A generalized code division multiplexing modulation and demodulation method

Abstract

This invention relates to a generalized code division multiplexing (GDM) modulation and demodulation method. GDM modulation is based on the superposition of codewords in a non-orthogonal codeword set through high-order modulation. The demodulation method fully utilizes the characteristics of GDM modulation, employing IFFT-FFT to achieve fast codeword detection, thereby reducing demodulation complexity. Simultaneously, frequency response estimation of the transmission channel can be achieved through a simple spectrum division method in the pilot channel, thereby achieving channel equalization and improving anti-multipath demodulation performance. Furthermore, by constructing an interference cancellation matrix within the data channel, mutual interference caused by the non-orthogonality of codewords within the data channel can be eliminated, thus enabling data demodulation even under non-orthogonal codeword conditions.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of spread spectrum communication, and relates to a generalized code division multiplexing modulation and demodulation method. BACKGROUND

[0002] Code division multiplexing is a communication mode for realizing multiplexing by using orthogonal code structure of each signal, and has been widely applied to fields such as mobile communication, satellite communication and wireless local area network.

[0003] In a traditional code division multiplexing scheme, the number of data layers (one orthogonal code word corresponds to one data layer) is limited by orthogonal resources, and the spectral efficiency is greatly restricted, so it is difficult to achieve the theoretical system capacity, and in order to reduce the mutual interference between data layers and guarantee the reliability of data demodulation, the code word is not easy to use high-order modulation mode, and usually 2 or 4 binary modulation is used, so it is difficult to realize super high speed data transmission. In addition, the traditional code division multiplexing modulation signal is easy to be affected by multipath interference, thereby affecting the reliability of demodulation. SUMMARY

[0004] Technical problems to be solved

[0005] In order to avoid the shortcomings of the prior art, the application provides a generalized code division multiplexing modulation and demodulation method, which is based on code words in a non-orthogonal code word set, and is obtained by high-order modulation and superposition, wherein the code word used for modulating data is not limited by orthogonality, so that the spectral efficiency is greatly improved, and super high speed data transmission can be realized, and the demodulation performance of the generalized code division multiplexing modulation can be guaranteed by designing a pilot channel and a balanced modulation channel.

[0006] Technical scheme

[0007] A generalized code division multiplexing modulation and demodulation method, characterized by comprising a modulation step and a demodulation step; the modulation step is:

[0008] Step a1: constructing a non-orthogonal code word set for generalized code division multiplexing modulation, wherein the non-orthogonal code word set is obtained by oversampling expansion and cyclic shift of an m sequence with a period of N:

[0009]

[0010] Wherein: is a code word with a phase of i, and i is the number of cyclic shifts;

[0011] Step a2: dividing the code words of the non-orthogonal code word set into U data modulation channels, one pilot channel and one balanced modulation channel;

[0012] The data modulation channel is used for modulating data; the pilot channel is used for receiving end to realize signal synchronization, channel estimation and equalization; the balanced modulation channel is used for eliminating mutual interference between the data modulation channel and the pilot channel;

[0013] Step a3: sequentially modulating the code words in the data modulation channel, the pilot channel and the balanced modulation channel in amplitude and phase;

[0014] Step a4: superimposing the code words after joint modulation in amplitude and phase of all channels to obtain a baseband signal containing modulation information Then superimposing the baseband signal And the carrier is I / Q modulated, and the real part is taken to obtain the generalized code division multiplexing modulation signal;

[0015] The demodulation steps are:

[0016] Step b1: constructing an interference repair matrix A in the data channel;

[0017] Step b2: down-converting the generalized code division multiplexing modulation signal to obtain a baseband signal, synchronizing the baseband signal by using the pilot channel code word, and obtaining the synchronized baseband signal denoted as r; performing cyclic shift correlation detection on the code word by using r to obtain a detection result R;

[0018] Step b3: performing channel estimation H on the frequency response of the transmission channel of the code word detection result R;

[0019] Step b4: performing channel equalization on the channel estimation H to obtain Using the interference repair matrix A to perform data demodulation on to obtain a demodulation result.

[0020] The non-orthogonal code word set of step a1 is obtained by oversampling and expanding the m sequence with a period of N and performing cyclic shift: the m sequence with a period of N is denoted as s0=[s0(0), s0(1), …, s0(n), …, s0(N-1)], wherein s0(n)∈{-1, +1} represents the nth chip of the m sequence; s0 is expanded by oversampling to obtain Wherein is composed of c e s0(n), c e is called a chip expansion number, representing the number of sub-chips obtained by oversampling and expanding a single chip of the m sequence; i is cyclically shifted to obtain the code word with a code word phase of i The non-orthogonal code word set for generalized code division multiplexing modulation is constructed by ​​

[0021] In step a2, each of the U data modulation channels contains c d Each code character, used Let represent the codeword phase corresponding to the j-th codeword in the u-th data modulation channel, and define . Where 1≤j≤c d 1≤u≤U; the balanced modulation channel contains a codeword phase of p ICI The code word, specifying p ICI =U(c d +c e -1)+c e The pilot channel contains a codeword phase of p. chan The code word, specifying p chan =p ICI +(Δ max +1)c e +c e , where Δ max p represents the maximum multipath delay difference in the transmission channel; chan p should be satisfied chan +Δ max c e +c e ≤N·c e -1, i.e., U, c d c e Should meet

[0022] When performing joint amplitude and phase modulation on the data modulation channels in step a3: the j-th codeword in the u-th data modulation channel of the U data modulation channels undergoes joint amplitude and phase modulation, and the modulated codeword is denoted as... in The amplitude modulation and phase modulation are represented respectively, and i represents the imaginary unit; when the pilot channel performs joint amplitude and phase modulation: the codeword of the pilot channel is modulated with fixed data "1", and the modulated codeword is recorded as The codeword of the balanced modulation channel is modulated, and the modulated codeword is recorded as follows:

[0023] In step a4, a baseband signal containing modulation information is superimposed to obtain the signal. The superposition of the three channels modulated as obtained in claim 4 is expressed as follows:

[0024] The interference repair matrix is ​​composed of c d 1×c d Composed of 1×c row vectors, where each 1×c d A 1×c row vector is a 1×cd 3D row vector After cyclic shifting, we obtain:

[0025]

[0026] in: α λ Let c be the λ-th element in vector α. e For chip extension number, c d The number of codewords in each data modulation channel.

[0027] The detection result R is the baseband signal r and Perform IFFT-FFT operations:

[0028]

[0029] Where: R = [R(0),R(1),…,R(n),…,R(N·c)] e -1)] represents the codeword detection result; conj(·) represents finding the conjugate of a complex number.

[0030] The channel estimation H = [H(0),H(1),…,H(n),…,H(N·c)] e -1)], the calculation process is as follows:

[0031] Let h1=[h1(0),h1(1),…,h1(n),…,h1(N·c e -1)], and stipulate: when p chan -c e ≤n≤p chan +c e Δ max When n is n, let h1(n) = R(n); when n is n, let h1(n) = 0;

[0032] Let h2=[h2(0),h2(1),…,h2(n),…h2(N·c e -1)], and stipulate:

[0033]

[0034] Dividing the spectra of h1 and h2, we get H = FFT(h1) / FFT(h2).

[0035] The channel equalization of the channel estimation H in The vector form is

[0036] The interference repair matrix A pair Data demodulation is performed, and c in the u-th data modulation channel d Joint demodulation of amplitude and phase for each codeword:

[0037]

[0038] in, Representing vectors The Middle There are 1 element; i represents the imaginary unit, and exp(·) is an exponential function with the natural constant e as the base. and The amplitudes of the j-th codeword in the u-th data modulation channel are respectively and phase The demodulation results.

[0039] Beneficial effects

[0040] The present invention proposes a generalized code division multiplexing modulation and demodulation method. The generalized code division multiplexing modulation is based on the codewords in the non-orthogonal codeword set being modulated by higher order and superimposed. The codewords used for modulating data are not constrained by orthogonality, which greatly improves the spectral efficiency.

[0041] The pilot channel designed in the generalized code division multiplexing modulation provided by this invention can not only realize signal synchronization, but also realize channel estimation and equalization, thereby reducing the impact of multipath interference on demodulation.

[0042] The balanced modulation channel designed in the generalized code division multiplexing modulation provided by this invention can eliminate mutual interference between channels. It not only utilizes the codewords in the data channel to adopt high-order modulation, but also utilizes the pilot channel to play a role in channel estimation and equalization.

[0043] The demodulation method for generalized code division multiplexing modulation provided by this invention fully utilizes the characteristics of generalized code division multiplexing modulation. It uses IFFT-FFT to achieve fast codeword detection, thereby reducing demodulation complexity. At the same time, it uses a simple spectrum division method in the pilot channel to estimate the frequency response of the transmission channel, thereby achieving channel equalization and improving anti-multipath demodulation performance. Furthermore, by constructing an interference cancellation matrix in the data channel, it can eliminate mutual interference caused by codeword non-orthogonality in the data channel, thereby achieving data demodulation even in the case of codeword non-orthogonality. Attached Figure Description

[0044] Figure 1 This is a schematic diagram illustrating the implementation process of the generalized code division multiplexing modulation provided by the present invention. Detailed Implementation

[0045] The present invention will now be further described in conjunction with the embodiments and accompanying drawings:

[0046] The present invention adopts the following technical solution:

[0047] Step 1: Construct a non-orthogonal codeword set for generalized code division multiplexing modulation. The codewords in the codeword set are obtained by oversampling and cyclic shifting of an m-sequence.

[0048] Step 2: Divide the codewords in the codeword set into several data modulation channels, one pilot channel and one balanced modulation channel, wherein each data modulation channel contains several codewords;

[0049] Step 3: The codewords in the data modulation channel, pilot channel and balanced modulation channel are sequentially subjected to joint amplitude and phase modulation, and then superimposed to obtain the baseband signal containing modulation information;

[0050] Step 4: Modulate the baseband signal with the carrier using I / Q modulation and take the real part to obtain the generalized code division multiplexing modulated signal;

[0051] The data modulation channel is used to modulate data; the pilot channel is used at the receiving end to realize signal synchronization, channel estimation and equalization; the balanced modulation channel is used to eliminate mutual interference between the data modulation channel and the pilot channel;

[0052] Step 1 is implemented as follows:

[0053] Step 1.1: Select an m-sequence with a cycle number of N, denoted as s0, as follows:

[0054] s0=[s0(0),s0(1),…,s0(n),…,s0(N-1)] (1)

[0055] In the formula, s0(n)∈{-1,+1} represents the nth chip of the m-sequence;

[0056] Step 1.2, expand s0 by oversampling to obtain It is expressed as follows:

[0057]

[0058] In the formula, By c e Composed of s0(n), c e It is called the chip extension number, which represents the number of sub-chips obtained after oversampling expansion of a single chip of the m-sequence;

[0059] Step 1.3, regarding the... Perform i cyclic shifts to obtain a codeword with phase i, denoted as . From the above The nonorthogonal codeword set S for generalized code division multiplexing modulation is constructed as follows:

[0060]

[0061] Step 2 is implemented as follows:

[0062] Step 2.1, convert the U·c in the codeword set d Each codeword is divided into U data modulation channels, and each data modulation channel contains c d Each code character; used Let represent the codeword phase corresponding to the j-th codeword in the u-th data modulation channel, and define it as follows:

[0063]

[0064] In the formula: 1≤j≤c d ; 1≤u≤U.

[0065] Step 2.2, set the codeword phase of the codeword set to p ICI The codewords are divided into balanced modulation channels, and the following rules apply:

[0066] p ICI =U(c d +c e -1)+c e (5)

[0067] Step 2.3, set the codeword phase of the codeword set to p chan The codewords are divided into pilot channels, and the following rules apply:

[0068] p chan =p ICI +(Δ max +1)c e +c e (6)

[0069] In the formula, Δ max This represents the maximum multipath delay difference (discrete value) in the transmission channel; the p chan It should satisfy: p chan +Δ max c e +c e ≤N·c e -1, that is, the U and c d c e Should meet:

[0070] Step 3 is implemented as follows:

[0071] Step 3.1: Perform joint amplitude and phase modulation on the j-th codeword in the u-th data modulation channel, and denote the modulated codeword as follows. in These represent amplitude modulation and phase modulation, respectively, with i representing the imaginary unit;

[0072] Step 3.2, modulate the codeword of the pilot channel with fixed data "1", and record the modulated codeword as...

[0073] Step 3.3: Based on the real-time modulation results of the data modulation channel and the pilot channel, modulate the codeword of the balanced modulation channel, and record the modulated codeword as follows:

[0074] Step 4 is implemented as follows:

[0075] Step 4.1: Superimpose the codeword modulated by the data modulation channel, the codeword modulated by the pilot channel, and the codeword modulated by the balanced modulation channel to obtain the baseband signal containing modulation information. It is expressed as follows:

[0076]

[0077] Step 4.2, the baseband signal The generalized code division multiplexing modulated signal is obtained by performing I / Q modulation with the carrier and taking the real part.

[0078] The present invention also provides a demodulation method for the generalized code division multiplexing modulation, the implementation steps of which include:

[0079] Step a, construct the interference repair matrix within the data channel;

[0080] Step b: The input generalized code division multiplexing modulation signal is captured and synchronized, and then the codeword is subjected to cyclic shift correlation detection;

[0081] Step c: Perform channel estimation based on the codeword detection results;

[0082] Step d: Demodulate the data based on the interference repair matrix, codeword detection results, and channel estimation results.

[0083] Step a is specifically implemented as follows:

[0084] Step a.1, define 1×c d 3D row vector Where α λ We obtain the following formula:

[0085]

[0086] Step a.2, construct c based on α d ×cd The interference repair matrix A within the Viton data channel is represented as follows:

[0087]

[0088] Step b is specifically implemented as follows:

[0089] Step b.1: The input generalized code division multiplexing modulation signal is orthogonally downconverted to obtain the baseband signal;

[0090] Step b.2: Capture and synchronize the codewords of the pilot channel contained in the baseband signal; denote the synchronized baseband signal as r = [r(0), r(1), ..., r(n), ..., r(N·c)]. e -1)];

[0091] Step b.3, combine r with the following formula By performing IFFT-FFT operations, cyclic shift correlation detection of codewords can be achieved.

[0092]

[0093] In the formula: R=[R(0),R(1),…,R(n),…,R(N·c) e -1)] represents the codeword detection result; conj(·) represents finding the conjugate of a complex number;

[0094] Step c is specifically implemented as follows:

[0095] Step c.1, let h1 = [h1(0), h1(1), ..., h1(n), ..., h1(N·c e -1)], and stipulate: when p chan -c e ≤n≤p chan +c e Δ max When n is n, let h1(n) = R(n); when n is n, h1(n) = 0.

[0096] Let h2=[h2(0),h2(1),…,h2(n),…h2(N·c e -1)], and stipulate:

[0097]

[0098] Step c.2, define H = [H(0), H(1), ..., H(n), ..., H(n·c)] e ),…,H(N·c e-1)] represents the frequency response of the transmission channel. Based on the spectrum division of h1 and h2, we get H = FFT(h1) / FFT(h2), and then realize the channel estimation.

[0099] Step d is specifically implemented as follows:

[0100] Step d.1: Based on R and H, perform channel equalization on R according to the following formula to obtain...

[0101]

[0102] The The vector representation of is

[0103] Step d.2, according to the above And A, c in the u-th data modulation channel is modulated according to the following formula. d Joint demodulation of amplitude and phase for each codeword:

[0104]

[0105] in, Representing vectors The Middle There are 1 element; i represents the imaginary unit, and exp(·) is an exponential function with the natural constant e as the base. and The amplitudes of the j-th codeword in the u-th data modulation channel are respectively and phase The demodulation results.

[0106] The following is combined with Figure 1 The implementation process of the generalized code division multiplexing modulation and demodulation method is described based on the following assumptions: The selected m-sequence has a period N = 15 and a chip spread c. e =3, Number of data modulation channels U = 2, Number of codewords c in each data modulation channel d =2. Maximum multipath delay difference (discrete value) Δ in the transmission channel max =3.

[0107] Reference Figure 1 The implementation process of the generalized code division multiplexing modulation is as follows:

[0108] Step 1: Construct a non-orthogonal codeword set for generalized code division multiplexing modulation. The codewords in the codeword set are obtained by oversampling and cyclic shifting of an m-sequence.

[0109] Step 1.1: Select an m-sequence with N=15 code cycles, denoted as s0, as follows:

[0110] s0=[s0(0),s0(1),s0(2),…,s0(n),…,…,s0(14)] (14)

[0111] In the formula, s0(n)∈{-1,+1} represents the nth chip of the m-sequence;

[0112] Step 1.2, the s0 is expanded by oversampling, and the chip expansion number c e =3, thus obtaining It is expressed as follows:

[0113]

[0114] In the formula, By c e It consists of s0(n), that is The same principle applies here.

[0115] Step 1.3: Perform i cyclic shifts on the s0 to obtain a codeword with phase i. And thus construct a nonorthogonal codeword set S for generalized code division multiplexing, represented as follows:

[0116]

[0117] Based on the aforementioned assumptions, the codeword set S contains N·c e = 45 codewords, 0th codeword The codeword phase is 0 (no cyclic shift), the first codeword The codeword phase is 1 (i.e. Depend on After one cyclic shift, the i-th codeword is obtained. The codeword phase is i (i.e. Depend on (After i cycles of shifting, we get the result), and so on.

[0118] Step 2: Divide the codewords in the codeword set into several data modulation channels, one pilot channel and one balanced modulation channel, wherein each data modulation channel contains several codewords;

[0119] Step 2.1, convert the U·c in the codeword set S d = 4 codewords are divided into U = 2 data modulation channels; each data modulation channel contains c d = 2 codewords; using Let represent the codeword phase corresponding to the j-th codeword in the u-th data modulation channel, and define:

[0120]

[0121] In the formula: 1≤j≤c d ; 1≤u≤U.

[0122] Based on the aforementioned assumptions, we can obtain Soon Assigned to the first data modulation channel, It was assigned to the second data modulation channel;

[0123] Step 2.2, set the codeword phase of the codeword set to p ICI The codewords are divided into balanced modulation channels, and the following rules apply:

[0124] p ICI =U(c d +c e -1)+c e (17)

[0125] Based on the aforementioned assumptions, we can obtain p ICI =11, soon Divided into balanced modulation channels;

[0126] Step 2.3, set the codeword phase of the codeword set to p chan The codewords are divided into pilot channels, and the following rules apply:

[0127] p chan =p ICI +(Δ max +1)c e +c e (18)

[0128] Based on the aforementioned assumptions, we can obtain p chan =26, soon Divided into pilot channels;

[0129] Step 3: The codewords in the data modulation channel, pilot channel and balanced modulation channel are sequentially subjected to joint amplitude and phase modulation, and then superimposed to obtain the baseband signal containing modulation information;

[0130] Step 3.1: Perform amplitude and phase joint modulation on the j-th codeword in the u-th data modulation channel, and denote the modulated codeword as follows. in These represent amplitude modulation and phase modulation, respectively, with i representing the imaginary unit. If there is no amplitude modulation, then... If the value is constant at 1, the modulation mode of the codeword can be MPSK modulation; otherwise, it is MQAM modulation.

[0131] Here it is assumed that the modulation method is MPSK modulation, that is Then we have:

[0132]

[0133] Step 3.2, modulate the codeword of the pilot channel with fixed data "1", and record the modulated codeword as...

[0134] Due to the p chan =26, then we have

[0135] Step 3.3: Based on the real-time modulation results of the data modulation channel and the pilot channel, modulate the codeword of the balanced modulation channel, and record the modulated codeword as follows:

[0136] Due to the p ICI =11, and based on the aforementioned assumptions, we have:

[0137] Step 4: Modulate the baseband signal with the carrier using I / Q modulation and take the real part to obtain the generalized code division multiplexing modulated signal;

[0138] Step 4.1: The codewords modulated by the data modulation channel, the pilot modulation channel, and the balanced modulation channel are superimposed in parallel to obtain the baseband signal containing modulation information. And can be represented as follows:

[0139]

[0140] Due to the middle The p chan =26, the The p ICI =11, and based on the aforementioned assumptions, we have:

[0141]

[0142] Step 4.2, the baseband signal The generalized code division multiplexing modulated signal is obtained by performing I / Q modulation with the carrier and taking the real part.

[0143] The implementation process of the demodulation method of the generalized code division multiplexing modulation is as follows:

[0144] Step a, construct the interference repair matrix within the data channel;

[0145] Step a.1, define 1×c cn 3D row vector Where α λ We obtain the following formula:

[0146]

[0147] Based on the aforementioned assumptions, we can obtain α = [1 2 / 3];

[0148] Step a.2, construct c based on α d ×c d The interference repair matrix within the Viton data channel, denoted as A, is represented as follows:

[0149]

[0150] Since α = [1 2 / 3], therefore we have

[0151] Step b: The input generalized code division multiplexing modulation signal is captured and synchronized, and then the codeword is subjected to cyclic shift correlation detection;

[0152] Step b.1: The input generalized code division multiplexing modulation signal is orthogonally downconverted to obtain the baseband signal;

[0153] Step b.2, in step b.2, the codewords of the pilot channel contained in the baseband signal are captured and synchronized; the synchronized baseband signal is denoted as r=[r(0),r(1),…r(n),…,r(N·c e -1)];

[0154] Step b.3, combine r with the following formula Perform IFFT-FFT operations, that is, perform cyclic shift correlation detection on the codewords contained in r:

[0155]

[0156] In the formula: R=[R(0),R(1),…,R(n),…,R(N·c) e -1)] represents the codeword detection result; conj(·) represents finding the conjugate of a complex number;

[0157] Step c: Perform channel estimation based on the codeword detection results;

[0158] Step c.1, let h1 = [h1(0), h1(1), ..., h1(n), ..., h1(N·c e -1)], and stipulate:

[0159] p chan -c e ≤n≤p chan +c e Δ maxWhen n is n, h1(n) = R(n); when n is n, h1(n) = 0.

[0160] Let h2=[h2(0),h2(1),…h2(n),…h2(N·c e -1)], and stipulate:

[0161]

[0162] Step c.2, define H = [H(0), H(1), ..., H(n), ..., H(N·c)] en -1)] represents the frequency response of the transmission channel. Based on the spectrum division of h1 and h2, we get H = FFT(h1) / FFT(h2), and then realize the channel estimation.

[0163] Because p chan =26, c e =3, therefore we can conclude:

[0164]

[0165] Step d: Demodulate the data based on the interference repair matrix, codeword detection results, and channel estimation results within the data channel;

[0166] Step d.1: Based on R and H, perform channel equalization on R according to the following formula to obtain...

[0167]

[0168] The The vector representation of is

[0169] Step d.2, according to the above And A, c in the u-th data modulation channel is modulated according to the following formula. d Joint demodulation of amplitude and phase for each codeword:

[0170]

[0171] In the formula, and The respective and stated The demodulation results.

[0172] Based on the aforementioned assumptions and taking the demodulation result of the first data modulation channel as an example, the necessity of the existence of the balanced modulation channel and the pilot channel will be explained below.

[0173] Assuming the codewords in the data modulation channel use 64PSK modulation, and assuming the modulation phases of the first and second codewords in the first data modulation channel are respectively... and

[0174] Table 1 shows the demodulation results under the condition of no multipath interference (in which the receiver does not perform channel estimation); Table 2 shows the demodulation results under multipath signals; the multipath signals contain two path signals, as explained below:

[0175] (1) The first path signal is the The power of the signal from the other path is related to the above. same;

[0176] (2) The relative time delay difference between the two path signals is 1 (discrete value);

[0177] Table 1 shows that the data demodulation performance is significantly improved when using the balanced modulation channel compared to when the balanced modulation channel is not used. This is because the designed balanced modulation channel can effectively eliminate codeword mutual interference between the data modulation channel and the pilot channel.

[0178] Table 2 shows that in a multipath channel, even if the balanced modulation channel is used, reliable data demodulation cannot be achieved without the pilot channel, and may even degrade demodulation performance. With the pilot channel, the receiver can achieve channel equalization. In addition, the balanced modulation channel eliminates the mutual interference between channels (especially when the codeword in the data channel is modulated with a higher order, the mutual interference between the pilot channel and the data modulation channel is quite serious), thereby improving the reliability of data demodulation.

[0179] In summary, to ensure the reliability of data demodulation, the use of the balanced modulation channel and the pilot channel is necessary. Existing code division multiplexing modulation techniques do not have a pilot channel capable of channel equalization, nor do they employ a balanced modulation channel to eliminate mutual interference between channels.

[0180] Table 1 Demodulation results without multipath interference.

[0181]

[0182] Table 2 Demodulation results under multipath interference.

[0183]

[0184] Based on the generalized code division multiplexing modulation provided by this invention, several typical ultra-high data rate modulation schemes are designed below, as shown in Table 3, where: U d =U·c dIndicates the number of data layers (each codeword in the data modulation channel is a data layer); k = U d log2(M) represents the number of bits modulated by a single modulation symbol, and M represents the modulation base. In this embodiment, M is 128. R represents spectral efficiency; b This represents the data transmission rate when the signal bandwidth is 100MHz. Table 3 shows that in the ultra-high data rate modulation scheme designed using this invention, the number of data layers U... d It can be much larger than N, greatly improving spectral efficiency; in addition, it can be applied to N and c. e The c d The design of U is flexible to meet the diverse needs of different application scenarios.

[0185] Table 3 shows several typical ultra-high data rate modulation methods designed using this invention.

[0186]

[0187] The following is a comparative analysis of the generalized code division multiplexing modulation provided by this invention and the traditional code division multiplexing modulation in terms of spectral efficiency:

[0188] The spectral efficiency of traditional code division multiplexing modulation is denoted as... Where N d ≤N represents the data layer in traditional code division multiplexing modulation; let As can be seen from Table 3, U d The spectral efficiency of the generalized code division multiplexing modulation provided by this invention is significantly improved compared to traditional code division multiplexing modulation.

[0189] In traditional code division multiplexing modulation, the data layer is limited by orthogonal resources. Furthermore, to reduce mutual interference between data layers and ensure the reliability of data demodulation, it is difficult to employ high-order modulation, typically using binary or quaternary modulation, which severely restricts its spectral efficiency. However, the generalized code division multiplexing modulation provided by this invention, on the one hand, employs a non-orthogonal codeword set, greatly increasing the number of data layers; on the other hand, by setting a balanced modulation channel, mutual interference between channels can be eliminated, and at the receiving end, interference cancellation matrices within the data channels are constructed to eliminate mutual interference between codewords within the data channels. Therefore, high-order modulation can be used.

Claims

1. A method of generalized code division multiplexing modulation and demodulation, characterized by The modulation step and the demodulation step are included; The modulation step is: Step a1: constructing a non-orthogonal code word set for generalized code division multiplexing modulation, the non-orthogonal code word set is obtained by oversampling extension and cyclic shift of an m sequence with a period of N: wherein: is a code word with phase i, i being the number of cyclic shifts; Step a2: dividing the code words of the non-orthogonal code word set into U data modulation channels, one pilot channel and one balance modulation channel; The data modulation channel is used for modulating data; the pilot channel is used for signal synchronization, channel estimation and equalization at the receiving end; and the balance modulation channel is used for eliminating mutual interference between the data modulation channel and the pilot channel; Step a3: sequentially performing joint modulation of the amplitudes and phases of the code words in the data modulation channel, the pilot channel and the balance modulation channel; Step a4: superimpose the code words after joint modulation of amplitude and phase of all channels to obtain a baseband signal containing modulation information Then superimpose the baseband signal with the carrier for I / Q modulation, and obtain the generalized code division multiplexing modulation signal after taking the real part. The demodulation step is: Step b1: constructing an interference repair matrix A in the data channel; Step b2: performing frequency down-conversion on the generalized code division multiplexing modulation signal to obtain a baseband signal, synchronizing the baseband signal by using the pilot channel code word, obtaining a synchronized baseband signal denoted as r, and performing cyclic shift correlation detection on the code word by using r to obtain a detection result R; Step b3: performing channel estimation H on the frequency response of the transmission channel of the code word detection result R; Step b4: channel equalization is performed on the channel estimation H to obtain Data demodulation is performed on the received signal y using the interference repair matrix A to obtain a demodulation result. Data demodulation is performed on the received signal y using the interference repair matrix A to obtain a demodulation result.

2. The generalized code division multiplexing modulation and demodulation method of claim 1, wherein: The process of obtaining the non-orthogonal code word set of step a1 from the m sequence with a period of N through oversampling expansion and cyclic shift is: an m sequence with a period of N is denoted as s0=[s0(0), s0(1), …, s0(n), …, s0(N-1)], wherein s0(n)∈{-1, +1} represents the nth chip of the m sequence; the m sequence s0 is expanded through oversampling to obtain wherein consisting of c e s0(n), c e is referred to as a chip expansion number, and represents the number of sub-chips obtained through oversampling expansion of a single chip of the m sequence; the code word is cyclically shifted i times to obtain a code word with a code word phase of i consisting of is constructed to obtain a non-orthogonal code word set for generalized code division multiplexing modulation 3. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: The step a2 includes c d code words in each of the U data modulation channels, and the code word phase of the jth code word in the u th data modulation channel is represented by p , and it is defined that where 1≤j≤c d , 1≤u≤U; the balanced modulation channel includes a code word with a code word phase of p ICI , and it is defined that p ICI =U(c d +c e -1)+c e ; the pilot channel includes a code word with a code word phase of p chan , and it is defined that p chan =p ICI +(Δ max +1)c s +c e , where Δ max represents the maximum multipath delay difference in the transmission channel; the p chan should satisfy p chan +Δ max c e +c e ≤N·c e -1, i.e. U, c d , c e should satisfy 4. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: When the step a3 is the joint modulation of amplitude and phase to the data modulation channel, the jth code word in the u th data modulation channel is jointly modulated in amplitude and phase, and the modulated code word is denoted as Wherein respectively represent the amplitude modulation and phase modulation, i represents the imaginary unit; when the pilot channel is jointly modulated in amplitude and phase, the code word of the pilot channel is modulated with fixed data "1", and the modulated code word is denoted as The code word of the balanced modulation channel is modulated, and the modulated code word is denoted as 5. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: The superposition in step a4 results in a baseband signal containing modulation information For the superposition of the three channels modulated in claim 4, the expression is:

6. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: The interference repair matrix is composed of c d 1x c d dimensional row vectors, where each 1x c d dimensional row vector is composed of 1x c d dimensional row vectors After cyclic shift, we get: where: α λ is the λth element of the vector α, c e is the number of chips spread, c d is the number of codewords burst in each data modulation channel.

7. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: The detection result R is a baseband signal r and IFFT-FFT operation is performed: where: R = [R(0), R(l),..., R(n),..., R(N-c e -1)] represents the code word detection result; conj( ) represents the conjugate complex number of a complex number.

8. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: The channel estimate H = [H(0), H(l),..., H(n),..., H(N-c e -1)] is calculated by the process: Let h1 = [h1(0), h1(1),..., h1(n),..., h1(N-c e -1)], and define: h1(n) = R(n) when p chan -c e ≤ n ≤ p chan +c e Δ max ; h1(n) = 0 when n is otherwise. Let h2 = [h2(0), h2(1),..., h2(n),... h2(N-c e -1)] and define: Divide the spectrums by h1 and h2 to obtain H = FFT(h1) / FFT(h2).

9. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: channel equalization of the channel estimate H wherein the vector form of 10. The method of generalized space-time block coding modulation and demodulation according to claim 1, wherein: The interference repair matrix A is used to perform data demodulation, c d code words in the u-th data modulation channel perform joint demodulation of amplitude and phase: wherein represents a vector of the i-th element; i represents an imaginary unit, exp(·) is an exponential function with base of natural constant e; and are demodulation results of amplitudes and phases of the j-th codeword in the u-th data modulation channel, respectively.​

Images