equation (6), by performing the linear interpolation (method in the related art) on the complex plane to the propagation channel estimation value in the pilot subcarrier, a propagation channel estimation value in the data subcarrier is obtained and is set to a reference parameter.

[00003]$.Math.\left[\mathrm{Math}..Math.6\right]$$\begin{array}{cc}{\tilde{h}}_{i,j}^{\mathrm{IQ}}=\frac{1}{\left(l-1\right).Math.\left(m-1\right)}.Math.\left\{\left(l-1-i_m\right).Math.\left(m-1-j_m\right).Math.{\hat{h}}_{1+i_f,1+j_f}+{\mathrm{mi}}_m.Math.{\mathrm{lj}}_m.Math.{\hat{h}}_{l+i_f,m+j_f}+{\mathrm{mi}}_m\left(m-1-j_m\right).Math.{\hat{h}}_{l+i_f,l+j_f}+\left(l-1-i_m\right).Math.{\mathrm{lj}}_m.Math.{\hat{h}}_{1+i_f,1+j_f}\right\}& \left(6\right)\end{array}$

[0034] Step ST6: The Ich component and Qch component of the data subcarrier propagation channel estimation value (reference parameter) by the linear interpolation on the complex plane are compared with the direction vector of the data subcarrier propagation channel estimation value by the phase/amplitude separation linear interpolation, thereby discriminating whether or not the quadrants coincide (whether or not the phase connection is necessary) (expressions 7).

[Math 7]

Re{h.sup.IQ.sub.i,j}.Math.cos {tilde over ()}.sub.i,j<0

Im{h.sup.IQ.sub.i,j}.Math.sin {tilde over ()}.sub.i,j<0 (7)

[0035] Step ST7: If it is determined in step ST6 that the quadrants are different, the phase connection is cancelled by adding 2 to a smaller one of the values as shown in the following equation.

[00004]$\left[\mathrm{Math}..Math.8\right]$$\begin{array}{cc}\left({}_{p,q_{p.Math..Math.1}}^{},{}_{p,q_{p.Math..Math.2}}^{}\right)=\{\begin{array}{cc}\left({}_{p,q_{p.Math..Math.1}}+2.Math.,{}_{p,q_{p.Math..Math.2}}\right)& \left({}_{p,q_{p.Math..Math.1}}<{}_{p,q_{p.Math..Math.2}}\right)\\ \left({}_{p,q_{p.Math..Math.1}},{}_{p,q_{p.Math..Math.2}}+2.Math.\right)& \left({}_{p,q_{p.Math..Math.1}}>{}_{p,q_{p.Math..Math.2}}\right)\end{array}& \left(8\right)\end{array}$

[0036] Step ST8: Only the phase interpolation in the phase/amplitude separation linear interpolation method is performed again as follows. An interpolating process (equation 9) of step ST9 is executed after step ST8.

[00005]$.Math.\left[\mathrm{Math}..Math.9\right]$$\begin{array}{cc}{\stackrel{~}{}}_{i,j}^{}=\frac{1}{\left(l-1\right).Math.\left(m-1\right)}.Math.\left\{\left(l-1-i_m\right).Math.\left(m-1-j_m\right).Math.{}_{1+i_f,1+j_f}^{}+{\mathrm{mi}}_m.Math.{\mathrm{lj}}_m.Math.{}_{l+i_f,m+j_f}^{}+{\mathrm{mi}}_m\left(m-1-j_m\right).Math.{}_{l+i_f,l+j_f}^{}+\left(l-1-i_m\right).Math.{\mathrm{lj}}_m.Math.{}_{1+i_f,1+j_f}^{}\right\}& \left(9\right)\end{array}$

[0037] Step ST9: A data subcarrier propagation channel estimation value by the final phase/amplitude separation linear interpolation is calculated by the estimated amplitude fluctuation value and the estimated phase fluctuation value after the phase connection as shown by the following equation (10). Even if it is determined in step ST6 that the quadrants are not different, the process of step ST9 is executed.

[Math 10]

{tilde over (h)}.sub.i,j={tilde over (R)}.sub.i,j exp(j{tilde over ()}.sub.i,j) (10)

[0038] The actual propagation channel fluctuation does not linearly fluctuate on the complex plane but the fluctuation occurs in each of the phase and the amplitude. Therefore, as a result of the foregoing process, by performing the equalization by using the phase/amplitude separation linear interpolation method according to the invention, the more actual propagation channel fluctuation can be interpolation traced as compared with the case of using only the linear interpolation on the complex plane as a method in the related art. Thus, the propagation channel estimation error can be further suppressed and even in the severe fading propagation channel as mentioned above, the good reception quality can be accomplished.

[0039] Although the embodiments of the invention have specifically been described above, the invention is not limited to the foregoing embodiments but various modifications based on the technical idea of the invention are possible.

REFERENCE SIGNS LIST

[0040] ST1ST9 . . . Steps of propagation channel estimation method