WO2004068761A1

WO2004068761A1 - A method for coding and applying void time spread spectrum multiple access codes

Info

Publication number: WO2004068761A1
Application number: PCT/CN2003/000116
Authority: WO
Inventors: Yongzhong Zou; Daoben Li; Yongsheng Zhang; Li Fang
Original assignee: Linkair Communications, Inc.
Priority date: 2003-01-31
Filing date: 2003-01-31
Publication date: 2004-08-12
Also published as: CN1640038A; AU2003211295A1

Abstract

The present invention provides a method for coding and applying the void time spread spectrum multiple access codes, by generating or selecting a pair or a plurality pairs of the basic orthogonal complemental code groups, in which each code length is N and width of the zero correlation windows is L; spreading the pair of generated or selected basic orthogonal complemental code groups, and obtaining the void time orthogonal complemental code groups kernel with the intergroup zero correlation windows; spreading the code length and number of the said void time orthogonal complemental code groups kernel with the intergroup zero correlation windows, and obtaining the cross-correlation function with the zero correlation windows between the void time access code groups; then the side lobes of the cross-correlation function don't exist in the zero correlation windows between the access code groups, so the multiple access interference between groups is eliminated; in addition the united testing technique, the interference offset technique and the balance technique can be utilized, so it is possible that the capacity of system is improved. Moreover the present invention solves the complication problem about the application of the united testing in the conventional CDMA system.

Description

Space-time spread-spectrum multi-address code encoding and application method

The present invention relates to the field of spread spectrum and code division multiple access (CDMA) wireless communication technology, and in particular to a spread spectrum multiple access code with space-time diversity characteristics in a wireless communication system. Address code encoding and application method.

BACKGROUND OF THE INVENTION With the advent of the information society and the era of personal communication, people have become more and more urgent to improve the spectral efficiency in wireless communication systems, because frequency resources are very limited. The so-called spectrum efficiency refers to the maximum number of users that can be accommodated by a system in a cell (cel l) or sector (sector) when given the user's transmission rate and system bandwidth. Its measurement unit is per cell (or sector). The total transmission rate supported by the system per unit bandwidth. Obviously, the higher the spectral efficiency, the larger the system capacity.

Traditional wireless multiple access technologies, such as frequency division multiple access (FDMA) and time division multiple access (TDMA), have system capacity limited by the time bandwidth product of the system, and it is impossible to add additional users. For example: The basic transmission rate of a user is 1 / T symbol second, and the bandwidth of the system (including the channel) is B Hertz (Hertz), then the time bandwidth product is BT. BT is the maximum number of users in the system, and one more is impossible. .

Code division multiple access (CDMA) is completely different. Its system capacity is only determined by the signal-to-interference ratio. It has the characteristics of large capacity and soft capacity. Increasing users will only reduce the signal-to-interference ratio and reduce communication quality, and will not be rejected. That is, the system capacity does not have an insurmountable BT value like frequency division multiple access (FDMA) or time division multiple access (TDMA).

The capacity of a code division multiple access (CDMA) system depends on the level of interference in the system. Therefore, whether or not the interference level in the system can be controlled will become the key to the success or failure of a code division multiple access system. Interference can be divided into four major parts: First, the local and internal noise levels, there is no other method except for using low-noise amplifiers; Second, inter-symbol also known as inter-symbol interference (ISI); Third, multiple-access interference ( MAI), that is, interference from other users in the cell; the fourth is adjacent cell or inter-channel interference (ACI). For ISI, MAI, ACI can be reduced or even eliminated by choosing a good address code. In a code division multiple access (CDMA) system, each user has its own unique address code for mutual identification. Not only this, the spreading address codes of each user should also be orthogonal to each other, and this requirement of orthogonality is consistent for any multiple-access system. If the channel is an ideal linear time-frequency non-proliferation system, and there is a strict synchronization relationship within the system, then the orthogonality between user address codes can be ensured. However, none of the actual channels is ideal, and strict synchronization is impossible for signals in time and frequency spread channels. Therefore, it is the life of a code division multiple access (CDMA) system to maintain orthogonality between address codes in a non-ideal time-frequency diffusion channel.

As we all know, mobile communication channels are typical random time-varying channels. There are random frequency spreads (produced by the Doppler effect) and random time spreads (produced by the multipath propagation effect). The former will cause time-selective fading of the received signal, that is, the level of the received signal will have different random fluctuations with time; the latter will cause the frequency of the received signal to be faded, that is, different spectral components of the received signal will have different random fluctuations. Variety. In addition to severely degrading system performance, fading will also significantly reduce system capacity. In particular, the time spread of the channel caused by multipath propagation prevents the signals from reaching the receiving point at the same time, so that signals between adjacent symbols of the same user overlap with each other, resulting in inter-symbol interference (ISI). In addition, the time spread of the channel will also worsen the multiple access interference, because when the relative delay between different user signals is zero, its orthogonality is easily guaranteed, and any orthogonal code can be used. But when the relative delay between signals is not zero, it will become very difficult to maintain their orthogonality.

In order to reduce the inter-symbol interference (ISI), the signal waveform selected by each user, that is, the autocorrelation function of its address code should be an ideal impulse function, that is, it should be zero except for the origin. In order to reduce the multiple access interference (MAI), the signal waveform selected by each user, that is, the cross-correlation function between the address codes should be zero for various relative delays. From the perspective of orthogonality, apart from the relative zero delay, each spreading address code should be orthogonal to any non-zero relative delay, and any relative delay (including (Zero delay) should be orthogonal to each other.

For the sake of clarity, the value of the autocorrelation function at the origin is called the main peak of the correlation function. The value of the autocorrelation or cross-correlation function outside is called the peak of the correlation function. The peaks of the auto-correlation and cross-correlation functions between ideal multiple access codes should all be zero. Unfortunately, the Welch bound of the theory states that there are no multiple access code groups with zeros everywhere in the binary, finite, and even complex domains. In particular, the peaks of the auto-correlation function and the peaks of the cross-correlation function are a contradiction. When one is required to decrease, the other is necessarily increased. In addition, NASA also announced that it has exhaustively calculated various codes and proved that the We 1 ch boundary cannot be broken.

In fact, NASA calculates only the group codes exhaustively, and the Welch bound is only valid for the sub-plural domains. Outside of this, ideal address codes are possible. For example, in 1971, BP Schweizzer of the University of California, Los Angeles (UCLA) in his doctoral dissertation "Generalized Complementary Complementary Code Sets," (Generalized Complementary Complementary Code Sets), has found an encoding that can achieve the performance of ideal address code sets. Then, in 1993, Leppanen, pent ti, and others of European NOKIA MOBILE PHONES LTD .; NOKIA TELECOMMUNICATIONS applied their ideas to a time division / code division (TDMA / CDMA) hybrid system and applied for a European patent. Its patent publication number is EP 0600713A2, and the application number is 93309556. 4. This encoding method is actually encoding in a high-dimensional space. The high-dimensional space has broken through the conditions for the establishment of the Welch world. However, the spectral efficiency of this encoding method is extremely high. It is low and has no practical value, which is why it was proposed that no one has used it for nearly thirty years. Because for a communication system that requires N addresses, this coding method needs to use N ² basic codes and each code needs at least N bits, i.e. total of N ^3-bit addresses supported e.g. N: If N is the number of addresses 128, a 16QAM modulation mode, then the Department

ts / Hz (bit / hertz). It can be seen that the larger the number of addresses, the lower the spectral efficiency of this coding method. However, this encoding method gives a good revelation, that is, a "complementary" method can be used to construct a high-performance address code group, but the total number of code bits required by Dr. BP Schweitzer must be avoided with the number of addresses. The shortcomings of growing cubic.

In addition, if the two-way synchronization technology is used, the relative delays in or between each address code in a random time-varying channel will not exceed the maximum time spread of the channel (at the maximum multipath time). Delay) plus the maximum timing error. Let this amount be Δ seconds. Then, as long as the correlation function and cross-correlation function between address codes in (-Δ, Δ) have no peaks, the inter-symbol interference (ISI) and multiple access interference (MAI) can be guaranteed to be zero. . An address code with this property is called an address code with a "zero correlation window". Obviously, as long as the correlation characteristics of the address code have a "zero correlation window" and the window width is greater than the maximum time spread of the channel (the maximum multipath delay difference) plus the maximum timing error, the corresponding code division multiple access (C ^ MA) system The performance will be ideal, and the fatal "far and near effects" in traditional code division multiple access (CDMA) systems will disappear. The "far-near effect" is caused by the unsatisfactory auto-correlation and cross-correlation characteristics of the address code, because a sub-peak of a close-range signal may drown the main peak of a long-range signal. In order to overcome the "far and near effect", the signals of the users at each address must be made substantially equal in strength when they arrive at the base station. This leads to the need to adopt accurate, complex and fast power control algorithms, thereby complicating the system.

In 1997, Professor Li Daoben proposed a new type of spreading code with zero correlation window in PCT / CN00 / 00028. Assume the maximum time spread of the channel (the maximum multipath delay difference) plus the maximum timing error. Let this amount be △, this codeword guarantees the correlation characteristics within (-Δ, Δ) is ideal. The system using this codeword eliminates ISI and MAI and greatly increases the capacity of the system.

Assume that the complementary code group is {C,., SJ, 1≤ ≤M, the code length of part C or S is N, and the width of the single-sided zero correlation window is Γ, then according to the zero correlation window bound: M≤ ^{2N + 2T} . So given the code length and zero window size,

T + 1

The maximum possible number of codes has been determined, and it is impossible to find more code words.

In order to eliminate the influence of interference, in addition to directly constructing a zero-window code when designing the codeword, another method is to send a non-zero window code at the transmitting end, and use joint detection technology, interference cancellation technology, and equalization technology at the receiving end. To achieve optimal reception. Assuming that there are M code channels in total and using g-ary modulation, the total detection amount when the optimal joint detection is used is

The complexity of this detection method increases exponentially with the number of users M. When the number of users M increases, the receiver becomes powerless, which limits the increase in system capacity. In order to reduce the complexity of the receiver, some sub-optimal joint detection algorithms are used, such as decorrelated multi-user detection, interference cancellation techniques, etc., but these algorithms will bring a loss in performance, and the number of users is very high. Complexity in big time still Of course very big.

Diversity technology is an effective technology to combat fading. The most basic diversity technologies are space diversity, frequency diversity and time diversity. Space-time coding technology (Space-ime Coding) research started in Bell Labs in 1996. It combines space diversity, frequency diversity and time diversity. There are two main types of space-time coding: grid (Trel l is) space-time code and layered space-time code. The existing space-time coding can be regarded as a technology combining spatial diversity, frequency diversity, and time diversity technologies with channel coding, or with forward error correction coding technology.

Summary of the Invention

The object of the present invention is to provide a space-time spread spectrum multi-address code encoding and application method, so that the formed group of space-time spread spectrum multi-address codes has a "zero correlation window" for correlation characteristics between groups, that is, at zero The cross-correlation function between the address codes of each group in the correlation window has no peaks, thereby eliminating multiple access interference (MAI) between groups. Although there is multiple access interference (MAI) between each address code in the same group, but You can use joint detection technology to achieve optimal reception. The space-time spreading code coding method with inter-group zero correlation window characteristics proposed by the present invention. This new space-time spreading code coding with inter-group zero correlation window characteristics combines both diversity technology and zero-correlation window characteristics. In addition, joint detection, interference cancellation technology, and equalization technology can be used, which provides the possibility for increasing system capacity. At the same time, the present invention solves the complexity problem of applying joint detection in the traditional C application A system.

The above-mentioned space-time spread spectrum multi-address code with "zero correlation window" between groups has the following five characteristics:

(1) The cross-correlation function between each group of space-time spread-spectrum address codes has a zero correlation window near the origin. From the perspective of orthogonality, the space-time spread-spectrum address codes are completely orthogonal when the relative delay is smaller than the width of the zero correlation window.

(2) The autocorrelation function of each space-time spread-spectrum address code is not only zero at the two non-zero relative delays except for the origin in the above-mentioned inter-group zero correlation window, that is, it has zero Ideal characteristics.

(3) The cross-correlation function of each space-time spread-spectrum address code in the same group is only zero at the two non-zero relative delays in the above-mentioned inter-group zero correlation window, and is zero at other places. (4) For each of the above-mentioned space-time spread-spectrum address code divisions to transmit on two transmitters

(5) For each of the space-time spread-spectrum address codes described above, the size of the zero correlation window between groups can be increased by inserting a zero guard interval or time slot.

The technical solution of the present invention is:

A space-time spread-spectrum multi-address code encoding method, comprising the following steps: generating or selecting one or more pairs of basic orthogonal complementary code groups whose code lengths are each N zero correlation window width is L;

Extending the generated or selected pair of basic orthogonal complementary code groups to obtain a space-time orthogonal complementary code group kernel with a zero correlation window between the groups;

The code of the space-time orthogonal complementary code group core with zero correlation window between groups is extended by code length and number of codes, and the cross-correlation function between each group of space-time address codes obtained after expansion has a zero correlation window.

Processing of inserting a zero guard interval or a time slot on the space-time orthogonal complementary code group core with an inter-group zero correlation window may be performed to make the group-to-group zero correlation between each group of space-time spread-spectrum address codes formed after processing The size of the window is increased.

The selection of a pair of basic orthogonal complementary code groups where each code length is N zero correlation window width is L includes: can be set: (CS), (C, ₂ , S, ₂ ) as the basic orthogonal Complementary code set, and:

C ₂₂ "'C _2N ),

S ₂₂ * " ^E S _2N ),

The non-periodic autocorrelation and cross-correlation functions of c code and s code are oppositely formed except for the origin within the zero correlation window, and the values of the autocorrelation function and the cross-correlation function after addition are zero except for the origin.

The extending the pair of basic orthogonal complementary code groups includes: according to the maximum number of user addresses actually required, performing the code length and code number of the orthogonal complementary code group core in the spanning tree structure. Extended, the cross-correlation function between the extended groups of space-time spread-spectrum address codes has an objective window near the origin, and its window is less than or equal to 2L-1; according to the orthogonality, it can be obtained that The time-spreading address codes are completely orthogonal when the relative delay is less than the width of the zero correlation window; the autocorrelation function of each space-time spreading address code after expansion is divided by the original zero correlation window between the groups. Outside the point, only the two non-zero relative delays are not zero, and the rest are zero; the cross-correlation function of each space-time spread-spectrum address code in the same group is only two in the above-mentioned inter-group zero correlation window. Non-zero relative delays are not zero, and zero everywhere else.

The expanding the basic orthogonal complementary code group pair includes: expanding the selected basic orthogonal complementary code group pair (C, S), (C, ₂ , s, ₂ ), where:

C ₂₂ '"C _2N ),

S, i ⁼ (S _n S ₁₂ '"S _1N ), S, ₂ = (S ₂₁ S ₂₂ "' S _2N );

The obtained space-time orthogonal complementary code group kernel with zero correlation window between groups is:

The processing of inserting a zero guard interval or a time slot into the group core of space-time orthogonal complementary codes with zero correlation windows between groups or the extended space-time spreading address codes includes: The space-time orthogonal complementary code core with zero correlation window between groups or the extended space-time spread-spectrum address codes are inserted into the zero guard interval or time slot to increase the space between the groups of space-time spread-spectrum address codes formed after expansion. The size of the zero correlation window between groups.

When calculating the space-time spread-spectrum address code, it must be ensured that the C code operates only with the C code (including itself and other codes), and the S code operates with only the S code (including itself and other codes).

The pair of basic orthogonal complementary code groups (Cp S, (C ₂ , S ₂ ) refers to: their auto-correlation and cross-correlation functions are aperiodic auto-correlation between C codes and cross-correlation functions between S codes. The sum of non-periodic autocorrelation and cross-correlation functions, where within a zero correlation window, the non-periodic autocorrelation and cross-correlation functions of c code and S code are opposite to each other except the origin, and the added autocorrelation function value and cross-correlation function value It is zero everywhere except the origin.

The space-time orthogonal complementary code group kernel with inter-group zero correlation window refers to: the length of each code of the space-time orthogonal complementary code group core formed after expansion is 2N, and the width of the zero-correlation window between groups is greater than Or equal to 2L-1. The insertion of a zero guard interval or time slot into the space-time orthogonal complementary code group core of the zero-correlation window between groups refers to: firstly, the basic orthogonal complementary codes with the length N of each code and the width L of the zero-correlation window window L Group pair (C ^? _1? ) (C, ₂ , S, ₂ ), C ' ₁ = (C ₁₁ C ^ ^ -C ^), C, ₂ = (C ₂₁ C ₂₂ "· _2N ), S ⁹ ^ (S ₁₃ S ₁₂ ～ S _1N ;), S ' ₂ = (S ₂₁ S ₂₂ ... S _2N ) is expanded into a space with the code length of 2N and the width of the zero correlation window between the groups of 2L-1. Time orthogonal complementary code group kernel,

-C _2I C ₂₂ -C ₂₂ '.'C _2N -S _2t S ₂₂ -S ₂₂ ... s.

Then, a certain number of zero guard intervals or time slots can be inserted into the above-mentioned space-time orthogonal complementary code group cores, and the new space-time orthogonal complementary code group cores formed therefrom have an inter-group zero correlation window width greater than or equal to the original The inter-group zero correlation window width of the space-time orthogonal complementary code group kernel.

The insertion of a zero guard interval or time slot into the space-time orthogonal complementary code group core of the zero correlation window between groups refers to: inserting T zeros per L + 1 chips (chips), thereby forming a new space The width of the zero-correlation window between the groups of time orthogonal complementary code group kernels is greater than or equal to 2L-1, so that the new space-time orthogonal complementary code group kernels are continuously expanded according to the tree structure, and the space-time orthogonal complementary code groups obtained are The width of the zero correlation window between pairs is greater than or equal to 2L-1:

The expansion of the code length and code number of the space-time orthogonal complementary code group core in the spanning tree structure refers to: if (C _pt ), (C ₂ , S ₂ ) is a pair, each code length is N , The space-time orthogonal complementary code group kernel with a width L of zero correlation window window L can generate two pairs, that is, four groups of space-time orthogonal complementary code group pairs each having a code length of 2N in the following manner:

(C _v c ₂ , s _i; s ₂ )

Among them, the cross-correlation function between two pairs of space-time orthogonal complementary code group pairs formed by the two branches above and below the expansion has a zero correlation window near the origin, and the window width is greater than or equal to L. We can treat the two orthogonal complementary code pairs formed by the two branches above and below as two pairs of code lengths

2N. The orthogonal complementary code group kernel whose width of the zero correlation window window is greater than or equal to L is further expanded to obtain four pairs of space-time orthogonal complementary code group pairs. There is a zero correlation between the cross-correlation functions between the groups near the origin. Window whose window width is greater than or equal to

The expansion can be continued according to the spanning tree structure to generate 2 ⁿ pairs of space-time orthogonal complementary code group pairs with a coding length of N 2 ⁿ and a width of the zero correlation window between groups greater than or equal to L, where n = 0, 1 , 2, .. ·, is the number of expansions.

Said pair of extended space-time spreading address codes with inter-group zero correlation window can be inserted into the zero guard interval or time slot to refer to: For inter-group zeros generated by space-time orthogonal complementary code group core extension Each space-time spread-spectrum address code of the correlation window is inserted into a certain number of zero guard intervals or time slots, and the new space-time orthogonal complementary code group formed from this group has a zero-correlation window width greater than or equal to the original space-time positive Inter-group zero correlation window width of the complementary code group.

If each of the extended groups of space-time spread-spectrum address codes has a zero correlation window with an inter-group width of 2W-1, then every W chips (Chip) of each of the extended space-time spread-spectrum address codes (Chip) ) Insert T zeros; the width of the inter-group zero correlation window of the new space-time orthogonal complementary code group thus formed is greater than or equal to 21. The insertion of T zeros needs to satisfy the following conditions: that the width of the zero correlation window between groups of the new space-time orthogonal complementary code group to be formed is maximized.

The inserting of the T zeros includes: inserting T zeros at the tail of each L + 1 chip (Chip), inserting T zeros at the head of each L + 1 chip (Chip), and the like. The present invention also provides a space-time spreading multiple address code application method, which includes the following steps: determining the required according to the propagation conditions of the applied system, the basic spreading code rate used by the system, and the maximum timing error in the system. The width of the zero correlation window between groups;

Generate or select one or more pairs of basically orthogonal complementary code group pairs according to the width of the correlation window between the required zero groups;

Extending the basic orthogonal complementary code group pair to generate a space-time orthogonal complementary code group kernel with a zero correlation window between the groups;

According to the actual number of users, determine the maximum number of user addresses required, and use the selected space-time orthogonal complementary code group kernel with zero correlation window between groups as the origin, and expand the code length and number of codes in the spanning tree structure. , The cross-correlation function between the extended groups of space-time spread-spectrum address codes has a zero correlation window between the groups near the origin;

The groups of space-time spread-spectrum address codes with zero correlation windows between the groups after spreading are spread-modulated and transmitted on corresponding transmitters.

The generating and selecting a pair of basic orthogonal complementary code groups includes: selecting a pair of basic orthogonal complementary code groups (d, S,), (C ₂ , each code length N and zero correlation window window width L) S ₂ ); The auto-correlation and cross-correlation functions of (, S _t ) and (C ₂ , S ₂ ) are aperiodic auto-correlation between C codes and aperiodic auto-correlation between cross-correlation function and S code, respectively And the cross-correlation function, where within a zero correlation window of width L, the aperiodic autocorrelation of the C code and the S code is opposite to the cross-correlation function except the origin, and the value of the added autocorrelation function and the cross-correlation function It is zero everywhere except the origin.

The generating and selecting the basic orthogonal complementary code group pair includes: the basic orthogonal complementary code group pair (C _p S, (C ₂ , S ₂ )) can be generated by using a spanning tree structure. The step of expanding the basic orthogonal complementary code group pair to generate a space-time orthogonal complementary code group kernel with zero correlation windows between the groups includes: obtaining each code length N and the width of the zero correlation window window. The pair of basic orthogonal complementary code groups ((^, Si), (C ₂ , S ₂ ) for L) is expanded into a space-time orthogonal complementary code group kernel with a zero correlation window between groups as follows, where: CfCCu CfC ^ :), C ₂ = (C ₂₁

The length of each code of the extended space-time orthogonal complementary code group kernel is 2N, and the width of the zero correlation window between groups is greater than or equal to 2L-1.

According to the method of the present invention, a certain number of zero guard intervals or time slots are inserted into the space-time orthogonal complementary code group core,

The width of the close window is larger than the width of the zero correlation window between the original space-time orthogonal complementary code group kernels.

According to the method of the present invention, if the length of each code of the basic orthogonal complementary code group pair (, S,) and (C ₂ , S ₂ ) is N and the width of the zero correlation window is L, then the extension Each subsequent code length is 2N space-time orthogonal complementary code group cores, and every L + 1 chips (Chip) is inserted with T zeros; wherein, each L + 1 chip may be inserted at the tail of each chip (Chip) T zeros, and the width of the inter-group zero correlation window of the new space-time orthogonal complementary code group core formed by this is equal to or equal to 2L-1, so that the new space-time orthogonal complementary code group core continues according to the spanning tree structure By extension, the width of the inter-group zero correlation window of the space-time orthogonal complementary code group pair obtained is greater than or equal to 2L-1.

The insertion of T zeros needs to satisfy the following conditions, that is, the new space-time positive interaction is formed so that the width of the zero correlation window between groups of the group kernel is maximized.

The inserting T zeros includes: inserting T zeros at the tail of each L + 1 chip (Chip), inserting T zeros at the head of each L + 1 chip (Chip), and so on.

The expansion will jointly determine the number of expansion stages required in the spanning tree according to the required maximum number of users and the number of selected pairs of basic orthogonal complementary code groups. The spreading and modulation of each group of space-time spread-spectrum address codes with zero-correlation windows between groups on the corresponding transmitters includes: Corresponding to each group with zero-correlation windows between groups The same two transmitters.

The step of spreading the space-time spread-spectrum address codes of the extended groups with zero correlation windows between groups on the corresponding transmitters respectively includes: Corresponds to two different transmitters.

The spreading and modulation of each group of space-time spread-spectrum address codes with zero correlation windows between groups on a corresponding transmitter includes: When the following two sets of space-time orthogonal complementary codes are used:

_{_{_{S N S 12 S 12 • SIN}}} S 1N

"C _u C ₁₂ -C ₁₂ ... C _1N -C-S _N S ₁₂ -S ₁₂ , •• SIN _ S _1N

Each code division of the two sets of space-time orthogonal complementary code groups is transmitted on two transmitters, and the same two transmitters can be corresponded between the groups and the groups, or the two transmitters of the different two groups can be respectively corresponded to. The corresponding relationship between two space-time orthogonal complementary codes and two transmitters in the same group is as follows: For all the odd digital chips (Chips) of the first space-time orthogonal complementary code, they correspond to the first transmitter Corresponding odd-numbered chip (Chip) of the second transmitter, and all odd-numbered chip (Chip) of the second space-time orthogonal complementary code. Corresponding to the corresponding odd digital chips transmitted on the second transmitter, all the even digital chips (chips) corresponding to the corresponding even digital chips on the first transmitter are transmitted; or vice versa, that is, for the second space-time orthogonal All the odd-numbered chips (Chips) of the complementary code correspond to the corresponding odd-numbered chips that are transmitted on one transmitter, and all the even-numbered chips (Chips) correspond to the corresponding even-numbered chips that are transmitted on the second transmitter. Space-time orthogonal complement All the odd digital chips (Chip) of the code correspond to the corresponding odd digital chips transmitted on the second transmitter, and all the even digital chips (Chip) It should be transmitted on the corresponding even digital chip of the first transmitter.

The space-time orthogonal refers to: when each space-time spread spectrum address code of the two sets of space-time orthogonal complementary code groups is transmitted on two transmitters to form the following two sets of four space-time sequences in total , These two sets of space-time sequences remain orthogonal between groups and have zero correlation windows between groups.

among them:

t = a cos (2¾T ^ +, a ₂ = a ₂ cos (27 _c t + φ ₂ )? a ₃ = ₃ cos (27 _c t + φ _ζ , a ₄ = ₄ 008 (2 ^^ + ^ ₄ ) ? a _{1? 25 3 5 4} 6 [0 ₃ + oo) _? φ _ι , φ ₂ , φ _{3 7} φ ₄ € [0 ». Equivalent transformation can be performed on the generated multi-address code.

The equivalent transformation may include: exchanging positions of C and S codes, exchanging positions of C1 and C2 and S1 and S2 at the same time, inverting code sequence, inverting each code bit, interleaving the polarity of each code bit, Rotate and change each code position in the plane, change in the spanning tree, and so on.

The rotation of each code position in the complex plane includes: The basic complementary code group can be sequentially rotated by "degrees", and the properties of the autocorrelation function and cross-correlation function of each address code are unchanged after the rotation transformation. However, the peaks outside the "zero-correlation window" are related to the rotation angle; by properly selecting the rotation angle, the rotated code groups can be orthogonal, and multiple orthogonal codes can be generated from a set of orthogonal codes.

The generated multiple sets of orthogonal codes are suitable for networking requirements, handover requirements, capacity requirements, etc .; especially when the code length is long.

The orthogonal complementary code must ensure that the C code operates only with the C code (including itself and other codes), and the S code operates only with the S code (including itself and other codes). In practical applications, C code and S code should be separated.

The separation measures may include: modulating the C code and the S code on orthogonal polarized waves, respectively; or placing the c code and the S code in two time slots that do not overlap each other after transmission. .

In the method of the present invention, the transmission channel changes randomly with time. In order to ensure the realization of complementarity, the channel characteristics in the two polarized waves and the two time slots should be consistent during the transmission process, that is, their The fading should be synchronized; this requires that when using polarization separation, a frequency band that can guarantee the simultaneous polarization of orthogonal polarization waves, no depolarization, and corresponding measures must be used. When using time-division separation, the time between two time slots must be maintained. The interval is much smaller than the correlation time of the channel. When other separation methods are used, their synchronization fading must be guaranteed.

In the method of the present invention, since the C code and the S code should be transmitted separately and their complementarity should be used, the information bits modulated on them should be the same, and the C code and the S code should be despread and demodulated. The subsequent outputs should be added.

The beneficial effect of the present invention is that, by providing a space-time spread spectrum multi-address code encoding and application method, the correlation characteristics between the formed space-time spread spectrum multi-address code group and the group have a "zero correlation window", That is, the cross-correlation function between groups of address codes in the zero correlation window does not have peaks, thereby eliminating multiple access interference (MAI) between groups, while there is multiple access interference (MAI) between each address code in the same group. ), But joint detection techniques can be used to achieve optimal reception. The space-time spreading code coding method with inter-group zero correlation window characteristics proposed by the present invention. This new space-time spreading code coding with inter-group zero correlation window characteristics combines both diversity technology and zero-correlation window characteristics. In addition, joint detection, interference cancellation technology, and equalization technology can be used, which provides the possibility for increasing system capacity. At the same time, the invention solves the complexity problem of applying joint detection in the traditional CDMA system. The above-mentioned space-time spreading multiple address codes with "zero correlation windows" between groups have the following five characteristics: (1) A cross-correlation function between each group of space-time spreading address codes has a zero correlation window near the origin. From the perspective of orthogonality, the space-time spread-spectrum address codes are completely orthogonal when the relative delay is smaller than the width of the zero correlation window. (2) The autocorrelation function of each space-time spread-spectrum address code is not only zero at the two non-zero relative delays except for the origin within the above-mentioned inter-group zero correlation window, That is, it has ideal characteristics. (3) The cross-correlation function of each space-time spread-spectrum address code in the same group is only zero at the two non-zero relative delays in the above-mentioned inter-group zero correlation window, and is zero at other places. (4) For each of the above-mentioned space-time spread-spectrum spreading address codes transmitted on two transmitters; (5) For each of the above-mentioned space-time spread-spectrum spreading address codes, a zero guard interval or time slot can be inserted to increase the inter-group zero The size of the relevant window.

BRIEF DESCRIPTION OF THE DRAWINGS

Fig. 1 is the first "^^^" of the space-time orthogonal complementarity of the present invention with closed windows between groups.

FIG. 2 is a second schematic diagram of the space-time orthogonal complementarity with inter-group window closing in the present invention.

FIG. 3 is a kind of ^^^ of the orthogonal complementary ^ ¾a ^ of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below through embodiments and drawings.

The first step is the generation and selection of the basic orthogonal complementary code pair.

The pair of basic orthogonal complementary code groups (Cp Si (C ₂ , S ₂ ) in which each code length is N and the width of the zero correlation window is L according to the technical solution of the present invention refer to: their autocorrelation and crosscorrelation functions are respectively Is the sum of the non-periodic autocorrelation and cross-correlation functions between the C code and the S-code, and the non-periodic autocorrelation between the C code and the S code in a zero correlation window of width L It is opposite to the cross-correlation function except for the origin, and the added value of the auto-correlation function and the cross-correlation function are zero except for the origin. The pair of basic orthogonal complementary code groups (C, Si) (C ₂ , S ₂ ) One kind of generating method can use the basic orthogonal complementary code group pair generating method of Professor Li Daoben in PCT / CN00 / 00028.

It should be noted that for orthogonal complementary codes, when performing correlation or matching filtering operations on them, the C code only operates on the C code, the S code only operates on the S code, and the C code and S code do not meet during the operation.

Please refer to FIG. 3, which is a spanning tree diagram of a basic complementary code group pair. In the specific multi-address coding process, the basic orthogonal complementary code group pair in FIG. 3 can be used. In the figure, each pair of code groups in <> is a basic orthogonal complementary code group pair, and their complementary autocorrelation functions and cross-correlation functions have no peaks at all, that is, they have completely ideal characteristics. It should be noted that only the generated in Figure 3 It is only a basic complementary code group pair, and there are many equivalent forms, for example, swapping their order up and down, or left and right, reversing their order before and after, reversing the spacing, and rotating in the complex plane. Can get the equivalent bowl complementary code. Their autocorrelation and Λguanxian are also 4 * thought.

The second step is to generate a space-time orthogonal complementary code group kernel with a zero correlation window between groups.

For the pair of basic orthogonal 'complementary code pairs (CS (C, ₂ , S, ₂ )) where each code length is N and the width of the zero correlation window window is L (CS (C, ₂ , S, ₂ )), 0 第一步 = (C _n C ₁₂ .. .C _1N ), C, ₂ = (C ₂₁ C ₂₂ ... C _2N ), S = (S _n S ₁₂ ... S _1N ), s, ₂ = (s ₂₁ S ₂₂ ... S _2N ) Expand into a space-time orthogonal complementary code group kernel with inter-group zero correlation window as follows:

i

The space-time orthogonal means that each space-time spread-spectrum address code division of the above-mentioned space-time orthogonal complementary code group core is transmitted on two transmitters to form the following two sets of four space-time sequences. Group space-time sequences remain orthogonal and have zero correlation windows between groups

Where a _x = _x cos (2 / _c t + φ _χ ) _{7 2} = a ₂ cos (27 _c t + φ ₂ ) ₃ a ₃ = ₃ cos (2¾T _c i + φ ₃ ) ₃ a ₄ = a cos ( 27 _c t + φ ₄ ), α _λ , α ₂ , α, α e [0, + oo), cpm ≡ [0 ». We can insert a certain number of zero guard intervals or time slots into the above-mentioned space-time orthogonal complementary code group cores. The new space-time orthogonal complementary code group cores have a width of the zero correlation window greater than or equal to the original The inter-group zero correlation window width of the space-time orthogonal complementary code group kernel. If the code length of the selected pair of basic orthogonal complementary code groups (C _1? Si), (C ₂ , S ₂ ) in the first step is N and the width of the zero correlation window window is L, then we can Each L + 1 chip (Chip) of the 2N space-time orthogonal complementary code group core after the extended code length is inserted with T zeros. According to the following zero guard interval or time slot insertion method: T zeros are inserted at the end of each L + 1 chip and chip, and the new space-time orthogonal complementary code group core formed by the group has zero correlation. The window width is greater than or equal to 2L-1, so that the new space-time orthogonal complementary code group kernel continues to expand according to the tree structure of the third step, and the width of the zero-correlation window between groups of the space-time orthogonal complementary code group pair is obtained. Greater than or equal to 2L-1. The criterion for inserting the T zeros is to maximize the width of the zero correlation window between the new space-time orthogonal complementary code group kernels formed by the method. There are many methods for inserting the T zeros, for example, inserting every L The tail of +1 chips (chip) is inserted at the head of every L + 1 chips (chip), and will not be listed here.

For example, we select the following basic orthogonal complementary code groups whose code length is 1 and the zero correlation window window width is 3:

(C, S) = (++, +-) and (C, ₂ , S, ₂ ) = (-+, ―)

Then the space-time orthogonal complementary code group kernel with zero correlation window between groups generated according to the above generation method is:

_ / —— + + \ „I one one one one one \

C) S

\-+ +-/ \-+-+ / The third step is to expand the code length and number of codes of the space-time orthogonal complementary code group core with zero correlation window between the groups in the second step. The cross-correlation function between space-time spread-spectrum address codes has a zero correlation window near the origin.

Two or two pairs of space-time orthogonal complementary code groups can be generated in the following manner, that is, four space-time orthogonal complementary codes each having a code length of 2N: (C, C ₂ , S, S ₂ )

(C _2- , S ₂ -S _t )

Because a pair of space-time orthogonal complementary code groups can obtain two new pairs of two space-time orthogonal complementary codes, but the length of each code is doubled. Two new pairs or two groups of four new space-free orthogonal codes can be obtained. The time-orthogonal complementary code can be derived from four pairs or four groups of eight new space-time orthogonal complementary code groups, and then, eight pairs or eight groups of sixteen space-time orthogonal complementary code groups can be generated. There is a zero correlation window in the cross-correlation function between code groups, and its width is related to the original basic complementary code group. This process can be described by a lifetime spanning tree graph relationship. Figure 1 is one such spanning tree graph, and Figure 2 is another spanning tree. There are many other types of spanning trees, and the relationship between them is equivalent transformation. The equivalent transformation does not change the width of the zero correlation window between the groups, but it can sometimes change the height and distribution of the peaks outside the zero correlation window. For example, we select the following kernels with 4 space-time orthogonal complementary code lengths:

After performing an expansion according to the above expansion method, we obtain two pairs, that is, four sets of space-time orthogonal complementary code groups each with a code length of 8:

+ —— + one + For the two pairs generated above, that is, the four complementary codes of the four space-time orthogonal complementary code groups each having a code length of 8, the numbers are renumbered as follows:

(CI, S1) = (+ + + +--+ +, + +);

(C2, S2) = (+-+--+ +-, +--+-+-+);

(C3, S3) = (+ + + + + +--, + + —— + + + +);

(C4, S4) = (+-+-+ _ 一 +, + —— + + — + —);

(C5, S5) = (—— + + + + + +, + + ——);

(C6, S6) = (-+ +-+-+-, one +-+ + —— +);

(C7, S7) = (--+ +, + +);

(C8, S8) = (-+ + —a +-+,-+-+-+ +-);

When the eight complementary codes of the four groups of space-time orthogonal complementary code groups each having a code length of 8 are respectively formed in the four groups of two transmitters each forming the following four groups of a total of 8 space-time sequences, the four groups The sequence remains orthogonal and has a zero correlation window between groups:

(C, l, S'l) = ία _λ α ₂ α _χ α ₂ ~ α _χ ― α ₂ α α ₂ ? Α _χ α ₂ —α _χ —α ₂ —α —α ₂ - _χ ~ α ₂ ); (C, 2, S'2) = (α ₂ ~ α _λ α ₂ -α _χ —α ₂ α _λ α ₂ -α _{ ? Α ₂ ~ α _χ -α ₂ α _χ -α ₂ α _λ - ₂ α _λ );

(C, 3, S, 3) = (α ₃ ^α 4 ^α 3 ^α 4 ^α 3 ^α 4 ~ ^α 3 ~ ^{α 3 α} 3 ^β 4 ~ ^β 3 ~ ^U 4 ^β 3 ^{α fl} 3 ^Ω 4 ^

^{(C'4, S 3 4) ~} (α 4 ~ α 3 α 4 - 3 α 4 -α 3 ~ α 4 3, α 4 -α - 4 α 3 α ~ α 3 α 4 -α 3); ( C, 5, S'5) = ( _-5 — ₆ α ₅ α ₆ α ₅ α ₆ α ₅ α ₆ , -α ₅ -α ₆ -α ₅ —α ₆ α ₅ α ₆ -α ₅ — α ₆ ) ; (6, S'6) = (~ α _β α ₅ α _β ~ α ₅ α ₆ —α ₅ α ₆ ~ α ₅ , —α ₆ α ₅ —α ₆ α ₅ α ₆ -α ₅ —α ₆ α ₅ ); (C7, S7) = (-α _Ί —α ₈ α _Ί α ₈ -α _Ί — ₈ ~ α _Ί -α _% , one α _Ί — α & —α _Ί - ₈ -α _Ί — ₈ α _Ί α _Β ); (C, 8, S ⁵ 8) = (— ₈ α _Ί α — α _Ί -α ₈ α _Ί —α _Β α _Ί , —α _ζ α _Ί — ₈ α _Ί —α _% α _Ί α —α ,); Where = α _{ cos (27f _c t + φ.), Α · e [0 ₃ + οο) _? ^. E [0 ₇ 2π), 2, 3, 4, 5, 6, 7, 8, 8. It can be proven that the cross-correlation functions of the four groups of 8 space-time orthogonal complementary sequences are arbitrary for any of: ^ [0, + οο), _e [0, 2r) (ί = 1, 2, 3, 4, 5, 6, 7 _? 8) There is a zero correlation window near the origin, and the width of the window is greater than or equal to 5; the autocorrelation function of each spreading address code after the expansion is within the above-mentioned inter-group zero correlation window except the origin and The relative shift τ is 1 and −1, and it is zero everywhere. The cross-correlation function of two spread-spectrum address codes in the same group is zero-phased between the above groups. Only two relative shifts in the off-window are 1 and -1, which are not zero, and zero in other places.

In a fourth step, a new space-time orthogonal complementary code formed by periodically inserting a certain number of zero guard intervals or time slots for each space-time spread-spectrum address code with inter-group zero correlation window can be formed. The width of the zero correlation window between groups is larger than the width of the zero correlation window between the original space-time orthogonal complementary code groups.

If the inter-group zero correlation window width of the space-time orthogonal complementary code group obtained in the third step is

2L-1, then we can insert T zeros per L + 1 chips, and the new space-time orthogonal complementary code group formed by this group has a width of the zero correlation window of 2L-1. The criterion for inserting the T zeros is to maximize the width of the zero correlation window between the new space-time orthogonal complementary code groups thus formed. There are many methods for inserting the T zeros, for example, inserting every L + The tail of a chip is inserted at the head of every L + 1 chip. It will not be listed here.

For example, for the two pairs generated in the third step, that is, the eight complementary codes of the four space-time orthogonal complementary code groups each having a code length of 8:

(Cl, SI) = (+ + + +--+ +, + +);

(C2, S2) = (+-+--+ +-, +--+-+-+);

(C3, S3) = (+ + + + + + ——, + + —— + + + +);

(C4, S4) = (+-+-+ —— +, + —— + +-+-);

(C5, S5) = (—— + + + + + +, + + ——);

(C6, S6) = (-+ +-+-+-,-+-+ +-+);

(C7, S7) = (--+ +, + +);

(C8, S8) = (-+ +--+-+,-+-+-+ +-);

The width of the zero correlation window between the four groups of code length 8 space-time orthogonal complementary code groups is 5, if we insert 1 zero at the end of every 4 chips, a new two pairs are generated. That is to say, the eight complementary codes of the four groups of space-time orthogonal complementary code groups each having a code length of 10 (the effective length is zero if zero is not considered) are now renumbered and arranged as follows:

(Cl, Sl) = (+ + + + 0--+ + 0, + +--0 0);

(C2, S2) = (+-+-0-+ +-0, +--+ 0-+-+ 0);

(C3, S3) = (+ + + + 0 + +--0, + +--0 + + + + 0); (C4, S4) = (+-+-0 + —— + 0 + —— + 0 +-+-0);

(C5, S5) = (--+ + 0 + + + + 0 0 + +-0);

(C6, S6) = (-+ +-0 +-+-0-+-+ 0 + —— + 0);

(C7, S7) = (--+ + 0 0 0 —— + + 0);

(C8, S8) = (-+ +-0-+-+ 0-+-+ 0-+ +-0);

The eight complementary codes of the four groups of space-time orthogonal complementary code groups each with a code length of 10 (if no zero is considered, the effective length is 8), the space-time orthogonal of each code length of 4 can also be first Complementary code group core:

For this pair of space-time orthogonal complementary code group kernels, we can insert 1 zero every 4 chips, and insert the zero guard interval or time slot as follows: Insert 1 at the end of every 4 chips The number of zeros, and the new space-time orthogonal complementary code group kernel formed by this group of zero correlation window width is 7, as shown below:

Therefore, the new space-time orthogonal complementary code group kernel is expanded once according to the tree structure of the third step, and the obtained orthogonal complementary code group pair is obviously complementary to the above-mentioned four groups of space-time orthogonal complementary codes each having a length of 10. The 8 complementary codes of the code group are exactly the same. That is to say, we can insert the space-time spreading address code into the orthogonal complementary code group kernel and insert a zero guard interval or time slot to form a new space-time orthogonal complementary code group. Inter-group zero correlation window width of the original space-time orthogonal complementary code group.

It can be verified that the cross-correlation functions of these four groups of 8 space-time orthogonal complementary codes exist near the origin A zero correlation window, the width of which is greater than or equal to 7; the autocorrelation function of each space-time spread spectrum address code after expansion is in the above-mentioned group of zero correlation windows except the origin and the relative shifts τ are 1 and-1 Xibu, it is zero everywhere; the cross-correlation function of two space-time spread-spectrum address codes in the same group is only zero at two relative shifts τ of 1 and -1 in the zero correlation window between the groups. It is zero everywhere.

The following describes the generation process of the spread spectrum address code of the present invention:

First, determine the required width of the zero correlation window based on the propagation conditions of the applied system, the basic spreading code rate used by the system (referred to as the Chirp rate in engineering, in terms of MCPS), and the maximum timing error in the system. .

In the second step, one or more basic orthogonal complementary code group pairs are generated or selected according to the width of the required zero correlation window.

The pair of basic orthogonal complementary code groups (C _p Si (C ₂ , S ₂ ) with each code length N and the width of the zero correlation window window L according to the technical solution of the present invention refers to: its autocorrelation and crosscorrelation functions The non-periodic autocorrelation and cross-correlation function between the C code and the S-code are the sum of the non-periodic autocorrelation and cross-correlation function between the S code, where the non-periodic autocorrelation between the C code and the S code is within a zero correlation window of width L. Correlation and cross-correlation functions are opposite to each other except the origin, and the added value of the auto-correlation function and the cross-correlation function are zero except for the origin. The pair of basic orthogonal complementary code groups (C ₁₅ Si), (C ₂ , S ₂ ) A generation method can use the generation method of the basic orthogonal complementary code pair of Professor Li Daoben in PCT / CN00 / 00028. For example, the width of the zero correlation window greater than or equal to the required width can be selected from FIG. 3 Any one or more pairs of basic orthogonal complementary code group pairs are used as the basic orthogonal complementary code group pair.

In the third step, a pair of basic orthogonal complementary code groups (C ₁ (C ₂ , S ₂ )) having a code length of N and a zero correlation window window width of L obtained in the second step,

C ₂₂ ... C _2N ), S ₁ = (SS ₁₂ ... S _1H ), S ₂ = (S ₂₁ S ₂₂ ... S _2N ) is expanded to space-time with zero correlation window between groups as follows Orthogonal complementary code group kernel:

Q

-CJJ C ₁₂ -C _{] 2} ... C _M -C _IN /-S ₁₂ ... S _1N S ₂₁ S ₂₂ S ₂₂ ,

2 • C2N ** C _2N -S ₂₁ S ₂₂ -S ₂₂ ... s 2N ■ s 2N

We can insert a certain number of zero guard intervals or time slots into the above-mentioned space-time orthogonal complementary code group cores, and the new space-time orthogonal complementary code group cores have a larger width of the zero correlation window than the original space. The inter-group zero correlation window width of the time orthogonal complementary code group kernel. If the code length of the basic orthogonal complementary code pair in the second step (C ₁₅ Si). (C ₂ , S ₂ ) is N and the width of the zero correlation window is L, then we can Each code length is 2N space-time orthogonal complementary code group core. Each L + 1 chip (Chip) inserts T zeros, according to the following zero guard interval or time slot insertion method: Each L + 1 chip number The T (zero) is inserted at the end of the chip, and the new space-time orthogonal complementary code group kernel formed by the new space-time orthogonal complementary code group kernel has a width greater than or equal to 2L-1, and thus the new space-time orthogonal complementary code group kernel. As the tree structure shown in FIG. 1 continues to expand, the width of the zero correlation window between the obtained space-time orthogonal complementary code group pairs is greater than or equal to 2L-1. The criterion for inserting the T zeros is to maximize the width of the zero correlation window between the new space-time orthogonal complementary code group kernels formed by this. There are many methods for inserting the T zeros, for example, inserting every L + 1 The tail of each chip is inserted at the head of every L + 1 chip. It is not included here.

(4) According to the actual number of users, determine the maximum number of user addresses required, and use the selected space-time orthogonal complementary code group kernel with zero correlation window between groups as the origin in Figure 1 or Figure 1. In the figure, the code length and the number of codes are expanded. The cross-correlation function between the expanded groups of space-time spread-spectrum address codes has a zero correlation window between the groups near the origin.

The expansion will jointly determine the number of expansion stages required in Figure 1 or Figure 2 according to the maximum number of users required and the number of selected pairs of basic orthogonal complementary code pairs. For example, the maximum number of users required is 120. If there is only one pair of basic The orthogonal complementary code group meets the system design requirements. Because 2 ^fi = 64> 120/2, the number of stages to be expanded is 6, and the total number of 26 = 64 codes in the sixth stage in FIG. 1 or FIG. 2 is 128. The address code can then be used as the selected multiple address code. At this time, the actual maximum number of user addresses is 128, which is large. With the required number of users of 120, the requirements can be fully met. If there are two pairs of basic orthogonal complementary code groups that meet the system design requirements, the two pairs of basic orthogonal complementary code groups can be used as the origins in Figure 1 or Figure 2, respectively. Expand the code length and the number of codes in the tree map. The number of stages to be expanded is 5. The two pairs of basic orthogonal complementary code groups are expanded to a total of 64 codes and 128 address codes. Address code; if there are four pairs of basic orthogonal complementary code groups that meet the system design requirements, the four pairs of basic orthogonal complementary code groups can be used as the origin in Figure 1 or Figure 1, respectively, and the code length and code can be coded in the tree diagram. If the number of stages is expanded, the number of stages required to be expanded is 4, and the resulting 64 groups of codes with a total of 128 address codes can meet the system design requirements. If there are eight pairs of basic orthogonal complementary code groups that meet the system design requirements, the eight pairs The basic orthogonal complementary code groups are respectively used as the origins in FIG. 1 or FIG. 2, and the code length and the number of codes are respectively extended in the tree diagram. The number of stages to be extended is 3, and the resulting 64 groups of codes have a total of 128 address codes. Can meet the system design Requirements; when there are 16 pairs of basic orthogonal complementary code groups that meet the system design requirements, the 16 pairs of basic orthogonal complementary code groups can be used as the origin in Figure 1 or Figure 1, respectively, and the code length and number of codes can be performed in the tree diagram. Expansion, the number of stages required for expansion is 2, and the resulting 64 groups of codes with a total of 128 address codes can meet the system design requirements; when 32 pairs of basic orthogonal complementary code groups meet the system design requirements, the 32 pairs of basic codes can be Orthogonal complementary code groups are used as the origins in Figure 1 or Figure 2, respectively, and the code length and number II are extended in the tree diagram. The number of stages to be extended is 1, and the resulting 64 group codes have a total of 128 address codes. Can meet system design requirements. Other user design can be deduced by analogy.

In a fifth step, each group of the space-time spread-spectrum spreading address codes with the zero-correlation window between the groups is correspondingly spread-modulated and transmitted on the corresponding two transmitters.

We take the following two sets of space-time orthogonal complementary code groups as examples

S _n S ₁₂ S ₁₂ ... S _1N S _1N

-S _u S ₁₂ -S _12: 'S _1N -S _lN /

For each group of code divisions transmitted on two transmitters, the same two transmitters (such as the downlink) can be corresponding to each other between the groups, or they can correspond to the two transmitters of different two groups. (E.g. uplink).

The corresponding relationship between two space-time orthogonal complementary codes in the same group and two transmitters is as follows: For all the odd-numbered chips (Chip) of the first space-time orthogonal complementary code correspond to the first transmitter Corresponding odd-numbered chips are transmitted, and all even-numbered chips (chips) correspond to corresponding even-numbered chips transmitted by the second transmitter; all odd-numbered chips (chips) of the second space-time orthogonal complementary code correspond to The corresponding odd digital chips on the second transmitter are transmitted, and all the even digital chips (Chip) correspond to the corresponding even digital chips on the first transmitter. The reverse can also be the opposite, that is, the second space-time orthogonal complementarity All the odd digital chips (chips) of the code correspond to the corresponding odd digital chips transmitted at the first transmitter, and all the even digital chips (chips) correspond to the corresponding even digital chips transmitted at the second transmitter; All odd-numbered chips (chips) of a space-time orthogonal complementary code correspond to the corresponding odd-numbered chips transmitted at the second transmitter, and all even-numbered chips (chips) correspond to the corresponding odd-numbered chips at the first transmitter. Digital film launch.

The space-time orthogonal refers to when each space-time spread-spectrum address code division of the above two sets of space-time orthogonal complementary code groups is transmitted on two transmitters to form the following two sets of four space-time sequences, Two sets of space-time sequences remain orthogonal between groups and have zero correlation windows between groups

+ φ ₂ ) _{? 3} ~ ₂ cos (27 _c t + φ ₃ )? 4 = a ₄ cs (27 _c t + ₄ ), _lt> a ₂ , a ₃₉ a ₄ e [0, + oo), ^ ₁₅ ^ _2? ^ ₃ , ^ ₄ e [0, 2π).

In engineering practice, sometimes more variants of the address code are needed. This requires equivalent transformation of the generated multi-address codes. There are many types of these transformations, and they cannot be listed one by one. The basic equivalent transformations are listed below:

Exchange the positions of the C and S codes.

Swap the positions of C1 and C2 and S1 and S2 at the same time.

Code sequence is reversed.

Each code position is inverted.

Interleave the polarity of each code point: For example, (++-+, + ---), (+++-, +-++) can be used to interleave the polarity of each code point, that is, the first of each code, The polarity of the third-order odd digits does not change, and the second and fourth-order even digits change their polarity, so (+ ---, ++-+), (+-++, +++-), or odd The digital bit polarity changes while the even digital bit polarity does not change.

Rotate each code point in the complex plane: For example, you can rotate the basic complementary code group pair (++-+, +-), (+++-, +-++) to each code point in order. Get

Vq ψαχ + «) _ Ψα J (^ _Cf + 3α) _t _ ⁺⁰ ^ — +2") _ +3)

t y ~ ti

(c '^-i ⁺ ^ j ^ c ₂ ^{+ 2ff} ) — · / · (^ ₂ + 3α)) ψ — _ / (. ₂ + α) _ / (ρ ,,. ₂ + 2α) _ / ( ^ ₂ + 3α)

fci ci ti fci ■ & ~~ fci ί Here,, and can be any initial angle. It can be verified that the properties of the auto-correlation function and cross-correlation function of each address code are unchanged after the rotation transformation, but the peaks outside the "zero correlation window" are related to the rotation angle (become smaller or change polarity).

Proper selection of different rotation angles can make the rotated code groups orthogonal, and multiple sets of orthogonal codes can be generated from one set of orthogonal codes, which brings great convenience to engineering applications. Especially when the code length is long, sometimes wonderful results can be obtained, which can meet various practical engineering requirements, such as networking requirements, switching requirements, and even increased capacity requirements.

Make changes in the spanning tree: For example, Figure 2 is an equivalent transformation of Figure 1, that is, Figure 2 is formed by moving all the upper half of C1 and S1 in Figure 1 to the left and C2 and S2 to the right. C1 and S1 in all the lower half of FIG. 1 are moved to the right, and C2 and S2 are moved to the left. For another example, the code positions of the C code and the S code in the generated multi-address code group can be staggered according to a certain rule, or the polarity can be changed. This transformation is called an equivalent transformation in mathematics, and there are many types of equivalent transformations. Please forgive it is impossible to list here. The use of orthogonal complementary codes in engineering applications must ensure that the C code operates only with the C code (including itself and other codes), the S code operates with the S code (including itself and other codes), and the absolute difference between the C code and the S code Not allowed to meet. Therefore, special separation measures should be taken in practical applications. For example, the C code and the S code can be modulated on mutually orthogonal polarized waves (horizontal and vertical polarized waves, left-handed and right-handed polarized waves), and for another example, the C code and S code can be placed separately Within two time slots that do not overlap each other after transmission. Because the transmission channel changes randomly over time, to ensure the realization of complementarity, the channel characteristics in the two polarized waves and in the two time slots should be kept consistent during the transmission process. In other words, in engineering description languages, their decline should be synchronized. This requires that when polarized separation is used, a frequency band that can guarantee the simultaneous fading of orthogonally polarized waves without depolarization and corresponding measures must be used. When using time-division separation, the interval between two time slots must be much smaller than The correlation time of the channels must also ensure their synchronous fading when using other separation methods.

Since the C code and the S code should be transmitted separately and their complementarity should be used, it is obvious that the information bits modulated on them should be the same, and the output of the C code and the S code after despreading and demodulation should be comparable.

The present invention provides a new encoding method for space-time spreading multiple address codes, so that the formed group of space-time spreading multiple address codes has a correlation characteristic between groups with a "zero correlation window", that is, zero correlation Correlation functions and cross-correlation functions between groups of space-time address codes in the window have no peaks, so as to eliminate multiple access interference (MAI) between groups, while there is multiple access interference between each address code in the same group ( MAI), but joint detection techniques can be used to achieve optimal reception. The space-time spreading code coding method with inter-group zero correlation window characteristics proposed by the present invention. This new space-time spreading code coding with inter-group zero correlation window characteristics combines both diversity technology and zero-correlation window characteristics. In addition, joint detection, interference cancellation technology, and equalization technology can be used, which provides the possibility for increasing system capacity. At the same time, the invention solves the complexity problem of applying joint detection in the traditional CDMA system. In addition, joint detection, interference cancellation technology, and equalization technology can be used, which provides the possibility of increasing system capacity. At the same time, the invention solves the complexity problem of applying joint detection in the traditional CDM system.

Claims

Rights request

1. A space-time spread-spectrum multi-address code encoding method, comprising the following steps: generating a basic orthogonal complementary code group in which each code length is N zero correlation window width is L; and generating the generated basic orthogonality The complementary code group pair is extended to obtain a space-time positive interactive code group kernel with a zero correlation window between the groups;

2. The method according to claim 1, characterized in that: the space-time orthogonal complementary code group core with zero correlation windows between groups or each space-time spread spectrum address code after expansion can be protected by inserting zeros The interval or time slot increases the size of the inter-group zero correlation window between each group of space-time spreading address codes formed after the expansion.

3. The method according to claim 1, wherein the generating a set of basic orthogonal complementary codes whose code lengths are all N zero correlation window width is L comprises: one or more pairs of code lengths can be selected Basic orthogonal complementary code groups whose widths are all zero zero correlation windows;

It can be set that: (CS) and (C, ₂ , S, ₂ ) are the basic orthogonal complementary code groups, and there are:

C ₂₂ "'C _2N ),

S'i ~ (S _u S ₁₂ "'S _1N ), S, ₂ = (S ₂₁ S ₂₂ , ·· 8 _2N ),

4. The method according to claim 1, wherein said extending said pair of basic orthogonal complementary code groups comprises:

Extend the selected pairs of basic orthogonal complementary code (C 'S',), (C, ₂ , S ' ₂ ), where: C 1— (C _n C ₁₂ "" C _1N ) ₉ C 2 ~ ( _lt C ₂₂ * "C _2N ),

S'i ~ (Sn S _U '"S _1N ),

S ₂₂ '"S _2N );

5. The method according to claim 4, wherein the space-time orthogonal complementary code group kernels with zero correlation windows between groups are: each of the space-time orthogonal complementary code group kernels formed after expansion. The code length is 2N and the width of the zero correlation window between groups is greater than or equal to 2L-1.

6. The method according to claim 2, wherein the step of inserting a zero guard interval or a time slot into the space-time orthogonal complementary code group core with zero correlation window between groups comprises:

By inserting a zero guard interval or time slot for the group of space-time orthogonal complementary code group cores with inter-group zero correlation windows, the inter-group zero correlation window between each group of space-time spread-spectrum address codes formed after expansion can be increased. the size of.

7. The method according to claim 2, wherein the inserting a zero guard interval or time slot into the space-time orthogonal complementary code group core of the zero correlation window between groups refers to: first, each code length is N , The pair of basic orthogonal complementary code groups with width L of the zero correlation window window (C ' _1? S'j), (C, ₂ , S, ₂ ), ^ Ί ⁼ (Cii C ₁₂ ... C _1N ) , C ' ₂ = (C ₂₁ C ₂₂ ".C _2N ), S' ^ (S _n S ₁₂ ... S _1N ), S, ₂ = (S ₂₁ S ₂₂ ... S _2N ) expands to the following Space-time orthogonal complementary code group kernel with each code length of 2N and the width of the zero correlation window between groups of 2L-1:

n S ₁₂ S ₁₂ ... S _1N S _1N

S _n S ₁₂ -S ₁₂ ... S _1N -S _1N

S ₂₁ S ₂₂ S ₂₂ ... S ₂ NS _2N

"C ₂ i C ₂₂ -C ₂₂ · '· C _2N -C _2N I -S ₂₁ S ₂₂ -S ₂₂ ... S _2N -S _2N can then insert a certain number of the above-mentioned space-time orthogonal complementary code group cores The zero guard interval or time slot of the new space-time orthogonal complementary code group core has a larger inter-group zero correlation window width than the original space-time orthogonal complementary code group core.

8. The method according to claim 7, wherein the inserting a zero guard interval or a time slot into the space-time orthogonal complementary code group core of the zero correlation window between groups refers to: every L + 1 chips (Chip) Inserts T zeros, resulting in a new space-time orthogonal complementary code group kernel inter-group zero correlation window width Greater than or equal to 2L-1, whereby the new space-time orthogonal complementary code group kernel continues to expand according to a tree structure, and the width of the zero correlation window between the obtained space-time orthogonal complementary code group pairs is greater than or equal to 2L-1 :

is s „s 12 s ,. s L + l .L + l 0 ... 0 s

twenty two

s, • s · · S L + l

1-S L + l 0 ... 0 s,, L + 1 ,,-S L + l ,, Sl _N "S _1N 0 ... 0

1

twenty two

9. The method according to claim 8, wherein, if the base of the complementary orthogonal code group (C ,, S,), ( C 2, S 2) each code length N and the window IFW The width is L, then T zeros can be inserted for every L + 1 chips (Chips) of the 2N space-time orthogonal complementary code group kernel after the expansion of each code length; the new space-time orthogonality formed thereby The width of the zero correlation window between the complementary code group kernels is greater than or equal to 2L-1, so that the new space-time orthogonal complementary code group kernels are continuously expanded according to the spanning tree structure, and the obtained groups of space-time orthogonal complementary code group pairs are obtained. The width of the zero correlation window is greater than or equal to 2L-L

10. The method according to claim 9, wherein the insertion of T zeros needs to satisfy the following conditions: that the width of the zero correlation window between groups of the new space-time orthogonal complementary code group kernel formed is maximize.

11. The method according to claim 9, wherein the inserting T zeros comprises: inserting T zeros at a tail of every L + 1 chips and a head of each L + 1 chips Insert T zeros and so on.

12. The method according to claim 2, characterized in that each of the extended space-time spreading address codes with zero correlation windows between groups can be inserted by inserting zero guard intervals or time slots. Refers to: inserting a certain number of zero guard intervals or time slots for each space-time spread-spectrum address code with inter-group zero correlation window generated by the core-space extension of the space-time orthogonal complementary code group, thereby forming a new space-time orthogonality The width of the inter-group zero correlation window of the complementary code group is greater than or equal to the width of the inter-group zero correlation window of the original space-time orthogonal complementary code group.

13. The method according to claim 12, characterized in that, if each of the extended sets of space-time spread spectrum address codes has a zero correlation window with a width between groups of 2W-1, then the extended T zeros are inserted for every W chips of each space-time spread-spectrum address code; the inter-group zero correlation window width of the new space-time orthogonal complementary code group formed thereby is greater than or equal to 2W-1.

14. The method according to claim 13, wherein the insertion of T zeros needs to satisfy the following conditions: that the width of the inter-group zero correlation window of the new space-time orthogonal complementary code group to be formed is the largest Into.

15. The method according to claim 13, wherein the inserting T zeros comprises: inserting T zeros at a tail of each L + 1 chip and a head of each L + 1 chips Insert T zeros and so on.

16. The method according to claim 1 or 3 or 4, characterized in that, when the space-time spread spectrum address code is calculated: it must be ensured that the C code operates only with the C code, including itself and other codes; S code Operate only with S code, including itself and other codes.

17. The method according to claim 3 or 4, wherein the pair of basic orthogonal complementary code groups (C (C ₂ , S ₂ )) are: their auto-correlation and cross-correlation functions are C codes, respectively. The non-periodic autocorrelation and cross-correlation function between the S code and the non-periodic autocorrelation and cross-correlation function of the S code, in the zero correlation window, the non-periodic autocorrelation and cross-correlation function of the C code and the S code except the origin On the contrary, the values of the autocorrelation function and the crosscorrelation function after addition are zero everywhere except the origin.

18. The method according to claim 1, wherein extending the code length and the number of codes of the space-time orthogonal complementary code group core with inter-group zero correlation window is included:

According to the actually required maximum number of user addresses, the code length and number of codes are extended according to the orthogonal complementary code group core in the spanning tree structure, and the mutual interaction between the extended space-time spread-spectrum address codes is performed. A close window exists near the correlation point, and ¾t of its window is ^ ^ 2L-1; each group of space-time spread-spectrum address codes is completely orthogonal when the relative delay is smaller than the width of the zero correlation window; extension The autocorrelation function of each subsequent space-time spread-spectrum address code except for the origin within the above-mentioned inter-group zero correlation window is only zero at two non-zero relative delays and zero at other places; The cross-correlation function of the space-time spread-spectrum address code is only zero at the two non-zero relative delays within the above-mentioned inter-group zero correlation window, and is zero elsewhere.

19. The method according to claim 18, wherein the extending the code length and code number of the space-time orthogonal complementary code group core in the spanning tree structure refers to:

If (C ₁ S,), (C ₂ , S ₂ ) are a pair of space-time orthogonal complementary code group kernels whose code length is N and the width of the zero correlation window window is L, two can be generated as follows: Pairs are four pairs of space-time orthogonal complementary code pairs with code lengths of 2N:

(C, C ₂ , S, S ₂ )

(C ₂ -C ,, S ₂ -S,) where, expand

The correlation function has a zero correlation window near the origin, the window width of which is greater than or equal to L; the orthogonal complementary code group kernel whose width is 2N and the width of the zero correlation window window is greater than or equal to L, and continues to expand to obtain four pairs of space-time positive Cross complementary code group pairs, their cross-correlation function has a zero correlation window near the origin, and its window width is greater than or equal to ^

20. The method according to Burgundy was claim 12, wherein said extended tree structure may continue to produce a code length of N2 ^n, between the set of zero correlation window width is greater than or equal to 2 L, ⁿ pairs of space-time orthogonal complementary code group pairs, where n = 0, 1, 2, ..., are the number of extensions.

21. A method for applying space-time spread-spectrum multi-address codes, characterized in that it comprises the following steps: Determine the required width of the zero correlation window according to the propagation conditions of the applied system, the basic spreading code rate used by the system, and the maximum timing error in the system;

Generate a basic orthogonal complementary code group pair according to the width of the required zero correlation window;

Extending said basic orthogonal complementary code group pair to generate a positive space-time interactive code group kernel with a zero correlation window between the groups;

According to the actual number of users, determine the maximum number of user addresses required, and use the selected space-time orthogonal complementary code group kernel with zero correlation window between groups as the origin, and expand the code length and number of codes in the spanning tree structure. , The cross-correlation function between the extended groups of space-time spread-spectrum address codes has a zero-correlation window between the groups near the origin;

22. The method according to claim 21, wherein the generating a pair of basic orthogonal complementary code groups comprises: selecting a basic orthogonal complementary code whose code length is N and a width of a zero correlation window window is L. The pair (d, (C ₂ , S ₂ ); the auto-correlation and cross-correlation functions of (d, S, (C ₂ , S ₂ )) are aperiodic auto-correlation and cross-correlation functions between C codes and The sum of aperiodic autocorrelation and cross-correlation functions between S codes, where the non-periodic autocorrelation and cross-correlation functions of C codes and S codes are opposite to each other except the origin within a zero correlation window of width L. Correlation function values and cross-correlation function values are zero except for the origin.

23. The method according to claim 22, wherein the generating or selecting a pair of basic orthogonal complementary code groups comprises: the pair of basic orthogonal complementary code groups (C Sj) (C ₂ , S ₂ ) Can be generated using a spanning tree structure.

24. The method according to claim 21, wherein the expanding the pair of basic orthogonal complementary code groups to generate a space-time orthogonal complementary code group kernel with an inter-group zero correlation window comprises:

The pairs of basic orthogonal complementary code groups (, _Sl ), (c ₂ , s ₂ ) with each code length N and the width of the zero correlation window window L are expanded as follows with inter-group zero correlation windows. Space-time orthogonal Complementary code group core, where: C! C "C ₁₂ '" C _1N ),

s ₂₂

^ 2);

n S ₁₂ S ₁₂ ... S _1N S _1N 、

S _u S ₁₂ -S ₁₂ ... S _1N -S _m S ₂₁ S ₂₂ S ₂₂ --S _2N S _2N 、

"" S ₂₁ S ₂₂ -S ₂₂ ... S _2N -S _2N The length of each code of the extended space-time orthogonal complementary code group kernel is 2N, and the width of the zero correlation window between groups is greater than or equal to 2L- 1.

25. The method according to claim 24, characterized in that a certain number of zero guard intervals or time slots are inserted into the space-time orthogonal complementary code group kernel, thereby forming a new space-time orthogonal complementary code. The width of the inter-group zero correlation window of the group core is larger than the width of the inter-group zero correlation window of the original space-time orthogonal complementary code group core.

26. The method of claim 25, wherein, if the basic group of orthogonal complementary code _{(CS,), (C 2} , S 2) each code length N and a width of the zero correlation window is the window L , Then each L + 1 chip of the extended code length of 2N space-time orthogonal complementary code group core can be

(Chip) inserts T zeros; among them, T zeros can be inserted at the end of each L + 1 chip (Chip), and a new space-time orthogonal complementary code group kernel formed by the zero correlation between groups The window width is greater than or equal to 2L-1, so that the new space-time orthogonal complementary code group kernel continues to expand according to the spanning tree structure, and the width of the zero correlation window between the groups of the obtained space-time orthogonal complementary code group pair is greater than or equal to 2L. -1.

27. The method according to claim 26, wherein the insertion of T zeros needs to satisfy the following conditions: that the width of the inter-group zero correlation window of the new space-time orthogonal complementary code group kernel formed is maximize.

28. The method according to claim 26, wherein the inserting T zeros comprises: inserting T zeros at a tail of each L + 1 chip (Chip), and inserting T zeros at each L + 1 chip (Chip) inserts T zeros and so on into the head.

29. The method according to claim 21, wherein the expansion is to jointly determine the spanning tree according to the required maximum number of users and the number of groups of selected basic orthogonal complementary code group pairs. The number of expansion stages required in.

30. The method according to claim 21, wherein spreading and modulating each group of space-time spread-spectrum address codes with the zero-correlation window between groups on a corresponding transmitter after spreading comprises: The groups of space-time spread-spectrum address codes with zero correlation windows between groups can be spread-spectrum-modulated and transmitted on corresponding two transmitters respectively; and the same two transmitters can be corresponding between groups.

31. The method according to claim 21, wherein the spreading and modulating each group of space-time spreading address codes with zero correlation windows between groups on a corresponding transmitter after spreading modulation comprises: Each group of space-time spread-spectrum address codes with a zero-correlation window between groups can be spread-spectrum-modulated and transmitted on corresponding two transmitters respectively; and two different transmitters can be respectively corresponded between groups.

32. The method according to claim 21, wherein the spreading and modulating transmission of each group of space-time spreading address codes with zero correlation windows between groups on a corresponding transmitter after spreading comprises: When the following two sets of space-time orthogonal complementary codes are used:

Each code division of the two sets of space-time orthogonal complementary code groups is transmitted on two transmitters, and the same two transmitters can be corresponded between the groups and the groups, or the two transmitters of the different two groups can be respectively corresponded. ;

The corresponding relationship between two space-time orthogonal complementary codes in the same group and two transmitters is as follows: For all the odd-numbered chips (Chips) of the first space-time orthogonal complementary code, corresponding to the first transmitter The corresponding odd digital chips are transmitted, and all the even digital chips (Chips) correspond to the corresponding even digital chips of the second transmitter. For the second space-time orthogonal complementary code, all the odd digital chips (Chips) correspond to. The corresponding odd digital chips on the second transmitter are transmitted, and all the even digital chips (Chip) correspond to the corresponding even digital chips on the first transmitter. All the odd digital chips (chips) of the second space-time orthogonal complementary code correspond to the corresponding odd digital chips of the first transmitter, and all the even digital chips (chips) correspond to the corresponding second transmitter. The even-numbered chips for the first space-time orthogonal complementary code correspond to the corresponding odd-numbered chips for the second transmitter, and all the even-numbered chips correspond to the The first transmitter transmits the corresponding even digital chip.

33. The method according to claim 26, wherein the space-time orthogonality refers to: each space-time spreading address code of the two sets of space-time orthogonal complementary code groups is divided into two When the transmitter transmits the following two sets of four space-time sequences in total, the two sets of space-time sequences remain orthogonal between the groups and have a zero correlation window between the groups.

-S _n aj S _{12 2} -S _l2 ci ₁ , .. S _{iN 2} -S _1N j /

/

among them:

a _{ = a cos (2¾Tt + φ _λ ), a ₂ = a ₂ cos (2 ^ T _c t + ^? ₂ ), a ₃ = ₃ cos (2 t + φ _ζ ), a ₄ = a ₄ cos ( 2¾f " _c t + ^ ₄ ),, α ₂ , α ₃ , ₄ e [0, + oo), φ ₁ , φ ₂ , φ ₃ , φ e [0,2; r).

34. The method according to claim 1 or 21, wherein the generated multiple address codes can be equivalently transformed.

35. The method according to claim 34, wherein the equivalent transformation comprises:

Exchange the positions of C and S codes, exchange the positions of C1 and C2 and S1 and S2 at the same time, invert the code sequence, invert each code bit, interleave the polarity of each code bit, and rotate each code bit in the complex plane. Make changes in the spanning tree.

36. The method according to claim 35, wherein the rotating changes of each code position in a complex plane comprises:

The basic complementary code group can be sequentially rotated by "degrees" for each code position. After the rotation transformation, the properties of the autocorrelation function and cross-correlation function of each address code are unchanged, but the "zero correlation window, and the external peaks are related to the rotation angle. ; Proper selection of the rotation angle allows orthogonality between the rotated code groups, and multiple sets of orthogonal codes can be generated from a set of orthogonal codes.

37. The method according to claim 36, wherein the generated multiple sets of orthogonal codes are applicable to networking requirements, handover requirements, increased capacity requirements, etc., especially when the code length is long.

38. The method according to claim 1 or 15 or 16, wherein the orthogonal complementary code must ensure that the C code operates only with the C code (including itself and other codes), and the S code operates only with the S code (Including itself and other codes).

39. The method according to claim 38, characterized in that, in practical applications, separate measures should be taken for the C code and the S code.

40. The method according to claim 39, characterized in that said separating measures may include: modulating the C code and the S code on mutually orthogonal polarized waves, respectively; or separately dividing the C code and the S code Placed in two time slots that do not overlap each other after transmission.

41. The method according to claim 40, wherein the transmission channel changes randomly with time, and to ensure complementarity, channels in two polarized waves and in two time slots are transmitted during transmission. The characteristics should be consistent, that is, their fading should be synchronized; this requires that when using polarization separation, you must use a frequency band that can guarantee the simultaneous fading of orthogonal polarized waves, no depolarization, and corresponding measures, and use time division to separate In this case, the interval between the two time slots must be made much smaller than the channel correlation time. When other separation methods are used, their synchronization fading must also be guaranteed.

42. The method according to claim 41, characterized in that: since the C code and the S code should be transmitted separately and their complementarities are used, the information bits modulated on them should be the same, and the C code and the The output of S code despreading and demodulation should be added.