JPH0879320A - Constant envelope polyphase modulator - Google Patents

Abstract

PURPOSE: To obtain amplitude of each channel corresponding to a signal phase by allowing a read means to sequentially read an amplitude value at each sample point from an amplitude hold means depending on two consecutive symbol inputs. CONSTITUTION: A read means 114 reads I channel and Q channel amplitude values at each sample point corresponding to a current symbol and a just preceding symbol. Thus, the result of sampling transition of a signal phase from a just preceding symbol to a current symbol at the prescribed number of sample points is obtained. The amplitude values of I and Q channels obtained by the means 114 are given to an A/D converter means 115, in which an envelope of a modulation signal by an orthogonal transformation means 116 is made constant. Moreover, the read means 114 and the amplitude value hold means 113 are used to realize functions of a phase modulator, a phase arithmetic unit and a waveform generator with a simple configuration thereby improving the power efficiency.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital modulator for digitally modulating information in mobile communication, and more particularly,
The present invention relates to a polyphase PSK modulator using a polyphase phase keying (PSK) system.

A digital modulator provided in a mobile communication device such as a car telephone is required to satisfy the following conditions. The first condition is that the spectrum characteristic has a narrow band.

In recent years, the number of mobile communication devices has significantly increased with the spread of automobile telephones and the like, and a large number of mobile communication devices share a limited frequency band for use. Is an indispensable condition.

The second condition is that the signal is efficiently modulated and amplified with a limited electric power, that is, the transmission amplification has high power efficiency, and the third condition is that the receiving side has a high power efficiency. It has a high sensitivity characteristic that makes it possible to obtain a demodulated signal of quality.

These conditions impose two different approaches to adapting to a system where the available power is severely limited: an approach to obtain a stronger transmitted signal with less power, and a power of the transmitted signal. Even if it is small, it corresponds to the approach to enable the receiving side to handle it.

Recently, users of mobile communication devices have
There is a demand for downsizing and power saving of mobile communication devices, and in order to meet these demands, mobile communication devices are becoming a system in which available power is severely limited. Therefore, it has been emphasized as a new technical theme to develop a digital modulator that satisfies the above-mentioned second and third conditions and meets a strict power limit.

[0007]

2. Description of the Related Art A typical digital modulation method that satisfies the first condition is π / 4 shift quadrature phase modulation (π / 4
There is a Shift Quadri-Phase shift Keying method (Akaiwa
Y.andNagata Y.:"Highly efficient digital mobile c
ommunications with a linearmodulation method ", IEEE
Journal on Selected Areas in Commun. SAC-5,5, pp89
0-895 June 1987).

This π / 4 shift quadrature phase modulation system uses π
It is a kind of four-phase phase modulation method using four carrier waves each having a phase difference of / 2, and has a characteristic that it is relatively strong against nonlinear distortion because its peak power is small.

In addition, an amplitude / phase modulation method in which an amplitude modulation method and a quadrature phase modulation method are combined has been studied, and in particular, multi-valued quadrature amplitude modulation (Quadrature Amplitude Modulat).
Ion) method is being actively researched.

However, since the output signal of the modulator to which these methods are applied has large fluctuation in amplitude, the modulator output is amplified by the linear amplifier, and the efficiency of this linear amplifier is the limit of improvement in power efficiency. Has become.

On the other hand, as a digital modulation method satisfying the second condition, since a non-linear amplifier capable of obtaining high power efficiency can be used, the constant envelope modulation method in which the amplitude of the output signal does not change is noted. Has been done.

A typical example of the constant envelope modulation method is TFM (Tamed) which is a kind of digital FM method.
Frequency Modulation) method (de Jager F. and Dekker
CB: "Tamed frequency modulation, a novel method to
achive spectrum economy indigital transmission ", I
EEE Trans.Commun., COM-26,5.pp.534-542 May 1978)
And GMSK system (Murota K. and Hirade K .: “GMSK M
odulation for digitalmobile radio telephony ", IEEE
Trans.Commun., COM-29,8.pp.1044-1050 July 1981).

As a digital modulation method satisfying the third condition, a method in which coding and modulation are combined has been studied. As a typical example, trellis coding 8-phase P
The SK (Trellis-Coded 8 Phase Shift Keying) method has been proposed.

This trellis coded 8-phase PSK system is a combination of convolutional coding and 8-phase PSK, and it has been reported that a sensitivity gain of about 3 dB can be obtained when compared with QPSK in the same band. (Ungerb
oeck G .: "Channel coding with phase signals", IEEE Tr
ans.Inf.Thoery, IT-23,1, pp55-67 Jan.1982).

FIG. 11 shows the configuration of a trellis-coded 8-phase PSK modulator. In FIG. 11, the serially input data is output by the serial-parallel converter 301.
The data is converted into 2-bit parallel data, and further converted into a 3-bit code by the trellis encoder 302.

In response to the input of this 3-bit code, the code conversion circuit 303 adds an appropriate 1 bit to form a 4-bit code, and the 2-bits allocated to the I channel are each converted into two binary transversal filters ( BT
F) It is sent to 304 _I1 and 304 _I2, and 2 bits assigned to the Q channel are sent to two binary transversal filters 304 _Q1 and 304 _Q2 in the same manner.

The outputs from these binary transversal filters 304 _I1 , 304 _I2 , 304 _Q1 , 304 _Q2 respectively correspond to weighting circuits 305 _I1 , 305.
The weights of the constant K ₁ or the constant K ₂ are given by _{I 2} , 305 _Q1 and 305 _Q2 , and the results are added by the adders 306 _I and 306 _Q corresponding to the respective channels to obtain the signals of the I channel and the Q channel. Each is configured to limit the band.

The I-channel and Q-channel signals thus band-limited are converted into analog signals by the corresponding digital-analog conversion circuits (D / A) 307 _I and 307 _Q , and the low-pass filter (L
After being smoothed by PF) 308 _I and 308 _Q , it is quadrature-modulated by a quadrature modulator 309 and output.

As described above, in this system, 2-bit input data is converted into 3-bit at the trellis encoding stage,
Then, after being converted into 4 bits for distribution to the I and Q channels, the band is limited and the quadrature modulation is performed. That is, it is intended to obtain high sensitivity characteristics on the receiving side by modulating and transmitting including redundant information.

In the above-mentioned trellis-coded 8-phase PSK system, since the band is limited by the binary transversal filter 304 shown in FIG. 11 irrespective of the amplitude variation, the envelope of the output signal varies greatly. Therefore, improvement is needed in terms of power efficiency.

Accordingly, trellis coded 8-phase PSK
Techniques aimed at improving power efficiency by controlling the envelope of the output signal of the system have been proposed (Shigeru Tomisato, Suzuki.
Hiroshi, "Envelope-controlled digital modulation system with improved power efficiency in transmission amplification-Trellis coding 8P for mobile radio
Application to SK- "Transactions of the Institute of Electronics, Information and Communication Engineers B-II Vol.J7
5-B-II No.12 pp912-928 December 1992).

In the above-mentioned paper, a smoothed envelope linear trellis coding for suppressing fluctuation of the envelope by setting an upper limit or a lower limit on the amplitude values of the I channel and the Q channel at the midpoint of the envelope connecting the two signal points. An 8PSK modulator and a constant envelope linear trellis coded 8PSK modulator that realizes a constant envelope by controlling the amplitude values of the I channel and the Q channel by using a cosine roll-off function have been proposed. Since the envelope linear trellis coded 8PSK modulator can eliminate the amplitude fluctuation of the output signal, a significant improvement in power efficiency is expected.

FIG. 12 shows the configuration of the constant envelope linear trellis coded 8PSK modulator proposed in the above-mentioned paper. 12, the constant envelope linear trellis coded 8PSK modulator is a phase converter 321, instead of the code conversion circuit 303, the binary transversal filter 304 and the weighting circuit 305 of the trellis coded 8PSK modulator shown in FIG. The phase calculator 322 and the waveform converter 323 are provided, and the output of the waveform generator 323 is used as the quadrature modulator 309.
It is configured to input to.

In this constant envelope linear trellis coded 8PSK modulator, the phase converter 321 converts the 3-bit symbols obtained by the trellis encoder 302 every symbol time T _s into the corresponding signal phase φ at 8PSK.
And is input to the phase calculator 322.

Further, the phase calculator 322 determines the n-1th symbol based on the signal phase φ _n corresponding to the nth symbol and the signal phase φ _n-1 corresponding to the n-1th symbol. each time t _n-1 + t included in the period from the corresponding time t _n-1 to time t _n corresponding to the n th symbol
The signal phase P _nm at _m (0 <t _m <T _s ) is obtained, and the waveform converter 323 determines the time t based on the signal phase P _nm.
It is configured such that I (t) and Q (t) are obtained as a continuous function for and are sent to the quadrature modulator 309.

Here, the phase calculator 322 first obtains the phase difference θ _n between the signal phase φ _n and the signal phase φ _n−1, and uses this phase difference θ _n and the cosine roll-off function shown in the equation. Then, the signal phase change amount δ _m at time t _n-1 + t _m with reference to the signal phase φ _n-1 is obtained, and this signal phase change amount δ _m is added to the signal phase φ _n-1 to obtain the signal phase P _nm should be calculated.

[0027]

[Equation 1]

However, in the equation, the constant α is the roll-off rate. By obtaining the signal phase change amount δ _m using this roll-off function, the phase change can be smoothed to narrow the band, and as will be described later, it is obtained based on this phase change amount δ _m. As the cosine and sine of the signal phase P _nm to be generated, the amplitude values I (t), Q of the I channel and Q channel
By obtaining (t), it is possible to eliminate the amplitude fluctuation of the output signal and obtain the output signal of the constant envelope.

Further, the waveform converter 323 determines that the time t _n-1 +
From t _m to time t _n−1 + t _{m + 1} , the amplitude values I (t) and Q (t) of the I channel and Q channel are constant values cos.
The waveforms of (P _mn ) and sin (P _mn ) may be generated.

[0030]

The above-mentioned constant envelope control type trellis coded 8PSK modulator has a high sensitivity characteristic on the receiving side obtained by adopting trellis coding and a constant envelope control. It also has high power efficiency on the transmission side.

However, in the above-mentioned paper,
Although the principle of the constant-envelope linear trellis coded 8PSK modulator has been described, no example has been published in which this modulator is actually realized by a circuit or the like.

Therefore, how is the constant envelope control type trellis coded 8PSK modulator shown in FIG.
A future issue is how to realize it as an actual circuit. An object of the present invention is to provide a constant-envelope linear polyphase modulator with a simple hardware configuration.

[0033]

FIG. 1 is a principle block diagram of a constant-envelope linear polyphase modulator according to claims 1 to 3.

According to a first aspect of the present invention, the conversion means 111 converts the information to be transmitted into a series of symbols corresponding to the signal phase and outputs each symbol at a predetermined symbol time.
The delay means 112 for delaying a series of symbols by the symbol time and all combinations considered as symbol transitions are obtained by sampling signal phase transitions represented by a predetermined roll-off function at a predetermined number of sample points. Corresponding to a combination of corresponding symbols according to the input of the amplitude value holding means 113 for respectively holding the amplitude values of the I channel and the Q channel, and the current symbol obtained from the conversion means 111 and the immediately preceding symbol obtained from the delay means 112. Read means 114 for reading the amplitude values of the I channel and Q channel at each sample point held in the amplitude value holding means 113, and analog conversion means 115 for converting the amplitude values of the I channel and Q channel into analog signals. , The I channel obtained by the analog conversion means 115 and Characterized in that a orthogonal transform unit 116 to orthogonal transform an analog signal of Q channel.

According to a second aspect of the invention, in the constant envelope linear polyphase modulator according to the first aspect, the amplitude value holding means 113 is provided.
Is an I channel obtained by sampling the signal phase transitions represented by the roll-off functions respectively defined by a plurality of roll-off ratios at all the combinations considered as the symbol transitions at a predetermined number of sampling points. And the amplitude values of the Q channel are respectively held, and the reading means 114 at each sample point held in correspondence with the corresponding roll-off function from the amplitude value holding means 113 according to the input of the roll-off rate. It is characterized in that the amplitude values of the I channel and the Q channel are read out.

According to a third aspect of the present invention, in the constant envelope linear polyphase modulator according to the first aspect, the positions of a predetermined number of sample points on the circumference corresponding to the signal phase changing at a constant angular velocity are set to I. The quadrature modulator 1 is based on the test data holding unit 121 that holds the test data of the amplitude values of the channel and the Q channel, and the carrier leak included in the spectrum of the modulated signal obtained as the output of the quadrature modulator 116.
Offset detection means 122 for detecting 16 offsets
And an amplitude ratio deviation detecting means 123 for detecting an amplitude ratio deviation of the quadrature modulator 116 based on an image included in the spectrum of the modulation signal, and the reading means 114 performs a test in response to the input of the test mode instruction. Data holding means 12
It is characterized in that the amplitude values of the I channel and the Q channel corresponding to each sample point are read out from 1.

FIG. 2 is a block diagram showing the principle of the constant envelope linear polyphase modulator of claim 4. According to a fourth aspect of the present invention, in the constant envelope linear polyphase modulator according to the third aspect, I read from the amplitude value holding means 113 in response to the input of the offset value detected by the offset detection means 122. The offset correction means 131 for correcting the amplitude values of the channel and the Q channel, and the amplitude value holding means 113 according to the input of the amplitude ratio deviation detected by the amplitude ratio deviation detection means 123.
And an amplitude ratio deviation correcting means 132 for correcting the amplitude values of the I channel and the Q channel read from.

[0038]

According to the invention of claim 1, the reading means 114 reads the amplitude values of the I channel and the Q channel at each sample point corresponding to the current symbol and the immediately preceding symbol from the amplitude value holding means 113, and The result of sampling the signal phase transition to the current symbol at a predetermined number of sample points can be obtained.

Since the number of symbols is finite in the polyphase modulation method, the combination is also finite. Therefore, if the roll-off rate is determined, it is possible to calculate in advance the amplitude value corresponding to each sample point of the transition between symbols corresponding to each combination of symbols.

That is, the functions of the phase converter 321, the phase calculator 322, and the waveform generator 323 proposed in the above-mentioned paper can be realized with a simple structure by the above-mentioned reading means 114 and the amplitude value holding means 113. And the amplitude values of the I and Q channels obtained by the reading means 114 are passed through the analog converting means 115 to the quadrature modulating means 1.
By sending it to 16, the modulation signal by the quadrature modulation means 116 can be made into a constant envelope.

According to the second aspect of the invention, the reading means 114 selectively reads the amplitude value corresponding to the corresponding roll-off rate from the amplitude value holding means 113 in response to the input of the roll-off rate, so that various applications can be achieved. It is possible to limit the band of the modulation signal by changing the roll-off rate according to the above.

According to the third aspect of the present invention, the reading means 114 reads the test data from the test data holding means 121, so that a signal changing at a constant angular velocity can be input to the quadrature modulating means 116.

At this time, the spectrum of the output signal of the quadrature modulating means 116 is observed by the offset detecting means 122 and the amplitude ratio deviation detecting means 123, and the carrier leak and the image contained in the modulated signal are observed, so that the quadrature modulating means are respectively obtained. The offset of 116 and the deviation of the amplitude ratio can be detected.

Further, according to the invention of claim 4, the offset correcting means 131 and the amplitude ratio deviation correcting means 132 are configured to read the reading means 1 according to the detected offset and amplitude ratio deviation.
By correcting the amplitude values of the I channel and the Q channel read out by 14, the quadrature modulation means 116
It is possible to correct the offset and the deviation of the amplitude ratio.

[0045]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 3 is a block diagram of an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of the present invention is applied.

3, the constant envelope linear trellis coded 8PSK modulator of the present invention is the phase converter 321, the phase calculator 322, and the waveform generator 323 of the constant envelope linear trellis coded 8PSK modulator shown in FIG. Instead, the counter circuit 210, two flip-flop circuits 220 _I and 220 _Q corresponding to the I channel and the Q channel, and two look-up tables (LUT) 230 _I and 230 _Q are provided.

Also, this constant envelope linear trellis coding 8P
In the SK modulator, the 3-bit code obtained by the trellis encoder 302 corresponding to the conversion unit 111 is the lookup table 230 described above as the current symbol S _n.
I , 230 _Q and after being delayed by one symbol time by the flip-flop circuit 220 corresponding to the delay means 112, they are sent to the look-up tables 230 _I , 230 _Q as the immediately preceding symbol S _n-1. It is composed.

These look-up tables 230 _I ,
230 _Q is equivalent to the amplitude value holding means 113, and holds a value obtained by sampling the change in the amplitude value of each channel at 2 ^m sample points when transitioning between any two symbols. It is configured to do.

Here, the lookup table 230 _I corresponding to the I channel corresponds to each combination of the current symbol S _n and its immediately preceding symbol S _n-1 as shown in FIG. 4 (a). as the amplitude value I in 2 ^k sample points _{(t m) (m = 1~2} k), the corresponding signal phase P _nm
It is configured to hold the cosine value corresponding to. Also,
Look-up table 230 _Q, as shown in FIG. 4 (b), in response to each combination of the current symbol S _n and the immediately preceding symbol S _n-1, the amplitude value of 2 ^k sample points Q ( t _m ) (m = 1 to 2 ^k ) holds the sine value corresponding to the corresponding signal phase P _nm .

Further, a clock signal for dividing the symbol time Ts into 2 ^k sample points is input to the counter circuit 210 shown in FIG.
10 counts in synchronization with this clock signal,
The k-bit count value obtained as a result is used as the lower part of the address, and the lookup tables 230 _I , 23
It is configured to input to both 0 _Q.

That is, by inputting the current symbol S _n from the trellis encoder 302 and the immediately preceding symbol S _n-1 from the flip-flop circuit 220 as the upper 6 bits of the address, the transition from the immediately preceding symbol to the current symbol is performed. At this time, a function representing the change in the amplitude value of each channel is specified, the corresponding sample point is specified by the count value of the counter circuit 210 input as the lower k bits of the address, and the look-up tables 230 _I , 230. _The configuration is such that each appropriate amplitude value is read from _Q.

That is, the counter circuit 220 and the structure of the data stored in the look-up tables 230 _I and 230 _Q perform the function of the reading means 114, and look up the amplitude value of each channel at each point of transition between symbols. The tables 230 _I and 230 _Q are read from the respective tables.

Further, in FIG. 3, digital-analog converters (D / A) 307 _I and 307 _Q and low-pass filters 308 _I and 308 _Q are analog conversion means 11 respectively.
The output of the lookup tables 230 _I and 230 _Q corresponding to each channel is converted into an analog signal and is input to the quadrature modulator 309 corresponding to the quadrature modulator 116. .

As described above, the signal phase P _nm corresponding to the m-th sample point on the transition path from the immediately preceding symbol S _n-1 to the current symbol S _n is the signal of the current symbol S _n as expressed by the equation. It is obtained by adding the signal phase change amount δ _m obtained as a function of the phase φ _{n to} the signal phase φ _n-1 of the immediately preceding symbol S _n-1 .

In addition, in the 8PSK modulation method, the number of symbols is 8, and the combinations of transitions between symbols are finite. Therefore, a predetermined cosine roll-off function is set in advance for each combination of transitions between symbols. The signal phase P _nm corresponding to each sample point can be obtained by using this.

If this signal phase P _nm is used, the amplitude value I (tm) = cos (P _nm ) of the I channel and the amplitude value Q (tm) = sin (P _nm ) of the Q channel are calculated, It can be stored in the lookup tables 230 _I and 230 _Q , respectively.

Here, the time t _m indicating each sample point is
Period T _s / 2 ^{k of} clock signal corresponding to sample time
And the number m that identifies the sample point, as shown in the equation.

T _m = m · T _s / 2 ^k ... Using this equation, an appropriate roll-off rate α (for example, α
= 0.5) is set and the formula is divided into cases in the range of the sample point number m, and the formula is applied according to this case,
As shown in the equations, the amplitude values I (tm) and Q (tm) of the I channel and Q channel at each sample point may be calculated respectively.

[0059]

[Equation 2]

[0060]

(Equation 3)

As can be seen from the equations, the amplitude values I (tm), Q (tm) of each channel calculated as described above.
Is the signal phase φ corresponding to the immediately preceding symbol S _{n-1.
From n-1} to the signal phase φ _n corresponding to the current symbol S _n ,
It is the cosine value and sine value of the signal phase P _nm that changes smoothly according to the roll-off function. Therefore, if the transition of the signal point at this time is plotted in the signal space, it is clear that the locus moves on the arc connecting the current symbol S _n and the immediately preceding symbol S _n-1 .

For example, as shown in FIG. 5, when the symbol indicated by the symbol and the symbol indicated by the symbol in FIG. 5 are input as the immediately preceding symbol S _n-1 and the current symbol S _n , respectively, the counter circuit 210 counts. Depending on the numerical value, the look-up tables 230 _I and 230 _Q output amplitude values I (tm) and Q (tm) corresponding to the locus moving on the arc connecting the signs.

That is, by sequentially reading the amplitude values I (tm) and Q (tm) at the corresponding sample points from the above-mentioned lookup tables 230 _I and 230 _Q , the phase converter 321 proposed in the above-mentioned paper is read. , Phase calculator 3
22 and the waveform generator 323 can be realized, and the modulation signal of the trellis coded 8PSK modulator can be a constant envelope signal.

As a result, it is possible to realize with a simple hardware a modulator having both high reception sensitivity and high power efficiency of the constant envelope modulation method, which are features of the multi-phase PSK modulator adopting trellis coding. Becomes

The number of sampling points when sampling transitions between symbols is determined by the symbol time T _s , the speed of the clock signal, and the look-up tables 230 _I and 230 _Q.
It may be determined in consideration of the processing speed of the circuit such as.

In reality, the lookup table 230 _I ,
Since the read speed of the ROM used to construct the 230 _Q is increasing, it is considered possible to secure a sufficient number of sample points within one symbol time.

Further, the lookup table 230 _I , 2
The accuracy of the amplitude values I (tm) and Q (tm) stored in 30 _Q is
It may be determined in consideration of the phase accuracy required for the modulator. For example, the phase accuracy required by the 4PSK modulator is about 2 °, but the bit accuracy of the amplitude value corresponding to this phase accuracy is 5 bits, and if the bit accuracy of the amplitude value is 8 bits, it becomes 0. Since a phase accuracy of about 0.5 ° can be obtained, it is considered possible to secure sufficient phase accuracy. Further, the number of bits representing the amplitude value can be increased to further improve the accuracy.

Further, in the case where various roll-off rates are applied to the cosine roll-off function as shown in the equation, a plurality of I-channel amplitude value tables and a plurality of Q-channel amplitude value tables are prepared respectively. Good.

FIG. 6 shows a block diagram of an embodiment of a trellis coded 8PSK modulator to which the invention of claim 2 is applied. In FIG. 6, the trellis coded 8PSK modulator is configured by adding the roll-off rate input unit 241 to the trellis-coded 8PSK modulator shown in FIG. 3, and the roll-off rate input unit 241 is used by the user. In response to an instruction from, the code corresponding to the specified roll-off rate is output, and this code is used as a part of the address in the lookup table 230 _I ,
It is configured to input to 230 _Q.

In this case, when various roll-off rates are applied, the amplitude values at the time of sampling the transitions of all combinations of symbols are calculated, and the amplitude values are calculated in accordance with the code indicating the roll-off rate. It may be stored in advance in the lookup tables 230 _I and 230 _Q.

That is, in this case, the look-up tables 230 _I and 230 _Q may be configured to have a plurality of tables corresponding to codes showing various roll-off rates, as shown in FIG.

In addition, the code indicating the roll-off rate is included in the address, and the lookup tables 230 _I and 230 _{Q are included.}
The function of the reading means 114 described in claim 2 may be realized by adopting a configuration in which the corresponding amplitude value is read from each of the above.

In this case, the phase transition between the symbols can be smoothed by the cosine roll-off function to which an appropriate roll-off rate is applied according to the user's instruction, and according to the designated roll-off rate. Occupied bandwidth can be controlled.

As a result, depending on the application of the modulator, etc.
Bandwidth can be effectively limited. Further, it is also possible to detect the offset and the amplitude ratio shift of the quadrature modulator by utilizing the characteristics of the constant envelope modulation method.

FIG. 8 is a block diagram showing an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of claim 3 is applied. In FIG. 8, trellis coded 8PSK
As the modulator, a mode switching signal input unit 251 and a selector 252 are added to the trellis coded 8PSK modulator shown in FIG. 3, and the mode switching signal input unit 251 is set in the test mode in response to an instruction from the user. The test mode code to that effect is output, and the selector 252 inputs this test mode code to the lookup tables 230 _I and 230 _Q instead of the immediately preceding symbol and the current symbol.

Here, the mode switching signal input section 251
A bit string that does not appear as a combination of the immediately preceding symbol and the current symbol may be generated as the test mode code, and in this case, the lookup table 23
For 0 _I and 230 _Q , the test data may be stored in the storage area indicated by the test mode code described above in correspondence with the count value of the counter circuit 210.

For example, from phase 0 (rad) to phase 2π (rad)
Are equally divided into 2 ^k sections, and the cosine and sine corresponding to the phase of each partition point are used as the test data of the I channel and the Q channel, respectively, and the lookup table 230
_It should be stored in _I , 230 _Q.

In this way, the test data can be read from the look-up tables 230 _I and 230 _Q according to the instruction from the user, and the test data holding means 121 can be read by the look-up tables 230 _I and 230 _Q. The function of can be realized.

Here, the test data as described above is
Since a state in which the test data is rotated on the circumference at a constant angular velocity ω ₀ is shown, the spectrum of the modulation signal obtained when this test data is input to the ideal quadrature modulator is shown in FIG.
As shown in (a), one spectrum should appear at the position represented by the carrier frequency f ₁ and the frequency f ₀ corresponding to the angular velocity ω ₀ .

However, if the actual quadrature modulator 309 has an offset or an amplitude ratio deviation, the spectrum of the modulated signal will have the spectrum of the original modulated signal as shown in FIG.
The presence of the offset causes the quadrature modulator 3 to
Carrier leak occurs at the position of carrier frequency f ₁ of 09, and due to the presence of the amplitude ratio shift, carrier frequency f
_An image is generated at a position symmetrical to the original frequency of the modulated signal with respect to ₁ (that is, at a position of frequency f ₁ −f ₀ ).

Therefore, as shown in FIG. 8, the output of the quadrature modulator 309 is configured as a spectrum analyzer 254 by using a measuring device such as a multi-channel analyzer, and the carrier leak and the image are observed to detect the offset. It is possible to realize the functions of the means 122 and the amplitude ratio deviation detecting means 123, detect the offset and the amplitude ratio deviation, and measure the magnitude thereof.

For example, when there is an offset on the I channel side, the amplitude value I (t) of the I channel is set to the above-mentioned angular velocity ω.
It is equivalent to the case where it is expressed as an expression using ₀ . I (t) = (I _of + COSω ₀ t) Therefore, the output signal S (t) of the quadrature modulator 309 includes the component of the original modulation signal and the component generated by the offset as shown in the equation. It will be.

S (t) = I _of sin ω ₁ t + sin (ω ₁ t + ω ₀ t) ... The first term on the right side of the equation shows the carrier leak generated by the offset, and in this case, it is observed as the carrier leak. The power P _{CL to} be generated is expressed by an equation using the offset value I _of .

P _CL = 20log (I _of / 1) [dB] Therefore, the spectrum analysis unit 254 measures the carrier leak power P _CL and calculates the offset value I _of from the above formula. Good.

When the amplitude ratio C is deviated from the normal value "1" in the quadrature modulator 309, the modulation signal S (t)
Splits into two frequency components, as shown in the equation.

[0086]

[Equation 4]

The second term on the right side of the equation shows an image generated by the deviation of the amplitude ratio, and the image amount P _IM observed in this case is represented by the equation using the amplitude ratio C.

[0088]

(Equation 5)

Therefore, the spectrum analysis unit 254
It suffices to measure the image amount P _IM , calculate the amplitude ratio C conversely from the above formula, and evaluate the amplitude ratio deviation. As described above, according to the present invention, it is possible to easily detect the offset and the amplitude ratio deviation and evaluate the magnitude thereof.

If the output device 255 such as a display device or a printer is provided and the evaluation result obtained by the spectrum analysis section 254 is output and provided to the user, the performance of the modulator can be evaluated precisely. This makes it possible to efficiently carry out modulator development and improvement work.

Further, the amplitude values I (t) and Q (t) of each channel can be corrected by using the values of offset and amplitude ratio deviation obtained by the above-described measurement. Figure 10
FIG. 9 shows a configuration diagram of an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of claim 4 is applied.

In FIG. 10, trellis coded 8PSK
The modulator adds a correction code conversion unit 261 to the trellis coded 8PSK modulator shown in FIG. 8, and the correction code conversion unit 261 inputs the offset value and the amplitude ratio measured by the spectrum analysis unit 254 described above. In response, the corresponding correction code is output and is input to the lookup tables 230 _I and 230 _Q as a part of the address.

In this case, the lookup table 230
_I and 230 _Q are the amplitude value I (t which is corrected by the offset value I _of or the amplitude ratio C corresponding to each correction code.
It is sufficient to store m) and Q (tm) for each sample point.

For example, the lookup table 230
_I is provided with a table of amplitude values I (tm) corresponding to all combinations of symbols corresponding to each of the correction codes indicating various offset values I _of , and each table has a corresponding roll-off function. normal may be stored the value obtained by subtracting the appropriate offset values I _of the respective amplitude value I (tm) calculated based.

Similarly, the lookup table 230
_{If Q} is provided with a table corresponding to each of the correction codes indicating various amplitude ratios C, and a value obtained by multiplying each normal amplitude value Q (tm) by the amplitude ratio C is stored in each table. Good. Further, in this case, the correction code conversion unit 2
The reference numeral 61 may send the correction code corresponding to the offset value to the lookup table 230 _I and the correction code corresponding to the amplitude ratio to the lookup table 230 _Q.

In this way, the lookup table 230
By storing the corrected data in _I and 230 _Q in advance and reading the corresponding data according to the correction code indicating the analysis result by the spectrum analysis unit 254, the lookup tables 230 _I and 230 _Q With the correction code conversion unit 261, the function of the correction unit 131 can be realized.

As a result, the amplitude value at each sample point can be corrected according to the detection result of the offset and the amplitude ratio shift using the test data to suppress the carrier leak and the image, and the effective use of the electric power can be achieved. Go ahead,
It is possible to improve power efficiency.

[0098]

As described above, according to the present invention, the reading means sequentially reads the amplitude value at each sample point from the amplitude value holding means in response to the input of two consecutive symbols, so that an appropriate roll-off is performed. Since the amplitude value of each channel corresponding to the signal phase that changes according to the function can be obtained, a constant-envelope linear polyphase modulator can be realized with a simple configuration, and a polyphase modulator excellent in band characteristics. It is possible to improve the power efficiency by making the modulation signal of the constant envelope.

Further, according to the second aspect of the present invention, since it is possible to switch a plurality of roll-off functions according to the input of the roll-off rate, effective band limitation is performed according to the application of the modulator. Is possible.

According to the third aspect of the invention, the offset and amplitude ratio deviation of the quadrature modulation means can be detected by examining the output spectrum of the quadrature modulation means when the test data is input, and the modulator performance. Can be evaluated easily.

Further, according to the invention of claim 4, since the offset and the amplitude ratio deviation detected by the offset detecting means and the amplitude ratio deviation detecting means can be corrected by the correcting means, the carrier leak and the image are suppressed, Power efficiency can be improved.

[Brief description of drawings]

1 is a principle block diagram of a constant-envelope linear polyphase modulator of claims 1 to 3;

FIG. 2 is a principle block diagram of the constant envelope linear polyphase modulator of claim 4;

FIG. 3 is a configuration diagram of an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of claim 1 is applied.

FIG. 4 is a diagram showing a configuration example of a lookup table.

FIG. 5 is a diagram illustrating phase transition according to the present invention.

FIG. 6 is a configuration diagram of an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of claim 2 is applied.

FIG. 7 is a diagram showing a configuration example of a lookup table.

FIG. 8 is a configuration diagram of an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of claim 3 is applied.

FIG. 9 is a diagram illustrating a spectrum of a modulation signal in a test mode.

FIG. 10 is a configuration diagram of an embodiment of a trellis coded 8PSK modulator to which the constant envelope linear polyphase modulator of claim 4 is applied.

FIG. 11 is a diagram showing a configuration example of a conventional trellis coded 8PSK modulator.

FIG. 12 is a diagram showing a configuration example of a constant envelope linear trellis coded 8PSK modulator.

[Explanation of symbols]

 111 conversion means 112 delay means 113 amplitude value holding means 114 reading means 115 analog conversion means 116 quadrature modulation means 121 test data holding means 122 offset detection means 123 amplitude ratio deviation detection means 131 correction means 210 counter circuit 220 flip-flop (FF) circuit 230 Look-up table (LUT) 241 Roll-off rate input unit 251 Mode switching signal input unit 252 Selector 254 Spectral analysis unit 255 Output device 261 Correction code conversion unit 301 Serial-parallel converter (S / P) 302 Trellis encoder 303 Code conversion circuit 304 Binary transversal filter (BTF) 305 Weighting circuit 306 Adder 307 Digital-analog converter (D / A) 308 Low-pass filter (LPF) 309 quadrature modulator 321 phase converter 322 phase calculator 323 waveform generator

Claims (4)

[Claims] 1. A conversion means for converting information to be transmitted into a series of symbols corresponding to a signal phase and outputting each symbol at a predetermined symbol time, and a delay means for delaying the series of symbols by the symbol time. And for all combinations considered as the transitions of the symbols, hold the amplitude values of the I channel and the Q channel obtained by sampling the transitions of the signal phase represented by a predetermined roll-off function at a predetermined number of sample points. Corresponding to the combination of the corresponding symbol in accordance with the input of the amplitude value holding means, the current symbol obtained from the conversion means and the immediately preceding symbol obtained from the delay means. Reading means for reading the amplitude values of the I channel and the Q channel at the sample point; A constant envelope comprising: an analog conversion means for converting the amplitude value of each channel into an analog signal; and an orthogonal conversion means for orthogonally converting the analog signals of the I and Q channels obtained by the analog conversion means. Linear polyphase modulator. 2. The constant envelope linear polyphase modulator according to claim 1, wherein the amplitude value holding means has roll-offs defined by a plurality of roll-off rates for all combinations considered as symbol transitions. Each of the I-channel and Q-channel amplitude values obtained by sampling the transition of the signal phase represented by each function at a predetermined number of sampling points is held, and the reading means responds to the input of the roll-off rate. And a constant-envelope linear multi-phase modulator that reads out the amplitude values of the I channel and the Q channel at each sample point held corresponding to the corresponding roll-off function from the amplitude value holding means. . 3. The constant envelope linear polyphase modulator according to claim 1, wherein the positions of a predetermined number of sample points on the circumference corresponding to the signal phase changing at a constant angular velocity are set to the I channel and the Q channel. Test data holding means for holding as the test data of the amplitude value, offset detection means for detecting the offset of the quadrature modulator, based on the carrier leak contained in the spectrum of the modulation signal obtained as the output of the quadrature modulator, An amplitude ratio deviation detecting means for detecting an amplitude ratio deviation of the quadrature modulator based on an image included in a spectrum of a modulation signal; and a reading means, in response to an input of a test mode instruction, the test data holding means. Constant-envelope linear polyphase characterized by reading the amplitude values of the I channel and Q channel corresponding to each sample point from Phase modulator. 4. The constant envelope linear polyphase modulator according to claim 3, wherein the I channel and the Q channel read from the amplitude value holding means in response to the input of the offset value detected by the offset detecting means. And an amplitude correction unit for correcting the amplitude value of the I channel and the Q channel read from the amplitude value holding unit according to the input of the amplitude ratio deviation detected by the amplitude ratio deviation detecting unit. A constant-envelope linear multi-phase modulator, comprising: ratio deviation correction means.