JP2506748B2 - Digital signal transmission method - Google Patents

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital signal transmission method for transmitting digital signals in a multipath fading transmission line such as wireless transmission in urban areas.

2. Description of the Related Art In recent years, even in the field of mobile communication, digitization is progressing due to improvement of confidentiality, sophistication of communication, and compatibility with peripheral communication networks. However, in urban areas where such demand is considered to be most concentrated, communication quality is significantly deteriorated due to multipath due to reflection and diffraction by buildings and other structures. In the case of digital transmission, if the propagation delay time difference between the waves forming the multipath cannot be ignored with respect to the time slot length, the waveform error and the tracking error of the synchronous system significantly deteriorate the code error rate characteristic.

Hereinafter, a first example of the above-described conventional digital signal transmission method will be described with reference to the drawings.

FIG. 17 is a phase transition waveform diagram showing the phase transition of the transmission signal of the digital signal transmission method in the first conventional example.
T indicates a time slot length which is the minimum unit for transmitting one data symbol. When the data is 1, the phase transitions by π, and when the data is 0, the phase transition does not occur. This signal format is called differentially encoded binary phase modulation.

To detect such a transmission signal, for example, differential detection having a delay line of one time slot can be performed. Now, as a typical example of multipath, consider what the detection output signal looks like under a two-wave multipath having a propagation delay time difference τ that cannot be ignored compared to the time slot length. In addition, the wave that precedes in time is a direct wave,
The delayed wave is called the delayed wave.

FIG. 18 is a diagram for explaining what a detection output signal looks like when the transmission signal as shown in FIG. 17 is subjected to delay detection under a two-wave multipath. Figure 18 (a)
Shows the phase transition of the direct wave. On the other hand, the phase transition of the delayed wave delayed by the propagation delay time difference τ is as shown in FIG. 18 (b). The detection output at a certain time point is the vector inner product of the combined phase of the two waves at that time and the combined phase of the two waves one time slot before. For example, the 18th
In FIG. 6C, the detection output in the area B becomes the value of the vector inner product of the two-wave combined phase in the case of B ′ and that in the case of B.

FIG. 19 is a vector diagram showing the combined phase of the direct wave and the delayed wave in order to obtain the detection output at each time point A to C. The amplitude ratio of the direct wave and the delayed wave was ρ, and the phase difference was α. For example, the absolute value of the detection output at time B is the inner product of vector OB 'and vector OB in FIG. 19, that is, the square of line segment OB. Therefore, using the cosine theorem or the like, the detection output at each time point A to C in FIG. 18 (c) is as follows.

A: undefined B: a _n (1 + ρ ² +2 ρ cos α) C: undefined where a _n (a _n = ± 1) is the transmitted data string.

In areas A and C, it becomes undefined due to the data values of the preceding and succeeding time slots, respectively. After differential detection, a low-pass filter is usually inserted to remove harmonic components and unnecessary noise components, so the final detection output signal waveform is filtered by the solid line waveform of FIG. 18 (c). First
The waveform is as shown by the dotted line in FIG. 18C, and constitutes a part of the eye pattern. Where ρ is close to 1 and α is π
In the vicinity, the detection output in the area B, which is an effective detection output, becomes almost zero. Therefore, the eye is closed and the code error rate characteristic deteriorates. Further, at this time, since the invalid detection output of the areas A and C is much larger than the effective detection output of the area B, the eye greatly fluctuates in the time axis direction, the reproduced clock cannot follow, and the code error rate is further increased. Remarkably deteriorates (eg, Onoe et al., "Code Error Rate Characteristics in Rayleigh Fading with Propagation Delay Time Difference", IEICE Tech.
168, 1982, or Takai et al., "Analysis of Instantaneous Code Errors Due to Multipath Propagation and Error Generation Mechanism Based on Bit Synchronous System," IEICE Tech. Report, CS83-158, 1984).

In this way, in order to reduce the deterioration of the error rate characteristics due to the deterioration of the eye pattern and the fluctuation of the eye in the time axis direction, the phase transition waveform of the transmission signal is devised so as to generate a plurality of types of detection outputs, A method of improving the diversity effect by combining these plural types of detection outputs has been proposed. An example of the second conventional digital signal transmission method will be described below with reference to the drawings.

FIG. 20 is a phase transition waveform diagram showing the phase transition of the transmission signal of the digital signal transmission method in the second conventional example.
One time slot of data is divided into a first half portion and a second half portion, and has a stepwise waveform. The time of one time slot is T, the time of the first half is T ₁ , the time of the _second half is T ₂ , and the phase transition between the first half and the second half is φ. Similar to the first conventional example, the information to be transmitted is in the phase difference between adjacent time slots. For example, 0 and π are used as possible values of this phase difference, and 0 and 1 are associated with each other.
1 bit information is transmitted by allocating 1 bit.

Next, it will be explained that the digital signal transmission method in the second conventional example exhibits a good error rate characteristic under multipath fading.

The digital signal transmission method of the second conventional example is also a kind of differential encoding phase modulation, so that it is detected by differential detection using a delay line of one time slot. Figure 21 shows 2
FIG. 21 is a diagram illustrating what a detection output signal looks like when the transmission signal of FIG. 20 is detected by a delay detector under wave multipath. FIG. 21 (a) shows the phase transition of an arbitrary time slot of a direct wave and its adjacent time slot. On the other hand, the phase transition of the delayed wave delayed by the propagation delay time difference τ is as shown in FIG. 21 (b). Similar to the first conventional example, the detection output at a certain time point is the vector inner product of the combined phase of the two waves at that time and the combined phase of the two waves one time slot before.

FIG. 22 is a vector diagram showing the combined phase of the direct wave and the delayed wave in order to obtain the detection output at each time point A to E. The amplitude ratio between the direct wave and the delayed wave is ρ, and the phase of the delayed wave carrier viewed from the direct wave carrier is α. No. 22
From the figure, if there is no deformation of the waveform by the low-pass filter after detection, or if the cutoff frequency is sufficiently higher than the data transmission rate, the detection output at each time point A to E in FIG. become that way.

A, E ... Indefinite B, D ... 1 + ρ ² + 2ρ cos α C ... 1 + ρ ² + 2ρ cos (α-φ) In the areas A and E, the data values of the time slots before and after each become indefinite. In practice, the cut-off frequency of the low-pass filter is selected to be low enough not to cause intersymbol interference, and the detection output signal after passing through the low-pass filter has the waveform shown by the solid line in FIG. 21 (c). Takes the first
21 A part of the eye pattern is formed as shown by the dotted line in FIG. The detection outputs of areas B and D and area C are complementary,
No ρ or α will be zero at the same time and the eye will never close. Further, at least one of these effective detection outputs does not become smaller than the invalid detection output in the area A or E, so that fluctuations of the eye in the time axis direction are reduced, and the code due to poor reproduction clock tracking does not occur. There is little deterioration in the error rate. Therefore, the code error rate characteristic is remarkably improved, and high-speed digital transmission becomes possible.

In general, the detection output in each region of B to D under the two-wave multipath is a transmission data string a _n (a _n = ± 1), a polymorphization number m (m = 2,4,8 ...), and fading. Let S be the complex product noise that represents the received vector of the accompanying direct and delayed waves.
_The following can be expressed as ₁ (t) and S ₂ (t).

B, D ... a _n sin (π / m) ・ (| S ₁ + S ₂ | ² ) C ... a _n sin (π / m) ・ (| S ₁ exp (jφ) + S ₂ | ² ) ... The detection output is obtained by further shifting the carrier wave phase of the direct wave by φ. Therefore, the principle of improvement of the digital signal transmission method in the second conventional example is a kind of diversity for combining such different types of detection outputs. Incidentally, assuming an appropriate diversity model, if the fading of the direct wave and the delayed wave is independent, and the average error rate P _e is calculated when the averages of both are equal, Therefore, the optimum value of φ when there is no band limitation is π (for example, Takai, “A Proposal of Multi-wave Modulation and Demodulation-Resistant Method”, IEICE Technical Report, SAT86-23, 1986).

However, since the digital signal transmission method in the second conventional example further has a phase discontinuity point in the time slot, the envelope variation remarkably occurs when the band is limited, resulting in a non-linear distortion. weak. If the phase transition φ is made smaller than π in order to suppress the envelope variation, the improvement effect decreases, and the non-linearity resistance and the improvement effect are not compatible. Further, in the digital signal transmission method in the second conventional example, when T ₁ = T ₂ , when the delay time difference τ exceeds τ / T and exceeds 0.5, the regions B and D disappear and the improvement effect is lost. . By setting T ₁ ≠ T ₂ ,
Although it is possible to improve even larger τ, if the occupied bandwidth is further expanded and the band is limited, the deterioration of the error rate characteristic becomes large. Further, there is a problem that the envelope variation becomes larger and becomes weak against nonlinear distortion.

In view of the above problems, the present invention provides a digital signal transmission method that is strong against band limitation and non-linear distortion, and exhibits good characteristics even with a larger τ / T.

Means for Solving the Problems In order to solve the above problems, in the digital signal transmission method of the present invention, the phase transition waveform within one time slot of data is a parabolic waveform, and the phase within an arbitrary time slot is The transition waveform and the phase transition waveform in the time slot after the predetermined time slot have the same shape regardless of the information to be transmitted, and the same position of both time slots separated by the predetermined time slot. It uses a transmission signal that has information to be transmitted due to the phase difference between them.

Action The present invention uses the above-mentioned transmission signal and eliminates the phase discontinuity point in the time slot, so that the envelope variation at the time of band limitation can be suppressed. In addition, it is possible to obtain multiple types of detection output even for a larger delay time τ, it is resistant to band limitation and nonlinear distortion, and moreover, it shows good error rate characteristics even for a larger τ / T. Become.

Embodiment Hereinafter, a digital signal transmission method according to an embodiment of the present invention will be described with reference to the drawings.

FIG. 1 is a phase transition waveform diagram showing an example of a phase transition waveform of a transmission signal in the digital signal transmission method of the present invention.
Phase transition waveform ψ (t) within one time slot of data
(0 <t <T) is different from the conventional phase modulation method in that it has a parabolic waveform as shown by the equation.

The phase transition waveforms in the first time slot and the (n + 1) th time slot apart from each other by the predetermined n time slots have the same shape, and are entirely shifted by θ according to the transmitted information. . That is, differential encoding of n time slots is performed. For example, if a two-phase system of 0 and π is used as θ, 1 bit per time slot and 0, π / 2, π, 3 as θ.
If a π / 2 four-phase system is used, 2-bit information can be sent per time slot. If θ is generally shown, the following equation is obtained.

However, the value of i indicates the gray-encoded data value to be transmitted, and 0 ≦ i ≦ m, iεInteger. Therefore, if the phase transition waveform of the first time slot is ψ (t), the phase transition waveform of the (n + 1) th time slot is ψ (t).
It is expressed as (t−nT) + θ.

If the phase shift amount carrying information is expressed as the phase shift amount θ _a (t) from the absolute phase, the phase shift amount θ _a
(T) is a step-like function having a constant value in each time slot, and is a gray-coded data value sequence i _q to be transmitted.
(Qε Integer) can be expressed as the following equation using a data value sequence id _q obtained by differentially encoding n time slots.

On the other hand, there may be a plurality of types of phase transition waveform ψ (t) in the time slot. In the case of n time slot differential encoding, a maximum of n types of phase transition waveforms in time slots ψ
₁ (t), ..., ψ _n (t) can be selected.

If ψ _r (t) = 0 (t ≦ 0, t ≧ T, r = 1 to n) ..., the general formula of the phase transition waveform ψ (t) of the transmission signal in the digital signal transmission method of the present invention is Using The characteristic of the phase transition waveform of the transmission signal in the present invention represented by is in the first term of the equation, and the second term is the same as the conventional differential encoding phase modulation. In addition, the phase transition waveforms ψ ₁ (t), ψ ₂ (t), ..., ψ in the time slot
_{Some of n} (t) may be the same, or as a special case, all of them may be the same. Anyway, n
It suffices that the phase shift waveforms φ (t) within the time slot separated by the time slot match. The value of n may be 1, and in this case, there is one type of phase transition waveform ψ (t) within the time slot, and the phase transition waveforms within the time slot of all the time slots have the same shape. When there is one type of phase transition waveform ψ (t) in the time slot, the phase transition waveform ψ (t) of the transmission signal is expressed by the following equation.

There may be a plurality of types of the phase transition waveform ψ (t) in the time slot as described above. FIG. 2 shows a case where the maximum phase transition amount ψmax of ψ (t) has a plurality of types, and FIG. 3 shows a case where the phase transition directions alternate between advanced and delayed phases. However, in the latter case, the distance n between the corresponding time slots is an even number. Also, among these multiple types, other than the parabolic waveform,
For example, a staircase waveform may be included.

FIG. 4 shows, as an example, the phase transition waveform ψ (t) in the time slot is one kind of ψmax = 225 ° parabolic waveform,
n = 1, that is, 1 time slot differentially encoded, a specific example of a phase transition waveform of a transmission signal of a digital signal transmission method of the present invention capable of transmitting 2 bits per 1 time slot with a polyphase number m = 4 is shown. FIG. 3 is a phase transition waveform diagram.

Next, a method of obtaining the transmission signal as described above will be described with reference to an embodiment.

FIG. 5 is a block diagram of a transmission signal generation circuit of the digital signal transmission method in the first embodiment of the present invention. In FIG. 5, 501 is a data input terminal, 502 is a differential encoding circuit, 503 is an oscillator, 504 is a waveform generating circuit, 505 is a quadrature modulator, and 506 is a transmission signal output terminal. The transmitted digital data is input from the data input terminal 501 and differentially encoded by the differential encoding circuit 502. Then, in the waveform generation circuit 504, according to the differentially encoded data, I
A modulated signal for each of the axis and the Q axis is generated. On the other hand, the oscillator
At 503, a carrier is generated, and this carrier is the quadrature modulator 505.
Are modulated by the above-mentioned I-axis and Q-axis modulation signals to form a transmission signal, which is output from the transmission signal output terminal 506.

FIG. 6 shows an example of a circuit configuration diagram inside the quadrature modulator 505 in FIG. In FIG.
601 is a 90 ° phase shifter, 602 and 603 are balanced modulators, and 604 is a combiner. The carrier wave signal supplied from the oscillator 503 is
The balanced modulator 602 is used to modulate the I-axis modulated signal from the waveform generation circuit 504 to form an I-axis modulated signal. On the other hand, the above-mentioned carrier signal is 90 ° phase-shifted by the 90 ° phase shifter, and is modulated by the Q-axis modulation signal from the waveform generation circuit 504 using the balanced modulator 603 to become a Q-axis modulated signal. The I-axis and Q-axis modulated signals thus obtained are combined by the combiner 604 to be a transmission signal which is a modulated signal, which is output from the transmission signal output terminal 506.

FIG. 7 shows an example of a circuit configuration diagram inside the differential encoding circuit 502 in FIG. 701 and 704
Is a Gray code conversion circuit, 702 is an adder, and 703 is a delay device. In the case of the polymorphization number m (m = 2, 4, 8), that is, in the case of m phases, as shown in the equation, it is input to the Gray code conversion circuit 701 as a p-bit parallel data value sequence. The Gray coded data value sequence i _q enters the adder 702 and
In the delay unit 703, the output of the adder 702 is delayed by n time slots, that is, by n clocks, and addition is performed using m as a modulus. Then, the output of the adder 702 is further converted by the Gray code conversion circuit 704, whereby the input p-bit parallel data value sequence is Gray-coded and n
A differentially encoded p-bit parallel data value sequence id _{q of the} time slot is obtained.

In FIG. 8, the phase transition waveform Ψ (t) is expressed by the equation 4
FIG. 7 is a diagram showing an example of a circuit configuration diagram inside the waveform generation circuit 504 of FIG. 5, taking the case of a phase system as an example. 801 is an I-axis data input terminal, 802 is a data clock output terminal, 803 is a Q-axis data input terminal, 804 and 806 are shift registers,
805 is a binary counter, 807 is a read only memory (hereinafter abbreviated as ROM), 808 is a clock generator, 809 and 810 are digital / analog converters (hereinafter abbreviated as D / A converter), 811 and Reference numeral 812 is a low-pass filter, 813 is an I-axis modulation output terminal, and 814 is a Q-axis modulation output terminal. In the case of a four-phase system, the output id _q of the differential encoding circuit 502 is 2-bit parallel data, and its upper bit and lower bit are input from the I-axis data input terminal 801 and the Q-axis data input terminal 803, respectively. . The respective input data strings are delayed by shift registers 804 and 806 to obtain the modulation data of the current time slot and the modulation data of the time slots before and after it. That is, in the example of FIG. 8, Qd of the shift registers 804 and 806 is the modulation data of the current time slot, and the modulation data of three time slots before and after Qe to Qg and Qa to Qc can be obtained. On the other hand, RO
In the M807, I-axis and Q-axis modulation waveforms are written according to the modulation data, and in the example of FIG. 8, each one time slot is composed of 16 sample points. Addresses A4 to A17 of the ROM 807 are used as a select signal for determining which modulation waveform to select, and the above-mentioned present and preceding and following three time slot modulated data are input. To the addresses A0 to A3 of the ROM 807, a reference clock generated by the clock generator 808, which is frequency-divided by the binary counter 805, is added and becomes a read signal of a modulated waveform. ROM807 output X0 ~
X7 and Y0 to Y7 are D / A converters 809 and 810 and low-pass filters 811 and 812 that remove aliasing components, respectively.
Is converted into an analog signal and becomes an I-axis and Q-axis modulation signal. In the case of multi-phase modulation such as 8-phase system, the number of shift registers is equal to the number of p in the equation, and the ROM address corresponding to the shift registers is required.

Next, the modulation waveform for each time slot written in the ROM 807 will be described. Basically, based on the phase transition waveform Ψ (t) of the transmission signal obtained from the equation from the differentially encoded data value sequence id _q to be transmitted, the modulation waveform M _I (t) of the I axis and the Q axis is calculated by the following equation. , or if you get the M _Q (t).

M _I (t) = cos Ψ (t) M _Q (t) = sin Ψ (t) However, since it is a wide band signal as it is, the impulse response of the band limiting filter is set to h (t) and band limiting is performed by this filter. When you do, the formula becomes as follows.

As the frequency characteristic of the band limiting filter, various types such as a cosine square type and a Gauss type can be used as long as they are low pass type.
The impulse response h (t) is also known accordingly. As an example, an impulse response h (t) of a surplus square filter having a cutoff angular frequency ω ₀ and a rolloff coefficient γ is shown.

The I-axis and Q-axis modulation waveforms M _I (t) and M _Q (t) for one time slot are written in the ROM 807 of FIG. 8 according to the equation. The integration range (-t ₀ , t ₀ ) of the equation is selected to be about the spread range of the impulse response h (t), and in the example of FIG. 8, there are three time slots before and after, and the phase transition waveform Ψ (t) is calculated from the equation. In order to calculate, the modulation data of three time slots before and after is required. Therefore, the ROM 807 has calculated and written all the modulation data patterns of the present and the preceding and following three time slots from the formula, and is the modulated data of these present and the preceding and following three time slots.
Which modulation waveform is selected is selected by the addresses A4 to A17 of M807.

As shown in the equation, the phase transition waveform Ψ (t) is almost the same when there are a plurality of types of the phase transition waveform ψ (t) in the time slot, and the equations modulate the I axis and the Q axis for one time slot. RO waveforms M _I (t) and M _Q (t)
Write to M. However, when obtaining Ψ (t) in the equation from the equation, the phase transition waveform ψ in the current time slot
_It is further necessary to determine what value r (1 ≦ r ≦) of _r (t) has. Therefore, the waveform data written in ROM is
Not only the modulation data patterns of the current time slot and several time slots before and after the current time slot but also r indicating the order of the phase shift waveform ψ _r (t) in the current time slot is calculated and written. In accordance with this, the internal circuit configuration diagram of the waveform generation circuit 504 in FIG. 5 must be as shown in FIG. In FIG. 9, 801 is an I-axis data input terminal, 802 is a data clock output terminal, 803 is a Q-axis data input terminal, 804 and 806 are shift registers, 805 is a binary counter, 808 is a clock generator, and 809 and 810. Is a D / A converter, 811 and 812 are low-pass filters, 813 is an I-axis modulation output terminal, and 814 is a Q-axis modulation output terminal. The above is exactly the same as the configuration of FIG. The difference from the configuration of FIG. 8 is that a binary counter 901 indicating the current value of r is added, and in order to select the waveform by the value of r, the addresses of A18 and A19 are stored in the ROM of 902. It has been added. The period of the binary counter 901 is n,
In the example of FIG. 9, n = 4.

Next, a method of detecting a transmission signal in the digital signal transmission method of the present invention as described above will be described.

In the digital signal transmission method of the present invention, the detection method is a delay detector having a delay line of n time slots. The following is a brief description.

FIG. 10 shows a circuit configuration diagram of the differential detector in the case of a two-phase system. In FIG. 10, 1001 is an input terminal,
1002 is a multiplier, 1003 is a low pass filter, 1004 is an n time slot delay device, and 1005 is an output terminal. In the n time slot delay device 1004, the signal is delayed by n time slots, but the phase of the carrier wave is in phase with the input and output. The low-pass filter 1003 is a carrier wave generated by the multiplier 1002.
It not only removes the double frequency component, but also serves to combine a plurality of types of detection outputs described later. The frequency characteristic of the low pass filter 1003 is half of the symbol transmission rate 1 / T,
That is, a so-called Nyquist filter having a cutoff frequency of 1 / 2T and an attenuation characteristic that is axisymmetric with respect to this frequency is desirable.

FIG. 11 shows a circuit configuration diagram of the differential detector in the case of a four-phase system. In FIG. 11, 1101 is an input terminal,
1102 and 1106 are multipliers, 1103 is −45 ° phase shifter, 1105 is + 45 ° phase shifter, 1104 is n time slot delay device, 1107 and 1108 are low pass filters, 1109 is output terminal A, 1110.
Is an output terminal B. The difference from the case of FIG. 10 is that a −45 ° phase shifter 1103 and a + 45 ° phase shifter 1105 are used,
This is that differential detection is performed on two axes that are orthogonal to each other and 2-bit parallel data is demodulated. Other operations are the same as in the case of FIG.

FIG. 12 shows a circuit configuration diagram of the differential detector in the case of an 8-phase system. In FIG. 12, 1201 is an input terminal,
1202 to 1205 are multipliers, 1206 is n time slot delay device,
1207 is −22.5 ° phase shifter, 1208 is 22.5 ° phase shifter, 1209 is +
67.5 ° phase shifter, 1210 −67.5 ° phase shifter, 1211 to 1214 low pass filter, 1215 comparator, 1216 output terminal A, 12
Reference numeral 17 is an output terminal C, and 1218 is an output terminal B. In this case, the phase shifters 1207 to 1210 further perform differential detection on the three axes shifted by 45 ° to demodulate the 3-bit parallel data. In the comparator 1215, the polarities of both inputs match,
Detects discrepancies.

Next, it will be described that the digital signal transmission method of the present invention exhibits excellent error rate characteristics under multipath fading.

First, the method introduced as the second conventional digital signal transmission method is the phase transition waveform ψ (t) in the time slot.
In the case of a stepwise waveform, the improvement principle is shown to be applied to an arbitrary phase change waveform.

Figure 13 shows the phase transition waveform ψ in an arbitrary time slot.
FIG. 22 is a diagram for explaining how the detection output signal becomes under two-wave multipath, similar to FIG. 21, for (t). Similar to the case of Fig. 21, the detection output is roughly divided into F,
Areas F and H are classified into three areas G and H, and areas F and H are areas of invalid detection output whose polarities do not necessarily match the data values to be transmitted. The area G of the areas B, C, and D in FIG. 21 corresponds to this area, which is an area of effective detection output whose polarity always matches the transmitted data value. However, different types of detection outputs appear as shown by the solid line in FIG. 13 (c). Similar to the case of FIG. 21, the solid line waveform of FIG. 13 (c) is further filtered to form a part of the eye pattern as shown by the dotted line of FIG. 13 (c).

Detection output in the area G, as in the formula, the z as a parameter, the region _{G; a n sin (π /} m) · (| s 1 exp {jψ (z)} + s 2 exp {j
ψ (z−τ) | ² ) = a _n sin (π / m) ・ (| s ₁ exp {jψ
(Z) −ψ (Z−τ)}] + s ₂ | ² ) where τ ≦ z ≦
It can be expressed as T ... Therefore, the phase transition waveform ψ in the time slot satisfying the condition of ψ (z) −ψ (Z−τ) ≠ const. (Τ ≦ z ≦ T) ...
As long as (t) is used, the equation is not a constant value,
It can be seen that the error rate performance is improved under multipath fading by a kind of diversity effect by combining different detection outputs. The condition of the formula indicates that the phase transition waveform ψ (t) in the time slot is a waveform in which the phase change rate changes or is discontinuous.

Next, a representative example of the digital signal transmission method of the present invention will be taken and an example of an average error rate characteristic under two-wave Rayleigh fading having a delay time difference will be shown.

FIG. 14 shows the average error rate characteristics with respect to the S / N ratio in the case where the phase transition waveform in the time slot is a four-phase system with ψmax in FIG. 1 or the equation as a parameter.
For comparison, the same graph also shows the case of four-phase phase modulation, which is a conventional digital signal transmission method. As shown in FIG. 14, quadrature phase modulation causes an irreducible error that cannot be alleviated even if the S / N ratio is increased. However, such a phenomenon does not appear in the digital signal transmission method of the present invention, and a significant error occurs. It can be seen that the rate characteristic is improved.

Similarly, FIG. 15 shows the average error rate characteristics in the case of a four-phase system with respect to the delay time difference τ, using ψmax as a parameter. When ψmax is 180 ° to 360 °, 0 <τ / T <0.7
It can be seen that the maximum τ / T is improved to about 0.8 when ψmax is 225 to 270 °. Since increasing ψmax leads to an increase in occupied bandwidth, ψmax is 225
It is appropriate to select around °. It should be noted that when τ / T = 0 or τ / T ≧ 0.7, the improvement effect disappears and the characteristics are close to those of four-phase phase modulation.

As described above, according to the present embodiment, by making the phase transition waveform in the time slot parabolic, the improvement effect can be obtained even for a larger τ, and the phase discontinuity point is present in the time slot. Since there is no band limitation, envelope fluctuation at the time of band limitation can be reduced, and the characteristics against band limitation and nonlinear distortion are improved.

Hereinafter, a second embodiment of the present invention will be described with reference to the drawings.

FIG. 16 is a block diagram of a transmitter circuit of a digital signal transmission method in the second embodiment of the present invention. In FIG. 16, 501 is a data input terminal, 1601 is a transmission signal generation circuit, and the above is exactly the same as the configuration of FIG. 5 in the first embodiment. 1602-1604 is the first antenna of the k system
The kth antenna, 1605 to 1607 are k system level adjusters, 1608
˜1610 are the (k−1) th system first delay device to the (k−1) th delay device. The level adjusters 1605 to 1607 may have an amplifying function. In addition, the detection method on the receiving side is the delay detection of n time slots as shown in FIGS. 10 to 12 shown as the first embodiment.

The digital signal transmission method configured as above will be described below with reference to FIGS. 15 and 16 and equations.

FIG. 15 shows the average error rate characteristic when the transmission signal of the present invention, which is the output signal of the transmission signal generation circuit 1601, propagates through two Rayleigh fading paths having a delay time difference τ and is received and detected. As mentioned above. Now, assume that the delay time difference τ of the propagation path, that is, the so-called delay dispersion is smaller than the time slot length T. This condition corresponds to a case where the delay dispersion is small on the premises or a case where the transmission speed is slow. Thus, when τ / T is close to 0, the left side of the equation is z
Changes less, and the diversity effect due to combining different types of detection outputs as described in the first embodiment is reduced. Therefore, as shown in Fig. 15, τ
As / T approaches 0, the error rate performance does not improve. Therefore, if a delay that is within the range of 0 to 0.8, which is the improvement range of τ / T, is given in advance on the transmitting side, the error rate characteristic is improved rather due to the diversity effect.

In FIG. 16, the delay units 1608 to 1610 provide the delay on the transmitting side as described above.The first arriving wave and the last arriving wave on the receiving side including the delay due to the path difference from each antenna. It is necessary to set the time difference τ _{m of the} waves to be τ _m / T so as not to exceed the maximum improvement range (about 0.8) of τ / T determined by the phase transition waveform ψmax in the time slot. The level adjusters 1605 to 1607 set the average level of the wave with fading from each antenna to be substantially equal at the receiving point. The first to kth antennas need to be installed separately from each other or use antennas having different polarization planes so that fadings of the paths from the respective antennas to the receiving point are mutually uncorrelated. The simplest and most useful case is k = 2. In this case, the time difference τ _m between the waves arriving from the two antennas is τ _m / T.
Is the best error rate determined by ψmax, 0.3
It is desirable to select ~ 0.4.

As described above, in the second embodiment of the present invention, by transmitting the same transmission signal from different antennas with a time difference, the diversity effect can be obtained even when τ / T is small, and the error rate characteristic can be improved. Can be improved. Since this diversity requires only one antenna on the receiving side, it is advantageous for downsizing and portability of the receiving side device.

As described above, the present invention eliminates the phase discontinuity point in the time slot by using the parabolic waveform for the phase transition waveform in the time slot of the transmission signal, and suppresses the envelope variation at the time of band limitation. The characteristics are improved with respect to the limit and the nonlinear distortion. In addition, it is possible to obtain multiple types of detection output for a larger delay time τ, and while achieving compatibility with the required bandwidth or band limitation, a good error rate characteristic can be obtained for a larger τ / T. Obtainable.

[Brief description of drawings]

1 to 4 are phase transition waveform diagrams showing an example of phase transition waveforms of a transmission signal in the digital signal transmission method of the present invention, and FIG. 5 is a transmission of the digital signal transmission method in the first embodiment of the present invention. 6 is a circuit configuration diagram of the signal generation circuit, FIG. 6 is a circuit configuration diagram of the quadrature modulator 505 of FIG. 5, FIG. 7 is a circuit configuration diagram of the differential encoding circuit 502 of FIG. 5, FIG. 8 and FIG. FIG. 9 is a circuit configuration diagram of the waveform generation circuit 504 of FIG. 5, FIGS.
FIG. 2 is a circuit configuration diagram of a detector of a digital signal transmission method according to an embodiment of the present invention, and FIG. 13 is an explanatory diagram explaining a detection output signal under a two-wave multipath of the digital signal transmission method of the present invention. FIG. 15 to FIG. 15 are characteristic diagrams showing the average error rate characteristics of the digital signal transmission method of the present invention under two-wave Rayleigh fading, and FIG. 16 is a transmission circuit of the digital signal transmission method of the second embodiment of the present invention. FIG. 17 is a phase transition waveform diagram showing the phase transition of the transmission signal of the digital signal transmission method in the first conventional example, and FIG. 18 is a two-wave multiwaveform of the digital signal transmission method in the first conventional example. Explanatory diagram explaining the detection output signal under the path, FIG. 19 is a vector diagram showing the combined phase of the direct wave and the delayed wave in order to obtain the detection output of FIG. 18, and FIG. 20 is the second conventional example. Dizzy FIG. 21 is a phase transition waveform diagram showing the phase transition of the transmission signal of the digital signal transmission method, and FIG. 21 is an explanatory diagram explaining the detection output signal under the two-wave multipath of the digital signal transmission method in the second conventional example, and FIG. Is the
FIG. 22 is a vector diagram showing a combined phase of a direct wave and a delayed wave for obtaining a detection output of FIG. 21. 501 ... Data input terminal, 502 ... Differential encoding circuit, 503
...... Oscillator, 504 …… Waveform generation circuit, 505 …… Quadrature modulator, 506 …… Transmission signal output terminal, 601 …… 90 ° phase shifter, 60
2,603 ... balanced modulator, 604 ... combiner, 701 ... Gray code conversion circuit, 702 ... adder, 703 ... delay device, 704 ...
... Gray code conversion circuit, 801 ... I-axis data input terminal, 8
02 …… Data clock output terminal, 803 …… Q axis data input terminal, 804,806 …… Shift register, 805,901 …… Binary counter, 807,902 …… Read only memory (R
OM), 808 ... Clock generator, 809,810 ... Digital
Analog converter (D / A converter), 811,812,1003,1107,110
8,1211 to 1214 ... Low-pass filter, 813 ... I-axis modulation output terminal, 814 ... Q-axis modulation output terminal, 1001,1101,1201
...... Input terminal, 1002,1102,1106,1202 to 1205 …… Multiplier, 1004,1104,1206 …… n time slot delay device, 1005
...... Output terminal, 1109,1216 …… Output terminal A, 1110,1218…
… Output terminal B, 1217 …… Output terminal C, 1103 …… −45 ° phase shifter, 1105 …… + 45 ° phase shifter, 1207 …… −2.25 ° phase shifter, 1208 …… + 22.5 ° phase shifter Device, 1209 …… ＋ 67.5 ° phase shifter, 1210 …… −67.5 ° phase shifter, 1215 …… comparator, 1601…
... Transmission signal generation circuit, 1602 ... First antenna, 1603 ... Second antenna, 1604 ... Kth antenna, 1605 to 1607 ... Level adjuster, 1608 ... First delay device, 1609 ... Second delay Bowl, 16
10 ... k-1th delay device.

Claims (3)

(57) [Claims] 1. A transmission method for transmitting digital data, wherein a phase transition waveform within one time slot of data is a parabolic waveform, and the phase transition waveform within an arbitrary time slot is followed by a predetermined time slot. The phase transition waveform in the time slot of the above has the same shape regardless of the information to be transmitted, and the phase difference between the same positions of these two time slots separated by the predetermined time slot. A digital signal transmission method, characterized in that a transmission signal having information to be transmitted is used, and the transmission signal is detected by differential detection using a delay line capable of obtaining a delay corresponding to the predetermined time slot. 2. A digital signal transmission method as set forth in claim 1, wherein the phase difference is one of angles obtained by equally dividing 2π by a power of 2. 3. A transmission signal is such that the information to be transmitted has a phase difference between a phase transition waveform in an arbitrary time slot and the same position of a phase transition waveform separated by one time slot. The digital signal transmission method according to claim (1), characterized in that: