KR100186290B1 - Phase compensation type phase shift modulation/demodulation method - Google Patents

Abstract

The present invention relates to a novel method of phase shift modulation and demodulation having better phase error compensation characteristics than the conventional method. The PC-PSK inverts and synthesizes the second half of the received signal so that phase errors caused by fading during transmission, mutual interference, and non-linearity of the receiving amplifier cancel each other out.

The received signal power is smaller as the error phase to be compensated for is larger and the bandwidth is increased due to the inversion of the latter signal during transmission. To reduce bandwidth, PC-PSK must be polyphased. When the PC-PSK is polyphased, the crystal region is narrowed, so the signal power is small in the relatively low E _b / N _o region, thereby weakening the effect of phase compensation and being affected by white noise. On the other hand, in the relatively high E _b / N _o region, the congestion of bit error rate due to phase error or fading is reduced and the performance is greatly improved.

Description

Phase compensated phase shift demodulation method

1 is a block diagram of a PC-PSK modulator according to the present invention.

2 is a diagram showing a signal vector spatial diagram of a gray-coded PC-QPSK.

3 is a signal waveform diagram of an input signal, I (t) and Q (t) according to the PC-QPSK modulation method according to the present invention.

4 is a diagram for explaining a PC-PSK modulation method according to the present invention.

5 is a block diagram of a PC-PSK demodulator according to the present invention.

6 is a diagram for explaining a PC-PSK demodulation method according to the present invention.

7 is a graph for comparing bit error rates of QPSK and PC-16PSK in AWGN channel.

8 is a graph for comparing the bit error rates of QPSK and PC-16PSK in slow fading Rician channels.

9 is a graph for comparing bit error rates of QPSK and PC-16PSK in a time-varying Rician channel.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital phase modulation demodulation method, and more particularly, to a phase shift modulation and demodulation method of a new method having better phase error compensation characteristics than the conventional method.

Phase Shift Keying (PSK) is a carrier modulated phase modulated by digital information, and is widely used for relatively easy phase synchronization detection. However, in the mobile communication environment, errors occur due to the movement of the transmit / receive antenna, fading due to multipath, instability of the receiver local oscillator, and nonlinearity of the amplifier.

In order to improve the synchronous detection performance of PSK in noisy communication environment, channel error correction code, diversity transmit / receive method, or double phase shift keying (DSK) have been studied. The introduction of the channel error correcting code uses an error correcting code such as a block or convolutional code and adds a redundant bit to correct the error bit. The diversity transmission / reception method has been used since 1927, and is a technique that suppresses fading distortion by reducing the probability of a drop in reception field strength for received signals via several independent transmission paths. This method includes spatial diversity, polarization diversity, temporal diversity, and frequency diversity depending on the transmission path selection.

Modulation and demodulation methods that overcome the characteristics of recently developed multipath include DSK, MC-PSK (Manchester-coded PSK), and Phase Shift Keying with Varied Phase (PSK-VP). Among them, DSK is a method proposed by S.Ariyavisitakul, S.Yoshida, F.Ikegami, T.Takeuchi, etc. of Japan University of Japan, and is a modulation method for overcoming multiple waves. This method prevents the closing of the eye pattern between codes by multipath to the entire 1-bit interval, and performs a phase shift of _± 90 degrees twice for binary information of 1 and 0 of data. . The double _± 90 degree phase shift reduces the differential detector's effect on the delay element, even with the worst multi-wave effects of equal magnitude and 180 degrees out of phase. This method is mainly used in DBPSK's differential detection receiver, and minimizes the interference by complementarily canceling the influence of inter-signal interference (ISI) due to multipath within 1 bit interval. MC-PSK and PSK-VP apply the principle of DSK and have the characteristics of widening the multiple wave overcoming than DSK.

Analog error phase compensation includes PAL (Phase Alternation by Line) TV. PAL TV system was developed by W.Bruch of Telefunken Co., Ltd. in 1962. Compared to other methods such as NTSC (National Television System Committee) and SECAM (Sequencial Couleur A Memoire) TV, It has excellent robustness in compensating for distortion of receiver tints due to poor propagation environment and phase distortion of transceiver circuitry. In the PAL TV system, the color of the color signal is phase-modulated and the chromaticity is amplitude-modulated. This color signal is represented in a signal space that is very similar to the amplitude phase modulation (QAM) signal coordinates of digital communications. In addition, the PAL TV method assumes that the color information of adjacent scan lines is almost the same. The color information of odd and even scan lines is inverted and transmitted only with signals of orthogonal components. In the receiver, color information of odd and even scan lines is sequentially inverted and averaged and displayed on the screen. By doing this, color distortion generated in the channel is removed and the original color is found.

An object of the present invention is to provide a new phase-compensated PSK modulation / demodulation method for compensating for errors occurring in digital communication based on the excellent color distortion compensation principle of PAL TV. To provide.

The phase shift modulation method according to the present invention for achieving the above object is composed of m binary data to convert an input binary signal having a predetermined information symbol interval into an output waveform having one phase of a predetermined phase state. A digital phase shift modulation method for changing, comprising: a first step of dividing an information symbol section of an input binary signal into two parts; In generating the 1-axis signal to be modulated by the in-phase carrier signal and the Q-axis signal to be modulated by the orthogonal carrier signal in the information symbol section, the I-axis signal by the phase vector according to the input binary signal in the first half of the information symbol section. And a second step of generating a Q-axis signal and generating an I-axis signal and a Q-axis signal by a phase vector in which the phase vector of the input binary signal is inverted with respect to the I-axis in the second half of the information symbol section. It is characterized by.

The phase shift demodulation method according to the present invention for achieving the above object is a modulation consisting of m binary data, the input binary signal having a predetermined information symbol interval is a phase shift modulated and transmitted by a predetermined modulation method A digital phase shift demodulation method for restoring a signal to an original signal, the method comprising: recovering a reproduction carrier signal from the transmitted modulation signal to generate an in-phase carrier signal and a quadrature carrier signal; Generating a first I-axis signal detected by the in-phase carrier signal and a first Q-axis signal detected by the quadrature carrier signal in the first half of an information symbol section; Generating a second I-axis signal detected by the in-phase carrier signal and a second Q-axis signal detected by the quadrature carrier signal in a second half of an information symbol section; Generating an I-axis detection signal by adding the first I-axis signal component and the second I-axis signal component, and subtracting the second Q signal component from the first Q-axis signal component to generate a Q-axis detection signal, Synthesizing the I-axis detection signal and the Q-axis detection signal with each other; And generating a demodulated signal for the transmitted modulated signal by the synthesized signal vector.

Hereinafter, the error phase generated by the conventional PSK modulation and demodulation method will be described first. Next, the PC-PSK modulation and modulation method according to the present invention will be described. Finally, the method according to the present invention is superior to the conventional method. The compensation characteristics will be described in detail.

In general, since the height of the receiver antenna used in mobile communication is much lower than the height of the surrounding structure, the received radio waves are reached through various paths due to reflection, diffraction, and scattering, and thus a closed edging phenomenon occurs. In addition, time-varying fading occurs as the transceiver moves relatively quickly. Synchronization detection performance is degraded by an irregular phase variation, and a sampling error occurs when the signal-to-noise ratio of a received signal is reduced by an amplitude variation. In addition, if the phase shift is severe, the detection output signal vector goes to the error-determining region, and even if the signal-to-noise power ratio is increased, error rate stagnation does not improve. The higher the transmission speed of the digital signal, the shorter the bit section per unit information time, so that the multiple delayed signals overlap and the signal spreads. Interference between symbols occurs when this spread cannot be ignored compared to the bit transmission time.

The interference includes cochannel interference (CCI) generated by using the same frequency band, adjacent channel interference (AC1) caused by overlapping use of adjacent frequency bands, and jamming. Among them, the effect of co-channel interference due to the overlap of frequency bands is the biggest. Such interference and interference of the same channel result in phase error of the received signal.

In digital communication systems, phase errors can occur not only in transmission lines but also in transceivers. For example, most power amplifiers, including traveling waveguides used for ultra high frequency (UHF), exhibit a phenomenon in which the output phase changes according to the amplitude of the input, that is, an amplitude phase shift (AM / PM). This phenomenon is somewhat different depending on the type of amplifier, but increases greatly when the non-linear part of the class C amplifier is used to increase power efficiency.

The error phase factors of PSK listed so far are synthesized and integrated into the received signal. The conventional PSK received signal considering the error phase ψ _c (t) and the additive white noise n (t) can be expressed as follows.

Here, E _s is symbol energy, T _s is symbol duration, ω _c is the angular frequency, ψ _K (t) is the modulation phase, n (t) is zero mean and both sides onions power density of carriers N _0/2 [ W / Hz] is noise.

If the reproduction carrier signals for detecting in-phase and quadrature components in the carrier generator of the synchronous detector are U _c (t) and U _s (t), they can be expressed as follows.

Here, ψ _p (t) is the phase of the reproduction carrier and normalizes the energy of the reproduction carrier. In other words,

The synchronous detector receiver multiplies the received signal R (t) by the reproduction carrier signal U _c (t) and U _s (t), respectively, and detects and outputs the in-phase signal component ^ I and the quadrature signal component ^ Q, which are integrals during the T _s period. . If the phase of the reproduction carrier is equal to ψ _c (t), that is, ψ _c (t) = ψ _p (t), the detected signal is given by the following equation.

Here, it can be said that the modulation phase in MPSK is as follows.

And a noise component n _I n and _Q are mutually independent and the signal component is also a pinggyun zero and variance N _0/2 of a Gaussian random variable. If the phase error ψ is zero, only white Gaussian noise is affected. However, in a real communication environment, the received phase ψ _c (t) varies randomly between 0 degrees and 360 degrees due to fading, interference, ISI and noise. Therefore, in a channel whose phase synchronization is unstable, ψ _c (t) ≠ ψ _p (t), and an error phase occurs as follows.

The detected signal is as follows.

That is, the error phase of the conventional PSK is represented by ψ = ψ _c (t) -ψ _p (t), and the detected signal is applied to the quadrature signal and the in-phase signal as shown in equations (8) and (9). Affects performance.

Next, a description will be given of a phase compensation PSK modulation and demodulation method according to the present invention.

Phase-compensated PSK (PC-PSK) is a modulation / demodulation method that accurately compensates for a phase error on a channel without using a channel error correction code.

The PC-PSK generates digital signals having codes opposite to each other based on the center point of one information symbol period and transmits them sequentially, and the receiving side compensates by performing vector synthesis. Comparing the same transmission rate, the bandwidth of PC-QPSK is twice that of conventional QPSK. In order to improve the shortcomings of the frequency characteristics, we consider the modulation and demodulation method of M-ary PC-PSK. The bandwidth of 16-ary PC-PSK is equal to the bandwidth of QPSK.

The formula of the grayed M-ary PC-PSK (PC-MPSK) signal is as follows.

Here, A and ω _c are the amplitude and frequency of the carrier, respectively, and ψ _k (t) is the modulation phase mapped to M (= 2 ^m ) information symbols composed of m binary bits. T _s is the information symbol period, and if the second bit of the binary cycle is T _b T _s = mT _b.

In the conventional MPSK, the modulation phase of the information symbol is kept the same for the symbol period T _s . However, in PC-MPSK, the I (t) value remains the same in the 0≤tT _s interval, but the Q signal is divided into 0≤tT _s / 2 interval and T _s / 2≤tT _s interval to invert at the midpoint of the symbol interval. Has the difference. This signal structure has a function of canceling an error phase generated during detection as can be seen in the following description.

1 shows a block diagram of a modulator of a PC-MPSK according to the present invention.

The input binary bit string is composed of m-bit symbols during the T _s period in the serial to parallel converter (S / P) (11), and corresponding in the I (Inphase) Q (Quadrature) generator (12). The symbol is converted into an in-phase component I (t) and an orthogonal component Q (t) value.

The I (t) signal has the same value during the symbol period, and defines a signal in a section 0≤tT _s / 2 as 1 ₁ (t) and a signal in a section T _s / 2≤tT _s as I ₂ (t), respectively. . The Q (t) signal passes through the Manchester coder 13. The Manchester encoder 13 inverts the polarity of T _s / 2 at the symbol center point. Therefore, the Q (t) signal defines a signal in a section 0 ≤ tT _s / 2 as a Q ₁ (t) signal, a signal in a section T _s / 2 ≤ tT _s as Q ₂ (t), and Q ₁ ( t) and Q ₂ (t) have opposite values. I (t) and Q (t) are amplitude modulated (14, 15) on the carriers cosw _c t and -sinw _c t, respectively, and then synthesized (16) and output.

2 shows a signal vector of PC-QPSK with M = 4. Here, if the signal vector (1,1) is transmitted, the modulation phase information of (1,1) is 0 during t≤2 _s / 2, and the modulation phase information of the signal vector (1,0) is stored during T _s / 2≤tT _s . Is sent. It can be seen that the I (t) value is the same for T _s during the modulation process, while the sign of the Q (t) value is reversed in the symbol period.

FIG. 3 shows the signal waveform relationship between the input binary signal, I (t) and Q (t), for the signal vector shown in FIG. In the signal constellation diagram of FIG. 2, the signal vector during Q ≦ tT _s / 2 and the signal vector during T _s / 2 ≦ tT _s are configured symmetrically about the in-phase axis.

Next, a method of receiving a signal modulated and transmitted by the PC-MPSK modulation method according to the present invention and demodulating the signal will be described.

In the PC-MPSK receiver, a received signal considering phase error? C (t) and additive white noise n (t) due to fading influence, channel interference, ISI, and nonlinear amplification characteristics of the receiver is shown in Equation (1). In other words,

In PC-PSK, the received signal is detected in units of T _s / 2, so for convenience, T _c is defined as follows.

If the bit energy is E _b and the information symbol energy consisting of m bits is E _s ,

Becomes

In the carrier regeneration circuit of the PC-PSK synchronous detector, the in-phase component and the quadrature component have to be synchronously detected in the T _c section, so the reproduction carrier signals r _c (t) and r _s (t) can be expressed by the following equation.

Here, ψ _p (t) is the phase of the reproduction carrier, and the energy of the reproduction carrier is normalized as follows.

The in-phase segment and orthogonal component of PC-PSK detected through the correlator of the reproducing carriers r _c (t) and r _s (t) are as follows.

Here, the subscripts 1 and 2 mean detection output values of the first half section 0 ≦ tT _c and the second half section T _c ≦ tT _s of the symbol period, respectively. In the quadrature detection, the orthogonal component is inverted and transmitted in the second half of the symbol period. Therefore, the orthogonal component should be inverted and synthesized again.

The detection signals of the in-phase components of the first half-symbol section and the second half-symbol section are as follows.

The error phase ψ affecting the detection performance is assumed to be a constant constant for at least one period T _s because it is a very small value that can be ignored compared to the time when the communication environment can be changed during the number information symbol interval.

The noise component n _Ij is expressed as Equation (24) below, which is a Gaussian random variable that is independent of each other and signal components and has a mean of 0 and a variance of No / 2.

The detection signals of the orthogonal components of the first half-symbol section and the second half-symbol section are as follows.

And n _Qj (j = 1,2) are mutually and the signal component and is also independent and zero mean and variance N _0/2 of a Gaussian random variable.

4 shows an example of signal vectors of in-phase and quadrature components for PC-QPSK when the influence of white noise is not taken into account. If we want to send the signal vector (1,1), the signal vector (1,1) is transmitted by Eq. (13) for 0≤tT _c and the orthogonal component is inverted for T _c ≤tT _s Send. And by the phase error ψ for a signal vector transmitted to 0≤tT _c interval (I ^ _l, ^ Q ₁₎ and the detection by the phase error ψ for a signal vector transmitted to the T _c ≤tT _s interval (^ I ₂ , ^ Q ₂ ) is detected. Therefore, the final synthesized signal vectors according to equations (18) and (19) are given by the following equations, where the error phase is compensated, while the power of the signal can be seen to be as small as cosψ.

5 is a block diagram illustrating a demodulation process of the PC-MPSK signal described so far. Here, the synchronization part of the switch to be considered in hardware realization is omitted.

On the receiving side, the carrier recoverer 51 recovers the carrier, multiplies the carrier r _c (t) and r _s (t) by the received signal R (t) and passes through the integrator in the interval T _{c and} then passes ^ I _j , ^ Q _j An in-phase component and an orthogonal component having (j = 1,2) are detected (52, 53).

After delaying ^ I ₁ and ^ Q ₁ by T _c , the in-phase component combines ^ I ₁ and ^ I ₂ , and the orthogonal component synthesizes ^ Q ₁ and inverted ^ Q ₂ . The signal vector detector 54 calculates a phase value from the two components thus obtained, detects a modulation phase having the smallest difference, and outputs a corresponding signal vector.

FIG. 6 is a diagram for describing in detail a signal detection process for PC-MPSK. In this case, it is assumed that a signal vector of (1,1) is transmitted and an error phase is generated 55 degrees in a channel. This 55-degree error phase is compensated for because the absolute value is within 90 degrees. The effect of white Gaussian noise is not considered here.

6a, b, and c show a modulation process of a signal vector, 6d and e show a signal vector showing a phase error, and 6f, g, and h show demodulation. It can be seen that the last detected modulation phase is the same as the modulation phase of the transmission signal vector.

The signal received in the interval 0≤tT _c is ^ ψk = + 100 Is detected. The detected signal appears as an error region as shown in FIG. At this time, the detected signal components ^ I ₁ and ^ Q _l are delayed for T _c as shown in FIG. 6f.

Assuming the same error phase as the interval 0≤tT _{c in the} interval T _c ≤ tT _s , the detected signal is detected in the error region with ^ ψk = + 10 ° as shown in FIG. At this time, the detected signal ^ Q ₂ is inverted back to-^ Q ₂ as shown in FIG. 6g about the I axis. Add the delayed ^ I ₁ and ^ Q ₁ signals to the ^ I ₂ and-^ Q ₂ signals, respectively. As a result, as shown in Fig. 6h, it is represented by vector synthesis, and the vector synthesized in-phase signal and quadrature signal are as follows.

The phase of the final detected signal is the phase ^ ψk = + 45 ° of the final vector synthesized signal. Therefore, the PC-QPSK detection signal represents a signal having accurate phase information but slightly attenuated in amplitude as shown in FIG. 6h even in a communication channel having a phase shift of ψ = -55 °.

Hereinafter, the principle of phase compensation by the PC-MPSK method according to the present invention will be described in more detail with reference to the following equation.

Ideally, the error phase ψ in Eq. (7) is equal to 0 if the phase of the reproduction carrier ψ _p (t) is equal to the error phase ψc (t) of the received signal due to fading, channel interference, ISI and nonlinear characteristics of the receiving amplifier. In this way, an accurate transmission phase can be detected. However, there is an error phase under the actual communication environment.

Using equations (20)-(26) and (3),

Can be obtained. Where I and Q are transmission signal vectors as follows.

And n _I and n _Q are independent of each other and signal components and are Gaussian random variables with mean 0 and variance N ₀ .

In the equations (27) and (28), if the error phase range is │ψ│∏ / 2 and the effect of white Gaussian noise is ignored, the signs of the detected in-phase and quadrature components are inverted, respectively, to show the original signal vector I and Q values. It is the opposite of the sign. This means that a symbol error occurs due to the error area. Thus, in this case, the PC-PSK cannot compensate for the error phase.

If the error phase range is │ψ│∏ / 2 and also ignore the effect of the white Gaussian noise, the following relation can be obtained and the error phase is completely compensated.

In a communication environment having multiple propagation paths, a signal having a constant amplitude is transmitted, but the amplitude of the received signal is not constant, but the fading phenomenon is increased or decreased. This is because there is an increase or decrease in amplitude due to the phase difference between the signals received through the various paths. In addition, signal interference of the same or adjacent channels is a factor of fading. In such a communication environment, a cellular mobile communication channel may be a practical example.

The Rayleigh fading channel in the form of a Rayleigh distribution and the Rician fading channel in the form of a Rician distribution according to the probabilistic change characteristic of the amplitude. The direct fading channel is a case where the power of the direct wave received by the signal is directly transmitted is relatively greater than the power of the indirectly received wave such as a multipath signal, and the case where no direct wave exists is modeled as a Rayleigh fading channel. do. The Rayleigh fading channel can be interpreted as a special case of the Rician fading channel.

Hereinafter, the PC-PSK performance in the Rician fading channel is analyzed and compared with the conventional PSK performance. The type of fading is divided into slow fading and time varing fading, respectively. If there is a Doppler frequency shift due to rapid movement of the transmitting or receiving side or other changes in the propagation path, this means that there is fast time-varying fading, which is the case. The degree of change in time-varying fading is represented by the coefficient of the Lysian distribution function.

The carrier transmitted through the fading channel has a severe phase error, and since the phase becomes more unstable due to signal interference or nonlinearity of the amplifier, the carrier reproduced by the synchronous detection receiver is not accurately synchronized. That is, the synchronization of the phase locked loop (PLL) of the receiver becomes unstable. The probability density function of this unstable phase variable ψ is determined according to the channel model.

To evaluate the performance of the PC-MPSK modulation and demodulation method, we compare the bit error rate of the conventional QPSK and PC-16PSK with the same bandwidth. A study on the error rate equation of QPSK considering phase error in AWGN channel has already been obtained and leads to the conditional bit error rate of QPSK and PC-16PSK considering phase error.

First, a bit error rate of QPSK is derived in consideration of the phase error in the AWGN channel. Assuming that the error phase is generated by ψ, the received signal is given by Equation (1). In other words,

The modulation phase is given by the following equation (40).

The in-phase signal detected by the AWGN channel is as follows.

Here, n _I denotes a Gaussian random variable that is noise of an AWGN channel.

The quadrature signals detected in the AWGN channel are as follows.

Here, n _Q represents a Gaussian random variable of the AWGN channel.

The conditional bit error rate of QPSK considering phase error can be obtained as follows.

The bit error rate of the final QSK is averaged for the error phase as follows.

Next, a bit error rate of PC-16PSK is derived.

Since PC-MPSK is m = 4, it can be expressed as follows from Formula (15) and Formula (13).

Assuming the error phase ψ as a conditional constant, the normalized signal distance (vector size) for the synchronous detection signal vector is obtained as

The exact conditional bit error formula considering the phase error of PC-16PSK is as follows.

Here, A _j represents the probability that the received signal vector enters R _j when X = 0, and is defined as follows.

Here, the bit error rate may be calculated by dividing the upper limit value and the lower limit value as follows.

Therefore, the conditional bit error rate equation of PC-16PSK considering the phase error can be obtained by using the above equation as shown in the following equation.

The average of the error phases is as follows.

The PC-PSK inverts and transmits the second half phase of the transmitted signal, and the receiver inverts the second half of the signal to perform vector synthesis. This vector synthesis cancels phase errors generated during transmission and finds the original information signal. This PC-PSK performance is applied to the AWGN channel, slow-dried Rician channel and time-varying Rician channel.

First, in the case of α ₀ = 8, 10, 15 [dB] in the AWGN channel, the performance of QPSK and PC-16PSK is shown in FIG. As a specific example, when α ₀ = 10dB, QPSK shows error congestion after having a bit error rate of about 10 ⁻² at E _b / N ₀ = 14 [dB]. On the other hand, PC-16PSK improves performance due to lower bit error rate as E _b / N ₀ increases. However, the performance of PC-16PSK is worse than QPSK in the low range of E _b / N ₀ 12 [dB]. This is because the crystal region is multiplied to 16, narrowed, and the reception signal power is minute to reduce the effect of phase compensation and to be more affected by white noise.

In addition, QPSK performance is congested, even if E _b / N ₀ increases, bit error rate does not improve. This is because the signal vector has already passed to the error domain due to the phase error.However, the PC-16PSK method compensates for the error phase, so that the error congestion decreases as the E _b / N ₀ increases, resulting in improved performance. And as shown in FIG. 7, the basic performance is greatly affected by the value of α ₀ .

Second, in the case of α ₀ = 15 [dB] in the slow fading Rician channel, if the bit error rate of QPSK and PC-16PSK is shown for ζ = 0, 5, 10, 20, and 30 with different degree of Rician fading, As shown in the figure. When ζ = 0, it is a Rayleigh fading channel, and when ζ = 30, it is a Rician channel having strong straight waves. It is almost the same as that of the AWGN channel.

In areas where E _b / N ₀ is low, PC-16PSK performs worse than QPSK. This is because even if the phase error is compensated as in the AWGN channel, the crystal region is polyphased and narrowed to 16 regions, and thus is affected by white noise. QPSK degrades due to error congestion in the high E _b / N ₀ region. This is because the signal vector is moved to the error region. Even if E _b / N ₀ is increased, the performance is not improved. However, the PC-16PSK method compensates for the phase when E _b / N ₀ is increased, which reduces error congestion and improves performance. The performance of PC-PSK for the slow-moving Rician fading channel is determined by the values of α ₀ and ζ, and the changing trend is the same as that of the AWGN channel.

Third, in the time-varying Rician channel, if B ₀ / B _{L 0} = 0.5, 0.8, 1.0, and α ₀ = 10dB, the bit error rates of QPSK and PC-16PSK for ζ = 10 are as shown in FIG.

The PC-16PSK is almost unaffected by fading because it compensates for error phases in the range of B ₀ / B _L0 1. For QPSK, when B ₀ / B _L0 = 0.1, the performance is almost the same as that of a slow lysian fading channel. This means that the PLL can sufficiently track the phase of the carrier and does not affect performance. However, since B ₀ / B _{L 0} 0.5, the performance of QPSK is degraded. This is because the signal vector gets into the error region due to the fading effect, and even if B ₀ / N ₀ is increased, the performance decreases due to the error congestion phenomenon. On the other hand, PC-16PSK performance is hardly affected by time-varying lysian fading. This indicates that the PC-16PSK scheme compensates for the error phase due to fading in the high B ₀ / N ₀ region. Here, as in the slow-moving lysian fading channel, it can be seen that the basic change characteristics are determined by the values of α ₀ and ζ. As the value of B ₀ / B _L0 increases, the performance deteriorates.

As compared with the above analysis, PC-PSK can reduce the performance degradation of bit error rate caused by AWGN, slow-dried lysian closed fading and time-varying lysian fading effects. This indicates that good performance can be obtained in a relatively high B ₀ / L ₀ region. On the other hand, in the low B ₀ / N ₀ region, performance degradation occurs. This is because the crystal region is polymorphized and narrowed so that it is affected by white noise.

The main band width of PC-16PSK is 1 / 2T _b [HZ] if the bit period is T _b , which is the same as the main band width of QPSK. In particular, the practical width of PC-8PSK is 1 / 1.5T _b [HZ], which is about 4/3 times larger than that of QPSK. Can be.

As described above, the present invention relates to a new phase compensation PSK (PC-PSK) using a phase inversion method that corrects an error phase between +90 degrees and -90 degrees. The PC-PSK inverts and synthesizes the second half of the received signal so that phase errors caused by fading during transmission, mutual interference, and non-linearity of the receiving amplifier cancel each other out. At this time, the signal component is inverted and transmitted so that it is not canceled. The received signal power is smaller as the error phase to be compensated for is larger and the bandwidth is increased due to the inversion of the latter signal during transmission. To reduce bandwidth, PC-PSK must be polyphased.

For numerical consideration of PC-PSK performance, the conventional QPSK and the PC-16PSK according to the present invention were compared for the AWGN channel, the slow-moving lysian fading channel and the time-varying lysian fading channel.

When the PC-PSK is polyphased, the crystal region is narrowed, so the signal power is small in the relatively low E _b / N ₀ region, thereby weakening the effect of phase compensation and being affected by white noise. On the other hand, in the relatively high E _b / N ₀ region, the congestion of bit error rate due to phase error or fading is reduced and the performance is greatly improved.

Therefore, the PC-PSK according to the present invention may be a new digital modulation and demodulation method that can transmit data with high quality even in a communication environment with a high phase error such as mobile communication.

Claims (3)

A digital phase shift modulation method for converting an input binary signal composed of m binary data with an information symbol section into an output waveform having one of a predetermined phase state, the method comprising: A first step of dividing the information symbol section into two parts; In generating the I-axis signal to be modulated by the in-phase carrier signal and the Q-axis signal to be modulated by the quadrature carrier signal in the information symbol section, in the first half of the information symbol section, the I-axis signal by the phase vector according to the input binary signal. And a second process of generating a Q-axis signal and generating an I-axis signal and a Q-axis signal by a phase vector in which the phase vector of the input binary signal is inverted with respect to the I-axis in the second half of the information symbol section. Phase shift modulation method characterized in that. 2. The method of claim 1, wherein in the first process, the input signal is divided into two parts at an intermediate point of an information symbol period of the input binary signal. In the digital phase shift demodulation method for restoring an input binary signal composed of m binary data and having a predetermined information symbol section is phase shifted by a predetermined modulation method and then transmitted to the original signal. Restoring a reproduction carrier signal from the transmitted modulation signal to generate an in-phase carrier signal and a quadrature carrier signal; Generating a first I-axis signal detected by the in-phase carrier signal and a first Q-axis signal detected by the quadrature carrier signal in the first half of an information symbol section; Generating a second I-axis signal detected by the in-phase carrier signal and a second Q-axis signal detected by the quadrature carrier signal in a second half of an information symbol section; Generating an I-axis detection signal by adding the first I-axis signal component and the second I-axis signal component, and subtracting the second Q signal component from the first Q-axis signal component to generate a Q-axis detection signal, Synthesizing the I-axis detection signal and the Q-axis detection signal with each other; And generating a demodulated signal for the transmitted modulated signal based on the synthesized signal vector.