JPH10271072A - Peak reduction device and transmission device using the same - Google Patents

Abstract

(57) Abstract: A peak reduction device for reducing a code amplitude of a transmission code sequence and a transmission device using the device reduce the spread of a band. A peak reduction device sets a code sequence of a predetermined length of a pattern whose amplitude increases when a transmission code sequence is band-limited by a transmission filter in a register, and shifts the code sequence to a shift register. A comparison unit 2 for comparing the transmission code sequence to be transmitted and a comparison match signal from the comparison unit 2, and a code amplitude reduction unit for reducing the amplitude of at least one code at the center of a predetermined length of the transmission code sequence. The transmission device includes a peak reduction device 1, a transmission filter 6, a modulator 7, and a transmission amplification unit 8.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a peak reduction device for reducing a peak factor of a transmission signal in a digital radio communication system, and a transmission device using the same. 2. Description of the Related Art In a digital radio communication system, it is necessary to reduce interference to an adjacent channel, and it is desired that a transmission amplifier be operated efficiently.
Therefore, the band is limited by the transmission filter, and the transmission amplifier is operated in the linear characteristic portion immediately before the saturation characteristic. However, there are problems such as an increase in a peak factor due to band limitation by a transmission filter and an expansion of a band due to distortion characteristics of a transmission amplifier.

[0002]

2. Description of the Related Art A transmitting apparatus in a digital radio communication system includes a transmission filter for limiting a band, a modulator for modulating a digital signal through the transmission filter, and a transmission amplifier for amplifying and transmitting the modulated signal. Have. Although it is desirable that the transmission amplifier be operated in a linear region where distortion is small, it is general that the transmission amplifier is operated in a non-linear region in terms of efficiency. In such a non-linear region, distortion is increased, a transmission band is widened, and a problem of an influence on an adjacent channel occurs.

Therefore, the distortion component is extracted and input to the transmission amplifier via a linearizer that cancels the distortion of the transmission amplifier using a feedback or feedforward circuit or a circuit having the inverse characteristic of the distortion characteristic of the transmission amplifier. A pre-distortion method for canceling distortion of a transmission amplifier is known. FQPSK, TFM, and the like are known as modulation schemes in a modulator to reduce the influence of distortion of a transmission amplifier by reducing envelope dispersion.

[0004]

In a digital radio communication system, various coding systems are applied.
The peak factor is increased by a specific pattern of the code sequence. Further, since the transmission amplifier operates in the nonlinear region with high efficiency as described above, there is a problem that when the peak factor increases, the distortion increases and the transmission band increases. In order to suppress such spread of the transmission band, the above-described linearizer and pre-distortion method are applied.

In such a linearizer or pre-distortion system, if the characteristics change due to temperature change or aging of each part of the transmission amplifier and the like, it may be configured to follow the distortion and perform distortion suppression. However, the circuit scale becomes very large. For example, a mobile device in a mobile radio communication system is required to be downsized. It will be difficult. The above-mentioned FQPSK and TFM are hardly affected by the distortion of the transmission amplifier because the envelope of the signal is small, but they have a large required bandwidth and are disadvantageous in terms of frequency utilization efficiency. An object of the present invention is to suppress the spread of the band by relatively simple means.

[0006]

According to the present invention, there is provided a peak reducing apparatus comprising: (1) a predetermined length of an input transmission code sequence and a pattern whose amplitude increases when the transmission code sequence is band-limited by a transmission filter 6; And a code amplitude reduction unit 3 that reduces the amplitude of at least one code at the center of a predetermined length of the transmission code sequence by a comparison match signal of the comparison unit 2. I have. Thereby, a peak factor corresponding to intersymbol interference is reduced.

(2) Comparison section 2 in the peak reduction device
The first and the first channel corresponding to the I and Q channels respectively compare the orthogonal I and Q channel transmission code sequences with a predetermined length code sequence of a pattern in which the transmission code sequence is band-limited by the transmission filter and the amplitude increases. The code amplitude reduction unit 3 includes a second comparison unit, and the code amplitude reduction unit 3 uses a comparison match signal of both the first and second comparison units to determine a center of a predetermined length of one or both of the I and Q channel code sequences. May be provided to reduce the amplitude of at least one of the codes.

(3) The transmitting apparatus of the present invention further comprises a comparing section 2 for comparing the transmission code sequence with a predetermined length code sequence of a pattern whose amplitude increases when the transmission code sequence is band-limited by the transmission filter. A peak reduction device 1 having a code amplitude reduction unit 3 for reducing the amplitude of at least one code at the center of a predetermined length of a transmission code sequence by the comparison coincidence signal of the comparison unit 2, and the output of the peak reduction device 1 A transmission filter 6 for inputting a signal to limit a band, a modulation unit 7 for inputting and modulating an output signal of the transmission filter 6,
A transmission amplification unit 8 for receiving and amplifying the output signal of the modulation unit 7;

(4) The peak reducing device 1 in the transmitting device determines a transmission code sequence of the orthogonal I and Q channels and a pattern in which the transmission code sequence is band-limited by the transmission filter 6 to increase the amplitude. First and second comparison units corresponding to I and Q channels respectively for comparing the long code strings with each other;
By the comparison match signals of both the first and second comparison units,
A code amplitude reduction unit 3 for reducing the amplitude of at least one code at the center of the predetermined length of one or both of the I and Q channel code sequences can be provided.

(5) Transmission filter 6 in transmission device
Can be configured to have a moderate attenuation characteristic.

[0011]

DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 is an explanatory diagram of a first embodiment of the present invention, showing a main part of a transmitting apparatus, wherein 1 is a peak reducing apparatus, 2 is a comparing section, and 3 is a code amplitude reducing section. Reference numeral 4 denotes a shift register, 5 denotes a register, 6 denotes a transmission filter, 7 denotes a modulation unit, 8 denotes a transmission amplification unit, 9 denotes an antenna, and 10 denotes a code processing unit.

The peak reduction device 1 has a configuration including a comparison unit 2, a code amplitude reduction unit 3, a shift register 4, and a register 5, and the shift register 4 transmits the signal processed by the code processing unit 10. The code sequence is sequentially shifted and input to the code amplitude reduction unit 3. The register 5 stores a code sequence pattern whose amplitude is increased by band limitation by the transmission filter 6.

The comparing section 2 compares a code sequence of a predetermined length set in the register 5 with a transmission code sequence of a predetermined length shifted to the shift register 4, and applies a comparison coincidence signal to the code amplitude reducing section 3. . The code amplitude reduction unit 3 inputs the transmission code sequence output from the shift register 4 to the transmission filter 6 as it is when the comparison coincidence signal is not input, and sets it in the register 5 when the comparison coincidence signal is input. The amplitude of at least one code at the center of the length of the generated code sequence.

The transmission filter 6 applies a roll-off filter, a root Nyquist filter, or the like. The solid curve in FIG. Is shown. When the impulse response of the transmit filter 6 and g (t), bandlimited signal v by the transmit filter (t) may be expressed by _{v (t) = Σa n g} (t-nT) ... (1) it can. However, a _n data, T is the symbol length.

[0015] In this case, a BPSK signal if the data a _n is represented as binary data of 1 or -1, and 1
When it is represented by a complex number of + j, 1-j, -1 + j, -1-j, it is a QPSK signal. As described above, when an impulse response is expressed by an equivalent baseband system such as a roll-off filter or a root Nyquist filter, a filter having a real number is often used as the transmission filter as described above. Therefore, it is assumed that the impulse response g (t) in Expression (1) is a real number. In this case, the I channel signal and the Q channel signal do not depend on each other even in the QPSK signal.

Therefore, the I-channel signal v _I (t) and the Q-channel signal v _Q (t) can be independently expressed as follows. v _I (t) = Σa _I, ng (t−nT) (2) v _Q (t) = Σa _Q, ng (t−nT) (3)

The maximum value v (t ₀ ) of the signal at a certain time t ₀
Can be expressed as: v (t ₀ ) = Σ | g (t ₀ −nT) | (4) As described above, the combination of data that maximizes the signal amplitude is a data sequence that cancels the sign of the impulse response and a data sequence that is the reverse of the data sequence. That is, ｛…, sgn (g (t ₀ + T)), sgn (g (t ₀ )), sgn (g (t ₀ −T)), ……｝ (5) ｛…, -sgn (g ( t ₀ + T)), -sgn (g (t ₀ )), -sgn (g (t ₀ −T)),...｝ (6) However, sgn (x) is 1 when x ≧ 0 and −1 when x <0. In addition, since the equation (4) is a periodic function of T, the timing of t ₀ which is the maximum at 0 <t <T is the timing at which the signal is maximized.

FIG. 2B shows the impulse response of the root Nyquist filter. When t = T / 2, the points a ₃ g (t−3T) and a ₂ g (t−2T) indicated by arrows are shown. ),
a ₁ g (t−T), a ₀ g (t), a ₋₁ g (t + T),
a ₋₂ g (t + 2T) is the output signal of the filter. Note that the after a _4, but also present data prior a _-3,
It is omitted because the amplitude is small.

In this case, _assuming that an is a binary value of 1 and −1, the absolute value of the output signal becomes maximum because of a _2,.
a ₋₁ , a ₀ , a ₁ , a ₂ ,.
1,1, -1,1, ... or = ...- 1,1,-
1, -1, 1, -1,..., A ₁ g
Even if the amplitude of (t−T), a ₀ g (t) is about 0.6, the amplitude becomes about 1.3 due to intersymbol interference.

Therefore, as a code sequence of a predetermined length, for example, four (-1, 1, 1, −1) or (1, −1, −)
1, 1) (in the case of this pattern, the impulse response waveform of FIG. 2B is inverted) or both are set in the register 5. If you set both,
Set as the first and second registers, respectively, and
Is configured to compare the contents of the shift register 4 with the contents of the first and second registers, respectively. In this case, a configuration in which comparison is performed in parallel or a configuration in which comparison is performed sequentially in series can be adopted.

The transmission code sequence of a predetermined length of the shift register 4 is compared with the code sequence of a predetermined length set in the register 5 by the comparing unit 2. As described above, the predetermined length is 4 symbols. When (−1, 1, 1, −1) or (1, −1, −1, 1), the comparison coincidence signal is sent to the code amplitude reduction unit 3 when the transmission code sequence has any one of these patterns. Added.

The code amplitude reducing section 3 sets 0 <a <1
The transmission code sequence (-1, 1, 1, -1) is changed to (-1, a,
a, -1). Also, the transmission code sequence (1, -1,-
(1, 1) is (1, -a, -a, 1). Assuming that the value a when the code amplitude is not reduced is 1 in this case, for example, values such as 0.8, 0.6, and 0.4 can be adopted.

In the case described above, the amplitude of the two central codes of the transmission code sequence of a predetermined length is reduced, and the transmission code sequence sequentially shifted and input via the shift register 4 is compared with the transmission code sequence. 2nd and 3rd from the time when the coincidence signal is generated
The absolute value of the amplitude of the sign of the number is reduced from 1 to a. Such processing can be easily performed on the basis of the shift clock signal of the shift register 4, and the amplitude reduction is controlled by a variable attenuator for analog processing or whether or not via an attenuator is determined by a switch circuit. It can be easily realized by a switching configuration or the like. The code amplitude can also be reduced by expressing each code as a multi-valued level and reducing the level. The functions of the comparison unit 2 and the code amplitude reduction unit 3 are, for example, a digital signal processor (D
SP).

Therefore, even if the band is limited by the transmission filter 6, the peak factor does not increase, so that the expansion of the band due to the non-linear characteristic of the transmission amplifier 8 can be suppressed. After shifting by one bit in the shift register 4, the condition that the comparison match signal is not output from the comparison unit 2 is not satisfied, so that the comparison match signal is not continuously output. Therefore, the transmission code sequence is not subjected to the double amplitude reduction processing in the code amplitude reduction unit 3.

Further, the transmission amplifier 8 outputs the output (dBm) and the phase (degre) corresponding to the input (dBm) shown in FIG.
Assuming that the input / output characteristics have the relationship with e), the aforementioned code amplitude reduction rate a is 1.0, 0.8,
The output power Po of the transmission amplifying unit when 0.6 and 0.4 are set.
ut (dBm) and the adjacent channel leakage power ratio D / U (d
B) is shown in FIG.

When the code amplitude reduction rate a is 1.0, the code amplitude is not reduced. When the code amplitude reduction rate is 0.8, the code amplitude is reduced from 1.0 to 0.8. Means that. Then, as shown in FIG.
The D / U is improved as the code amplitude is reduced.

FIG. 4 is an error rate characteristic curve diagram, in which the horizontal axis represents E _b / N _o and the vertical axis represents the bit error rate BER, and the code amplitude reduction rate a corresponds to FIG. 1.0,0.
The bit error rate BER when 8, 0.6, and 0.4 are set is shown. E _b is the energy per bit at the filter input, and N _o is the power density of Gaussian noise at the filter input. This error rate characteristic deteriorates as the code amplitude decreases.

FIGS. 3A and 3B and FIG.
When the direct spread spectrum method is applied, a symbol rate of 64 ksps and a chip rate of 4.096 M
cps, a root Nyquist filter having a roll-off rate of 0.22 is provided on each of the transmitting side and the receiving side, and the offset QPSK modulation method is used. FIG. This is the result of calculating the leakage power ratio D / U of the channel.

Therefore, as can be seen from FIGS. 3B and 4, the code amplitude reduction rate a is selected by a trade-off between the adjacent channel leakage power ratio D / U and the bit error rate BER. become.

A case where the amplitudes of the two codes at the center of a transmission code sequence of a predetermined length are reduced will be described. For example, (−1, 1, 1, −1) of the transmission code sequence described above is changed to (−1). ,
(1, a, -1), and (1, -1,-
(1,1) can be changed to (1, -1, -a, 1). That is, by reducing the amplitude of one code near the center of the predetermined length, the peak factor when the band is limited by the transmission filter 6 can be reduced. When the amplitude of one code is reduced as compared with the case where the amplitude of two codes is reduced, the value of a is reduced.

FIG. 5 shows the frequency characteristics of the transmission filter and D /
FIG. 9 is an explanatory diagram of the U characteristic. By reducing the attenuation characteristic of the transmission filter 6 (see FIG. 1), the dispersion of the envelope of the signal after the band limitation can be reduced.
That is, as shown in FIG.
(B) shows the D / U characteristics in the case of the root Nyquist filters of 0.22, 0.3, and 0.4. Again,
This is obtained under the same conditions as the modulation method and the like in the case of the characteristics shown in FIGS. 3A, 3B and 4.

As can be seen from FIG. 5B, as the roll-off rate k increases, the adjacent channel leakage power ratio D
/ U is improved. In that case, k for k = 0.22
= 0.3, there is almost no deterioration of the error rate, but k = 0.
In the case of 4, deterioration of 0.005 dB or less occurred.

FIG. 6 is an explanatory view of a main part of a second embodiment of the present invention, wherein 2A and 2B denote first and second comparison sections, 3A and 3A, respectively.
3B is a first and second code amplitude reduction unit, 4A and 4B are shift registers, 5A and 5B are registers, 6A and 6B are transmission filters, 11 is an AND circuit, and 12 is a modulation unit. , And orthogonally modulates transmission code sequences of the I channel and the Q channel.

The transmission code sequence I of the orthogonal I and Q channels
The channels ch and Qch are input to the shift registers 4A and 4B and shifted. The registers 5A and 5B have
A code string having a predetermined length similar to the code string set in the register 5 is set. The first and second comparison units 2A and 2B
The shift register 4A, 4B and the register 5A, 5B compare the code string of a predetermined length.

For example, in the registers 5A and 5B, (-1, 1, 1, 1) as a code sequence having a predetermined length, as in the case described above.
Assuming that both -1) and (1, -1, -1, 1) are set respectively, the transmission code sequences I and Q channels of predetermined lengths shifted to the shift registers 4A and 4B.
When the ch and Qch are (-1, 1, 1, -1) or (1, -1, -1, 1) at the same time, a comparison match signal is simultaneously output from the first and second comparison units 2A and 2B. Is output. When the transmission code sequence is in such a state, the envelope of the band-limited signal tends to be particularly large.

Then, the AND circuit 11 detects that the transmission code sequences Ich and Qch on the I and Q channels coincide with the code sequences set in the registers 5A and 5B at the same time. One or both of the code amplitude reduction units 3A and 3B are controlled. For example, as shown by the solid line, the code amplitude reduction unit 3A on the I channel side can be controlled, or as shown by the dotted line, the code amplitude reduction unit 3B on the Q channel side can be controlled. Or the first and second
Can be controlled at the same time.

The first and second controlled by the comparison coincidence signal
The code amplitude reduction units 3A and 3B reduce the code amplitude for a transmission code sequence having a predetermined length, and for example, convert the transmission code sequence (-1, 1, 1, -1) into (-1, a, a,-
1) or (-1,1, a, -1) and (1,-
(1, -1, 1) is (1, -a, -a, 1) or (1, -1, -a, 1).

By reducing the code amplitude, the envelope of the band-limited signal becomes substantially constant, and the spectrum does not spread due to the nonlinear distortion of the transmission amplifier. Therefore, the adjacent channel leakage power ratio D / U is improved. be able to.

As described with reference to FIG. 5, the transmission filters 6, 6A, and 6B in FIG. 1 or FIG. 6 have a gradual attenuation characteristic to further reduce the adjacent channel leakage power ratio D / U. Can be improved. Also, a case where a binary code of 1 or -1 is used as a transmission code sequence is shown. However, the present invention can be applied to a case where other multi-value codes are used. Since the number of patterns in which the variance of the envelope becomes large may increase, a code sequence having a predetermined length is set in the registers 5, 5A, and 5B.

[0040]

As described above, according to the present invention, the code sequence set in the register 5 is compared with the transmission code sequence by the comparison unit 2.
Since the variance of the envelope of the signal becomes large due to the band limitation when the comparison and coincidence are performed, the code amplitude reduction unit 3 determines that one code at the center of the predetermined length or an arbitrary number of codes in the vicinity thereof. This is for reducing the amplitude of the code, and has the advantage that the variance of the envelope of the signal band-limited by the transmission filter 6 can be reduced and the band expansion due to the nonlinear distortion of the transmission amplifier 8 can be suppressed. Thereby, the adjacent channel leakage power ratio D / U can be improved, and there is an advantage that the radio frequency can be effectively used.

[Brief description of the drawings]

FIG. 1 is an explanatory diagram of a first embodiment of the present invention.

FIG. 2 is an explanatory diagram of characteristics of a transmission filter.

FIG. 3 is an explanatory diagram of input / output characteristics and D / U characteristics of a transmission amplifier.

FIG. 4 is an error rate characteristic curve diagram.

FIG. 5 is an explanatory diagram of a frequency characteristic and a D / U characteristic of a transmission filter.

FIG. 6 is an explanatory diagram of a main part of a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Peak reduction device 2 Comparison part 3 Code amplitude reduction part 4 Shift register 5 Register 6 Transmission filter 7 Modulation part 8 Transmission amplification part

 ──────────────────────────────────────────────────続 き Continuing on the front page (72) Inventor Takeshi Inoue 2-3-9 Shin-Yokohama, Kohoku-ku, Yokohama-shi, Kanagawa Prefecture Fujitsu Digital Technology Co., Ltd. In-house (72) Inventor Yoshiharu Tozawa 4 Ueodanaka, Nakahara-ku, Kawasaki City, Kanagawa Prefecture Chome 1-1 Fujitsu Limited

Claims (5)

[Claims] 1. A comparison unit for comparing an input transmission code sequence with a predetermined-length code sequence of a pattern whose amplitude increases when the transmission code sequence is band-limited by a transmission filter, and a comparison between the comparison units. A code amplitude reducing unit configured to reduce the amplitude of at least one code at the center of the predetermined length of the transmission code sequence by a coincidence signal. 2. The comparison section compares a transmission code sequence of orthogonal I and Q channels with a code sequence of a predetermined length of a pattern in which the transmission code sequence is band-limited by a transmission filter and has a large amplitude. There are first and second comparison units corresponding to the I and Q channels, and the code amplitude reduction unit determines which one of the I and Q channels according to the comparison coincidence signal of both the first and second comparison units. 2. The peak reduction apparatus according to claim 1, further comprising a configuration for reducing the amplitude of at least one code at the center of the predetermined length of one or both code sequences. 3. A comparison unit for comparing a transmission code sequence of a predetermined length of a pattern whose amplitude becomes large when a transmission code sequence is band-limited by a transmission filter with the transmission code sequence, and a comparison match signal from the comparison unit. A peak reduction device having a code amplitude reduction section for reducing the amplitude of at least one code at the center of the predetermined length of the transmission code sequence; and the transmission filter for inputting an output signal of the peak reduction device and performing band limitation. A transmission unit that receives and modulates an output signal of the transmission filter, and a transmission amplification unit that receives and amplifies the output signal of the modulation unit. 4. The apparatus as claimed in claim 1, wherein the peak reducing device comprises orthogonal I, Q
A first and second comparing unit corresponding to the I and Q channels for comparing a transmission code sequence of a channel with a code sequence of a predetermined length of a pattern in which the transmission code sequence is band-limited by a transmission filter and has a large amplitude. And the amplitude of at least one code at the center of the predetermined length of one or both of the I and Q channels in accordance with both of the comparison coincidence signals of the first and second comparison units. The transmission device according to claim 3, further comprising a code amplitude reduction unit. 5. The transmission device according to claim 3, wherein the transmission filter has a moderate attenuation characteristic.