JPH07147552A - Tip base selection reception system for spread spectrum signal - Google Patents

Abstract

PURPOSE:To eliminate the effect due to color noise mixed only at a receiver side during transmission by receiving a signal spread spectrum modulated by multiplying a predetermined spread code sequence with data to be sent, and multiplying the signal with a skipped spread code sequence obtained by omitting a predetermined tip from the spread code series used at modulation. CONSTITUTION:When eISMi is not sufficiently larger than an e*ISMi, it can be estimated that fading occurred in the frequency band having an i-th pattern and the signal power is reduced. Therefore, deterioration in the S/N due to fading is blocked by utilizing this fact. That is, the correlated demodulation output for the received input and both the e*SMi and eSMi is obtained and the result is fed to a maximum likelihood discrimination circuit MLD so that the reception input in a fading band does not appear in the demodulated output. In order to accumplish this, a circuit for multiplying a skipped M sequence is provided with an analog gate 15, and the usual multiplication output is given to an input terminal of the gate 15 and the skipped pattern output eg from an SM sequence generator 16 is given to the gate, respectively, and the final data is obtained by an integration device 17 and a discrimination circuit 18.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a spread spectrum communication receiving system capable of improving resistance to noise and the like mixed in during a transmission process.

[0002]

2. Description of the Related Art In spread spectrum communication, the data having a spectrum of a relatively narrow band is spread and transmitted in a wide frequency band by modulating a spread code sequence with data to be transmitted. It is an excellent communication system that exhibits many features such as low transmission power per unit frequency, limiting interference with other communications, and essentially having strong resistance to environmental noise mixed in the transmission process.

FIG. 2 is a block diagram showing a general configuration of a system for performing spread spectrum communication via a wireless communication path. A transmitter TX ₁ is generated by a sequence generator 1 by data D to be transmitted. Product modulation of the spread code sequence
Further, the carrier wave having the frequency f ₀ generated in the oscillator 2 is modulated by the modulated signal to spread the spectrum of the carrier wave including the data D, and then the carrier wave is transmitted to the receiver RX ₁ through the wireless communication path. As the spread code sequence, pseudo noise (PN) having the same bit cycle length as the cycle length of the data D is used.
It is common to use a sequence), and the M sequence, which is the most widely used of the PN sequences, will be described below as an example.

The receiver RX ₁ guides the spread spectrum modulated signal to an amplifier 3 through an antenna (not shown) and amplifies it to a required level, and a local signal f _L (≈f ₀ ) of the local oscillator 4 and a frequency. The mixed signal is taken out through the low-pass filter 5 to demodulate into a spread signal in the baseband. This baseband spread spectrum signal and the same M as the transmitter TX ₁ generated from the sequence generator 6
The sequence code and the sequence code are input to the multiplier 7, where mutual correlation is obtained, and the correlation output is integrated by the integrator 8 for the period of one frame of the M sequence, and the detector 9 is used at the end of the frame. The original data D is demodulated by detecting. The demodulated data is input to the control terminal of the sequence generator 6 via the synchronous detector 10, and the generation timing of the M sequence is controlled so that the phase is synchronized with the received signal.

FIG. 3 is a diagram schematically showing the spectrum of a signal in the process of transmission, in which 11 is a spread spectrum modulated signal and 12 is a spectrum of mixed environmental noise.
When this is demodulated (inverse conversion) by the M sequence in the receiver, the spread spectrum modulation signal 11 which has been spread over a wide frequency band is converted into a narrow band signal 13 and the environmental noise 12 is wide as shown in FIG. Since the signals 14 are distributed in the frequency band, they are attracting attention as a communication system that can essentially limit the influence of environmental noise.

However, since the spread spectrum modulation signal is spread over an extremely wide frequency range, there are many colored noises having a strong power concentration in a narrow band such as the environmental noise 12 in FIG. 3 within the frequency range. When a signal arrives at the receiver via such a transmission system, there is a drawback that the noise level 14 after inverse conversion by the M sequence increases and the S / N deteriorates.

[0007]

SUMMARY OF THE INVENTION The present invention has been made to eliminate the drawbacks of the conventional spread spectrum communication system as described above, and reduces the influence of colored noise having a relatively narrow band spectrum that is mixed in during the transmission process. By receiving S
An object of the present invention is to provide a receiving method for spread spectrum communication with improved / N characteristics.

[0008]

SUMMARY OF THE INVENTION To achieve the above object, the present invention receives a signal which is spread spectrum modulated by multiplying a data to be transmitted or a carrier signal which is modulated by the data by a predetermined spreading code sequence. Then, this is a spread spectrum signal reception method in which spread spectrum demodulation is performed by multiplying this by a skipped spread code sequence in which a predetermined chip is omitted from the spread code sequence used at the time of modulation. Furthermore, in order to reduce the influence of colored noise mixed in the received signal, the skipped spreading code sequence is one in which a predetermined chip is omitted from the spreading code sequence so that the frequency component of the colored noise is not demodulated and output. . Furthermore, the spread spectrum demodulation is performed in parallel with a plurality of skipped spreading code sequences with different omitted chips, and the correlation demodulation outputs are compared with each other by the maximum likelihood determining means to estimate the incoming strong noise frequency range. Then, a plurality of skipped spread sequences that do not demodulate this noise frequency component are selected, and a plurality of demodulated outputs are obtained by demodulating with this, and the one with the maximum likelihood is the demodulated data. It is a thing.

[0009]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Prior to the description of the embodiments, the principle and characteristics of the skipped M-sequence demodulation reception (SMD) system will be described to facilitate understanding of the present invention. In order to avoid the influence of the colored noise from the natural world in the communication environment, a method of removing the colored noise by placing an appropriate filter on the receiver side can be considered. However, when the filter removes the specific frequency component F _E of the received signal, the spread spectrum signal included in the F _E region of the received signal is also blocked. 1 "("
0 ") corresponding to a uniform + as shown in FIG.
The voltage output is not (-), and the demodulation output obtained by integrating the voltage output in which the non-uniform + and-components are mixed is significantly deteriorated. As a result, in general, the error rate could not be sufficiently improved.

The skipped M-sequence demodulation reception proposed by the present inventors is an effective method for realizing such adaptive avoidance of colored noise, and FIG. 5 is a diagram for explaining its principle. A spread spectrum modulated signal is received in the transmission process without being affected by noise, and the same 7-chip (0010111) M sequence e _M as shown in FIG. In the conventional reception demodulation method for performing demodulation by means of the above, the multiplication output e _{MM for} 7 chips has a flat waveform as shown in FIG. 5B. On the other hand, for a received signal transmitted under the same conditions, as shown in FIG. 5C, a skipped M sequence having a missing waveform with a part of the chip omitted from the M sequence used for modulation on the transmitter side. when subjected to demodulation using e _SM, multiplication output e _MSM has a waveform in which part is omitted chip as shown in FIG. 5 (d) missing. That is, the level of the colored noise can be reduced by matching the frequency component corresponding to this missing portion with the frequency component of the colored noise.

When the code length of the skipped M sequence is M and the number of chips to be used is u, the chip utilization coefficient α is α
= U / M, and in this case M = 7, α = 4/7
Becomes Therefore, the powers of the outputs e _MM and e _MSM after spread demodulation are a ² and αa ² , respectively. There are many numbers in the case of selecting the skip position of such skipped M-sequence (SM) signal. It is more efficient to use a sequence group in which the frequency characteristics are complementary to each other.
Therefore, the i-th sequence is e _SMi (i = 0, 1, 2, ...)
Then, it is only necessary to confirm whether or not they are complementary to each other by taking the cross-correlation between the continuous frame which is non-inverted, non-inverted and inverted and the j-th sequence e _SMj . For signals modulated by the M series e _M of 7 chips (0010111) shown in FIG. 5A, two complementary patterns e _SM0 (3, 4, 5) and e _SM1 (1, 6, 7)
The skipped M sequence of is used. Here, the numbers in parentheses show the positions of the skipped chips when the positions of the 7 chips forming the series are given by sequence numbers. FIGS. 6A to 6C show spectrums of waveforms in which the M sequence and the above two skipped M sequences repeat in a data cycle, and the present invention mixes by utilizing this difference in frequency characteristic. The purpose is to avoid the influence of colored noise.

Next, in order to compare the frequency characteristics of this skipped M-sequence demodulation reception with the conventional M-sequence, the demodulation (multiplication) output when a sine wave is input to this model is calculated by the simulation model shown in FIG. Ask. Sequence e to multiply
_M , e _SM0 and e _SM1 are shown in parentheses in the figure. Where e
The input to the integrator is cut off in the chip section where _SM = 0, and the output is binary. 8 (a) to 8 (c) show e _M , e _SM0 (3, 4, 5) or e, respectively.
The demodulation frequency characteristic of the demodulation output e _I obtained by integrating the output obtained by multiplying the sine wave by the waveform of _SM1 (1, 6, 7) every data cycle is obtained. Here, the phase θ of the sine wave of the frequency f has a uniform distribution, and the average value of the absolute value of the demodulation voltage output with respect to θ is e _I (f). This characteristic corresponds to the response when the white noise coming in a uniform amplitude on the use frequency band (0 to F _c) was passed through a matched filter M-sequence or skipped M-sequence. Where f _c
Is the chip rate, and in this example, the length of the M sequence is 7
Since it is a chip, if the data rate is f _d , f _c = 7
It becomes f _d , and in FIG. 8, the horizontal axis shows f _d as a unit. Therefore, e _I (f) is used in the used frequency band (0 to f
Integrating over _c ) gives the following in-band noise power P _ND:

[Equation 1] This is equivalent to obtaining the area on the lower side of the curve of FIG. 8, and the value of the equation shown on the lower side of each graph of FIG. 8 is the simulation result of this area, and the coefficient of the last term of each equation. If is equal to 1.00, it is equal to the theoretical value. Although the length of the M sequence is relatively short, the obtained coefficient agrees with the theoretical value with a high accuracy of 10% or less with respect to 1.00. As a matter of course, there is a strong correlation between the spectrum of FIG. 6 and the demodulation frequency characteristic of FIG. Further, it should be noted that the signal input to the integrator is blocked in the skipped chip section of the skipped M sequence.

The signal-to-noise ratio is estimated in consideration of the above characteristics. The idealized natural environment noise characteristic and the demodulation characteristic of the virtual skipped M sequence corresponding to this characteristic are schematically shown in FIG. In FIG. 9A, N ₀ is the power density of basic noise such as thermal noise and semiconductor noise added during reception amplification, and represents white noise having a flat characteristic over a wide band.
In the figure, pN ₀ is the power density of colored noise with N ₀ as a reference. If the respective noise band groups corresponding to these are B _Li and B _Hi , B _L and B _H are the sums of the respective i. Since it can be assumed that the portion of the skipped M sequence corresponding to the demodulatable chip substantially matches B _Li and the portion corresponding to the skipped chip matches B _Hi , the following approximate expression holds. B _L ≈αf _c = α (B _L + B _H ) ... (2) In such an ideal model, the skipped M-sequence demodulation reception system is not affected by colored noise at all. (Fig. 9 (b))

Now, for convenience, the total white noise power N in the f _c band
Assume that Wc is equal to the signal power a ² . N _Wc = N ₀ f _c ≡a ² (3) Then, the noise in the f D band (one chip of the M sequence, that is, the band corresponding to the data rate) when demodulated by the M sequence is

[Equation 2] Becomes On the other hand, when demodulating with the skipped M sequence, only the band B _L has to be taken into consideration, so that N _PDSM = αN ₀ f _D / 2 = αa ² / (2M) (5). That is, in the M sequence demodulation, + a and −a are obtained as non-inverted and inverted outputs, respectively, and the signal power is S = (2a) ² from the amplitudes thereof, but in the skipped M sequence demodulation, the signal power is (2a) ² . It becomes α times. Therefore, the S / N ratio after demodulation is

[Equation 3]

[Equation 4] Becomes That is, the expression (7) corresponds to the case where p = 1 in the expression (6). FIG. 10 shows the S / N ratio given by the equations (6) and (7) with respect to p corresponding to the level of colored noise, and the skipped M-sequence demodulation reception system has an influence of colored noise on the ideal model. It will be more obvious that it has the characteristic of not receiving.

Next, the effect of colored noise on a realistic model is evaluated based on the result of the above simulation. The demodulation output e _I (f) of FIG.

[Equation 5]

[Equation 6] Then, the demodulation characteristic will be as shown in FIG. Here, (a) and (b) are the models [NM1] and [NM2] demodulated in the M sequence assuming P = 10, and (c) and (d) are the skipped M sequence in the same model. It was demodulated in. The in-band noise output P _NPD is obtained by integrating this e _IP (f) using the equation (1), and the S / N ratio after demodulation is (S / N) = 4a ² / P _NPD ·・ Given in (10). Therefore, M sequence e _M and skipped M
The S / N ratio calculated for the series e _SM0 and e _SM1 is shown in FIG.
The characteristics are as shown in. According to the figure, it can be confirmed that even in the case of a realistic model, that of the present method is significantly higher than the S / N ratio of the conventional method in the region of large p. For example, the difference ΔS / N between the S / N ratios when the conventional method e _M and the skipped M sequence e _SM0 are used for the noise model (NM1) shown in FIGS. 11A and 11C is shown in FIG. That is the case, and an extremely large improvement effect can be obtained.

FIG. 1 shows a block diagram of an embodiment of a spread spectrum system using a skipped M-sequence demodulation receiver embodied based on the above principle. The most important issue in realizing this method is to find an effective means for detecting the noise component contained in the received wave.
For example, a method using a so-called DFT transform is generally known, but since the configuration is complicated and processing time is required, here, the correlation demodulation by the different type M sequence e ^* _M and the different type skipped M sequence e ^* _{SM is performed.} We propose a simpler method that uses output.

The different type M sequence e ^* _M has the same sequence length as the M sequence e _M and has a different code, and the different type skipped M sequence e ^* _SM is a skipped M sequence derived from the different type M sequence e ^* _{M. Yes,} it is assumed that e ^* _SMi and e _SMi have almost equal demodulation characteristics. The received modulated wave is e _IM , e ^* _IM by performing correlation demodulation by e _M , e ^* _SM and e ^* _SMi.
And get e ^* _ISMi . However, e ^* _ISMi is multiplied by α ^-1 to match the power level with other outputs. These values are compared with each other by the maximum likelihood determination circuit MLD ₁ and the outputs are ranked in descending order. For example, when the maximum output is obtained as follows, the respective judgment results are shown on the right side. Maximum output _{^{= e * ISM0: 0 ≦ B}} H ≦ 2.5f D e * ISM1: 2f D ≦ B H ≦ 4.5f D e * IM: ( no need to use a skipped M sequence) flat noise e _IM: (S / N) _M >> 1 (same as above)

A sequence to be used for correlation demodulation is determined based on the result of the ranking (when it is determined that the _{bands in which} the output of e ^* _ISMK is particularly large and noise is present are, for example, H ₁ and H _2. This is equivalent to selecting a skipped M sequence e _ISMK whose output characteristic e _I (f) for signals in the H ₁ and H ₂ bands is close to 0. Therefore, e ^* _ISMK and e _ISMK have complementary characteristics and the skip position However, the selection circuit PS selects a desired generally plural series. The demodulation output by these sequences is
To the maximum likelihood decision circuit MLD ₂ for making a final decision. Here, the reason for using e ^* _M in particular is to avoid the influence of a relatively large signal component included in the received wave as much as possible. Also, as a matter of course, the code length M
It is predicted that a large number of sequences with more ideal S / N characteristics will be found by increasing If Figure 9
If there is a sequence that can realize the ideal characteristics as shown in (b), the S / N ratio hardly depends on the increase in the level of colored noise, and the M sequence has a low power density N.
The S / N when the white noise of ₀ is added can be retained.

Further, in the above-described embodiment, the multiplication of the received input by the skipped M sequence results in multiplication of the analog input by three values (+1, -1, 0), resulting in a complicated configuration. Instead of this ternary multiplication, as in the case of normal M-sequence demodulation, a method of first performing multiplication by binary (+1, -1) and processing by the skipped pattern can be used. FIG. 13 (a) is a diagram showing the embodiment thereof, in which reference numeral 15 is an analog gate (AG) which can be easily realized by an FET transistor, and a normal multiplication output is input to the input terminal (source) of the AG15. If the skipped pattern binary output e _g from the SM sequence generator 16 (showing the value corresponding to e _SM0 in the figure) is added to the gate terminal of the AG 15, On the output side, the outputs corresponding to the third, fourth and fifth chips are blocked and do not appear. Then, this output is added to the integrator 17, and the decision circuit 18 makes a decision to obtain the final data output.

As described above, by using a binary or ternary skipped pattern, it is possible to reduce a desired frequency component of received data and reduce unnecessary components such as noise, but the above-mentioned simulation is performed. It is difficult to say that the ideal characteristics as shown in FIG. 9 (b) are obtained, for example, by increasing the code length M, the sequence in which the S / N characteristics become more ideal. Although it is predicted that a large number of colored noise characteristics can be found, it is actually not easy to deal with various colored noise characteristics. Therefore, the inventors of the present application have found a method of providing a multi-valued pattern or an analog pattern in place of the binary or ternary skipped pattern. For example, in the case of e _g shown in FIG.
In addition to the level e _ON and the OFF level e _OFF , there is an intermediate level (however, the same voltage in the chip). The higher the voltage, the smaller the attenuation amount when the signal passes through the AG 15.
In this case, e _g is not a simple skip operation, but a higher-level control operation, so this is called a control pattern. By performing such multi-value (or analog) control, the frequency characteristic of the demodulation output e _I (f) obtained in FIG. 8 can be brought close to the ideal characteristic shown in FIG. 9 (b). Therefore, the effect of improving the S / N ratio can be further enhanced. In the following, spread spectrum demodulation using an attenuated spread code sequence exhibiting a control pattern will be described in some detail.

FIG. 14 (a) shows M used for spreading modulation on the transmitting side.
The waveform of the sequence e _M is shown, and the code length is M = 15. (B) to (d) are skipped M sequences e _SM and shaping sequences obtained from this e _M , respectively.

[Outer 1] (Hereinafter, it is expressed as (eΔ) only in the specification), showing an example of the waveform of the attenuated M-sequence e _{SS according to} the present invention. The attenuated M sequence e _SS is obtained by adding and synthesizing the skipping M sequence e _SM and the shaping sequence (eΔ). Figure 1
FIG. 5 shows frequency spectra of these sequences by DFT analysis, and shows that of the M sequence e _M (a).
4f _D and 5f of skipped M series e _SM of (b) compared to
It can be seen that the residual component appears although the component in _D is small. On the other hand, in the figure (c) showing the spectrum of the attenuated M-sequence e _{SS according} to the present invention, 4f _D and 5f _D
The component at is almost zero. That is, in the above-mentioned skipped M sequence e _SM , a slight residual appears in the spectrum component to be removed, whereas in the preferred shaping sequence (e
This residual component can be canceled by adding Δ).

In practice, a configuration as shown in FIG. 13 (a) can be considered. In the figure, reference numeral 15 is an analog gate (AG) that can be easily realized by an FET transistor. The received signal (LPF output in FIG. 6) is multiplied by the M sequence used during modulation, and the output is input to the AG15. It is led to a terminal (source) and, for example, a skipped pattern binary output r _g from the SM sequence generator 16 as shown in FIG.
₍ Indicating the value corresponding to _SMO ) is added to the gate terminal of AF15, the outputs corresponding to the third, fourth and fifth chips are blocked and do not appear on the output side. Therefore, binary ON / OF
For example, instead of F skipped patterns, the e _g such multilevel pattern shown in FIG. (C), or the analog pattern can be realized demodulation by damping M-sequence by providing the gate terminal of AG15. This output is then used by the integrator 1
Then, the final data output can be obtained. e _g is a ON level e _ON, an intermediate level is present in addition to the OFF level e _OFF, (provided that the same voltage in each chip) to reduce the attenuation amount when the passing signal AG15 higher voltage . By performing such multi-value (or analog) control, the frequency characteristic of the demodulation output e _I (f) obtained in FIG. 8 can be brought close to the ideal characteristic shown in FIG. 9 (b). Therefore, the effect of improving the S / N ratio can be further enhanced.

Here, the polarities of the waveforms of the chip numbers 3, 6, and 8 of the attenuated M series e _SS of FIG. 14 and the polarities of the waveforms of the same chip numbers 3, 6, and 8 of the original M series e _M are different from each other. This occurred when the shaping sequence (eΔ) was introduced. Therefore, no signal input corresponding to e _SS at the chip point is input to the input side of the analog gate AG15 of FIG. 13 (a). Therefore, a circuit that accurately obtains an e _SS compatible output is actually required. FIG. 16 (a) is a block diagram showing an embodiment of a system that enables strict shaping correction. Figure 1
This is a structure in which the circuit of the dotted line is added to 3 (a). That is, the received baseband prepares an output multiplied by the M-sequence and inverted M-sequence with respect to the input signal, these chips output AG ⁺ and AG ^- respectively is added. Gate control inputs e _g ⁺ and e _g ⁻ are added to AG ⁺ and AG ^{−, respectively} , and the amount of transmission is controlled by the multilevel pattern shown in FIGS. 16 (b) and 16 (c). As a result, Fig. 14 (d)
The transmission amount of e _SS shown in can be accurately realized. e _g ⁺ and e _g ^-
Comparing the two, the former is dominant. e _g ^- is used exceptionally. In this case, the frequency spectrum 4 of FIG.
It was introduced for shaping in order to make f _D and 5f _D as small as possible. e _g ^- transmission amount by the use exceptionally so has the effect of reducing the integration output to the signal. Therefore, if the desired spectrum can be made small enough,
It is not necessary to use the amount of transmission by e _g ^{−, and} the circuit of FIG. 13 is sufficient. For reference, FIG. 16 (d) shows FIG. 13 (b).
Gate control input e _g corresponding to the same skipped M sequence as
showed that. A shaping analog M sequence A (t) having the same effect as the shaping correction by such multivalues will be described. A
(T) is formed by shaping the M sequence e _M (t) used on the transmitting side. The preparation method is as follows. First, e _M (t) is analyzed by DFT, and as a result, the DC component S ₀ and M frequency components S
_{It is} assumed that _j (j = 1, 2, ... M) is obtained. Figure 14 (a)
Is a method of providing a constant demodulation level indicated e _SM, e _SS in each chip (b). Here, each chip is divided into J, a demodulation waveform that gives an appropriate level to each time position is defined as e _F, and a method that uses e _F to realize the ideal characteristics shown in FIG. 9 (b) is used. Think Now, assume that the ideal characteristic is, for example, a spectrum obtained by removing f = 4f _D and 5f _D from the characteristic shown in FIG. When the number of chips (code length) of the M series is M and the number of divisions of each chip is J, the number of sample points of the M series is N = MJ. Therefore, the values V _i of N sample points
(I = 0, 1, 2, ... N-1) is a sample value ± 1 of the original M series of M chips, and is the same for every J chips from the beginning. (Here, if J >> 1, the aliasing noise generated in the DFT analysis can be ignored.) When the DFT analysis is performed using this set of V _i , the real part α _k and the imaginary part β _k (k = 0, 1, 2, ... N-1) is obtained. Between V _i and γ _k = α _k −jβ _k are connected by DFT transformation. That is,

[Equation 7] In FIG. 15A, the vertical axis _represents the absolute value of γ _k and the horizontal axis represents kf _D. Now, from all frequency components of this M sequence, q, q ',
i of the waveform corresponding to the components excluding the q "th frequency component
If the th sample value is V _i ', using the inverse DFT,

[Equation 8] The values of N sample points are determined. This weighted waveform e _W
If (t) is used as the despread waveform on the demodulation side, the desired purpose can be achieved. That is, the frequency qf included in the received signal
Noise components in the vicinity of _D , q′f _D , q ″ f _D ... Are hardly generated in the demodulation output obtained by multiplying the received signal by e _W (t), and the influence thereof can be removed. Here, for example, using the M series of FIG. 14 (a) and considering the components obtained by removing f = 4f _D and 5f _D from the spectrum shown in FIG. 15 (a), this is q = 4, q ′ = 5, q ″ = 0, ...
Is equivalent to J = 1 using the above formula (3 ′)
FIG. 20 (b) shows the result of performing the inverse DFT with 0, M = 15 and N = 150. Each chip of the waveform is composed of 180 sampled values and becomes an analog waveform. Note that FIG. 20 (a) shows the same original M series as in FIG. 14 (a). This method has the advantage that the frequency component to be removed can be made completely zero, and that its waveform is always required. Since an analog waveform is used, high precision is required for circuit operation, but a great effect is obtained when the noise component is large.

FIG. 17 is a diagram showing another embodiment of the present invention, in which the received signal is NBE (nois) for each frame.
e band estimation) to detect in which frequency range the disturbing colored noise exists. Based on the NBE analysis result, an attenuation M sequence required to realize a spectrum suitable for removing colored noise is selected by a control unit (not shown). Based on this information, CCC (chip cha
The nnel controller) demodulates with a skipped M sequence so as to exhibit a function equivalent to the selected attenuated M sequence, and controls the resistance value of the analog gate by a shaping sequence. Can be maintained.

Further, in the embodiment of FIG. 1, when e _ISMi is not sufficiently _larger than e ^* _ISMi , fading occurs in the frequency band of the i-th pattern, and the signal power decreases. Since it can be presumed that this is the case, it is possible to use this phenomenon to avoid deterioration of the S / N ratio due to fading. In this case, in FIG. 1, the reception input and the correlation demodulation outputs for both e ^* _SMi and _eSMi are obtained, and these can be added to the MLD having the expanded function to perform the above determination. That is, the skipped M sequence may be selected so that the received input in the fading band does not appear in the demodulated output, and the final correlation modulation and the determination by MLD2 may be performed.

Although the present invention has been described above by taking as an example the case where a signal subjected to spread modulation by an M sequence is demodulated by a skipped M sequence, the present invention is not limited to this and other than the M sequence. For example, it can be universally applied to other code sequences such as GOLD code. For example,
A case where a discrete chirp signal is used as a transmission signal will be described. When the length of the signal to be transmitted is N as shown in FIG. 18 (a), it is expressed by different equations depending on whether N is an even number or an odd number. When N is an odd number,

[Equation 9] When N is an even number,

[Equation 10] The instantaneous frequency of this signal (obtained by differentiating the phase with respect to time) is as shown in FIG. 18 (b). Therefore, assuming sufficient synchronization between the transmitter and receiver, the frequency of the signal will be time dependent. As shown in FIG. 19, considering a window function lacking a time range corresponding to a certain frequency range, if the signal obtained by multiplying the transmission signal S ₀ is S _i , then S _i
The matched filter with respect to has a small correlation with the frequency corresponding to the missing portion in the figure. Therefore, when the colored noise in the transmission line uses the frequency component of S _i corresponding to the missing portion in the figure as the main energy, the matched filter outputs S _{1 to} S _i-1 and S _{i + 1} ... The output of s _i should be smaller than that of, and this means that the output of the noise component is most removed. Here, S _{1 to} S
_i−1 , S _{i + 1,} ... Are signals multiplied by a window function different from S _i, and parts corresponding to specific frequency ranges are lacking. With such a method, it is possible to improve the transmission characteristics when colored noise is added to the communication path.

Further, although only the wireless transmission is mentioned in the embodiment, it may be applied to the one which performs the transmission via a wire, for example, in the case of using a distribution line as a transmission line, 50 Hz ( Alternatively, since it is predicted that colored noise will occur at a frequency of 60 Hz) and its harmonic frequencies, the skipped M sequence may be selected in advance so as to skip and demodulate the frequency.

[0028]

Since the present invention is configured as described above, no influence is applied to the transmitting side, and the influence of the colored noise mixed during transmission on the receiving side regardless of the transmitting side is exerted. It has a remarkable effect on the spread demodulation with reduced.

[0029]

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of a spread spectrum system using a skipped M-sequence demodulation receiving apparatus according to the present invention.

FIG. 2 is a block diagram showing a configuration of an example of a conventional spread spectrum system.

FIG. 3 is a diagram schematically showing a spectrum of a spread signal in a transmission process.

FIG. 4 is a diagram schematically showing a spectrum of a signal subjected to spread demodulation.

5A to 5D are diagrams for explaining the principle of a skipped M-sequence demodulation reception system.

6A to 6C are diagrams showing spectra of an M sequence and a skipped M sequence, respectively.

FIG. 7 is a simulation model diagram of the skipped M-sequence demodulation receiver.

8A to 8C are frequency characteristic diagrams when spread demodulation is performed on an M sequence and a skipped M sequence, respectively.

9A and 9B are diagrams schematically showing idealized natural environment noise characteristics and demodulation characteristics of a virtual skipped M sequence corresponding to these characteristics.

FIG. 10: S / for p corresponding to the level of colored noise
The figure which shows N ratio.

11 (a) to (d) are demodulation characteristic diagrams when spreading and demodulating with a skipped M sequence.

FIG. 12 is a diagram showing an S / N ratio when spread demodulation is performed on an M sequence and a skipped M sequence.

13A to 13C are explanatory diagrams of an embodiment in the case of multiplying a binary or multivalued skipped M sequence.

14A to 14D are diagrams for explaining the principle of spread spectrum demodulation using an attenuated M sequence.

15A to 15C are diagrams showing spectra of an M sequence, a skipped M sequence, and an attenuated M sequence, respectively.

16A to 16D are diagrams showing another embodiment of spread spectrum demodulation using an attenuated M sequence.

FIG. 17 is a diagram showing yet another embodiment of spread spectrum demodulation using an attenuated M sequence.

18 (a) and 18 (b) are diagrams showing a signal length and a signal instantaneous frequency, respectively.

FIG. 19 is a conceptual diagram illustrating a window function.

20A and 20B are diagrams illustrating a shaped analog M sequence.

[Explanation of symbols]

 1, 6 ... Sequence generator 8 ... Integrator 9 ... Wave detector 10 ... Sync detector TX ... Transmitter RX ... Receiver

Claims (11)

[Claims] 1. A skipped spread code sequence obtained by multiplying data to be transmitted by a predetermined spread code sequence to obtain a spread spectrum modulated signal, and omitting a predetermined chip from the spread code sequence used at the time of modulation. A chip-based selective reception method for spread spectrum signals, which is characterized by performing spread spectrum demodulation by multiplication. 2. The spread code sequence is an M sequence code, and a skipped M sequence code in which a predetermined chip is omitted from the M sequence code is used as the skipped spread code sequence. Chip-based selective reception method for spread spectrum signals. 3. A method in which a predetermined chip is omitted from the spreading code sequence as the skipped spreading code sequence so that a component of colored noise mixed in a received signal is not demodulated and output. 3. The chip-based selective reception system for spread spectrum signals according to claim 1 or 2. 4. Spread spectrum in parallel with a different spreading code sequence having the same sequence length as the spreading code sequence and a small cross-correlation coefficient, and a plurality of different skipped spreading code sequences formed based on the different spreading code sequence. The demodulation is performed, and the respective correlation demodulation outputs are compared with each other by the maximum likelihood determining means,
4. The spread spectrum signal according to claim 1, wherein the spread spectrum demodulated data is the spread spectrum demodulated by the skipped spread code sequence corresponding to the frequency characteristic of the different type skipped spread code sequence having the maximum likelihood. Chip-based selective reception method. 5. The chip output corresponding to the omitted chip among the chip outputs obtained by multiplying the received signal by the same spreading code sequence as that on the transmitting side in advance as means for demodulating with the skipped spreading code sequence on the receiving side. 5. A chip-based selective reception system for spread spectrum signals according to claim 1, wherein transmission is blocked and remaining chip output is integrated over the period of the data. 6. The skipped spreading code sequence,
6. The spread spectrum signal according to claim 1, wherein spread spectrum demodulation is performed using an attenuated spread code sequence created by applying appropriate attenuation or amplification to these chips instead of omitting the predetermined chips. Chip-based selective reception method. 7. A chip of the sequence created by multiplying the data to be transmitted by a predetermined spreading code sequence to receive a spread spectrum modulated signal and multiplying this by the spreading code sequence used at the time of modulation. A non-inverted output and an inverted output are given appropriate attenuation or amplification, and then integrated for one frame period (code sequence length) and added to a decision circuit. Chip-based selective reception method. 8. A chirp signal whose frequency changes corresponding to chips forming a frame is used as a spreading code sequence to be multiplied by the data to be transmitted, and when the reception side multiplies the chirp signal to perform correlation detection, the correlation is calculated. 7. Proper attenuation or amplification of the output of the detection chip.
A chip-based selective reception method for the spread spectrum signal described. 9. A chirp signal whose frequency changes corresponding to chips forming a frame is used as a spreading code sequence to be multiplied with data to be transmitted, and when the reception side multiplies the chirp signal to perform correlation detection, the reception is performed. 7. A chip-based selective reception system for a spread spectrum signal according to claim 6, wherein demodulation is performed by multiplying each chip of the side chirp signal by an appropriate attenuation or amplification and an attenuated spread chirp signal created. 10. A colored noise frequency among spectrum components obtained by receiving a signal which is spread spectrum modulated by multiplying data to be transmitted by a predetermined spread code sequence and performing DFT analysis on the spread code sequence. A reference time waveform consisting of a plurality of sampling points per chip obtained by inverse DFT transforming spectral components excluding a plurality of neighboring frequency components, or a time waveform obtained by applying a predetermined modification to the reference time waveform, A chip-based selective reception method for spread spectrum signals, characterized by multiplying a received frame signal and integrating it to determine transmission information. 11. A chip signal produced by multiplying a received signal by a spreading code sequence used on the receiving side is attenuated or amplified corresponding to the reference time waveform and then integrated over one frame period. 11. The chip-based selective reception system for spread spectrum signals according to claim 10, wherein the output is added to a decision circuit.