JP3244434B2 - Spread spectrum communication reception method - Google Patents

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 method, and more particularly to a spread spectrum communication method using direct spread in wireless communication or wired communication, and is widely used for digital data transmission. Reception method.

[0002]

2. Description of the Related Art In recent years, a spread spectrum communication system has attracted attention as a new communication system. The modulation system used for general data communication is a narrow band modulation system, which can be realized with a relatively small circuit. However, such as a room (office, factory, etc.), for multipath or narrow band colored noise. Has the disadvantage of being weak.

[0003] On the other hand, the spread spectrum communication system has an advantage that these drawbacks can be eliminated because the spectrum of data is spread by a spread code and transmitted in a wide band.

[0004] However, on the other hand, the spread spectrum communication system requires a wide band with respect to the data transmission speed, so that high-speed data transmission has been difficult. For example, when transmitting by spreading with an 11-chip spreading code,
Considering the case of transmission using PSK modulation, 2MBP
A 22 MHz band is required for S data transmission. If sending 10 MBPS data, 11
A band of 0 MHz will be required. However, since the band that can be transmitted wirelessly is limited, it has been difficult to transmit high-speed data.

The inventor of the present application has proposed a method of delaying and multiplexing a spread signal (hereinafter referred to as a delay multiplexing method) as a method of performing high-speed transmission in a limited band in Japanese Patent Application No. 7-206159. Proposed. By using this method, high-speed transmission can be performed in a limited band. In this proposed method, 2 multiplexing and 4 MBPS
When 5 data are multiplexed, 10 MBPS data can be communicated.

FIG. 1 shows a configuration in which a parallel system of a multiplier and a delay element is increased by one in the proposed configuration. In FIG. 1, data generated by a data generator 1 is differentially encoded by a differential encoder 2, and then parallel-converted by a serial-parallel converter (SP converter) 5 into a number to be multiplexed. You. Each parallel signal is applied to multipliers 4 to 8 and is multiplied by the PN code from PN generator 14 and spread. Thereafter, the signals are respectively delayed by the delay elements 9 to 13 and multiplexed by the multiplexer 15 to become a multilevel digital signal. This digital signal is modulated by the modulator 16, frequency-converted by the frequency converter 18, power-amplified by the power amplifier 19, and transmitted via the antenna 20.

[0007] Here, as an example, consider a case where five multiplexes are performed using an 11-chip Barker code as a spread code. In this case, a Barker code is prepared in the PN code generator 14. The Barker code is (101101110)
00), which are generally well-known codes.

Further, in the delay elements 9 to 13, considering that 11 chips are divided into five multiplexes, four are delays of two chips and one is delay of three chips. Here, assuming that the chips have a delay difference of 2, 2, 2, 2, and 3 chips, respectively, the first delay element 9 may have a delay of 0 chip (that is, no delay element is required), and , 3, 4 and 5th delay elements 10 to 13 have delay times of 0, 2, 4, 6, and 9 . As a result, each is 2,2,2,3
Chip delay.

FIG. 2 is a block diagram showing a configuration of a receiver for receiving a signal delayed and multiplexed as described above. In FIG. 2, a signal received by an antenna 21 is
After being converted into an intermediate frequency signal by 2, the frequency converter 23 converts the signal into a baseband signal based on the local oscillation signal from the local oscillator 24. The baseband signal is correlated by the correlator 25, distributed by the distributor 26, and latched by the latch units 27 and 28 based on the signal from the latch controller 29. The differential portion of the latch output is extracted by the differential section 30 and is discriminated and demodulated by the discriminating section 31. Here, according to the above-described specific example, the data is latched by the latch units 27 and 28 in two chips or three chips. This makes it possible to demodulate a multiplexed signal in a delay-multiplexed communication system, thereby performing high-speed data communication.

[0010]

1 and 2 described above.
By using the example shown in (1), high-speed transmission is possible in a limited band, but there is a disadvantage that the error rate characteristic is deteriorated. This will be described in detail below.

Since the Barker code is composed of 11 chips, when a single non-multiplexed signal is considered, if a correlation is obtained, a correlation value of 11 in a correlation spike can be obtained. Here, the correlation spike refers to a correlation signal or timing when a correlation is obtained in spread spectrum.
However, when delay multiplexing is performed, there is a problem that the correlation value varies. The reason will be described with reference to FIG. FIG. 3 is a diagram showing a case where the Barker code is given 1 as 1 and 0 as -1 as digital values, multiplied by arbitrary data, multiplexed by 5 and correlated, and the absolute value of the correlation value at each correlation spike The horizontal axis indicates the time sequence, and the vertical axis indicates the correlation value at the correlation spike. As is clear from FIG. 3, the one that should originally obtain 11 correlation values has five values of 7, 9, 11, 13, and 15. The reason will be described.

FIG. 4 is a diagram showing the autocorrelation characteristics of the Barker code. Here, the data is taken as (1, 1),
Four types of (-1, -1), (1, -1), and (-1, 1) can be considered. In this case, each shows a different autocorrelation value. Here, EVEN is an even correlation, and ODD
Is an odd correlation. Considering the case of delay multiplexing, spreading and multiplexing are performed. However, since linear sum is maintained in spectrum spreading, when demultiplexing a multiplexed signal,
Each correlation value is also a linear sum. That is, when multiplexed, a linear sum of the respective autocorrelations can be obtained as a correlator output. In other words, an uncorrelated portion other than 11 or -11 called an autocorrelation side lobe has an adverse effect.

FIG. 5 is a diagram showing an example. FIG. 5 (a)
To (e) are correlation outputs originally obtained. Each is delayed by two or three chips. And
FIG. 5F shows the sum of the linearly summed signals.
It is. Therefore, the signal in the demodulator in the case of delay multiplexing corresponds to this signal. As a result, it can be seen that the signal at the time of the correlation spike is 7, 9, 11, 13, 15 (absolute value).

FIG. 6 is a diagram showing a signal point distribution when QPSK modulation is used when the correlation values vary as described above. This shows the signal points after despreading on the phase plane on the I and Q axes, of which only the first quadrant is shown.

[0015] Originally, if the correlation value is 11, the signal should come at the double circle point. However, as described above, in the proposed example, the signal point is broad as shown in FIG. It will spread. The spread at this time is 39.96 °
become.

Now, when demodulating the phase modulation,
The determination of data is performed based on the phase angle. For example, in QPSK, when the phase angle is 0-90 °, the data is determined as (1, 1), when the phase angle is 90-180 °, the data is determined as (-1, 1). Therefore,
Originally, when the signal point is (1, 1), there is a signal at a point of 45 °, so the noise margin Δθ is from + 45 ° to −45 °. However, when multiplexing is performed in the example proposed earlier, for example, in the case of a signal point having a correlation output of (15, 7), the angle has a signal point at 25.02 °, and the noise margin Δθ is − The side is -25.
The angle was 2 °, causing deterioration of characteristics. In the example shown in FIG. 1, since the DQPSK method is used, the angular spread after the differential is 39.96 × 2 = 79.92 °,
The angle margin is further reduced.

FIG. 7 is a diagram showing the deterioration of the error rate in that state. The deterioration at this time is, for example, BER = 10
In E-4, (17.2-3.4 =) becomes 13.8 dB.
Of these, 7 dB corresponds to the power per signal that is reduced to 1/5, so that the signal points are spread, that is, the deterioration due to the correlation output changing to 7-15 due to the deterioration due to the autocorrelation is 6. .8 dB. As described above, the delay multiplexing method in the above-mentioned proposed example has a problem that although high-speed transmission is possible, the error rate is deteriorated.

Therefore, a main object of the present invention is to provide a spread spectrum communication receiving method capable of high-speed transmission and having a low error rate.

[0019]

The invention according to claim 1 is
In spread spectrum communication that performs direct spreading, a system that delays and multiplexes a signal spread with the same spreading code by an arbitrary number of chips at a time and transmits it. Regardless, in a system using a code uniquely determined by one of the preceding data and the succeeding data, the correlation of the data timing for each (multiplex number -1) is held before and after the correlation to cancel. , Autocorrelation by selecting and adding the correlation values of the previous data and the subsequent data that are uniquely determined by selecting the previous data and the subsequent data, dividing by the spreading factor, and adding / subtracting the correlation value to be canceled. Is configured to cancel side lobes.

According to a second aspect of the present invention, in the case where the correlation value of the first aspect is blocked by the number of multiplexes and one of the blocked data is canceled, at the receiver side, other preceding data, The transmitter must ensure that all subsequent data combinations have a delay that can be processed with the data in the block.
Control the delay amount of the delay multiplexing, hold the correlation only for the block, select and add the correlation values other than the signal to be canceled from the correlation values, divide by the spreading factor, and cancel the cancellation. The side lobe of the autocorrelation is canceled by adding or subtracting the correlation value to be applied.

According to the third aspect of the present invention, when using a 11-chip code in which a side lobe of an autocorrelation is used as a reference signal used for canceling a correlation when using a Barker code, a value to be divided is k. 1 for multiple
1−k + 2, the side lobe of the autocorrelation has the same sign as the reference signal used for canceling the correlation 13
When the code of the chip is used, the value to be divided is divided by 13 + k-2 when k multiplexing is performed, so that the correlation value of the cancel code becomes a constant value, and the characteristics can be further improved.

According to a fourth aspect of the present invention, in the first or second aspect of the present invention, when demodulation using PDI, cancellation is performed even at the demodulation timing used for PDI in order to cancel side lobes of autocorrelation. Maintaining the correlation of the data timing of each (multiplex number) before and after the correlation centering on the correlation corresponding to the PDI to be applied, selecting the preceding data and the succeeding data, and determining the preceding data and the succeeding data which are uniquely determined. By selecting and adding the correlation value, dividing by the spreading factor, and adding or subtracting the correlation value at the timing of the PDI to be canceled, the side lobe of the autocorrelation is canceled, thereby improving the characteristics at the time of the PDI.

According to a fifth aspect of the present invention, in the case where the correlation value of the first aspect is divided into blocks by the number of multiplexes and one of the blocked data is canceled, all the data of the block to which the canceled data belongs is included. Is added, divided by the spreading factor, and added to the correlation value to be canceled to cancel the side lobe of the autocorrelation.

In the invention according to claim 6, the division of claim 5
, When used Barker marks <br/> No. 11 chips autocorrelation sidelobes become different signs in signal as a reference to be used for cancellation of the correlation value division in the case of k multiple 1
Divide by 1-k + 1, and the side lobe of the autocorrelation has the same sign as the reference signal used for canceling the correlation 13
When using the Barker code of the chip, the value to be divided is k
In the case of multiplexing, division is performed by 13 + k-1.

According to the seventh aspect of the present invention, when an 11-chip Barker code is used, the value to be divided is set to 8 and the lower 3
The division operation is performed by removing the bits and shifting the data by 3 bits. When a 13-chip Barker code is used, the value to be divided is set to 16, the lower 4 bits are removed, and the data is shifted by 4 bits. .

In the invention according to claim 8, the gain is further varied to control the amplitude level of the received signal, the output is digitized and quantized, the gain is controlled using the correlation output, and the reference correlation is used. As the output, the correlation output after canceling the side lobe of the autocorrelation according to claim 1 is used.

[0027]

FIG. 8 is a diagram showing a receiving system according to a first embodiment of the present invention. The transmission system shown in FIG. 1 is used. In FIG. 8, a signal received by an antenna 21 is converted by a frequency converter 23 into a local oscillator 24.
After being frequency-converted into a baseband signal based on the local oscillation signal from, a correlation is obtained by a correlator 25. This correlation is latched by the latch unit 32 at the timing of the correlation spike, and then the deterioration due to the autocorrelation is canceled by the correlation processing unit 33. The correlation output is given to the distributor 26 and distributed, and the control signal from the latch controller 29 Is latched by the latch units 27 and 28. Here, according to the above-described specific example, the data is latched by the latch units 27 and 28 in two chips or three chips.
The outputs of the latch units 27 and 28 are differentiated by the differential unit 30, and thereafter are discriminated by the discriminating unit 31 and demodulated.

FIG. 9 is a block diagram showing a specific example of the correlation processing section shown in FIG. Correlation processing unit 3 shown in FIG.
3 is an example in the case of five multiplexes. Also, in FIG.
Although only the sequences are shown, two sequences are required to make the example shown in FIG.

The input signal is input to a shift register 34 corresponding to the number of input bits, and four correlation spikes before and after the correlation spike at a desired data demodulation timing are held. These signals are calculated by the calculator with select function 35 based on the timing signal generated from the timing generator 36. Note that the timing generator 36 adjusts the timing of the input signal and the output signal by a signal of a correlation synchronization circuit (not shown).

FIG. 10 is a diagram showing a specific configuration of the arithmetic unit with select function 35 shown in FIG. In FIG. 10, the arithmetic unit with select function 35 includes selectors 351 to 35.
4, an adder 355, a divider 356, a timing controller 357, a latch unit 358, and an adder / subtractor 359.

FIG. 11 is a timing chart for explaining the operation of the operation unit with select function shown in FIG. FIG.
In FIG. 1, assuming that the correlation spike to be obtained is the signal of FIG. 11E, four side lobes changing the value of the correlation spike of this signal are interposed between the four desired signals of the signals 1-4. It is. That is, A and F of the signal shown in FIG. 11A, B and G of the signal shown in FIG.
These are C and H of the signal shown in FIG. 11C, and D and I of the signal shown in FIG. 11D. A combination of these four signals (E
(VEN, ODD), the autocorrelation side lobes (E1, E2, E3, E4) at the timing of E are determined. As a result, these signals are added to E to become the signal of E 'shown in FIG. . Here, each signal E1, E2, E
Let us consider 3 and E4 in more detail.

FIG . 4 shows the state of the autocorrelation of the Barker code.
You. In this invention, data can only take 1, -1
So, two connected data can take four kinds of data
You. That is, (1,1), (-1, -1), (1,-
1), (-1, 1)
The autocorrelation values are shown here. Correlated points (phase
The correlation value at the spike timing) is
Since the marker code is used, it becomes 11 or -11.
If you look carefully at these autocorrelation value data,
Of the correlation spike timing data
The correlation value after the even chip from the data (previous data) at
Is the data at the later correlation point (post data) 11 or -11?
Regardless, the value of the previous data is different sign and the absolute value is 1
The correlation value after the odd chip is
Regardless of whether the data is 11 or -11,
The correlation value has a different sign and an absolute value of 1.

Considering this for the signal E1, E
Since 1 is a signal after an even number of chips (8 chips) from the correlation spike of A, it is an opposite sign to the data of A and -1.
Becomes Considering the signal E2, since E2 is a signal after even-numbered chips (six chips) from the correlation spike of B, it has a different sign to the data of B and becomes -1. Considering this for E3, since E3 is a signal after an even number of chips from the correlation spike of C, it has a different sign to the data of C and becomes -1. Considering this with respect to E4, since E4 is a signal after an even number of chips from the correlation spike of D, it is an opposite sign to the data of D and becomes -1.

From the above, to return the signal of E 'to E, subtract E1, E2, E3, and E4 from E', that is, 1 / E, A, B, C, and D with different sign components. You just have to subtract the value of 11. This means that A, B, C, D
Is equivalent to the result obtained by adding 1/1 to the result of adding 1 '.

Referring again to FIG. 10, A and F are provided in selector 351, B and G are provided in selector 352, and selector 353 is provided.
, And signals of D and I are input to the selector 354. In the above-described example, the selectors 351 to 354 select the signals A, B, C, and D. Then, signals A,
B, C, and D are added by an adder 355, then a 1/11 division is performed by a divider 356, and a process of adding the signal to E ′ by an adder / subtractor 359 is performed. Latched and output.

In the example shown in FIG. 11, it is sufficient to use the preceding data in all cases because the difference is every two chips, but if this is an odd chip, the succeeding data is used. For example, if the standard is F, B and F
Are 9 chips apart, so that the subsequent data, that is, G, can be used. As described above, whether to use the previous data or the subsequent data depends on the delay relationship between the data to be obtained and the four data before and after the data to be obtained.

FIG. 12 is a flowchart showing how to determine the timing for switching the selector shown in FIG.
In FIG. 12, first, N = 0 is initialized, and then N + 1
Then, in the correlation spike position determination processing, it is determined whether the chip is an odd chip or an even chip, and in the next processing, it is determined whether to use the previous data or the subsequent data. Decide what to do. After this is determined for all selectors, an operation of switching the selector is performed.

This timing is performed by an external timing generator by delay multiplexing of the current received data. The delay multiplexing information is determined depending on whether the number of multiplexes and the amount of delay are uniquely determined by the system or whether the delay is an even chip or an odd chip. Although the determination of the selectors 351 to 354 is shown in the flowchart of FIG. 12, it is not actually necessary to perform the discriminating process every time. For example, since the delay amount at the time of 5-multiplexing is a fixed value determined by the system, A circuit that switches the selectors 351 to 354 in an order determined by repetition is practical.

FIG. 13 is a view showing the absolute value of the correlation output of the correlation spike when the processing is performed as described above. This figure 1
As is clear from FIG. 3, the value approaches 11 as compared with the example shown in FIG. Here, the reason why it is not completely 11 is that the data used for adding the preceding and following data is
Actually, not A, B, C, D, but A ', B', C ',
D ', which also includes an error due to the fact that it has changed from 11. Next, the present invention is called D
A case where the present invention is applied to QPSK will be described.

FIG. 14 is a diagram showing vectors on a phase plane used for describing the present invention, and FIG. 15 is a diagram showing a vector change at the time of multiplexing.

Assuming that there is no influence of the side lobe of the autocorrelation, the data is represented by respective vectors A, B, C, D and E as shown in FIG. Actually, as shown in FIG. 15, since there is an autocorrelation side lobe. Here, as a precondition, a case is considered in which delay is caused to an even chip affected by only previous data as in the above-described example. For simplicity, vectors A, B, C, and D have 11 outputs, and it is ignored that each of them has changed. With each of the vectors A, B, C, and D, the vector generated by the side lobe of the autocorrelation is EA, EB, EC,
ED, which is a composite vector of E and E, and appears as a vector of E '.

Here, A is (1) with respect to the I and Q axes, respectively.
Since (1,11), B, C, and D are (-11, 11) signals, the vectors generated by their side lobes are also (-1, -1), (1, -1). From this,
As shown in FIG. 8, it can be seen that I and Q can be operated independently, and the vector E 'can be returned to the vector E by the independent operation of the resulting I and Q components.

FIG. 16 is a diagram showing vectors on the phase plane in one embodiment of the present invention, and is a vector diagram when the phase plane is inclined. FIG. 17 is a multiplex diagram when the phase plane is inclined. 16 is a diagram showing a vector change at the time, FIG. 16 corresponds to FIG. 14, and FIG. 17 corresponds to FIG. This is when the phase plane is tilted when receiving in an asynchronous system.

In the example shown in FIGS. 16 and 17,
Although the axis of the signal is inclined, it can be seen that it is also on the same axis as A, B, C, and D when considered as a vector. In the vector calculation, if on the same axis, I,
If Q is separately encoded to be 1/11, then each vector E
A, EB, EC, and ED indicate that the influence of the side lobe of the autocorrelation can be canceled in the circuits of FIGS.

FIG. 18 is a diagram showing the improvement of the error rate according to one embodiment of the present invention. FIG. 18 shows the difference between the proposed example and the present invention in a DQPSK system obtained by simulation. The horizontal axis indicates C / N, and the vertical axis indicates the error rate. By using one embodiment of the present invention, BER = 1.0E-0
It can be seen that there is a 6 dB improvement near 4. In the present embodiment, the processing is performed by utilizing that the side lobe of the autocorrelation is determined by only one of the preceding data and the following data. However, a case is considered in which the determination is not made by only one side lobe of the autocorrelation, that is, a case in which the determination is made by a combination of the preceding data and the following data. If the vector of F is post-data with respect to the vector of A, the vector to be canceled will be different for each of the four possible cases of the vector of F. Therefore, it is necessary to find the vector. In this case, since the vector may not be on the axis of the A vector, it is necessary to calculate from the A and F vectors to find a vector to be canceled. This is even more difficult when the axis of the signal is unknown, such as in asynchronous systems, because the axis must be further estimated and determined.

Also, when the axis has dispersion due to noise,
Similarly, the calculation becomes difficult. However, in one embodiment of the present invention, only the calculation on the axis of A is required, and FIGS.
0, and the circuit becomes simple.

As described above, by using the embodiment of the present invention, the change in signal amplitude due to the influence of the side lobe of the autocorrelation can be reduced, and the error rate can be drastically improved. Further, in one embodiment of the present invention, attention is paid to the characteristics of the spreading code, and either the preceding data or the succeeding data is determined, and the front and back of four overlapping data for processing from the vector component are added. It is not necessary to perform vector processing and only the signal of the received signal needs to be added, so that the circuit can be simplified.
Further, even in an asynchronous system, since it is only necessary to process the received signals before and after, it is possible to apply the signal as it is without estimating the axis.

In the above description, only the five multiplexing of the Barker code as the spreading code has been described. However, this can be performed using other codes or other multiplexing numbers. The code condition at this time may be such that the side lobe of the autocorrelation is uniquely determined by the preceding data or the following data. This need not be determined by the preceding and succeeding data over the delay amount of all the side lobes as in the Barker code, and the correlation spike of the delayed signal may be at a position determined only by the preceding and succeeding data. This example will be described using an m-sequence composed of 15 chips.

FIG . 19 shows the autocorrelation of an m-sequence of 15 chips.
The state of is shown. These autocorrelation value data are also
Looking at it in the same way as the
Of the spike timing data, 5 channels from the previous data
The absolute value of the correlation value of the
The correlation value becomes 1 and the correlation value of the 10th chip becomes the previous data.
On the other hand, the absolute value is a correlation value of 1 with a different sign.

FIG. 20 shows numerical values of the correlation values when the m-sequence signals are delayed and multiplexed, and shows the correlation outputs as shown in FIG. 5 as numerical values. When the 15 or -15 is the correlation value of the correlation spike Each of signals are multiplexed is 1 or -1, before data, uniquely since the code is determined from the rear data, the invention uses <br /> Thus, in the present invention,
It can be seen that a code satisfying the above conditions can be widely applied. Also, if in the example above, but had become different signs, using Barker codes 1 3 chip shown in following
It has the same sign.

[0051] FIG. 21 13 show how the autocorrelation of the chip Barker code. Timing of two correlated spikes
Of over data, the correlation value after the even chips before data back data 11, regardless of -11 before correlation value next to the absolute value with the same reference numerals with respect to data value 1, whereas the odd correlation value after the chip is pre-data 11, regardless of -11 absolute value with the same sign for the value of the rear data that has become a first correlation value <br/>. From this, it is understood that the subtraction may be performed in a manner opposite to the above-described example. At this time, the adder / subtractor shown in FIG. 10 performs a subtraction operation.

Next, a second embodiment of the present invention will be described. As an example, let us consider a case where five signals are multiplexed using an 11-chip Barker code as in the above case.

FIG. 22 is a block diagram of a correlation processing unit according to the second embodiment of the present invention. First, before describing the circuit configuration, data used for canceling autocorrelation when the delay amount is 2, 2, 2, 2, and 3 will be examined.
In FIG. 11, G ′, G ′ are used for canceling F ′.
The data of H ', I', J 'are used for canceling G', the data of F ', H', I ', J' are used.
F ', G', I ', J' are used for canceling H '.
Data.

On the other hand, the data of F ', G', H 'and J' are used for canceling I ', and the data of F', G ', H' and I 'are used for canceling J'. It is. That is, for a signal that requires each of the five data F ', G', H ', I', and J 'as one block,
It is sufficient to use the other four. Therefore, since the circuit configuration requires only 5 data, FIG.
As shown in FIG. 22, a circuit configuration for shifting data for 5 data is better than that of FIG. That is, as shown in FIG. 22, the correlation processing unit 43 includes a 5-bit shift register 4.
4, a latch circuit 45, a switch 45, an arithmetic unit 46 with a select function, and a timing generator 47.

FIG. 23 is a diagram showing the internal configuration of the arithmetic unit with select function 46 shown in FIG. In FIG.
The input five signals are switches 451 to 4 respectively.
55 is input. Here, 4 other than the required signal
One signal is turned on and the required signal is not passed off as off. As a result, the adder 461 adds four signals other than the required signal. Also, the selector 466
, Five signals are input, and only necessary signals are selected. Other operations are the same as those in FIG. Thus, in this embodiment, it is not necessary to select all of the preceding and following data by the selector 466, and the data can be processed as one block. To do so, it is necessary to select the code to be used and the amount of delay.

As described in the first embodiment, the present invention is characterized in that only one of the preceding data and the succeeding data is used. Therefore, for the first cancellation of the block data, the data after the remaining multiplex wave is used,
To cancel the second data, the first preceding data and the remaining subsequent data may be used. in this way,
It is sufficient that the third, fourth,... Sequentially satisfy the condition. For example, in the case of a Barker code of 11 chips, it may be composed of a delay amount of a continuous even chip and a delay amount of one odd chip. By satisfying such a condition, it becomes possible to cancel the side lobe of the autocorrelation with the circuit configurations shown in FIGS.

Next, a third embodiment of the present invention will be described. When using the 11-chip Barker code, the first
In the second embodiment, addition is performed after dividing by the code length 11 used for spreading. This is, as shown in FIG.
This is because E1 was 1/11 of A. But,
In an actual circuit, as shown in FIG. 10, since the calculation is performed from the correlation output A ', the value does not completely return to 11 as shown in FIG. 13, but a residual component exists.

Therefore, attention is paid to the added value of the changed correlation value and other data. As described above, considering the case of five multiplexes, the correlation values are 7, 9, 11, 13, 15
And -7, -9, -11, -13, and -15. The other four addition values at this time are 28, 12, and −
4, -20, -36 and -28, -12, 4, 20, 3
It becomes 6. From this, each is equal and the difference is 1
It turns out that it is 6. Therefore, since the change of data 2 is a change of -16, it can be understood that dividing by 8 converges to the same value. Therefore, in the above-described example of five multiplexing, instead of dividing by 11, dividing by 8 all results in a value of 10.5.

[0059] When this mathematically described, side lobes of the autocorrelation of the Barker code, Cor _k the correlation values of successive correlation spike, when Cor k + _1, is expressed by the following equation.

[0060]

(Equation 1)

Therefore, since the delay time difference between the five multiplexed signals is 2, 2, 2, 2, and 3, the correlation spike values of the interfering 1 to 5 sequences are as follows.

[0062]

(Equation 2)

Here, Cor _l , _k is the original correlation spike value. Therefore, when divided by 8 as in this embodiment, the following equation is obtained.

[0064]

(Equation 3)

As a result, it is found that the value is 10.5.
This can be applied to the forward synchronization method. In the forward synchronization method, the frequency between transmission and reception substantially coincides but does not coincide with the phase, so that the phase of the vector of the received signal with respect to the I and Q axes in the receiver is undefined.
Consider a case where white noise is superimposed on the received signal at this time.

FIG. 24 is a diagram showing a vector when white noise is superimposed on a received signal. When the received signal at this time is represented by a vector, the phase is θ with respect to the transmission phase.
The difference is the vector sum of the original autocorrelation signal vector, the white noise vector, and the side lobe vectors of the remaining four multiplexed waves. The side lobe vector is 1/11 of the autocorrelation component of the four blocked waves.

At this time, what is added is the four received signal vectors before and after, and each of them contains noise. Assuming that the phase shift is θ, the signals output to the I-phase and Q-phase correlators by the forward synchronization method are as follows.

[0068]

(Equation 4)

Using this, the correlation improvement method of the present invention is processed independently for each of the I phase and the Q phase.

[0070]

(Equation 5)

Similarly,

[0072]

(Equation 6)

As a result, the value is 10.5, and it is understood that the side lobe of the autocorrelation of the multiplex wave that causes interference can be completely removed even in the forward synchronous demodulation method.

FIG. 25 shows the absolute value of the correlation value of the correlation spike in the third embodiment.
It can be seen that all values are 10.5. As a result, the phase after the cancellation has one point in which the variation still remains in the first embodiment, and the phase error can be reduced. The result is shown in FIG. FIG. 26 corresponds to FIG. 18, and it can be seen that the error rate is further improved by using the third embodiment. Thus, the third
The embodiment has a feature that the error rate can be further improved by dividing by 8 instead of 11.

In the above description, only five multiplexes have been described. However, when this is four multiplexes, an equal difference of -18 is obtained.
You just have to divide by 9. Similarly, in the case of triple multiplexing and double multiplexing, if they are divided by 10, 11 respectively, they are all integrated into one phase.

Next, consider the case where a 13-chip Barker code is used. Here, considering the case of six multiplexing, the correlation values are 8, 10, 12, 14, 16, 18 and-
8, -10, -12, -14, -16 and -18 are taken.
The other five addition values at this time are -80 and -4, respectively.
6, -12, 22, 56, 90 and 80, 46, 12,
−22, −56, and −90. From this, it can be seen that each is an equal difference value and the difference is 34. Therefore, since the change of data 2 is 34 changes, the change of 1
It can be seen that dividing by 7 all converge to the same value. Thus, in the 6 multiplex example, instead of dividing by 13, 17
Divides everything into a value of 12.7. Also, when this is 5 multiplexing, it suffices to divide by 16, and similarly, when it is 4 multiplexing, 3 multiplexing, and 2 multiplexing, dividing by 15, 14, and 13 respectively brings them all into one phase. As described above, when side lobes of autocorrelation occur in different codes such as an 11-valued Barker code, the number of divisions is reduced by 1 as the number of multiplexes increases, while the same as in a 13-chip Barker code. If side lobes of autocorrelation occur in the code, the number of divisions may be increased by one each time the number of multiplexes increases.

Next, as a fourth embodiment of the present invention,
An example of application to a case where a technique called PDI is used in spread spectrum will be described.

FIG. 27 is a diagram showing the configuration of a PDI circuit, which is described in "Spread Spectrum Communication System" written by Mitsuo Yokoyama and published by Science and Technology Publishing Company. In FIG.
The received signal is input to the matched filter 51, and a pulse train having a plurality of peaks appears at the output according to the arrival time and the signal strength. These pulse trains are input to the transversal filter 52. The time length of the delay line of the transversal filter 52 is adjusted to the maximum delay spread. The output of the transversal filter 52 and the output of the control filter 51 are provided to a multiplier 53 where they are multiplied, and synchronous detection is performed. By this multiplication, the peak value of the pulse train is emphasized, and the low-level noise component is weakened. The output of the multiplier 53 is supplied to an integrator 54, where the signal is time-integrated by T _M , and signals that are temporally dispersed by delay spread are collected. This operation achieves diversity. Then, the signal component is determined by the determination circuit 55. Thus, in PDI technology,
This technique uses all of the signals spread by the delayed waves for demodulation, and has the characteristic that the error rate can be improved under fading conditions.

FIG. 28 is a diagram showing correlation outputs for explaining PDI. FIG. 29 shows a correlation waveform in a state where there is no delay wave in an ideal state shown by a dotted line in FIG. 28, and FIG. 29 shows a state where one delay wave is added thereto. FIG.
In FIG. 9, a is the original correlation waveform, b is the delayed wave, and c is the synthesized waveform. In the original PDI, FIG.
As shown in FIG. 7, integration is performed by the integrator 54. Here, if signals at two sample points are used for demodulation for simplicity, the original signal is demodulated in FIG. )
It can be seen that the delay wave component can be demodulated by using the PDI, and the performance is improved by using the PDI as compared with a single case without using the PDI.

Therefore, consider the case of multiplexing this time. In this case, the actual correlation output is as shown by the solid line in FIG. 27, which is different from the ideal state indicated by the dotted line. This is because the side lobe of the autocorrelation influences other than the correlation spike.

FIG. 30 shows the correlation output for explaining the PDI, and shows how it changes under the influence of the side lobe of the autocorrelation. B shown in FIG.
Since the correlation of a is not 0 at the time of the correlation of 1, the synthesized output c drops from the original output at both the sample point (1) and the sample point (2). Since the error rate is determined by the ratio (C / N ratio) between the output signal component and the noise component, there has been a problem that the error rate deteriorates.

In order to solve this problem, the side lobe of the autocorrelation in the present invention may be used not only for the original correlation spike but also for the correlation spike of the delay wave used in PDI.

FIG. 31 is a block diagram showing an example in which the present invention is applied to a PDI receiver. FIG. 31 is the same as FIG. 8 except for the following points. That is, when the output of the correlator 25 is delayed by the timing of the delayed wave, the PDI latch 6
1 and the output of the PDI latch 61 is P
The correlation output signal of the correlation spike input to the DI correlation processing unit 62 and latched by the latch unit 32 is also provided to the PDI correlation processing unit 62, and the timing signal from the correlation processing unit 33 is sent to the PDI correlation processing unit 62. Given. PDI
The output of the use correlation processing unit 62 and the output of the correlation processing unit 33 are combined by the combining unit 63 and provided to the distributor 20. Further, a PDI unit 64 is provided at the output of the differential unit 30,
After the PDI processing, it is provided to the determination unit 31.

FIG. 32 is a block diagram showing the structure of the PDI correlation processing section shown in FIG. The PDI correlation processing unit 62 includes shift registers 621 and 622 and an arithmetic unit 623 with a select function.

In the PDI, it is necessary to cancel at the timing of the correlation spike of the delay wave, but the effect is that the side lobe of the signal at the original correlation spike timing and the autocorrelation of the delay wave itself are affected. It is a side robe. Therefore, the criteria for cancellation are:
There are two timings: the correlation signal at the original correlation spike timing and the correlation spike timing of the PDI delayed wave. In this case, the side lobe received from the correlation spike of the delay wave for PDI is exactly the same as the way of canceling the side lobe of the correlation spike. That is, before and after the signal latched at the timing of the delayed wave centered on the delayed wave (k
-1) Cancel from correlated spikes of delayed waves.

On the other hand, in the case of canceling the original correlation spike, taking the aforementioned FIG. 4 as an example, it depends on the subsequent data after the odd delay, and depends on the previous data after the even delay. From this, it is possible to cancel at the PDI timing by determining whether the original signal affecting the PDI timing is after the odd delay or even delay, and switching the selector.

In this case, the difference from the first embodiment is that the addition is not performed only for the multiplexed signals before and after the correlation spike to be canceled, but for all the multiplexed signals (5 for 5 multiplexing). It is to be. This is because the PDI is affected by the side lobe of the signal itself of the correlation spike from which it is based.

As a result, the PDI correlation processing unit 62
As shown in FIG. 32, the correlation spike of the delayed wave signal is 2
(K−1) +1 is held in the shift register 621, and the original correlation spike 2 k is held in the shift register 622.

FIG. 33 is a block diagram showing the structure of the arithmetic unit with a select function shown in FIG. As shown in FIG. 33, the arithmetic unit with select function 623 is the arithmetic unit 62 for the delayed wave.
4 and a side lobe calculator 625 for the side lobe of the original correlation spike. The arithmetic unit 624 for a delayed wave is the same as that in FIG. With this configuration, after canceling the autocorrelation of the delayed wave,
The side lobe of the original signal can be canceled.

Although the two arithmetic units 624 and 625 are shown separately in FIG. 33, the timing control, the adder / subtractor, and the latch circuit can be shared by integrating them into the actual processing. . Further, the performance can be improved by using only one of the arithmetic units in order to reduce the size of the circuit.
Further, when two or more delayed waves are used for the PDI, it is only necessary to prepare circuits for the delayed waves by that number.

FIG. 35 is a block diagram showing another example of the correlation processing unit. In FIG. 35, D-type flip-flops 100 to 108 are serially connected to form a shift register, and the output of the D-type flip-flop 100 and the output of the D-type flip-flop 105 are the selector 1
12, the output of the D-type flip-flop 101 and the output of the D-type flip-flop 106 are input to the data selector 111, and the output of the D-type flip-flop 102 and the output of the D-type flip-flop 107 are input to the data selector 112. Input, the output of the D-type flip-flop 103 and the D-type flip-flop 1
08 is input to the data selector 113. The selected output of the data selectors 110 and 111 is
, And the selected output of the data selector 112 and the selected output of the data selector 113 are added by the adder 122.
The addition output of the adder 120 and the addition output of the adder 122 are added by the adder 121. The addition output of the adder 121 is divided by the divider 130, and the result of the division and the output of the D-type flip-flop 104 are added by the adder 123.

FIG. 36 is a time chart showing a clock signal for operating the correlation processing section of FIG. In particular, FIG. 36A shows the system clock signal CLK, and FIG. 36B shows B which indicates the head of the data constituting the block.
This is the SCLK signal, and (c) is the SCLK signal indicating the correlation.

Next, the operation shown in FIG. 35 will be described with an example of five multiplexes. In the case of five multiplexing, as shown in FIG. 36C, five pulses are output while the BSCLK signal is output as the SCLK signal indicating the correlation. The combination of the number of delay chips is (2, 2, 2, 2, 3). First two
The data of the chip delay is a D-type flip-flop 104
, That is, the BS that indicates the beginning of the block
When the CLK signal becomes active, the control signals of the data selectors 110 to 113 are reset, and the input sides of the selectors 110 to 113 are all connected. The selector 110 is switched to the side by the next correlation signal, and the selector 111 is switched to the next correlation signal. Further, the selector 112 is switched to the side by the next correlation signal, and the selector 113 is switched to the side by the next correlation signal. At this point, one block ends, and the selectors 110 to 11
3 is connected to the side. When the next correlation signal is input, the BSCLK signal is also activated, so that all the selectors 110 to 113 are connected. This repetition cancels the autocorrelation unit.

However, in the example shown in FIG.
A bit data selector is required, and the selectors 110 to 113 must be switched according to a correlation pattern.

Next, an embodiment in which the configuration of FIG. 35 is simplified will be described. FIG. 37 is a block diagram showing still another example of the correlation processing unit. The embodiment shown in FIG.
All the data constituting the block are added, and during the same block, the data to be demodulated is added with the sum of the data divided by an appropriate value.

As shown in FIG. 37, D-type flip-flops 200 to 205 are serially connected, and the output of D-type flip-flop 203 and the output of D-type flip-flop 204 are added by adder 223. And D-type flip-flop 202
Are added by the adder 222. The addition output of the adder 222 and the output of the D-type flip-flop 201 are added by the adder 221, and the addition output of the adder 221 and the output of the D-type flip-flop 200 are added by the adder 220. The addition output of the adder 220 is divided by the divider 230, and the division calculation power is latched by the D-type flip-flop 206. The latch output and the output of the D-type flip-flop 205 are added by the adder 224 and output. .

When the BSCLK signal becomes active, the adders 220 to 223 add the addition values latched in the D-type flip-flops 200 to 204, respectively. Then, the addition result of the adder 220 is divided by, for example, the spread code length by the divider 230 and latched by the D-type flip-flop 206. And D
The value latched by the type flip-flop 206 is added by the adder 224 to the correlation value of the D-type flip-flop 205 to be demodulated. As a result, the side lobe of the autocorrelation is canceled. And again BSCL
When the K signal becomes active, the data latched in the D-type flip-flops 200 to 204 is the data of the next block, so that the addition result is updated again and the same operation is repeated. This allows
There is no need to select data to be used for cancellation, and data selectors 110 to 11 as shown in FIG.
3 can be made unnecessary, and its control becomes unnecessary.

Here, the actual demodulated data will be verified for easy understanding. For example, the correlator output when there is no influence other than the side lobe of the autocorrelation such as noise at all is as follows. ..., -11, 13, -11, -11, 13, -9,-
9, 15, -9, -9, 7, 7, 7, 7, 7, ... In order to make these easy to understand, ()
…,) (-11, 13, -11, -11, 13,) (-
9, -9, 15, -9, -9,) (7, 7, 7, 7,
7,) (...) The elements for canceling the side lobes of the data described here are as follows.

The first block, that is, (−11, 1
3, -11, -11, 13,) are demodulated as shown in FIG.
The sum of the seven adders 220 is -7. For example, when divided by the spread code length of 11, the value is about -0.64. When this is added to each of the correlation values, the first block becomes (-11.64, 12.36, -11.64, -1
1.64, 12.36). Similarly, in the next block, the addition sum is -21 and the division result is about -1.91.
When this is added to the correlation value, (-10.91, -10.9
1, 13.09, -10.91, -10.91). Further, in the next block, after the operation (10.18, 1
0.18, 10.18, 10.18, 10.18), which clearly shows that an improvement effect can be obtained.

In the above description, the divider 230 divides the sum by the spread code length of 11. However, if the sum is divided by 7, the absolute value of all the canceled data becomes 12, and the autocorrelation value becomes 12. The effects of side lobes can be completely eliminated.

In the above-described embodiment, when the 11-chip Barker code is used according to the multiplexing number, the correlation value is divided by 11−k + 2, and when the 13-chip Barker code is used, the correlation value is 13 + k−. Divided by two. But,
There is a risk that the scale of the circuit for this division will increase.
For example, consider a case where when each data output from the correlator is represented by 8 bits, three data are added in order to cancel the interference of the multiplexed other station. At this time,
The number of bits after the addition is increased by 2 bits to 10 bits.
When dividing the data of this bit number, 11-3 + 2 =
This would result in division by 10, but this division has two problems.

One of the reasons is that dividing data by 10 requires a considerable amount of computation, for example, processing by a CPU, which increases the circuit scale and makes high-speed computation difficult.
Another reason is that an accurate calculation requires a calculation result of 10 bits or more. After that, the data demodulation operation is performed. If the operation uses the value of the division result as it is, a data demodulation unit that operates with multiple bits (10 bits or more) is required. As a result, the circuit scale of the demodulation unit becomes large. In order to prevent this, it is necessary to perform processing such as changing the number of digits to 8 bits or the like by a method such as rounding off the result of dividing by 10 again.

FIG. 38 is a diagram showing an embodiment for solving the above-mentioned problem. In FIG. 38, the k-bit signal output from the adder 250 is input to the adder / subtractor 359 shown in FIG. 34 for k-3 bits excluding the lower 3 bits.
In a 2-bit operation, removing the lower 1 bit is 2
Equivalent to dividing by 2 is removing two bits is equivalent to dividing by 4. Therefore, removing 3 bits is equivalent to dividing by 8. As described above, the provision of the adder 250 makes it possible to significantly reduce the size of the circuit. Although the cancellation performance is slightly degraded by removing the lower three bits, there is an advantage that the division operation can be greatly simplified. When a 13-chip Barker code is used, 4 bits may be deleted and divided by 16.

FIG. 39 is a diagram showing still another embodiment of the present invention. In FIG. 39, a spread spectrum signal received by an antenna (not shown) is
And the amplitude level of the received signal is kept constant.
The output of the gain control amplifier 302 is output from the frequency converters 303 and 3.
04, and is frequency-converted by the local signal from the oscillator 24 to become I and Q baseband signals.
Further, digital data is provided to the correlators 308 and 309 by the A / D converters 306 and 307, and a correlation peak is detected. The outputs of the correlators 308 and 309 are calculated by the arithmetic unit 310 to obtain the square root of the sum of squares, and supplied to the correlation synchronization and latch unit 311. Correlator 308,
The output of 309 is also provided to the correlation synchronization and latch section 311.

The data processing section 317 performs the above-described data processing, and its output is supplied to a comparator 313 to be compared with an optimum value. The output of the comparator 313 is provided to the control circuit 316 via the filter 315, and the gain of the gain control amplifier 302 is controlled.

In this embodiment, the variation of the correlation output, which conventionally varies from 7 to 15, can be reduced, or all the correlation outputs can be integrated into the same value, and the residual fluctuation related to AGC can be reduced or eliminated.

[0107]

As described above, according to the present invention, the correlation of the data timing of (multiple wave-1) is held before and after the correlation to be canceled, and the preceding data and the following data are stored. By selecting and adding, and dividing by the spreading factor, the inverse vector direction of the side lobe component can be calculated. Therefore, the side lobe of the autocorrelation can be canceled by adding or subtracting the correlation value to be canceled. As a result, the spread of the correlation output in the phase plane is reduced, and as a result, the error rate can be improved in the case of phase modulation.

Further, there is an advantage that block processing can be performed by enabling all data processing to be performed with data outside the block.

Also, when using the Barker code, the value to be divided when using a code of 11 chips is divided by 11−k + 2 in the case of k multiplexing, and the value to be divided when using a code of 13 chips is k. In the case of multiplexing, by dividing by 13 + k-2, the correlation value after cancellation becomes a constant value, phase spread is eliminated, and characteristics can be further improved by concentrating on one point.

In the case of demodulation using PDI, the characteristics at the time of PDI can be improved by canceling the side lobe of the autocorrelation at a point other than the correlation spike.

Blocking is performed by the number of correlation length multiplexes, and by canceling one of the blocked data,
There is no need to select data to be used for cancellation, so that a data selector can be made unnecessary and its control can be made unnecessary.

Also, when using the Barker code, the value to be divided when using the 11-chip code is divided by 11−k + 1 when using k multiplexing, and the value to be divided when using the 13-chip code is k. In the case of multiplexing, by dividing by 13 + k-1, the correlation value after cancellation becomes a constant value, the phase spread is eliminated, and the characteristics can be further concentrated on one point.

Further, when the 11-chip Barker code is used, the value to be divided is set to 8, the lower 3 bits are removed, the data is shifted by 3 bits, and when the 13-chip Barker code is used, the division is performed. By shifting the data by 4 bits by removing the lower 4 bits by setting the value to 16, the dividing circuit can be greatly simplified, and the circuit size can be reduced and the power consumption can be reduced.

Further, by controlling the amplitude level of the received signal by varying the gain using the correlation output, the residual error can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a transmitter of a spread spectrum communication system proposed by the present inventors previously.

FIG. 2 is a block diagram of a receiver of a spread spectrum communication system previously proposed by the present inventors.

FIG. 3 is a diagram showing an absolute value of a correlation output of a correlation spike in the receiver shown in FIG. 2;

FIG. 4 is a diagram illustrating values of autocorrelation of Barker codes.

FIG. 5 is a diagram showing a state of a correlation output at the time of multiplexing.

FIG. 6 is a diagram showing a phase plane at the time of multiplexing.

FIG. 7 is a diagram illustrating characteristics of an error rate.

FIG. 8 is a block diagram of a receiver showing the first embodiment of the present invention.

FIG. 9 is a block diagram of a correlation processing unit shown in FIG. 8;

FIG. 10 is a specific block diagram of a computing unit with a select function shown in FIG. 9;

FIG. 11 is a diagram showing a state of a correlation output at the time of multiplexing according to an embodiment of the present invention.

FIG. 12 is a flowchart for explaining the timing of switching the selector shown in FIG. 10;

FIG. 13 is a diagram showing an absolute value of a correlation output of a correlation spike according to the first embodiment of the present invention.

FIG. 14 is a diagram showing vectors on a phase plane according to the first embodiment of the present invention.

FIG. 15 is a diagram showing a vector on a phase plane according to the first embodiment of the present invention, showing a vector change at the time of multiplexing.

FIG. 16 is a diagram showing vectors on a phase plane used in the first embodiment of the present invention, and is a vector diagram when the phase plane is inclined.

FIG. 17 is a diagram showing a vector change at the time of multiplexing when a phase plane used in one embodiment of the present invention is inclined.

FIG. 18 is a diagram showing an improvement in an error rate when one embodiment of the present invention is used.

FIG. 19 is a diagram illustrating autocorrelation of an m-sequence of 15 chips.

FIG. 20 is a diagram showing numerical values of correlation values when m-sequence signals are multiplexed with delay.

FIG. 21 is a diagram illustrating the autocorrelation of a 13-chip Barker code.

FIG. 22 is a block diagram of a correlation processing unit according to the second embodiment of this invention.

FIG. 23 is a block diagram of an arithmetic unit with a select function shown in FIG. 22;

FIG. 24 is a vector diagram when white noise is superimposed on a received signal.

FIG. 25 is a diagram illustrating an absolute value of a correlation value of a correlation spike according to the second embodiment of the present invention.

FIG. 26 is a diagram showing an improvement in an error rate according to the second embodiment of the present invention.

FIG. 27 is a block diagram showing a configuration of a PDI.

FIG. 28 is a diagram showing a correlation output used for explaining PDI.

FIG. 29 is a diagram showing the correlation output used for explaining the PDI, and showing the correlation output when combined.

FIG. 30 is a diagram showing a correlation output used for explaining PDI and showing a state where the correlation output changes due to the influence of a side lobe of autocorrelation.

FIG. 31 is a block diagram showing an embodiment of the present invention in the case of PDI.

32 is a block diagram illustrating a configuration of a PDI correlation processing unit in FIG. 31.

FIG. 33 is a block diagram showing a configuration of an arithmetic unit with a select function in FIG. 32;

34 is a block diagram showing a configuration of an arithmetic unit with a select function in FIG. 33.

FIG. 35 is a block diagram showing another embodiment of the PDI correlation processing unit of the present invention.

36 is a timing chart of a clock signal required for demodulation of the correlation processing unit for PDI in FIG. 35.

FIG. 37 shows a PD according to still another embodiment of the present invention.
It is a block diagram which shows the structure of the correlation processing part for I.

FIG. 38 is a diagram showing a division circuit according to another embodiment of the present invention.

FIG. 39 is a block diagram showing an AGC configuration of a receiver according to still another embodiment of the present invention.

[Explanation of symbols]

Reference Signs List 21 receiving antenna 23, 303, 304 frequency conversion unit 24 local oscillator 25 correlator 26 distributor 27, 28, 32 latch unit 29 latch controller 30 differential unit 31 discriminating unit 33 correlation processing unit 34 shift register 35 arithmetic unit with select function 36 Timing generator 51 Matching filter 52 Transversal filter 53 Multiplier 54 Integrator 55 Judgment circuit 61 PDI latch unit 62 PDI correlation processing unit 63 Synthesizing unit 64 PDI unit 100-108, 200-206 D-type flip-flop 110-113 Data selector 120 to 123, 220 to 223, 250 Adder 130, 230 Divider 302 Gain control amplifier 306, 307 A / D converter 308, 309 Correlator 310 Operation unit 311 Correlation synchronization and latch unit 313 Ratio Vessel 315 filter 316 control circuit 317 the data processing unit

Continuation of the front page (56) References JP-A-6-276173 (JP, A) JP-A-9-153843 (JP, A) JP-A-4-328921 (JP, A) JP-A-9-247123 (JP) , A) IEICE Technical Report RCS 96-66 (1996-8-9), Naoki Okamoto et al., Study on 10Mbps transmission method in ISM band using DS-delay multiplexing, p. 1-8 (58) Field surveyed (Int. Cl. ⁷ , DB name) H04J 13/00-13/06 H04B 1/69-1/713

Claims (8)

(57) [Claims] 1. A spread-spectrum communication system for directly spreading a signal, wherein a signal spread with the same spreading code is delayed and multiplexed by an arbitrary number of chips and transmitted, wherein the spreading code has an autocorrelation side lobe. Irrespective of odd correlation or even correlation, in a system using a code that is uniquely determined by one of the preceding data and the following data, the (multiplexing number −1) Maintains data timing correlation,
The auto-correlation is performed by selecting the preceding data and the following data, selecting and uniquely determining the correlation values of the preceding data and the following data, dividing by the spreading factor, and adding / subtracting the correlation value to be canceled. A method for receiving spread spectrum communication, characterized by canceling side lobes. 2. In a case where the correlation value is blocked by the number of multiplexes and one of the blocked data is canceled, at the receiver side, other combinations of the preceding data and the succeeding data are included in all the blocks. It can be processed by the data in the
The amount of delay of delay multiplexing at the transmitter side is controlled so that the delay is equal to that of the transmitter.The correlation is held for the number of blocks, and a correlation value other than the signal to be canceled is selected from the correlation values. 2. The spread spectrum communication receiving method according to claim 1, wherein side lobes of the autocorrelation are canceled by adding, subtracting, and adding / subtracting the correlation value to be canceled. Wherein the division, when the autocorrelation sidelobe used Barker code of 11 chips to be different signs to serving as a reference signal used for canceling the correlation value by dividing the case of k multiple Is divided by 11−k + 2. When a 13-chip Barker code in which the side lobe of the autocorrelation has the same code as the reference signal used for canceling the correlation is used, the value to be divided is k. 1 for multiple
3. The spread spectrum communication receiving method according to claim 1, wherein division is performed by 3 + k-2. 4. In the case of demodulation using PDI, even at the demodulation timing used for PDI, in order to cancel the side lobe of the autocorrelation, the demodulation is performed centering on the correlation corresponding to the PDI to be canceled. The correlation of the data timing for each (multiplex number) is retained before and after, and the correlation values of the unambiguously determined preceding and following data are selected, added, divided by the spreading factor, and the timing of the PDI to be canceled is determined. The spread spectrum communication receiving method according to claim 1, wherein side lobes of autocorrelation are canceled by adding or subtracting the correlation value. 5. When the correlation value is blocked by the number of multiplexes and one of the blocked data is canceled, all the data of the block to which the data to be canceled belongs are added and divided by the spreading factor. 2. The spread spectrum communication receiving method according to claim 1, wherein the side lobe of the autocorrelation is canceled by adding to a correlation value to be canceled. Wherein said division, when the autocorrelation sidelobe used Barker code of 11 chips to be different signs to serving as a reference signal used for canceling the correlation value by dividing the case of k multiple Is divided by 11−k + 1. When a 13-chip Barker code, in which the side lobe of the autocorrelation has the same code as the reference signal used for canceling the correlation, is used, the value to be divided is k. 1 for multiple
6. The spread spectrum communication receiving method according to claim 5, wherein division is performed by 3 + k-1. 7. When the 11-chip Barker code is used, the division value is set to 8, the lower 3 bits are removed, and data is shifted by 3 bits to perform a division operation. 4. The spread spectrum communication receiving method according to claim 3, wherein when a code is used, the division value is set to 16, the lower 4 bits are removed, and the data is shifted by 4 bits to perform the division operation. 8. A method for controlling the amplitude level of a received signal by varying the gain, digitizing and quantizing the output, performing gain control using the correlation output, and including the self- 2. The spread spectrum communication receiving method according to claim 1, wherein a correlation output after canceling the correlation sidelobe is used.