Communication system and phase error estimating method thereof

Abstract

A communication system includes a receiving circuit and a phase error estimating circuit. The receiving circuit receives an input signal x, which has an input phase in a polar coordinate system. According to partial differentiation performed on the natural logarithm of a function f(x, ), the phase error estimating circuit generates an estimated phase error of the input signal x. f(x, ) represents a probability function of receiving the input signal x at the receiving circuit.

Claims

1. A communication system for estimating a phase error, comprising: a receiving circuit, that receives an input signal x, the input signal x having an input signal radius r and an input phase , wherein the input signal x is generated based on M-phase shift keying (M-PSK); and an error estimating circuit, that generates a first harmonic coefficient h.sub.1(r) and a second harmonic coefficient h.sub.2(r) according to the input signal radius r, generates a first sine function sin(M) and a second sine function sin(2M) according to the input phase , multiplies the first harmonic coefficient h.sub.1(r) by the first sine function sin(M) to generate a first operation result and multiplies the second harmonic coefficient h.sub.2(r) by the second sine function sin(2M) to generate a second operation result, and generates an estimated phase error {circumflex over ()} according to the first operation result and the second operation result; wherein, the estimated phase error {circumflex over ()} is associated with a Fourier series of $\frac{}{}\ln f\left(x,\right),$ and f(x, ) represents a probability function of receiving the input signal x at the receiving circuit, and wherein the estimated phase error {circumflex over ()} is used to compensate a phase of the input signal x.

2. The communication system according to claim 1, wherein the receiving circuit further comprises: an auto gain control (AGC) circuit, that adjusts a signal strength .sup.2 of the input signal, where represents a normalization factor; wherein, the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r) are associated with the normalization factor .

3. The communication system according to claim 2, wherein the phase error estimating circuit generates the first harmonic coefficient h.sub.1(r) according to an equation: $h_1\left(r\right)=\{\begin{array}{cc}0,& r<d_1\\ d_2+\frac{d_3}{}\left(r-d_4\right),& r{d}_5\\ \frac{d_6}{2}\left(r-d_7\right),& \mathrm{else}\end{array};$ wherein, each of d.sub.1 to d.sub.7 represents a constant value.

4. The communication system according to claim 2, wherein the phase error estimating circuit generates the second harmonic coefficient h.sub.2(r) according to an equation: $h_2\left(r\right)=\{\begin{array}{cc}0,& r<d_8\\ \frac{d_9}{}\left(r-d_{10}\right),& r{d}_8\end{array};$ wherein, each of d.sub.8 to d.sub.10 represents a constant value.

5. The communication system according to claim 2, wherein the phase estimating circuit generates the estimated phase error {circumflex over ()} according to an equation:
{circumflex over ()}=h.sub.1(r)sin(M)h.sub.2(r)sin(2M).

6. A phase error estimating method applied to a communication system to estimate a phase error of an input signal x, the input signal x corresponding to an output signal that a transmitting end generates based on M-PSK, M being a value of 2 raised to a power of any positive integer, the phase error estimating method comprising: a) receiving an input signal x, the input signal x having an input signal radius r and an input phase , wherein the input signal x is generated based on M-phase shift keying (M-PSK); b) generating a first harmonic coefficient h.sub.1(r) and a second harmonic coefficient h.sub.2(r) according to the input signal radius r; c) generating a first sine function sin(M) and a second sine function sin(2M according to the input phase ; d) multiplying the first harmonic coefficient h.sub.1(r) by the first sine function sin(M) to generate a first operation result, and multiplying the second harmonic coefficient h.sub.2(r) by the second sine function sin(2M) to generate a second operation result; and e) generating an estimated phase error {circumflex over ()} according to the first operation result and the second operation result; wherein, the estimated phase error {circumflex over ()} is associated with a Fourier series of $\frac{}{}\ln f\left(x,\right),$ and f(x, ) represents a probability function of the input signal x, and wherein the estimated phase error {circumflex over ()} is used to compensate a phase of the input signal x.

7. The phase error estimating method according to claim 6, before step (b), further comprising: adjusting a signal strength .sup.2 of the input signal, where represents a normalization factor; wherein, the normalization factor a is for generating the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r).

8. The phase error estimating method according to claim 7, wherein step (b) comprises generating the first harmonic coefficient h.sub.1(r) according to an equation: $h_1\left(r\right)=\{\begin{array}{cc}0,& r<d_1\\ d_2+\frac{d_3}{}\left(r-d_4\right),& r{d}_5\\ \frac{d_6}{2}\left(r-d_7\right),& \mathrm{else}\end{array};$ wherein, each of d.sub.1 to d.sub.7 represents a constant value.

9. The phase error estimating method according to claim 7, wherein step (b) comprises generating the second harmonic coefficient h.sub.2(r) according to an equation: $h_2\left(r\right)=\{\begin{array}{cc}0,& r<d_8\\ \frac{d_9}{}\left(r-d_{10}\right),& r{d}_8\end{array};$ wherein, each of d.sub.8 to d.sub.10 represents a constant value.

10. The phase error estimating method according to claim 7, wherein step (e) comprises generating the estimated phase error {circumflex over ()} according to an equation:
{circumflex over ()}=h.sub.1(r)sin(M)h.sub.2(r)sin(2M).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1(A) and FIG. 1(B) are functional block diagrams of a communication system according to an embodiment of the present invention; and

(2) FIG. 2 is a flowchart of a phase error estimating method according to an embodiment of the present invention.

DETAILED DESCRIPTION OF THE INVENTION

(3) The mathematical expressions in the disclosure are for illustrating principles and logics associated with the embodiments. Unless otherwise specified, these mathematical expressions do not form limitations upon the scope of the present invention. One person skilled in the art can understand that, there are many technologies capable of realizing physical expressions and forms corresponding to these mathematical expressions.

(4) Details of how a communication system and a phase error estimating method of the present invention provide an estimated phase error that satisfies a Cramer-Rao bound (CRB) rule are given below.

(5) It is assumed that a receiving end of a communication system receives an input signal x, which has an input phase when the input signal x is expressed by a polar coordinate system. An observable S(x) is defined as an unbiased estimator of the input phase , and an expected value of the observable S(x) is:
E[S(x)]=S(x).Math.f(x,)dx=equation (1)

(6) In equation (1), f(x, ) represents a probability function of the input signal x.

(7) An estimated difference (x) between the estimated phase error generated by the receiving end of the communication system and the real phase error is:
(x)=S(x).sub.actequation (2)

(8) In equation (2), .sub.act represents the real phase error of the input signal x. The main goal of the communication system and the phase error estimating method of the present invention is to render a variance of the estimated difference (x) to have a predetermined lower limit. Further, based on the assumption of an unbiased phase, an expected value of the estimated difference (x) is zero:
E[(x)]=[S(x).sub.act].Math.f(x,)dx=0equation (3)

(9) In a regularity condition,

(10) $\frac{}{}=\frac{}{}.$
By performing partial differentiation on the expected value E[(x)], an equation below is obtained:

(11) $\begin{array}{cc}\begin{array}{c}\frac{}{}E.Math.\mathrm{.Math.}\left(\right).Math.=\frac{}{}.Math.S\left(\right)-{}_{\mathrm{act}}.Math..Math.f\left(,\right)=0\\ =\left[S\left(\right)-{}_{\mathrm{act}}\right].Math.\frac{}{}f\left(,\right)+\\ \frac{}{}\left[S\left(\right)-{}_{\mathrm{act}}\right].Math.f\left(,\right)\\ =\mathrm{.Math.}\left(\right).Math.v\left(,\right).Math.f\left(,\right)-f\left(,\right)=0\end{array}& \mathrm{equation}\left(4\right)\end{array}$

(12) In the above, f(x, )dx is equal to 1, and so (x).Math.(, x).Math.f(x, )dx=1.

(13) By using mathematical equations:

(14) $\begin{array}{cc}v\left(,\right)=\frac{}{}\ln f\left(,\right)& \mathrm{equation}\left(5\right)\\ E\left[v\left(,\right)\right]=0& \mathrm{equation}\left(6\right)\end{array}$

(15) It is obtained that a covariance of the estimated difference (x) and(, x) is equal to 1.
E[(x).Math.(,x)]=E[S(x).Math.(,x)].sub.actE[(,x)]=E[S(x).Math.(,x)]=1equation (7)

(16) Next, by using a mathematical equation 1StDev()StDev(v), when a correlation coefficient between the estimated difference (x) and (, x) is 1, an equation defined by CRB rule is as below:

(17) $\begin{array}{cc}E{.Math.\mathrm{.Math.}\left(\right).Math.v\left(,\right).Math.}^2=1E\left[{\mathrm{.Math.}}^2\left(\right)\right].Math.E[v^2\left(,\right)& \mathrm{equation}\left(8\right)\\ E\left[{\mathrm{.Math.}}^2\left(\right)\right]\frac{1}{E\left[v^2\left(,\right)\right]}& \mathrm{equation}\left(9\right)\end{array}$

(18) It is known from the above equations that, given the estimated difference estimated difference (x) is correlated with(, x), the above condition that satisfies the CRB rule can be established. Therefore, in the communication system and the phase error estimating method of the present invention,(, x) is utilized to estimate the estimated difference (x), so as to provide an estimated phase error that satisfies the CRB rule.

(19) FIG. 1(A) shows a functional block diagram of a communication system according to an embodiment of the present invention. A communication system 100 includes a receiving circuit 12 and a phase error estimating circuit 14. The receiving circuit 12 receives an input signal x, which may be converted and represented by an input radius r and an input phase in a polar coordinate system. f(x, ) represents a probability function of receiving the input signal x at the receiving circuit. According to an equation below, the phase error estimating circuit 14 generates an estimated phase error {circumflex over ()} of the input signal x:

(20) $\begin{array}{cc}\hat{}=\frac{}{}\ln f\left(,\right)& \mathrm{equation}\left(10\right)\end{array}$

(21) It should be noted that, the scope of the present invention does not limit a particular configuration or structure for realizing the communication system 100, and the setting of the probability function f(x, ) may be determined by a circuit designer according to characteristics of an actual signal and the transmission environment. In practice, after the probability function f(x, ) is determined, equation (10) may be expanded into an equation of other forms to be further appropriately simplified. On the other hand, the error phase estimating circuit 14 that calculates the estimated phase error {circumflex over ()} may be implemented as a fixed and/or programmable digital logic circuit, including a programmable logic gate array, an application specific integrated circuit (ASIC), a microcontroller, a microprocessor, a digital signal processor (DSP), or other necessary circuits. Alternatively, the phase error estimating circuit 14 may be designed to complete various tasks through executing processor commands stored in a memory (not shown).

(22) A communication system based on M-phase shift keying (M-PSK) is taken as an example in the description below (where M is a value of 2 raised to a power of a positive integer, e.g., 2, 4, 8, 16 . . . ). Without considering shifts of the signal in the frequency and time domains, the signal received by the receiving circuit 12 may be represented as:
x[k]=a[k]e.sup.f+w[k], k=0,1, . . . ,(L.sub.01)equation (11)

(23) In equation (11), L.sub.0 represents the number of sampling points on the time axis, a[k] represents an ideal signal provided by the transmitting end, r[k] represents the signal received by the receiving circuit 12, and w[k] represents additive white Gaussian noise (AWGN). An angle of the signal x[k] in the polar coordinate system is an unknown value to be determined. Corresponding to the probability function f(x[0]|a[0];]) of the signal x[0] may be defined as:

(24) $\begin{array}{cc}f\left(\left[0\right].Math.a\left[0\right];\right)=\frac{1}{\sqrt{2{}^2}}\exp \left\{\frac{-{.Math.\left[0\right]-a\left[0\right]{}^j.Math.}^2}{2{}^2}\right\}& \mathrm{equation}\left(12\right)\end{array}$

(25) In equation (12), the symbol represents a standard deviation.

(26) Equation (12) may be re-written as:

(27) $\begin{array}{cc}f\left(x\left[0\right];\right)=f\left(x\left[0\right]|a_0\left[0\right]\right)p\left(a_0\left[0\right]\right)+\mathrm{.Math.}=\frac{1}{M\sqrt{2{}^2}}\underset{k=0}{\overset{M-1}{.Math.}}\exp \left\{\frac{-{.Math.x\left[0\right]-a_k\left[0\right]{}^j.Math.}^2}{2{}^2}\right\}& \mathrm{equation}\left(13\right)\end{array}$

(28) In equation (13), a.sub.k[0] is defined as:

(29) 0$\mathrm{equation}\left(14\right)$$\begin{array}{cc}\begin{array}{c}{a}_k\left[0\right]=\sqrt{\frac{B_s}{N_0}}{}^{\frac{\mathrm{j2}k}{M}}\\ =a{}^{\frac{\mathrm{j2}k}{M}}\\ =\frac{1}{M\sqrt{2{}^2}}\underset{k=0}{\overset{M-1}{.Math.}}\exp \left\{\frac{-1}{2{}^2}\left(\begin{array}{c}{.Math.x\left[0\right].Math.}^2+{.Math.a_k\left[0\right].Math.}^2-\\ 2\mathrm{Re}\left\{x\left[0\right].Math.a_k^{*}\left[0\right].Math.{}^{-j}\right\}\end{array}\right)\right\}\\ =T^{}.Math.\underset{k=0}{\overset{M-1}{.Math.}}\exp \left\{\frac{2}{N_o}\mathrm{Re}\left\{a.Math.x\left[0\right].Math.{}^{\frac{-\mathrm{j2}k}{M}.Math.{}^{-j}}\right\}\right\}\end{array}& \end{array}$

(30) In equation (14), the symbol T represents:

(31) $\begin{array}{cc}{T}^{}=\frac{1}{M\sqrt{2{}^2}}\exp \left\{\frac{-1}{2{}^2}\left({.Math.x.Math.0.Math..Math.}^2+{.Math.a.Math.}^2\right)\right\}& \mathrm{equation}\left(15\right)\end{array}$

(32) A result of performing a natural logarithm operation on equation (13) is:

(33) $\begin{array}{cc}\ln f\left(x\left[0\right];\right)=\ln T^{}+\ln \left\{\underset{k=0}{\overset{\frac{M}{2}-1}{.Math.}}\cosh \left(\frac{2}{N_0}\mathrm{Re}\left\{a.Math.x\left[0\right].Math.{}^{+\mathrm{j2}k/M}.Math.{}^{-j}\right\}\right)\right\}& \mathrm{equation}\left(16\right)\end{array}$

(34) In equation (16), ln f(x[0], ]) is In f(x, ) in equation (10). Thus, the estimated phase error {circumflex over ()} may be obtained by performing partial differentiation on equation (16):

(35) $\mathrm{equation}\left(17\right)$$\hat{}=\frac{}{}\ln f\left(x\left[0\right];\right)=\frac{2}{N_0}\frac{\underset{k=0}{\overset{\frac{M}{2}-1}{.Math.}}\sinh \left(\frac{2}{N_0}\mathrm{Re}\left\{\begin{array}{c}a.Math.x\left[0\right].Math.\\ {}^{+/2k/M}.Math.{}^{-j}\end{array}\right\}\right).Math.\mathrm{Im}\left\{\begin{array}{c}a.Math.x\left[0\right].Math.\\ {}^{+\mathrm{j2}k/M}.Math.{}^{-j}\end{array}\right\}}{\underset{k=0}{\overset{\frac{M}{2}-1}{.Math.}}\cosh \left(\frac{2}{N_0}\mathrm{Re}\left\{a.Math.x\left[0\right].Math.{}^{+/2k/M}.Math.{}^{-j}\right\}\right)}$

(36) In practice, simulation software such as Malab may be utilized to simplify the complexity of actual circuits to further represent equation (17) by a Fourier series, and harmonic items with lower contribution in the calculation of the estimated phase error {circumflex over ()} are omitted. In one embodiment, only the first harmonic item and the second harmonic item are taken into consideration. Equation (17) is simplified as:
{circumflex over ()}=h.sub.1(r)sin(M)h.sub.2(r)sin(2M)equation (18)

(37) In equation (18), the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r) are associated with the signal input radius r of the input signal x in the polar coordinate system, and the first sine function sin(M) and the second sine function sin(2M) are associated with the input phase of the input signal x in the polar coordinate system. In other words, after the signal radius r and the input phase of the input signal x in the polar coordinate system are determined, the phase error estimating circuit 14 may generate the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r) according to the input signal radius r, and generate the first sine function sin(M) and the second sine function sin(2M) according to the input phase . Next, the phase error estimating circuit 14 may multiply the first harmonic coefficient h.sub.1(r) by the first sine function sin(M) to generate a first operation result, and multiply the second harmonic coefficient h.sub.2(r) by the second sine function sin(2M) to generate a second operation result. According to the first operation result and the second operation result, the phase error estimating circuit 14 may generate the estimate phase error {circumflex over ()} of the input signal x.

(38) As shown in FIG. 1(B), the receiving circuit 12 may further include an auto gain control (AGC) circuit 12A. Before the phase error estimating circuit 14 generates the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r), the AGC circuit 12A adjusts the input signal x such that the input signal x has a signal strength .sup.2, where represents a normalization factor. In one embodiment, the phase error estimating circuit 14 also considers the normalization factor when generating the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r), and generates the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r) according to equations below:

(39) $\begin{array}{cc}{h}_1\left(r\right)=\{\begin{array}{cc}0,& r<d_1\\ d_2+\frac{d_2}{}\left(r-d_4\right),& r{d}_5\\ \frac{d_6}{2}\left(r-d_7\right),& \mathrm{else}\end{array}& \mathrm{equation}\left(19\right)\\ h_2\left(r\right)=\{\begin{array}{cc}0,& r<d_8\\ \frac{d_9}{}\left(r-d_{10}\right),& r{d}_8\end{array}& \mathrm{equation}\left(20\right)\end{array}$

(40) In equations (19) and 20, each of d.sub.1 to d.sub.10 represents a constant value. For example, when the value M is equal to 8 (i.e., the communication system 100 adopts a 8PSK modulated operation), and a symbol energy/noise density ratio (Es/No) equal to 5.6 dB is substituted into equation (17), following equations are obtained:

(41) $\begin{array}{cc}{h}_1\left(r\right)=\{\begin{array}{cc}0,& r<1.357488\\ 3.164957+\frac{6.208301}{}\left(r-2.377077\right),& r2.377077\\ \frac{6.208301}{2}\left(r-1.357488\right),& \mathrm{else}\end{array}& \mathrm{equation}\left(21\right)\\ h_2\left(r\right)=\{\begin{array}{cc}0,& r<2.546168\\ \frac{0.966715}{}\left(r-2.546168\right),& r2.546168\end{array}& \mathrm{equation}\left(22\right)\end{array}$

(42) It should be noted that, the actual values of the first harmonic coefficient h.sub.1(r) and the second harmonic coefficient h.sub.2(r) are associated with the parameters in equation (17) as well as the method by which equation (17) is simplified, and are not limited to particular values. Details of ACG technologies and a method for selecting the normalization factor are known to one person skilled in the art, and shall be omitted herein.

(43) A phase error estimating method for a communication system is provided according to another embodiment of the present invention to estimate a phase error of an input signal x. The phase error estimating method first performs a determining step to determine an input phase of the input signal x in a polar coordinate system. f(x, ) represents a probability function of receiving the input signal x. An estimated phase error {circumflex over ()} of the input signal x is then generated according to an equation:

(44) $\hat{}=\frac{}{}\ln f\left(x,\right).$

(45) A phase error estimating method for a communication system is provided according to another embodiment of the present invention to estimate a phase error of an input signal x. FIG. 2 shows a flowchart of the phase error estimating method. The input signal x corresponds to an output signal that a transmitting end generates based on M-PSK, where M is a value of 2 raised to a power of any positive integer. In step S21, the phase error estimating method performs a determining step to determine an input signal radius r and an input phase of the input signal x in a polar coordinate system. In step S22, a first harmonic coefficient h.sub.1(r) and a second harmonic coefficient h.sub.2(r) are generated according to the input signal radius r. In step S23, a first sine function sin(M) and a second sine function sin(2M) are generated according to the input phase . In step S24, the first harmonic coefficient h.sub.1(r) is multiplied by the first sine function sin(M) to generate a first operation result, and the second harmonic coefficient h.sub.2(r) is multiplied by the second sine function sin(2M) to generate a second operation result. In step S25, an estimated phase error {circumflex over ()} of the input signal is generated according to the first operation result and the second operation result.

(46) One person skilled in the art can understand that, in FIG. 2, the sequences of certain steps may be equivalently exchanged without affecting the overall effect of the phase error estimating method. Further, operation variations in the description associated with the communication system 100 are applicable to the phase error estimating method, and shall be omitted herein.

(47) It should be noted that, the drawings of the present invention include functional block diagrams of multiple functional circuits related to one another. These drawings are not detailed circuit diagrams, and connection lines therein are for indicating signal flows only. The interactions between the functional elements/or processes are not necessarily achieved through direct electrical connections. Further, functions of the individual elements are not necessarily distributed as depicted in the drawings, and separate blocks are not necessarily implemented by separate electronic elements.

(48) While the invention has been described by way of example and in terms of the preferred embodiments, it is to be understood that the invention is not limited thereto. On the contrary, it is intended to cover various modifications and similar arrangements and procedures, and the scope of the appended claims therefore should be accorded the broadest interpretation so as to encompass all such modifications and similar arrangements and procedures.