JPH10294715A - Spread spectrum radio communication system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent mutual interferences among channels with a simple circuit structure by using a specific an approximate synchronous CDMA code as a spread spectrum processing code, when a sending part side performs spread spectrum processing and sends a result. SOLUTION: A sending part generates an approximate synchronous CDMA code, which consists of a real part and a virtual part and multiplies the real part by a signal. On the other hand, it generates an orthogonal element of a carrier signal and multiplies the virtual part of the approximate synchronous CDMA code by the orthogonal component of a carrier signal. It adds an output of a multiplier for the real part with an output of a multiplier for the virtual part and sends them. Here, when a cyclic matrix (represented by an expression I) of cycle series of a cycle N is A and B is a diagonal matrix as codes for spread spectrum processing, A=F<-1> BF on the condition that F takes a series where a self correlation function becomes zero for all the terms, excepts for a multiple of the cycle as an orthogonal series under the condition where a DFT matrix is established. An approximate synchronous CDMA code is used in an expression II under the presence of B, D.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a spread spectrum wireless communication system, and more particularly to a spread spectrum wireless communication system which adopts a new code as a spread spectrum system, has no mutual interference between channels, and has a simplified circuit configuration. Things.

[0002]

2. Description of the Related Art In recent years, as a new data communication system accompanying the progress of communication technology, a communication system using a spread spectrum system has been studied and put into practical use. In the communication system based on the spread spectrum method, a voice signal or the like is modulated by spread spectrum by a terminal such as a mobile phone on the transmitting side to be converted into data, and is transmitted as a radio signal by an antenna (antenna) while demodulated by a receiving terminal. This is to make a call or the like.

The information society in the near future is composed of information sources such as databases, users who use information, and communication lines for transmitting information. FIG. 28 shows a near-future network image expected as described above. The communication line is a core wired network with a large-capacity transmission line, and users can exchange information such as voice, image, data, etc. `` anytime, anywhere, with anyone ''
It is divided into a wireless network by a portable terminal device.

[0004] Currently, fiber optics are used in wired networks.
Coaxial cables and the like are used, and their development and spread are remarkable as computer networks represented by ISDN and the Internet. The feature of this wired network is that it is highly reliable and capable of large capacity transmission. .

On the other hand, regarding wireless networks, IE
In recent years, as represented by the EE802.11 Standardization Committee, wireless LAN (Local Area Network)
Interest in k) has increased, and its practical use has already begun. It is no exaggeration to say that the use of mobile communications has been increasing rapidly these days, and it is becoming a social infrastructure. Such wireless networking not only solves the problem of wiring installation space, but also enables free arrangement and movement of terminals. PHS (Personal)
A technique of connecting a portable telephone such as a handy-phone system and a portable terminal device such as an electronic organizer and applying it to data communication has also been put to practical use.

The following characteristics are required for a wireless transmission technology supporting a wireless network. (1) The reliability in the wireless section is on the same level as that of a wired section, or on the order of a wired section due to error correction. (2) Security of communication contents must be maintained. (3) It has reliability that does not depend on the use environment, especially, fading resistance in a room.

As a communication system satisfying the above conditions, a spread spectrum communication system has attracted attention. The spread spectrum communication method is a method of performing communication by spreading a spectrum to a frequency band far wider than a minimum band required for information transmission. Normal information modulation (primary modulation, FS
K, PSK, etc.) and then the PN code (Pseu
Frequency spreading (secondary modulation) is performed using a do noise code (pseudo noise code). By spreading the frequency band, it has capabilities such as interference resistance, signal concealment, and fading resistance. Also, since the spreading code has the capability as a communication station identification code, CDMA (Code
Random access by Division Multiple Access (code division multiple access) is possible.

In the United States (USA) in 1990,
In 1993, a frequency band for spread spectrum communication was approved in Japan and its use is encouraged. Table 1 shows standards [3] of spread spectrum communication bands in the United States and Japan.

[Table 1] As is clear from the above table, in Japan, the frequency band is the 2.4 GHz band of the ISM band and the bandwidth is 26M.
Hz.

In the spread spectrum communication, the same PN as the transmitting side is applied to the signal spread over a wide band in the receiving unit.
Despreading processing for restoring information by multiplying the codes is required. An element or device for performing the despreading process is called a correlator, and is generally classified into a digital type and an analog type.

[0010] SAW (Surface Acoustic)
c Wave: a surface acoustic wave device is a type of analog correlator, and performs wideband signal processing while including carriers. The SAW correlator and the SAW convolver are devices capable of performing despreading in the RF and IF bands, and are optimal as correlators for spread spectrum communication. The SAW convolver is a programmable correlator that operates on an arbitrary code by switching a reference signal, and is most suitable for CDMA using various spreading codes and a central control station of a network. The SAW correlator has a very simple structure and is suitable as a correlator for a portable wireless terminal that requires a low power consumption though it is a fixed code.

FIG. 29 is a conceptual diagram of a spread spectrum communication system. The spread spectrum communication system is a communication system in which a signal subjected to information modulation (primary modulation) restricted to a certain band is spread (secondary modulation) over a wide band using a spreading code, and information is transmitted. Compared with the normal narrowband communication system, the transmitting side needs a function of spreading the frequency using a constraint code completely unrelated to the information to be transmitted. Generally, a pseudo-noise code (PN code) is used as the high-speed spreading code. On the receiving side, the frequency band is despread by the operation exactly opposite to the process performed on the transmitting side, and after returning to a normal information modulated signal, information demodulation is performed.

In FIG. 29, the transmitting-side apparatus performs general modulation on transmission data to perform transmission modulation shown in FIG.
A carrier signal as shown in (a) is obtained and output in the form of a radio signal. This is the primary modulation of the signal. FIG. 29 (a)
Are composed of channel signals A, B and C corresponding to the respective frequencies f1, f2 and f3, and have a level higher than the level of the noise 30 associated with the circuit of the transmitting apparatus.

In the transmitting device, the channel signals A, B,
C is further subjected to spread modulation by the SAW convolver method to obtain a spread signal (SS signal) as shown in FIG. This is the secondary modulation. The secondary modulation signal 31 obtained by the spread modulation is wirelessly transmitted to the receiving device. The receiving side device receives the secondary modulation signal (spread modulation signal) 31, demodulates the received signal into an original data signal after despreading, and obtains a demodulated signal. FIG. 29 (c) shows a state where a demodulated signal is obtained for the channel signal A in the receiving device.

[0014] The secondary modulation signal 31 obtained by the spread modulation in the control device 6 which is a radio transmission terminal has a noise 3
It has the characteristic that it has a level lower than the level of 0, and since the signal level is lower than the noise level, it is transmitted over the communication line in the form of simultaneous progress with other general signals. (For example, a call signal) does not cause trouble such as crosstalk and generation of communication noise.

The spread-spectrum communication system has the following features due to the operation of spreading and de-spreading. (1) Interference or interference is less likely to be given or received. Has anti-interference ability. (2) Signal concealment ability increases. (3) The confidentiality is improved. A system suitable for secret story can be configured. (4) Fading, in particular, resistance to frequency selective fading is improved. (5) Code division multiple access (CDMA) is possible and elegant quality degradation (Graceful Degradation)
n). (6) Distance measurement and time synchronization are possible simultaneously with communication.

The anti-interference capability is a process gain (Processing Gain, G) which is a capability unique to spread spectrum communication obtained by spreading and despreading.
p). The process gain is equal to the spreading factor when ideal spreading and despreading are performed.
In the direct spreading method, the PN code length N is used. This process gain maximizes its capability with respect to interference waves having finite power.

Broadbanding by spreading not only achieves resistance to interference, but also reduces power density. Therefore, for those who do not have information (or knowledge) on the spreading code, the signal concealment ability that does not know the existence of the signal itself is realized (FIG. 29B).
, The secondary modulated signals of channels A, B, and C, that is, the spread signal levels are lower than the noise levels. In general, the longer the code length of a PN sequence used for spread spectrum, the more difficult it is to decode the code.

Further, widening the band is very effective against frequency selective fading which is a problem particularly in an indoor radio environment. Fading occurs because a radio wave arriving at the receiver is a composite wave arriving from a plurality of paths. This composite wave is composed of the same signal with different arrival times. Therefore, it is considered that fading of a frequency corresponding to the reciprocal of a certain arrival time difference occurs. The frequency at which the fading occurs rarely covers the entire band of the spread spectrum signal, and a certain portion of the signal can be received by frequency diversity. Also, since it has a time resolution that is the reciprocal of the spread frequency band, diversity can be realized by dividing the multiplexed wave that has arrived, matching the phases, and recombining them.

The spread code for spread spectrum communication not only widens the band but also has a role as a station identification code for many users.
A) can be realized. FIG. 30 shows a conceptual diagram of various multiple access. The conventional co-occurrence modulation, such as FM and AM, is shown in FIG.
FDMA (Frequency Division) for allocating channels by frequency division as shown in FIG.
Multiple Access). In addition, European DECT (Digital Euro)
pean Cordless Telecomuni
camps) and Japanese PHS (Personal)
Handy-phone System) and the like are shown in FIG.
As shown in (b), a time slot is assigned to a spreading code channel, and TDs that communicate using the entire band within that time are used.
MA (Time Division Multiple)
Access). On the other hand, CDMA
As shown in FIG. 30 (c), all users use the entire band and time at the same time, and perform channel division by a high-speed spreading code.

In FDMA and TDMA, since each user communicates in a completely separated channel, interference is not caused by other users in an ideal state, and communication can be performed with a specified communication quality. is there. However, CDMA in which all users share the same frequency and the same time gradually deteriorates in quality as the number of users increases (graceful communication quality deterioration: Graceful Degradat).
ion). Whereas FDMA and TDMA cannot allow more than a certain number of users, CDMA allows users as long as code synchronization can be set due to elegant quality degradation. CDMA is a key technology for next-generation wireless communication, and various standards have been proposed as shown in Table 2.

[Table 2]

Although various systems of spread spectrum communication have been proposed, typical systems will be described. (1) Direct Sequence, D
S) Method After narrowband information modulation, a wideband spread signal is obtained by switching the carrier phase by multiplying by a high-speed spreading code.
BPSK (Binary PhaseShift)
In the case where Keying (two-phase modulation) is used, it can be simply configured with only a carrier oscillator, a code generator, and a mixer. On the receiving side, the timing of the transmission spreading code is extracted by correlation detection, and the phase is returned based on the same spreading code as that on the transmission side to obtain a narrowband information modulated signal. (2) Frequency Hopping (Frequency Hop)
ping, FH) system By using a frequency synthesizer capable of high-speed switching, the frequency band of the RF is discontinuously switched based on a hopping sequence within a predetermined band, thereby widening the band. On the receiving side, the local oscillator is switched in accordance with the timing of the change of the transmission frequency to obtain a narrow-band modulated wave. The system configuration requires a sophisticated frequency synthesizer,
Since frequency synchronization capture is basically a sliding correlation, synchronization delay becomes a problem. (3) M-ary spread spectrum communication (M-ary / SS)
System This system is a multi-valued system of the direct spreading system, in which several code sequences are prepared in one station and one of them is transmitted according to several bits of information. When an orthogonal code is used, the error rate characteristic becomes equal to that of the M-ary orthogonal modulation method, and by making the multi-level number infinite, Sha
The Nnon limit can be achieved.

However, since the spreading sequence is changed for each transmission symbol, the correlation function at the connection point of the spreading sequence does not become 0, and a synchronization error is likely to occur. Also, if the transmission sequence is m
Since there are different types, m correlators are required for one synchronization point search. Therefore, there is a disadvantage that synchronization supplementation is difficult. Besides, time hopping method, chirp modulation method,
A multi-carrier system, a hybrid system of a spreading code system, and the like are being studied.

As described above, the code used for frequency spreading is optimally a completely random sequence such as Gaussian noise, but a digital code called a pseudo-noise code (PN code) is used in an actual system. It is generally used. The main properties of the PN code are listed below. (1) Balance Property The number of occurrences of “1” and the number of occurrences of “0” in one cycle of a sequence are different from each other by at most one. (2) Run Property (Run Property) In the “run of“ 1 ”” and “run of“ 0 ”” included in one cycle, if the length of each run is k,
A run of length k exists at a rate of 1/2 ^k . The sequence is a continuous number of “1” or “0”.
For example, in the case of 01110, the length of “a series of“ 1 ”” is 3 and the number is 1. (3) Correlation Property
ty) When the sequence is circulated and the comparison for each item is performed in all states, the number of terms that do not match the number of terms that do not match is at most 1
Only one is different. Here, "term" indicates a single element of a code.

In spread spectrum communication, an m-sequence (Mximum L) having a particularly small cross-correlation is used as a PN code.
length Sequence: the longest code sequence) or Gol
The d series is often used. FIG. 31A shows an m-sequence generation circuit. The m-sequence generation circuit can be easily formed by a plurality of stages of shift registers and an exclusive-OR (Exclusive-OR) logic circuit. A code sequence is obtained by giving an initial value to this circuit and cycling. Among them, the one having the longest cycle is the m-sequence. m series is n
In the case of a stage shift register, the code length is 2 ⁿ −1. The greatest feature of the m-sequence lies in its correlation. FIGS. 31B and 31C show a code length of 127 chips m
5 shows the auto-correlation and cross-correlation properties of a sequence. In these figures, the code is circulated one chip at a time, and the difference between the matching number of each term and the non-matching number is displayed as a horizontal axis shift amount. In the autocorrelation characteristic, a peak appears only when the phases of the codes completely match, and otherwise, the amplitude is 1 / (code length) of the peak. The steep autocorrelation peak can be used as a synchronization code. A pair having a small cross-correlation value among combinations of m sequences is called a preferred pair. Table 3 summarizes m sequences of various lengths.

[Table 3]

Since the number of m-sequences is limited,
There are limitations on channel assignment. The Gold series has the property to compensate for this defect. FIG. 32 shows a method of generating a Gold sequence. In FIG. 32, (a) is Gold
FIG. 3B is a diagram illustrating an example of a configuration of a sequence generator, wherein FIG.
FIG. 4 is a diagram illustrating the principle of sequence generation. As shown in FIG. 32 (a), the Gold sequence generator includes a first m-sequence generator 1, a second m-sequence generator 2, and a first and second m-sequence generator 1, And an adder 3 for adding the two outputs. First m-sequence generator 1 and second m-sequence generator
A clock signal is input to sequence generator 2, code 1 is output from first m-sequence generator 1, and code 2 is output from second m-sequence generator 2. The adder 3 outputs a Gold code (code 1 XOR code 2) by adding code 1 and code 2.

The Gold sequence is, as explained above, 2
It is generated from the logical sum of m sequences of one preferred pair. For a code length of 127, the phase relationship between the two codes is
Since 127 patterns can be obtained, 127 Gold sequences can be generated from one set of m sequences. The nature of the Gold series has been well studied mathematically, and it is known that only three specific values of the autocorrelation value can be taken.
Further, since the upper limit of the cross-correlation value is given, the number of simultaneous connections and the like can be estimated, which is convenient.

Here, a SAW device as a correlator for spread spectrum communication will be described. Despreading (correlation detection) is the biggest problem in receiving spread spectrum communication. Correlators that perform despreading are roughly classified into digital and analog types. FIG. 3 shows an example of a digital correlator.
3 shows a digital sliding correlator, and FIG. 34 shows the principle of a digital matched filter. The digital sliding correlator circulates the PN sequence earlier than the received signal, and performs synchronization with a supplementary system such as a DLL (Deley Lock Loop). Since the synchronization mechanism uses a loop, stable synchronization can be maintained. However, operation instability due to the balance of the correlator, a long synchronization supplementary time that requires a cycle of a maximum code of one cycle, and the like become problems.

The digital matched filter detects a PN timing in the form of a correlation peak by performing correlation integration with a received signal of a known spread code. The existence timing of the correlation peak may cause ambiguity, and the direct driving of the despreading PN code generator becomes unstable.

The SAW device is an analog correlator,
This device is capable of despreading with RF or IF carriers by a combination of electrodes and delay lines. FIG. 35 is a diagram showing the characteristics of the SAW device and the digital correlator and the detection and demodulation operation procedures in comparison to show the usefulness of the SAW device. The above-described digital correlator has an operating frequency in a baseband region (up to 100 M).
Hz), in order to perform despreading in the 2.4 GHz band spread spectrum communication, it is necessary to first recover the carrier of the received signal, and then perform detection and demodulation to a baseband signal. Thereafter, since despreading is performed, the demodulation method can be named “correlation after detection”. In this digital correlator, even if an attempt is made to perform demodulation after detection, carrier recovery cannot be performed, and demodulation is impossible.

On the other hand, the SAW device performs an operation as shown in the middle part of FIG. 35 as a demodulation procedure in a demodulation operation when receiving a signal. That is, first, a correlation operation is performed on the received signal 11 in a state where the signal contains an RF or IF carrier, and despreading is performed to obtain a correlation processing signal 12. In this correlation operation, noise and the like included in the received signal are suppressed, and an improvement in process gain including expansion of transmission characteristics is obtained, so that the correlation processing signal 12 becomes a signal that can be demodulated. Then, the correlation processing signal 12
, The correlation peak after despreading is detected and demodulated to obtain a detected signal, that is, a baseband signal. As described above, in the SAW device, the noise of the process gain is suppressed by the correlation operation, so that the received signal can be demodulated.
It can be named. The difference between the post-detection correlation and the post-correlation detection is that the location where the process gain (Gp) is obtained during demodulation is different.

In the spread spectrum, since the signal is spread over a wide band, the maximum signal power (carrier power) may be attenuated below the noise power at the time of reception. As described above, in an environment where the C / N (carrier power / noise power) ratio is very poor, the digital correlator that performs correlation after detection cannot perform carrier recovery at the time of synchronous detection, and converts the baseband signal into a baseband signal. I can't get it. Therefore, despreading becomes impossible. Carrier regeneration can be performed only when the carrier power exceeds the noise power, so the C / N ratio needs to be larger than 0 dB.

On the other hand, a SAW device that performs post-correlation detection performs an improvement in the S / N (signal power / noise power) of the process gain in order to perform despreading of the received signal while including the carrier. As a result, demodulation is possible even in an environment with a poor C / N ratio. If the despread signal is larger than the noise power, detection and demodulation are possible, so the C / N ratio at the time of reception needs to be larger than -Gp (dB). This means that demodulation is possible as long as the noise power is not greater than the carrier power by the process gain statement. From this, it can be said that the SAW device is optimal as a correlator for spread spectrum communication. Another analog correlator is a CCD (Charg).
e Coupled Device).

S as a correlator for spread spectrum communication
AW devices include a SAW convolver type and a SAW correlator type. Both are analog correlators that perform detection after correlation. FIG. 36 is a diagram showing the structure, features and application fields of the SAW correlator and the SAW convolver in comparison. As shown in FIG. 36, the SAW convolver has an input electrode 14 for receiving an input signal (f (t)).
And the reference electrode 15 to which the reference signal (g (t)) is input
And an input signal (f (t)) and a reference signal (g (t))
Perform an integration processing operation based on the output signal (h
(T)), and performs a correlation operation on the input signal while including the carrier signal, and automatically performs a high-speed correlation operation completely asynchronously. Since the correlation code is determined by the signal pattern of the reference signal, a correlation operation can be performed on an arbitrary signal. The SAW convolver becomes a programmable correlator by having a configuration in which the reference signal can be switched, and therefore has high versatility and is suitable as a correlator for a CDMA base station.

The SAW correlator comprises an input electrode 17 receiving (f (t)) and a tapped delay line 18. The correlation code is determined by the tapping pattern on the tapped delay line 18. Since the code is formed at the time of device manufacture, it becomes a fixed code correlator,
Like the SAW convolver, high-speed correlation operation can be performed automatically completely asynchronously. Further, the SAW correlator has a simple structure and can be miniaturized, and thus is suitable as a correlator for a portable wireless terminal.

[0035]

The codes generally used as CDMA codes include the m-sequence and the Gold-sequence as described above. (M-sequence) The m-sequence has the following features. (1) The appearance probabilities of the signs “1” and “0” are almost the same. (2) The side lobe level of the autocorrelation characteristic of the sequence is -1
/ (Sequence length). (3) Generation of a series is easy.

When an m-sequence is used as a code for performing CDMA, there are two ways of thinking. One is a method in which a certain m-sequence is prepared and each user uses the m-sequence shifted in time. The correlation value between the m sequences shifted in time is -1 / (sequence length) as described above, so that a good number of combinations of correlation characteristics can be obtained as many as the sequence length. However, in this method, since each user must transmit a signal in synchronization, it is difficult to use for general multiple access. In addition, since the cross-correlation value is not 0, interference from each channel causes a decrease in noise tolerance and a deterioration in error rate characteristics.

The other of the two methods is that each user uses a different sequence, that is, m sequences generated from shift registers at different tap positions. In this case, the inter-channel interference increases and the number of sequences in the same cycle is small, so that the number of channels is limited (18 sequences in 127 chips).

(Gold sequence) The disadvantage that the number of codes that can generate the m sequence is small is compensated by the G sequence based on the m sequence.
Old series. The Gold sequence is obtained by preparing two types of m-sequence generators and adding their outputs. By changing the initial offset of the shift register of length N, a large number of codes are generated, which is incomparable to the m-sequence, such as 2 ^N +1 including the two m-sequences as the basis. However, the maximum value of the cross-correlation value slightly deteriorates. In the case of the m-sequence and the Gold sequence, the properties thereof are well studied mathematically, and an upper limit of the cross-correlation value is given. Therefore, although the interference is not zero, the number of simultaneous connections and the like can be estimated, which is very convenient.

(Orthogonal m-sequence) The orthogonal m-sequence has been devised as a method for solving the problem of the m-sequence. This series is
It is generated by adding one chip after the M types of sequences obtained by cyclically shifting the m sequence one chip at a time. At this time, the chips to be added are “1”, “
0 ”is selected to be equal. An m-sequence obtained by shifting the m-sequence of length M by one chip is shown in the following equation: M = ｛x ₁ , x ₂ , x ₃ ,. _{x M-1, x M}} M = {x 2, x 3, x, ···, x M-1, x M, x 1} ········ M = {x M, x 1 , X ₂ , x ₃ ,..., X _M-1 } A sequence obtained by adding a sequence obtained by adding one chip after each sequence of the above equation is M = {x ₁ , x ₂ , x _{3, ········, x M-1} , x M, x add} M = {x 2, x 3, ·····, x M-1, x M, x 1, x add ... M = {x _M , x ₁ , x ₂ , x ₃ ,..., X _M−1 , x _add }. 0 in the side lobe
However, the cross-correlation value can be set to 0 when the phase difference is 0, and M-multiplex communication can be performed.

(Orthogonal Gold Sequence) The orthogonal Gold sequence has the same construction method as the Gold sequence, and is orthogonal because the sequences are orthogonal, and is simply called an OG sequence. Here, two m-sequences, A = {a _i ; i = 0, 1, 2,..., N−2} B = {b _j ; j = 0, 1, 2,. .., N−2} (n = 2 ⁿ ) are generated from shift registers at different feedback tap positions. By adding one chip "0" after the sequence, a sequence having a sequence length of N = ²ⁿ can be written as follows. U = (a ₀ , a ₁ , a ₂ ,..., A _N−2 , 0) =
(A, 0) V = (T ^j (b ₀ , b ₁ , b ₂ ,..., B _N−2 ),
0) = (T ^j B, 0) Here, T B is obtained by cyclically shifting the chips of sequence B j times. U, V (j = 0, 1,..., N-2)
Forms a sequence set of N sequences as follows. OG (A, B) = ｛U, (U * V _j (j = 0,1,2,2
..., N-2)) Here, "*" is an operator representing exclusive OR. O
N sequences in a set G (A, B) of G sequences are orthogonal to each other and have better correlation characteristics than orthogonal m sequences.

(Walsh code) The Walsh code is Ha
It is represented as a row vector of a damard matrix (H matrix). The H matrix is composed of (1) a square matrix and (2) an element of the matrix is either +1 or -1. (3) All two arbitrary row vectors are orthogonal. And that condition is satisfied matrix called, 2 ⁱ number of Walsh codes is obtained from 2 ⁱ following H matrix. Below 2 ⁱ
A method for creating the following H matrix will be described.

(Equation 4) From the property (3) of the H matrix, the Walsh code also has a cross-correlation value of 0 at a phase difference of 0, and 2 ⁱ orthogonal channels are obtained. However, when the phases are shifted, cross-correlation interference appears.

As described above, when an m-sequence or a Gold sequence is used as a code generally used as a CDMA code, there is a problem that cross-correlation interference appears between channels.

An object of the present invention is to solve the above-mentioned conventional problems and to provide a spread spectrum radio communication system having no interference between channels and a simple circuit configuration.

Further, the present inventor focused on CDMA, which is a feature of the conventional spread spectrum communication, and conducted research and development with the following objectives. (1) Propose a CDMA code free of inter-channel interference. (2) A correlation system using a SAW convolver is designed, and guidelines for putting the proposed code into practical use are shown. (3) We propose a CDMA system using the CDMA code, and show that its performance is sufficient as a practical solution of the local CDMA cell technology.

[0045]

According to the present invention, in order to achieve the above object, an approximate synchronous CDM is used as a code for spreading processing.
The code for A is used. This code for approximate synchronous CDMA is a code of a polyphase sequence represented by a complex number, and has recently been proposed as a code that can be used in wireless communication and the like.
Here, the code for approximate synchronous CDMA will be described.

The codes (m-sequence, Gold-sequence,
Walsh code), the cross-correlation value is not 0, or even when the correlation value becomes 0, it is limited to a certain point in time. From the viewpoint of realizing a CDMA system, it is desirable that the range of orthogonality is as wide as possible. Below,
An approximate synchronous CDMA code having orthogonality over a wide range will be described.

(Orthogonal sequence) Periodic sequence of period N (..., A ₀ , a ₁ ,..., A _N−1 , a ₀ , a ₁ ,
, A _N-1 ,...) Can be expressed as a cyclic matrix as follows.

(Equation 5)

A is a cyclic matrix, and B is a diagonal matrix. Here, when the diagonal element of B is equal to the first column of A that has been subjected to DFT (Discrete Fourier Transform), the following equation holds. A = F ^-1 BF where F is a DFT matrix. In a periodic sequence, when the autocorrelation function becomes 0 in all terms other than a multiple of the period, the sequence is referred to as an orthogonal sequence. Cyclic matrix A
Is a unitary matrix, A represents a sequence, and B is also a unitary matrix. Therefore, the absolute value of the diagonal element of the diagonal matrix B is 1. If the periodic sequences A and C without cross-correlation are cyclic matrices representing the periodic sequences, there are diagonal matrices B and D such that A = F ^-1 BF and C = F ^-1 DF. The cross-correlation function of the periodic series represented by A and C can be expressed as follows.

(Equation 6)

As one example, an orthogonal sequence of period 2 (1
Consider the period sequence of period 6 obtained from j). A new orthogonal sequence is obtained from (1j) by Fourier transform as shown in equation (A).

(Equation 7) F ₂ is a 2-point DFT matrix. In addition,
(A) formula of the matrix elements on the right side for convenience respectively W ¹ _8, W
^{Notated as 7} ₈ Since ^{_{(W 1 8, W 7 8}} ) are orthogonal sequence, periodic sequence of periodic 6 is given by formula (B) below.

(Equation 8) Here, F ₆ is a sixth-order DFT matrix.

Each column on the left side of the equation (B) represents a signal peak in the frequency domain, while each column on the right side represents a signal peak in the time domain. The three periodic sequences in the time domain are obtained by the columns on the right side of the equation (B). Since the multiplication of the terms associated with those spectra is 0 in all terms, the cross-correlation functions of two columns obtained from the three periodic sequences are 0 in all terms. The first column on the right side of the equation (B) is a quaternary sequence. Even if others are multi-valued series,
A quaternary sequence can be obtained by the methods described in the equations (A) and (B).

Further, by replacing these multi-valued sequences, carriers are allocated so that the spectra do not overlap, and a four-valued sequence is used. FIG. 1 shows that any multi-level sequence can be used by shifting the carrier frequency. “Sequence 1” is a quaternary sequence from the first term on the right side of the equation (B). When “sequence 1” and “sequence 2” are used at the same carrier frequency f ₀ as channel A and channel B, the spectra of channel A and channel B are shown in FIGS. 1 (a) and 1 (b). . Here, FIG.
As shown in (c), since the two spectra of channel A and channel B do not overlap each other, the cross-correlation value of channel A and channel B does not appear. FIG. 1 (d)
, A carrier frequency of (f ₀ + Δf) can be used for “sequence 1” for channel B. The peak of the spectrum is shifted by Δf. FIG.
As shown in FIG. 1E, the spectrum in FIG.
Since it does not overlap with (c), the cross-correlation function between channel A and channel B also becomes zero. Therefore, all channels can be divided by using a quaternary sequence to which carriers are assigned.

In the above description, the orthogonal sequence (1j) is used as an example. However, other orthogonal sequences can be used for signal design even if (1j) is changed.

As another example, consider an orthogonal sequence (1,1, W ₃ ) having a period of 3.

(Equation 9) Further, F ₁₂ is 12-order DFT matrix.

From the column vector of the matrix on the right side of equation (2),
Four polyphase periodic sequences are obtained. In the matrix on the left side, the elements are arranged so as to be shifted from each other so that the columns are orthogonal to each other. Here, a row can be regarded as each channel, and a column can be regarded as a frequency axis. Therefore, the matrix on the left-hand side corresponds to designing so that the frequency spectrum of each channel sequence does not overlap each other. Therefore, no matter which two of the sequences on the right side obtained by transforming into the time domain by the inverse DFT, the cross-correlation function is 0 in all terms.

The four column vectors on the left side of equation (2) are
(1, 1, W ₃₎ but has a structure that inserts 0 into between elements, and has a (1, 1, W ₃₎ is to be the orthogonal sequence, each orthogonal sequence. In this case, (1,1,
An orthogonal sequence that performs the same function as W ₃ ) is called a basic orthogonal sequence. The autocorrelation function of the four polyphase periodic sequences obtained as described above is as follows. (100100100100) (100j00-100-j00) (100-100100-100) (100-j00-100j00)

In general, an orthogonal sequence created by extending a basic orthogonal sequence is converted into a polyphase periodic sequence by inverse DFT, and its autocorrelation function becomes 0 in all terms other than a multiple of the period of the basic orthogonal sequence. . The code created according to the equation (2) has a feature that the correlation value becomes 0 even when the code synchronization is shifted slightly back and forth in the autocorrelation as well as the cross-correlation.

The approximate synchronous CDMA system controls and operates so that all signals fall within a certain range. The code for approximate synchronous CDMA created by the procedure as in equation (2) is sometimes called a Suehiro code in the name of the proponent (Naoki Suehiro: Associate Professor at University of Tsukuba).

(Code for Approximately Synchronous CDMA with Small Phase States) As described above, the code for approximate synchronous CDMA obtained from the right side of the equation (2) is a multiphase periodic sequence.
The number of phase states of the series increases as the cycle of the series becomes longer (the number of possible phase states is 128 in the series cycle 128). If the number of phase states to be considered is too large, the operation of accurately generating a code becomes complicated and difficult in practice. Therefore, in the present invention, the approximate synchronous CD
We propose a method for generating MA codes with a small number of phases. This facilitates sequence control and facilitates the implementation of a code generation circuit.

Here, a method of reducing the number of phase states will be described. First, DFT is performed on the basic orthogonal sequence. Since the obtained sequence is also an orthogonal sequence, this sequence is newly set as a basic orthogonal sequence, and the period is extended to an arbitrary length by inserting 0 between the elements of the sequence similarly to the equation (2). And
The matrix is constructed so that each column vector is orthogonal, ie, the frequency spectra do not overlap, and then the inverse DFT
To obtain a signal on the time axis. As a result, in the first column of the obtained matrix, the first basic orthogonal sequence is continuous, and the number of phase states is equal to that of the basic orthogonal sequence.

As an example, a case where the basic orthogonal sequence is a sequence with a period of 3 (1, 1, W ₃ ) will be described as in the above description.

(Equation 10)

Looking at the first column of the matrix on the right side of the equation (3), it can be seen that the basic orthogonal sequence (1,1, W ₃ ) is repeated. Therefore, the possible phase states of the sequence is two kinds of "1" or "W _3". However, a sequence obtained from another column vector is the approximate synchronous CDMA described above.
It is a polyphase periodic sequence like the use code. However, the characteristics of these series can be expressed using the series in the first column.

Focusing on the spectrum of the series obtained from the right side of the equation (3), it can be said that each series has the same number of (in this case, 3) peaks at equal intervals as the period of the basic orthogonal sequence in a certain band. Are common, but are distinguished by different peak positions. In other words, the spectral characteristics of each series are obtained by shifting the first column vector. The orthogonality of the approximate synchronous CDMA code is realized by the fact that the code spectra do not overlap each other. Therefore,
The same characteristics as when other column vectors are used as spreading codes by using only the sequence with the least phase state and displacing the carrier frequency to be multiplied to shift the spectrum as a code for each channel. Is obtained. This is shown in FIG.

As the basic orthogonal sequence, a four-phase orthogonal sequence instead of (1,1, W ₃ ), (1,1,1,1,1, j, -1, -j, 1, -1,
(1, -1,1, -j, -1, j) or a two-phase sequence (1,1,1, -1) gives a four-phase or two-phase approximate synchronous CD, respectively.
A code for MA is generated. These can be easily realized by a circuit. And it can be applied to SS communication.

Therefore, the present inventor actually created a code generator, constructed a correlation system using a ZnO / Si type SAW convolver as a correlator, and observed the correlation characteristics of the code for approximate synchronous CDMA. In the construction of this correlation system, a ZnO / Si type SAW as a correlator is used here.
A convolver was used. The structure is shown in FIG. As shown in this figure, the ZnO / Si type SAW convolver 2
Numeral 0 denotes a laminated structure of a ZnO (zinc oxide) layer 21, a SiO ₂ (silicon dioxide) layer 22, and a Si (silicon: silicon) structure 23 in order, and a back connection electrode 24 on the back surface of the Si structure 23. Is arranged. The gate electrode 25 is provided on the ZnO layer 21 while the IDT 2
6 (26a, 26b) are provided. The above Z
Table 4 shows the main characteristics of the nO / Si type SAW convolver 20 in comparison with a conventional elastic type convolver.

[Table 4]

As is apparent from Table 4, ZnO / S
The operating center frequency of the i-type SAW convolver 20 is 215M
The product of Hz and BT (Band-Time) is 207.
The BT product is the product of the 3 dB bandwidth and the delay time, and when applied to spread spectrum communication, corresponds to the maximum process gain of a SAW device. ZnO / Si type SAW
The convolver 20 realizes a terminal efficiency of −42 dB due to the nonlinear capacitance of the Si depletion layer in the gate electrode 25. This value is about 15 dB better than that of the conventional elastic convolver, and is optimal as a correlator for a wireless terminal requiring a medium BT product. The feature of the ZnO / Si type SAW convolver is that it realizes higher efficiency than the elastic type convolver. Especially,
The SAW convolver can programmatically operate matched filters for several series. This is because the reference signal is determined by the filter characteristics of the SAW convolver.

[0066]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a first embodiment of the present invention will be described. FIG. 4 is a block diagram illustrating a circuit configuration of a correlation system configured using the ZnO / Si type SAW convolver 20. As shown in this figure, the correlation system includes a ZnO / Si type SAW convolver 20, a first oscillator 27 for generating a first signal having a frequency fa input to the ZnO / Si type SAW convolver 20,
PN code 1 for inputting PN code 1 for the first signal
A supply unit 28, a first multiplier 29 for multiplying the first signal by the PN code 1, and a ZnO / Si type SAW convolver 2
A second signal for generating a second signal of frequency fb input to 0
Oscillator 30, a PN code 2 supply unit 31 for inputting a PN code 2 to the second signal, a second multiplier 32 for multiplying the second signal by the PN code 2, a ZnO / Si type S
And a band-pass filter 33 for performing band separation on the output of the AW convolver 20. For PN codes 1 and 2, which are spreading codes, the code length is 1
The m-sequence of 27 chips and the code rate are 14 MHz.

The input signal and the reference signal are applied to respective digital converters. The output signal from gate electrode 25 is a related signal between the input and the reference signal.

FIG. 5 shows ZnO / O obtained by performing a correlation process using the correlation system having the circuit configuration shown in FIG.
FIG. 7 is a diagram illustrating a result of an envelope characteristic of a Si type SAW convolver 20 with respect to a spreading code and a frequency. In this case, two binary modulation (BPSK) signal center frequency is f _a and f _b are input to respective input terminals of the convolver 20. The PN code used for the spread modulation is an m-sequence. The code length is 127 chips and the code chip rate is 14 Mcp
s. FIG. 5A is a diagram illustrating a change in the correlation peak value when the PN code changes. When the same code is input to the SAW convolver, the autocorrelation output of the spread code is obtained as shown in FIG. 5A, and the magnitude of the cross-correlation is an output having a steep peak. When the preferred pair m-sequences m (7.1) and m (7.3) are input, an output reflecting the cross-correlation characteristic is obtained, and the correlation output is suppressed as compared with the magnitude of the auto-correlation characteristic. You can see that it is. On the other hand, the change of the correlation peak when the input center frequency of the SAW convolver is changed is shown in FIG.
(B). In this case, the frequency difference between f _a and f _b is 11
At 0 kHz or more, the correlation peak disappears. 110k
The value of Hz corresponds to the reciprocal of the propagation delay time of the SAW convolver, that is, the gate delay time of 9 μsec (microsecond), and is the mathematical orthogonal frequency of the integrator having this integration time.

At this time, the autocorrelation peaks appearing periodically are obtained aperiodically and automatically as shown in FIGS. 5A and 5B. Period of autocorrelation peak (4.5
μS) is half the period of the PN code (9 μS).
This is because the two inputs propagate from the gates in opposite directions, and their relative speed is twice the SAW propagation speed.

Experiments have shown that a sufficient channel separation can be ensured by utilizing this characteristic. Such a channel separation method is referred to as a minute frequency displacement type multi-channel.

The SAW convolver does not have code information inside the element and performs a completely analog operation. Therefore, the SAW convolver has a feature that there is no limitation on the spread code to be used and the maximum correlation operation can be performed asynchronously. In the configuration of the SAW convolver correlation system for the code for the approximate synchronous CDMA, this feature and the minute frequency displacement type multi-channel are positively applied. Table 5 shows the experimental specifications of the correlation system.

[Table 5]

In the example described below, a four-phase sequence having a period of 16 is used as a basic orthogonal sequence. From this four-phase series (A)
A four-phase sequence with a period of 128 is created by the method represented by the expression and the expression (B). Here, the following four-phase sequence is used (1 + j / √2, −1 + j / √2, −1−j / √2, 1
−j / √2) They are obtained by rotating (1, j, −1, −j) by π / 4 in the complex plane. With this processing, the real part and the imaginary part of the four-phase series can be expressed by only binary values of 1 and −1. Therefore, the real part and the imaginary part 2
Since it is a value series, it can be easily realized by a digital circuit.

In addition to the above processing, a DFT is performed on the basic orthogonal sequence having the period 16 to obtain a new orthogonal sequence.

[Equation 11] Next, an approximate synchronous CDMA code is generated in the same manner as in the equations (A) and (B).

(Equation 12)

As shown in the equation (4), since the number of channels is set to 8 in the design of the spectral domain, the period is extended to 128 by inserting seven 0s between the elements of the orthogonal sequence. The elements of this series are sequentially shifted into an eight-column matrix. By performing DFT on this matrix, a signal in the time domain is obtained.
(4) In the first column of the matrix on the right side of the equation, a sequence in which the basic orthogonal sequence is repeated appears. This series is ± 1 and ±
Since it is composed of four values of j, it becomes a four-phase series. On the other hand, the sequences obtained from the other vectors are polyphase periodic sequences as in the past, and it is necessary to control the phase with 128 levels of accuracy in order to generate these sequences by the circuit. Becomes difficult and sufficient characteristics cannot be expected.

Therefore, as described above, the characteristics of the other series are realized using the series of the first column having the least number of phase states. That is, the carrier frequency is set to 110 for each channel.
Assigned by displacing each kHz, instead of the respective signs. As a result, the position where the peak of the spectrum rises is shifted, so that the same characteristics as when using each polyphase periodic sequence can be realized.

By the way, the obtained four-phase approximate synchronous CDM
The phase of the code for A is located on the real axis and the imaginary axis when viewed in the phase space. Therefore, in consideration of controlling the sign by dividing the sign into a real part and an imaginary part, three values of ± 1 and 0 are required. Therefore, for further simplification, in this experiment, the code is rotated by 45 degrees in the phase space as shown in FIG. 6 so that one state comes to each quadrant. With this process,
The real and imaginary parts of the sign take only two values, ±
It is possible to perform easy control by a digital circuit using.

The code for the approximate synchronous CDMA is 14 Mcp.
When used in s, 16 spectrum peaks will be raised in a bandwidth of 14 MHz. Further, the autocorrelation characteristic of the code is 0 in all terms other than a multiple of the period 16 of the basic orthogonal sequence.
Becomes That is, eight correlation peaks appear every 16 chip times in one cycle of correlation. On the other hand, the cross-correlation characteristic with the code for another channel used by displacing the carrier is theoretically zero.

FIGS. 7 and 8 are block diagrams showing the configuration of a correlation system using a SAW convolver in this embodiment. Among these figures, FIG. 7 is a block diagram showing a configuration of a transmission unit of the correlation system, and FIG. 8 is a block diagram showing a configuration of a reception unit of the correlation system. In FIG. 7, reference numeral 37 denotes a code generator for generating an approximate synchronous CDMA code, 38 denotes _a carrier signal generator serving as an output source of _a transmission carrier signal (fa), and 39 denotes a real part of the approximate synchronous CDMA code. A multiplier for multiplying the signal; 40, an orthogonal component generator for generating an orthogonal component of the carrier signal generated by the carrier signal generator 38; 41, an orthogonal component of the carrier signal for the imaginary part of the approximate synchronous CDMA code; A multiplier for multiplication, an adder for adding the output of the multiplier 39 and an output of the multiplier 41, and an antenna 43 as an antenna for emitting a transmission signal are provided.

In FIG. 8, reference numeral 44 denotes a reference code generator for generating a reference code for approximate synchronous CDMA; 45, a carrier signal generator serving as an output source of a received carrier signal (f _b );
46 is a multiplier for multiplying the real part of the code for approximate synchronous CDMA by a signal; 47 is an orthogonal component generator for generating orthogonal components of the carrier signal generated by the carrier signal generator 45; A multiplier for multiplying the imaginary part of the code by the orthogonal component of the carrier signal;
An adder for adding the output of the multiplier 6 and the output of the multiplier 48;
Is a SAW that receives the output of the adder 49 and performs a correlation process
The convolver 51 is an antenna serving as an antenna for receiving a radio signal.

The code generators 37 and 44 are provided with a PLD (Pro
It is realized with a simple configuration centered on a grabable logic device. Code generator 37, 4
4 outputs the real part and imaginary part of the four-phase approximate synchronous CDMA code. In the transmitting section, the approximate synchronous CDMA
The real part of the use code is multiplied by the in-phase (one phase) component of the carrier, and the imaginary part is multiplied by the quadrature component (Q phase) and transmitted. In the receiving unit, a reference code is generated, the real part signal and the imaginary part signal of the approximate synchronous CDMA code are correlated by the SAW convolver 50, and the result is obtained by performing an arithmetic processing on both. SAW
Due to the use of the convolver, the correlation process is performed completely asynchronously and automatically. Note that the signal propagation path is configured by wire, but a part or all may be wirelessly connected.

The feature of the correlation system using the SAW convolver realized in this way, and furthermore, the spread spectrum radio communication system is that only one convolver 50 is used in the receiving section. The reason why the correlation can be achieved by providing only one convolver in the receiving unit will be described below.

(R + jI) of the transmission sequence and (R + jI) of the reference sequence
+ 'JI'). The multiplication of these two complex numbers can be expressed as in equation (C). (R + jI) (R + 'jI') = (RR'-II ') + j (RI' + R'I) Equation (C) Since the right side of equation (C) consists of four terms, the correlation is calculated. To do so, four correlators are needed. When calculating the autocorrelation, (R-jI) is used as a reference sequence. (R + jI) (R−jI) = (R ² + I ² ) (D) The autocorrelation function is obtained by integrating the equation (D) over one period. From the theory of the code for the approximate synchronous CDMA, the cross-correlation is zero. Therefore, the integral value of both (RR'-II ') and (RI' + R'I) is zero. Since the imaginary part of the correlation function is 0 for both cross-correlation and auto-correlation, it is necessary for the correlation system to realize the real part of equation (C).

One convolver 50 shown in FIG.
Is mathematically expressed as follows. (Rcosωt + Isinωt) (R′cosωt + I′sinωt) (E) When the DC component of the equation (E) is removed, the following equation (F) is obtained. (1/2) (RR′−II ′) cos (2ωt) + (1/2) (RI′−IR ′) sin (2ωt) Equation (F)

If the reference sequence is a complex conjugate of the transmission sequence, the autocorrelation has the form (R ² + I ² ) cos (2ωt). After performing numerical calculations with a computer,
It has been found that the value of the cross-correlation in the equation (E) is 10 ^-4 or less. Therefore, the experimental correlation system using one SAW convolver can output the auto-correlation and cross-correlation characteristics.

FIG. 9 is a diagram showing a power spectrum (transmission spectrum) of a transmission signal generated from the transmission unit in the above-described correlation processing. From the theory of signal design, the number of spectrum peaks appearing in the chip width is the number of periods of the basic orthogonal sequence. From this figure, it can be seen that 1 out of 14 MHz bandwidth
The peaks of the six spectra are observed, and it can be seen that the codes as in the theory are generated. Multiplexing is realized by displacing the carrier by a minute frequency so that the peaks do not overlap (a minute frequency displacement type multi-channel). It should also be emphasized that the 28 MHz spreading band is close to the SS band allowed in Japan. In the above experiment, the center frequency is 215 MHz. Since it is easy to up-convert the carrier frequency to 2.4 GHz by adding a mixer and a transmission circuit, it is sufficiently applicable to 2.4 GHz.

FIG. 10A shows the theoretical characteristics of the autocorrelation characteristics of the code for the approximate synchronous CDMA, and FIG. 10B shows the measured values of the autocorrelation characteristics of the code for the approximate synchronous CDMA. Theoretically, a correlation peak appears in a term of a multiple of the cycle of the basic orthogonal sequence in one cycle, and becomes 0 in other cases. In the case of the present embodiment, since the sequence period is 128 and the period of the basic orthogonal sequence is 16, eight peaks appear in the correlation of one period of the sequence. However, in the case of a convolver, the input signal and the reference signal travel in a direction opposite to each other (that is, the input signals propagate to the electrodes in the opposite directions to each other). A peak appears. From the empirically observed autocorrelation characteristics described above, it can be seen that the operation results as theoretically obtained. However, in the simulation, the off-id lobe between each peak is 0, but in the actual measurement, this is not the case. This can be caused by
Two SAW convolvers 49 and 50 which are analog elements
And that the orthogonality of the I and Q phases of the carrier is not sufficient (deviation from π / 2) or that the adjustment of the phase state of the two output signals is not optimized. Conceivable.

FIG. 11A shows the theoretical characteristics of the cross-correlation characteristics of the code for the approximate synchronous CDMA, and FIG. 11B shows the measured values of the cross-correlation characteristics of the code for the approximate synchronous CDMA. In this case, the center frequencies of the transmission signal and the reference signal are 215 MHz and 215.11 MHz, respectively. FIG. 11B shows that the cross-correlation characteristics sufficiently suppressed can be obtained experimentally from these figures.

Embodiment 2 CDMA System Using Approximately Synchronous CDMA Code Next, a communication system is designed using an approximately synchronous CDMA code. First, in order to perform data modulation while maintaining the characteristics of the approximate synchronous CDMA code, a concept of pseudo-periodicization of the code is introduced. Next, a result of actually creating a pseudo-periodic sequence generator and observing its characteristics will be described. Further, a CDMA system will be examined, and a cell radius and a data transmission rate will also be examined.

As described above, the approximate synchronous CDMA code has a feature that the cross-correlation value is 0 and the side lobe between the auto-correlation peaks is 0 in all sections when only the code is used. However, BP is used as the code for the approximate synchronous CDMA.
When data modulation such as SK is performed, its characteristics deteriorate. FIG. 12 is a diagram showing a simulation result of a correlation characteristic of an approximate synchronous CDMA code when BPSK is used as a data modulation method and data is "1010...". In FIG. 12, the upper left diagram shows the approximate synchronization C
The autocorrelation characteristic when only the code for DMA is used,
Similarly, the upper right diagram shows the cross-correlation characteristics when only the approximate synchronous CDMA code is used. The lower left figure shows the autocorrelation characteristics when the BPSK data modulation is performed on the code for the approximate synchronous CDMA. Similarly, the lower right figure shows the approximate synchronous C code.
7 shows cross-correlation characteristics when BPSK data modulation is performed on a DMA code. From these figures, it can be seen that the features listed above have been lost. That is, the side lobe value between the peaks in the autocorrelation does not become 0 as indicated by “P” in the left diagram in the lower part of FIG. This peak also appears in the cross-correlation. This is the data
This is because the code phase is switched to 0 or π with respect to 1 ”and“ 0 ”, so that the characteristic of the code is lost when the data is switched.

As a method of maintaining the characteristics of the periodic sequence even for data modulation, a concept called pseudo-periodicization has been proposed. First, this concept will be described. (Pseudo-periodic sequence) Assuming that a sequence of finite length N is A = (a ₀ , a ₁ ,..., A _N−1 ), a sequence of length N + 2L created based on A, A ′ = ( a _NL , ..., a _N-1 , a ₀ , a ₁ , ...
, A _N−1 , a ₀ ,..., A _L−1 ) are called a pseudo-periodic sequence of A. FIG. 13 is a diagram illustrating a method of creating a pseudo-periodic sequence. In this figure, the L chips from the beginning of A ′ are the same as the series of L chips after A. On the other hand, the L chip after A 'is the same as the series of the first L chip of A. In other words, A 'is a shape obtained by adding a preceding and succeeding L-chip sentence around A and cutting it out from the periodic sequence (... AAA ...). In this state, A to A '
Is called pseudo-periodicization. When the correlation between the pseudo-periodic sequence A ′ and the sequence A is obtained, an autocorrelation output for 2N + 2L−1 chips is obtained. Among them, the output of the central 2L + 1 chips is equal to the autocorrelation function of the periodic sequence of A. Further, consider that a sequence of finite length N different from A is used instead of A and B is used to correlate A ′ and B. Then, as a result of the correlation, the center 2L + 1 chips match the cross-correlation function between A and B.

As an example, consider an orthogonal sequence A = (111-1) having a period of 4 (N = 4). First, the autocorrelation function of the periodic sequence of A is obtained as (... 40004000...). Further, the correlation between one period of A is (−101410-1).

Next, pseudo-periodicization is performed on A. L = 2
Then, A ′ = (1-1A11) = (1-1111-111) is obtained. When the cross-correlation of one period of A ′ and A is obtained, the result is (−12−100400121). You can see that we are doing it.

(Pseudo-periodic approximated synchronous CDMA code)
Consider a pseudo-periodic sequence of an approximate synchronous CDMA code.
As described above, the code periodic correlation characteristic is maintained in the central 2L + 1 chip of the correlation output between the pseudo-periodic approximate synchronous CDMA code and the normal approximate synchronous CDMA code. That is, no side lobe between the autocorrelation peaks is detected for the desired signal, and no correlation value appears for the undesired signal. This is because the chip added by pseudo-periodicization of the code plays a role as a guard bit for data modulation. Therefore, the autocorrelation characteristic of the code is maintained for the signal shifted by L chips. That is, if all the signals fall within the shift within the L chip time (that is, if the approximate synchronous control is performed), the characteristics of the approximate synchronous CDMA code are exhibited.

By simulation, the approximate synchronous CDM
The effect of pseudo-periodicization of the A code on data modulation was examined. The code for the approximate synchronous CDMA of N = 128 is L = 1.
5 and FIG. 15 show the simulation results of the correlation between the signal subjected to pseudo-period and subjected to BPSK modulation and the reference approximate synchronous CDMA code not subjected to pseudo-period. FIG. 14 shows the correlation between the desired signal and the pseudo-periodic sequence. In this figure, the side lobe between the autocorrelation peaks is 0 at the center, and the characteristics of the approximate synchronous CDMA code are maintained. I have. FIG. 15 shows the correlation between the undesired signal and the pseudo-periodic sequence. In this case, the code characteristics are maintained similarly, and the cross-correlation value is 0 at the center in FIG. Thus, it can be seen that pseudo-periodicization of the code is effective as a measure against deterioration of the characteristics of the code due to data modulation.

(Correlation Characteristics of Pseudo-Periodic Approximate Synchronous CDMA Code) Next, a pseudo-periodic sequence generator is actually created, and the correlation characteristics of the pseudo-periodic sequence generator-generated approximate synchronous CDMA code are examined. . FIGS. 16 and 17 are block diagrams showing an example in which a pseudo-periodic sequence generator is attached to a correlation system using a SAW convolver in this embodiment. Among these figures, FIG. 16 is a block diagram showing a configuration of a transmission unit of the correlation system, and FIG. 17 is a block diagram showing a configuration of a reception unit of the correlation system. In FIG. 16, reference numeral 55 denotes a pseudo-periodic sequence generator for pseudo-periodicizing an approximate synchronous CDMA code to generate a pseudo-periodic sequence code; 56, a baseband data generator for generating baseband data;
7, a multiplier for multiplying the real part of the pseudo-periodic sequence code by the baseband data signal; 58, a multiplier for multiplying the imaginary part of the pseudo-periodic sequence code by the baseband data signal;
Is a carrier signal generator serving as an output source of a transmission carrier signal, 60 is a quadrature component generator for generating a quadrature component of the carrier signal generated by the carrier signal generator 59, and 61 is a carrier for an output signal of the multiplier 57. A multiplier for multiplying the output signal of the multiplier 58 by an orthogonal component of the carrier signal; a multiplier 63 for adding an output of the multiplier 61 to an output of the multiplier 62; Is an antenna as an antenna for emitting a transmission signal.

In FIG. 17, reference numeral 65 denotes an approximate synchronous CDMA.
A reference code generator for generating a reference code for use, 66 is a carrier signal generator serving as an output source of a received carrier signal (f _b ), and 67 is a multiplier for multiplying the real part of the approximate synchronous CDMA code by a signal. 68, a quadrature component generation unit for generating a quadrature component of the carrier signal generated by the carrier signal generator 66; 69, a multiplier for multiplying the imaginary part of the approximate synchronous CDMA code by the quadrature component of the carrier signal; Is an adder that adds the output of the multiplier 67 and the output of the multiplier 69;
1 is an SA for inputting the output of the adder 70 and performing a correlation process.
A W convolver 72 is an antenna serving as an antenna for receiving a radio signal. What is important here is that, in the receiving block of FIG. 17, the number of convolvers that perform the correlation is only one as in the case of FIG. The reason why the correlation operation can be performed by one convolver is as described in (0067).

Table 6 shows various parameters used for producing the pseudo-periodic sequence generator.

[Table 6]

In this embodiment, the sequence used in the first embodiment is pseudo-periodic by adding 15 chips before and after. Pseudo-periodic sequence generator 55
For example, Xiliting FPGA (Field
d ProgrammableGate Array) X
This was realized using C4010-4. Data modulation is performed by inverting and inverting the polarity of one period (158 chips) of the pseudo-periodic sequence for "1" and "0" of the baseband data (BPSK method). On the receiving side, on the other hand, a normal approximate synchronous CDMA code is generated as a reference signal, and the result is obtained by correlating with the modulated signal.

Also in this embodiment, the correlation system using the SAW convolver, and furthermore, the feature of the spread spectrum wireless communication system is that only one convolver 50 is used in the receiving section. The reason why the correlation can be achieved only by providing only one convolver in the receiving unit is the same as that described in the first embodiment, and the description is omitted here.

FIGS. 18 to 21 are diagrams showing the results of correlation characteristics of pseudo-periodic approximate synchronous CDMA codes obtained using the above communication system. Of these figures, FIG.
8 is a diagram illustrating a correlation result with respect to a desired signal when only a code without multiplying data is used, and FIG. 19 is a diagram illustrating a correlation result with respect to another station signal when only a code without multiplying data is used. FIG. 20 is a diagram showing a correlation result with respect to a desired signal when data is modulated by alternating "0" and "1", and FIG. 21 is a diagram when data is modulated by alternating "0" and "1". It is a figure showing the correlation result with respect to another station signal. 18 and 19, the correlation with respect to the desired signal (fa = fb = 215 MHz) and the correlation with respect to the signals of other stations (fa = 215 MHz, fb = 215.11).
It can be seen that all the correlations (MHz) substantially match the theoretical characteristics obtained by simulation. On the other hand, FIG.
In the case of FIG. 21 and FIG. 21, the modulation is performed with the data being alternately “0” and “1”, which is the result when the phase of the baseband data is switched most steeply.
Also in this case, the correlation (fa = f
b = 215 MHz) and correlation (fa
= 215 MHz and fb = 215.11 MHz) hardly change even when compared with the case of only the above-mentioned code, and it can be seen that they substantially match the theoretical characteristics obtained by simulation. That is, even when data modulation is performed, the characteristics of the approximate synchronous CDMA code are maintained without deterioration.

(CDM Using Approximately Synchronous CDMA Code)
A system) Next, a communication form that can exhibit the characteristics (advantages) of the approximate synchronous CDMA code having orthogonality over a wide range will be devised. FIG. 22 is a diagram schematically showing an example of a CDMA system (communication form) using an approximate synchronous CDMA code. This CDMA system assumes a small zone configuration having a base station 75 at the center, that is, a so-called cellular CDMA system. This CDM
The A system includes a plurality of mobile stations 77 within a cell 76 of radius R.
And communicate with other terminals (including mobile stations) through the base station 75. Base station 75 to mobile station 77
Down link (Down Link) 78
And the line from the opposite mobile station 77 to the base station 75 is called an uplink (Up Link) 79. In the downlink 78, since a plurality of signals are transmitted from the base station 75 to each mobile station 77 at the same time, synchronization control of each channel can be performed. Therefore, multiplex communication can be performed even using a conventional synchronous CDMA code such as a Walsh code or an orthogonal m-sequence.

On the other hand, in the uplink, a plurality of mobile stations 77 transmit to base station 75 at a timing independent of each mobile station. Therefore, it is difficult to control the synchronization between the mobile stations 77, and in the case of the conventional multiplexing using the synchronous CDMA code, interference occurs between the channels. As a countermeasure against this, in the current CDMA system, the base station 75 controls the transmission power of the mobile station 77 (Power Control).
l) is being performed. By this operation, the same power is used to reach the base station 75 regardless of where in the cell 76 all the uplink signals originate, thereby reducing the influence of interference. However, to realize this function, an expensive and complicated device configuration is indispensable.

On the other hand, if attention is paid to the pseudo-periodic approximate synchronous CDMA code described above, as described in the above description of this embodiment, even if data modulation is performed, the maximum 2L chip It has a wide orthogonality of time, and no inter-channel interference appears even for a signal shifted by L chips. Utilizing this characteristic, an approximate synchronous CDMA code is applied to the uplink 79.

Next, the radius R of the cell 76 employed in the above-described cellular CDMA system will be discussed. That is, the size of the cell 76 in which all signals naturally fall within the range of the approximate synchronization is determined without performing the approximate synchronization control. From Table 6, L =
Since 15 is obtained, if all signals are within 15 chips, communication free from the influence of interference becomes possible without performing synchronization control between the mobile stations 77. When 15 chips are converted into time, since the chip rate is 14 Mcps, the following equation is obtained. 15 (chip time) = (14M) ⁻¹ × 15 = 1.07 [μ
sec] That is, the synchronization of the CDMA system has a margin of about 1 μsec.

Next, the base station 75 and each mobile station 77 (here
Here, the communication method between the mobile stations 77A and 77B) is described.
Will be described. FIG. 23 is a diagram of the cellular C shown in FIG.
Base station 75 and two mobile stations 77 in a DMA system
A for explaining an example of a communication procedure between the A and 77B.
It is an imchart. In this figure, each mobile station 77
A, the signal to 77B is transmitted simultaneously to the down line 78
On the other hand, each mobile station 77A, 77B responds immediately after reception.
Shall be returned. Time when the signal of the down link 78 was issued
And the time received by the mobile station 77A is
t₁, The time received by mobile station 77B_TwoToss
Then, each from the mobile station 77A and the mobile station 77B
The response signals arrive at the base station 75 at 2
t₁And 2t _TwoBecomes Base station 75 and mobile station 77A,
If the distance between 77B is different, the arrival of the response signal is delayed
Time arises. The mobile station 77A is located near the base station 75
However, when the mobile station 77B is on the outer periphery of the cell 76,
The delay time becomes large, and this chip is within 15 chip hours
If so, all uplink signals 79 do not perform synchronous control
Even within the approximate synchronization range. At this time
When the radius R is calculated, (delay time)_max= (2t_Two-2t₁) = (R-0) / (speed of light) = 1.07 [μsec] (15 chip time), so that R = 160.71 (m) is obtained. This value is
Because the cell radius of PHS is 100-200m
Are also practical enough.

Further, a simulation was performed on the signal of the uplink line 79. FIG. 24 is a block diagram showing a communication form assumed in a simulation of a signal of the uplink 79. FIG. 25 is a diagram showing a correlation result obtained in the simulation for the signal of the uplink 79. A plurality of mobile stations 77 (77A, 77A)
B,... 77G) exist, respectively, τA, τB.
It is assumed that a signal is returned to the base station 75 with a delay time of τG. A total of seven mobile stations 77A, 77B, ...
77G transmit at independent timing. However, it is assumed that the deviation is within 15 chips at the maximum (approximate period is set). It is assumed that all of these signals are added and arrive at the base station 75, and a correlation process is performed on the signals. When the correlation with the mobile station is obtained, a portion where the nearby side lobe becomes 0 appears periodically as shown in FIG. The data is demodulated by taking out this part and removing the carrier. On the other hand, when correlation is performed for the mobile station 77H that is not actually transmitting, the result is as shown in FIG. From this figure, it can be seen that a portion where the cross-correlation value is 0 appears. That is, it is not affected by inter-channel interference.

(Study of Transmission Rate) The transmission rate of baseband data when a pseudo-periodic approximate synchronous CDMA code is used will be examined. Base station 75 and mobile station 77
Is a TDD (Time Division Double)
It is assumed that the full-duplex communication according to x) is performed. TDD
Is a method of dividing transmission / reception timing along a time axis and performing pseudo full-duplex communication. FIG. 26 is a block diagram showing an outline of TDD. There is a master transmitter (base station 75 in this case) with a certain absolute time, and the other slaves (mobile station 77) transmit during a predetermined time. Arbitrary data can be transmitted within the given frame.

Considering that voice is sent as data,
Since the allowable delay time of human voice (the maximum time that a human does not feel a delay) is about 5 msec, the maximum frame length of the line is 5 msec. This value is equivalent to 450 symbol times if one symbol is transmitted in one period of the pseudo-periodic approximate synchronous CDMA code, that is, 158 chips. When the data modulation method is BPSK, one symbol represents one
Since bits are transmitted, the maximum number of transmittable bits per frame is 450 bits. Therefore, assuming that frames are equally distributed in the upstream and downstream, the data rate of the line is 45 kbps.

Here, further enhancement of the performance of the CDMA system using the code for the approximate synchronous CDMA will be examined.
In the above description, the CDM using the approximate synchronous CDMA code is described.
The performance of the system A was estimated for the case where the basic method was used. In this section, we consider increasing the number of channels and increasing the transmission rate as guidelines for realizing higher performance systems.

(Increase in the number of channels) In the case of a code for approximate synchronous CDMA, although there is an advantage that the orthogonality is wide-ranging, the code studied this time requires a code length of 128, a multiplexing number of 8, and a large number of channels. Can not.
On the other hand, an orthogonal m-sequence or Wal
When a synchronous CDMA code such as an sh code is applied,
Ideally, the number of channels close to the theoretical limit can be set, and if the code length is 128, the number of multiplexes is 128. Accordingly, the network has a very strong asymmetry, with the up line having 8 channels compared to the 128 channels of the down line. In order to improve this balance and realize efficient communication, it is desired to increase the number of uplink channels.

The number of channels of 8 studied in the present invention is:
It depends on the current regulations of 26 MHz bandwidth and the conditions such as the 9 μsec integration time of the SAW convolver. If there is no bandwidth limitation, by increasing the chip rate, the peak interval of the spectrum of the approximate synchronous CDMA code is widened and many channels can be set. So, if more bandwidth is granted in the future,
An increase in the number of channels can be expected. 100 as an example
Assuming that the MHz band is established, the number of channels can be simply calculated as 32 since the bandwidth is about four times the current bandwidth.

If the integration time of the SAW convolver can be lengthened, processing of a long-period series becomes possible, and the number of channels can be increased even at the current chip rate of 14 MHz. However, as the code length increases, the data rate decreases. For example, if the integration time of the SAW convolver is doubled to 18 seconds, the code length that can be handled is doubled to 256 chips, and the number of channels is doubled to 1 times.
Can be 6. However, the data rate at this time is about 24
kbps, which is about half of the value estimated in the previous section.

As another method, approximate synchronous CDMA
The number of channels can be doubled by contriving the use code. Earlier, it was stated that the equation for calculating the complex correlation is given by the following equation. Again, the expression is given again. Real part: Rcosωt · R'cosωt + Isinωt · I'sinωt + Rsinωt · R'sinωt + Icosωt · I′cosωt (5) Imaginary part: Icosωt · R′cosωt + Isinωt · R′sinωt− (Rsinωt−Rsinωt−Rsinωt− (Rin sin) I′cosωt) (6) Since the correlation value of the approximate synchronous CDMA code used in the present invention can be obtained as a real number, the output of equation (6) becomes 0,
It was sufficient to consider only equation (5).

Here, the approximate synchronous CD multiplied by j times (imaginary number times)
Consider a code for MA (j-fold code). The spectral characteristic of the code multiplied by j is the same as that of a normal approximate synchronous CDMA code, that is, that of a 1-time code. Here, the 1-time code and the j-time code are respectively expressed as follows.

(Equation 13) Therefore, the cross-correlation value with the code for another channel that does not overlap the spectrum is zero. The problem is the correlation with the 1-time code of the same carrier whose spectrum overlaps, but the correlation value is an imaginary number, and an output appears in equation (6), but is not output in equation (5). Therefore, if the correlation processing is performed by the system corresponding to the equation (5), the 1-time code and the j-time code can be identified, and there is no interference between them. Naturally, if a j-fold code is prepared as a reference signal, detection can be performed by the equation (5). Applying this, 8 channels for 1-time code and 8 channels for j-time code
A total of 16 channels can be secured. In FIG.
13 shows a correlation characteristic of a pseudo-periodic sequence of a j-fold code. A correlation peak is detected in the correlation with the j-fold code (FIG. 27A), and it can be seen that no correlation value appears in the correlation with the 1-fold code (FIG. 27B). That is, the j-fold code and the 1-fold code can be identified without interference.

(Acceleration of Transmission Rate) In the above study of the transmission rate, transmission of only voice was assumed. However, in consideration of transmitting large data such as images in the future, the faster the transmission rate, the better. As means for speeding up, QPSK (Quadrature Phase)
e Shift Keying (orthogonal PSK) is conceivable. In BPSK, one symbol = 1 bit, whereas in QPSK, one symbol = 2 bits of communication is performed, so that a data rate within 2, that is, 90 kbps is obtained. This value is 64 kb of B channel of ISDN.
It is a good value even when compared with ps.

As described above, a CDMA system using an approximate synchronous CDMA code has been designed. Using the pseudo-periodic approximate synchronous CDMA code, it was shown that the code characteristics were maintained even for data modulation. In addition, a prototype pseudo-periodic code generator was actually manufactured, and the correlation characteristics were observed. Furthermore, the performance of the system was estimated, and it was shown that the cell radius was about 160 m and the data rate was 45 kbps, which was practically sufficient. Especially,
In the uplink within the cell, no synchronization control between the mobile stations 77 is required. As a guideline for further improving the performance of the system, it was shown that the number of channels and the data rate can be easily doubled.

[0117]

As described above, according to the present invention,
In a spread-spectrum wireless communication system that includes a transmission unit and a reception unit and performs transmission by performing spread-spectrum processing on the transmission unit side, a predetermined approximate synchronous CDMA code is used as a code for the spread-spectrum processing. In addition, a spread spectrum wireless communication system having no cross-correlation interference between channels and having a simplified circuit configuration can be realized.

The present invention aims at realizing a portable information wireless terminal "Tele Pad", which is considered to be the mainstream in personal C & C in the 21st century. CDMA system can be easily designed and prototyped.

Further, since an approximate synchronous CDMA code having no inter-channel interference is used as the CDMA code, the CDMA code which has conventionally been a polyphase sequence can be expressed in a small number of phase states. This allows two phases or four
A phase code can be generated, and practical application is facilitated. In addition, the code for each channel can shift the spectral characteristics of the code by allocating a carrier frequency that is slightly displaced for each channel to the code with the reduced number of phase states, thereby eliminating interference between channels. The effect that it becomes possible is obtained.

Further, by realizing a code generator for using the above-mentioned approximate synchronous CDMA code and a correlation circuit using a SAW convolver, a four-phase approximate synchronous CDM is realized.
The code for A is rotated by 45 degrees in the phase space and can be directly controlled by a digital circuit.

Also, C using an approximate synchronous CDMA code
With the realization of the DMA system, pseudo-periodicization of the code is performed as a countermeasure against code characteristic deterioration due to data modulation, thereby enabling communication without inter-channel interference. And
A pseudo-periodic sequence generator was actually created, and the characteristics that were consistent with the theory were obtained. In addition, the system performance was estimated, and a practically sufficient cell radius of about 160 m and a data rate of 45 kb were used.
ps can be realized. In particular, in the uplink, the synchronization control between the mobile stations 77 can be made unnecessary. further,
The doubling of the number of multiplex channels and the data rate can be facilitated.

From the above, the CDMA system using the code for the approximate synchronous CDMA has a sufficient performance as a practical solution of the local CDMA cell technology, and is effective for the development of communication technology.

[Brief description of the drawings]

FIG. 1 is an approximate synchronous CD used in a spread spectrum wireless communication system using an orthogonal sequence having a period of 2 in the present invention.
FIG. 4 is a diagram of a polyphase periodic sequence by frequency shift, showing that any multilevel sequence can be used by shifting the carrier frequency from the orthogonality of the MA code.

FIG. 2 describes a method of expressing a polyphase periodic sequence by frequency shift from the orthogonality of an approximate synchronous CDMA code used in a spread spectrum wireless communication system using an orthogonal sequence having a period 3 different from that of FIG. FIG.

FIG. 3 shows a ZnO / Si type SAW implemented to verify the effectiveness of the spread spectrum wireless communication system of the present invention.
It is a perspective view which roughly explains the structure of the correlation system using a convolver.

FIG. 4 shows ZnO / S according to the first embodiment of the present invention.
FIG. 4 is a block diagram illustrating a circuit configuration of a correlation system configured using an i-type SAW convolver.

5 (a) shows a ZnO / Si type S obtained by performing a correlation process using a correlation system having the circuit configuration shown in FIG. 4;
FIG. 9 is a diagram illustrating a characteristic result when the same code is input to the SAW convolver as a characteristic of the AW convolver with respect to the spreading code and the frequency. (B) As a characteristic of the ZnO / Si type SAW convolver obtained by performing the correlation processing with the correlation system having the circuit configuration shown in FIG.
FIG. 11 is a diagram illustrating a characteristic result when a preferred pair m-sequence is input to the AW convolver.

FIG. 6 is a diagram for explaining an operation of rotating an approximate synchronous CDMA code used in spread spectrum of the present invention by 45 degrees in a phase space so that one state comes to each quadrant.

FIG. 7 is a block diagram illustrating a configuration of a transmission unit of a correlation system using a SAW convolver according to the first embodiment.

FIG. 8 is a block diagram illustrating a configuration of a receiving unit of the correlation system using the SAW convolver according to the first embodiment.

FIG. 9 is a diagram showing a power spectrum of a transmission signal generated from a transmission unit in the correlation processing of the correlation system.

FIG. 10 (a) is a diagram showing theoretical characteristics of autocorrelation characteristics of an approximate synchronous CDMA code in a processing operation experiment of the correlation system. FIG. 4B is a diagram showing measured values of the autocorrelation characteristic of the code for the approximate synchronous CDMA in the processing operation experiment of the correlation system.

FIG. 11A is a diagram showing theoretical characteristics of cross-correlation characteristics of an approximate synchronous CDMA code and a spread code of a signal multiplexed by a minute frequency displacement type multi-channel. FIG. 4B is a diagram illustrating measured values of cross-correlation characteristics of an approximate synchronous CDMA code of a signal multiplexed by a minute frequency displacement type multi-channel.

FIG. 12 is a diagram illustrating a simulation result of a correlation characteristic of an approximate synchronous CDMA code when BPSK is used as a data modulation method and data is “1010...” In the second embodiment.

FIG. 13 is a diagram illustrating a method of creating a pseudo-periodic sequence used in the second embodiment.

FIG. 14 is a diagram showing the configuration of the second embodiment, where N = 128;
The pseudo-periodic sequence of the desired signal among the simulation results of the correlation between the BPSK-modulated signal and the reference pseudo-synchronous CDMA code not subjected to pseudo-periodization is calculated by pseudo-periodicizing the approximate synchronous CDMA code of L = 15. FIG. 9 is a diagram showing a correlation characteristic with the と.

FIG. 15 is a diagram showing a configuration of the second embodiment, where N = 128;
Of the approximation synchronous CDMA code is pseudo-periodized as L = 15, and among the simulation results of the correlation between the BPSK-modulated signal and the reference approximation synchronous CDMA code not subjected to pseudo-periodization, the pseudo period of the undesired signal is FIG. 4 is a diagram illustrating a correlation characteristic with a sequence.

FIG. 16 is a block diagram illustrating a configuration of a transmission unit of a correlation system using a SAW convolver according to the second embodiment.

FIG. 17 is a block diagram illustrating a configuration of a receiving unit of a correlation system using a SAW convolver according to the second embodiment.

FIG. 18 (a) Pseudo-periodic approximated synchronous CDMA obtained using the communication system shown in FIGS. 16 and 17
FIG. 7 is a diagram illustrating theoretical characteristics of a correlation result with respect to a desired signal in a case where only a code that does not multiply data is included in a result of a correlation characteristic of a use code. (B) Correlation result with respect to a desired signal in the case of only a code that does not multiply data among the correlation characteristic results of the pseudo-periodic approximated synchronous CDMA code obtained using the communication system shown in FIGS. 16 and 17 It is a figure showing the measured value of.

FIG. 19 (a) Pseudo-periodic approximate synchronous CDMA obtained using the communication system shown in FIGS. 16 and 17
FIG. 10 is a diagram illustrating theoretical characteristics of a correlation result with respect to another station signal in the case of only a code that does not multiply data among the results of the correlation characteristic of the use code. (B) Correlation with other station signals in the case of only a code that does not multiply data among the results of the correlation characteristics of the pseudo-periodic approximated synchronous CDMA code obtained using the communication systems shown in FIGS. It is a figure showing the actual measurement value of a result.

FIG. 20 (a) Pseudo-periodic approximate synchronous CDMA obtained using the communication system shown in FIGS. 16 and 17
The data among the results of the correlation characteristics of the code for use are “0”, “1”
It is a figure showing the theoretical characteristic of the correlation result with respect to a desired signal when performing modulation alternately. (B) Modulation is performed by alternately setting data to "0" and "1" among the correlation characteristic results of the pseudo-periodic approximate synchronous CDMA code obtained using the communication systems shown in FIGS. FIG. 9 is a diagram showing actual measurement values of a correlation result with respect to a desired signal in the case of performing the above.

FIG. 21 (a) Pseudo-periodic approximate synchronous CDMA obtained using the communication system shown in FIGS. 16 and 17;
The data among the results of the correlation characteristics of the code for use are “0”, “1”
It is a figure showing the theoretical characteristic of the correlation result with respect to the other station signal at the time of performing modulation as alternation. (B) Modulation is performed by alternately setting data to "0" and "1" among the correlation characteristic results of the pseudo-periodic approximate synchronous CDMA code obtained using the communication systems shown in FIGS. FIG. 13 is a diagram illustrating actual measurement values of correlation results with respect to other-station signals in the case of performing the above-described operation.

FIG. 22 is a diagram schematically showing a cellular CDMA system which is an example of a CDMA system (communication mode) using an approximate synchronous CDMA code according to the present invention.

23 is a time chart illustrating an example of a communication procedure between a base station and a mobile station in the cellular CDMA system illustrated in FIG. 22.

24 is a block diagram showing a communication form assumed in a simulation of an uplink signal between a base station and a mobile station in the cellular CDMA system shown in FIG.

25 is a diagram illustrating a correlation result obtained in a simulation of an uplink signal between a base station and a mobile station in the cellular CDMA system illustrated in FIG. 22;

FIG. 26: TDD (Time Division Du)
FIG. 1 is a block diagram illustrating a schematic configuration of a spread spectrum wireless communication system that performs full-duplex communication according to (plex).

FIG. 27 (a) is a diagram illustrating a correlation characteristic between a pseudo-periodic sequence of a j-fold code and a j-fold code. (B) is a diagram showing a correlation characteristic between a pseudo-periodic sequence of a j-fold code and a 1-fold code.

FIG. 28 is a diagram showing a near future communication network configuration expected to reach a practical stage in the near future.

FIG. 29 (a) is a diagram showing, in a spread spectrum communication system which is a premise of the present invention, a model representing a form of a carrier signal obtained by performing general modulation on transmission data by a transmission side control device. It is. (B) In the spread spectrum communication method, FIG. 7 is a diagram illustrating a model of a mode of a secondary modulation signal obtained by further performing spread processing of the channel signals A, B, and C of (a) by a SAW convolver method. . (C) is a diagram modeling and representing a mode in which signals subjected to different modulations are mixed and transmitted on a line in the spread spectrum communication system.

FIG. 30 (a) Among the multiple access communication systems, FM, AM
FIG. 2 is a diagram illustrating an FDMA communication system for performing channel allocation by frequency division used in a communication system. (B) Among multiple access communication systems, a time slot is assigned to a spreading code channel used in a DECT or PHS communication system, and communication is performed using the entire band within that time.
FIG. 2 is a diagram illustrating a DMA communication system. (C) Of the multiple access communication systems, this is a diagram illustrating a CDMA communication system in which all users simultaneously use the entire band and time and divide channels by a high-speed spreading code.

FIG. 31A is a diagram illustrating an m-sequence generation circuit used as a PN code in spread spectrum communication. FIG. 4B is a diagram illustrating an autocorrelation characteristic of an m-sequence having a code length of 127 chips. (C) is a diagram illustrating a cross-correlation characteristic of an m-sequence with a code length of 127 chips.

FIG. 32 (a) is a block diagram illustrating a configuration of a Gold sequence generation circuit used as a PN code in spread spectrum communication. (B) is a diagram for explaining the principle of Gold sequence generation.

FIG. 33 is a block diagram illustrating a configuration of a digital sliding correlator that performs despreading by digital processing when receiving spread spectrum communication.

FIG. 34 is a diagram showing the principle of a digital matched filter used in the digital sliding correlator shown in FIG.

FIG. 35 is a diagram showing the characteristics of the SAW device and the digital correlator and the detection and demodulation operation procedures in comparison, and showing the usefulness of the SAW device.

FIG. 36 is a diagram showing the structure, features, and application fields of a SAW correlator and a SAW convolver in comparison.

[Explanation of symbols]

 Reference Signs List 20 SAW convolver (ZnO / Si type) 27, 30 Oscillator 28, 31 PN code supply unit 29, 32 Multiplier 33 Bandpass filter 37, 44 Code generator 38, 45 Carrier signal generator 39, 41, 46, 48 Multiplication Devices 40, 47 orthogonal component generators 42, 51 adders 43, 52 antennas 49, 50 SAW convolvers 75 base stations 76 cells 77 mobile stations 78 downlinks 79 uplinks

Claims (3)

[Claims] 1. A spread-spectrum wireless communication system comprising: a transmitting unit that performs a spread-spectrum process to transmit a signal; and a receiving unit that performs de-spreading of the received signal to convert the received signal into data. A code generator for generating an approximate synchronous CDMA code including a real part and an imaginary part, a carrier signal generator serving as an output source of _a transmission carrier signal (fa), and a signal for a real part of the approximate synchronous CDMA code , A quadrature component generator for generating a quadrature component of the carrier signal generated by the carrier signal generator, and an approximate synchronous CDMA
A multiplier for multiplying the imaginary part of the carrier code by the orthogonal component of the carrier signal, and an adder for adding the output of the real part multiplier and the output of the imaginary part multiplier. The unit outputs a reference code generator for generating a reference code for the approximate synchronous CDMA, a carrier signal generator serving as an output source of the received carrier signal (f _b ), and a signal for the real part of the approximate synchronous CDMA code. A multiplier for multiplying, a quadrature component generator for generating a quadrature component of the carrier signal generated by the carrier signal generator, and a multiplier for multiplying the imaginary part of the approximate synchronous CDMA code by the quadrature component of the carrier signal. , An adder that adds the output of the real part multiplier and the output of the imaginary part multiplier, and one SAW convolver that receives the output of the adder and performs a correlation process. Now, the code for spread spectrum processing , A period sequence of period N (..., A ₀ , a ₁ ,..., A _N−1 , a ₀ , a ₁ ,
, A _N-1 ,...) Where A is a diagonal matrix and B is a diagonal matrix, A = F ⁻¹ BF where F: DFT matrix is satisfied, and in a periodic sequence, the autocorrelation function is obtained for all terms other than multiples of the period. A sequence that is 0 is an orthogonal sequence, and similarly, A and C are cyclic matrices representing a periodic sequence, and the existence of diagonal matrices B and D such that A = F ⁻¹ BF and C = F ⁻¹ DF The cross-correlation function of the periodic sequence represented by A and C under A spread spectrum wireless communication system using an approximate synchronous CDMA code represented by the following formula: 2. A code having a period of 3 as a code for the spread spectrum processing when the spread spectrum processing is performed on the transmission side.
For the orthogonal sequence (1,1, W ₃ ) of F ₁₂ is a 12th-order DFT matrix. 2. The spread spectrum wireless communication system according to claim 1, wherein an approximate synchronous CDMA code represented by the following formula is used. 3. The transmitting unit further performs spread spectrum processing by a SAW convolver method to spread-modulate a signal and transmit a spread-modulated signal on a line, while the receiving terminal transmits the spread-modulated signal via a SAW device. 3. The spread spectrum wireless communication system according to claim 1, wherein a signal is received and demodulated, and data is transmitted and received.