Flexible spectral efficiency chirp multicarrier signal generation method based on zero insertion and zero padding

Abstract

The invention discloses a flexible spectral efficiency chirp multi-carrier signal generation method based on zero insertion and zero padding, and belongs to the technical field of chirp multi-carrier signal generation. The invention solves the problems that the spectrum efficiency of the existing non-orthogonal chirp multicarrier multiplexing technology is limited, and the applicability and universality of the structure of the existing non-orthogonal chirp multicarrier multiplexing system are limited. According to the invention, the operations of zero insertion and zero padding are added after the first phase rotation is carried out on the transmitting and receiving ends, and the number of points of DFT and IDFT can be adjusted by controlling the number of zero insertion and zero padding so as to adjust the bandwidth compression factor in the chirp multicarrier multiplexing system, thereby providing flexible and variable spectrum efficiency, and improving the performance of the chirp multicarrier multiplexing system. The problem that the spectrum efficiency range is limited due to the fact that the number of operation points must be an integer and is the power of 2 in the chirp multi-carrier signal generation process is solved, the structural application range of a chirp multi-carrier multiplexing system is further expanded, and the chirp multi-carrier multiplexing system has higher universality. The method provided by the invention can be applied to chirp multi-carrier signal generation.

Description

Technical Field

[0001] The present invention belongs to the technical field of chirp multi-carrier signal generation, and in particular relates to a flexible spectral effect chirp multi-carrier signal generation method based on zero insertion and zero padding. Background Art

[0002] With the development of wireless technology, mobile communications have made great progress. Nowadays, people's demand for communication is no longer limited to basic services such as calls and text messages. Various diversified data and multimedia services are increasing day by day, and emerging smart terminal devices are being updated day by day. People's high dependence on communication has led to a rapid increase in communication demand, and the huge amount of data has overwhelmed the limited spectrum resources. On the one hand, there is an increasing demand for communication, and on the other hand, there are increasingly tight frequency resources. Therefore, it is necessary to seek and explore communication technologies with higher spectrum efficiency. Under the premise of limited bandwidth, more efficient transmission technology should be used to meet people's demand for mobile communications as much as possible.

[0003] Orthogonal Frequency Division Multiplexing (OFDM) technology has been widely used and studied as a mainstream communication technology in recent years due to its excellent performance and low implementation complexity. In recent years, with the surge in user data volume, OFDM technology has gradually been unable to meet the huge needs of users. Due to the limitation of subcarrier orthogonality in OFDM, the only way to further improve spectrum efficiency is to increase the modulation order, but high-order modulation systems will be more sensitive to noise and interference. Therefore, the way to improve spectrum efficiency by relaxing orthogonality has gradually become the focus of scholars' research. Spectrally Efficient Frequency Division Multiplexing (SEFDM) system is a multi-carrier waveform system that reduces the spacing of subcarriers in the OFDM system in exchange for higher spectrum efficiency. I. Darwazeh and MRD Rodrigue first proposed the system idea of ​​SEFDM, which is based on the Fast-OFDM system and further improves spectrum efficiency. Analogous to the SEFDM system, which improves spectrum efficiency in the frequency domain by reducing the subcarrier spacing, Faster than Nyquist (FTN), as a single-carrier technology, uses a similar concept in the time domain to break the Nyquist orthogonality criterion and use a faster symbol rate than the orthogonal system, thereby achieving higher spectrum efficiency without increasing the modulation order. Subsequent communication technicians also expanded the research on single-carrier FTN to the multi-carrier field, further improving the theoretical research on FTN.

[0004] Although the above-mentioned non-orthogonal system transmission improves the spectrum efficiency, it is inevitable that there will be interference between subcarriers or between codes, which will affect the overall performance of the system. In order to solve the problem that the inter-subcarrier interference in the SEFDM system affects the system bit error rate performance, some researchers have recently started from the signal basis and replaced the sine and cosine function basis in the traditional non-orthogonal system with chirp signals. Taking advantage of the chirp signal's anti-multipath, anti-Doppler, and low sensitivity to frequency changes, a non-orthogonal chirp multi-carrier multiplexing technology was proposed, which further alleviated the interference between subcarriers and provided better bit error rate performance. However, in the existing non-orthogonal chirp multi-carrier multiplexing technology, due to the limitation of the number of transformation points, the spectrum efficiency it can provide is limited, and the applicability and versatility of the existing system structure are also limited, which is not conducive to digital implementation. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that the spectrum efficiency of the existing non-orthogonal chirp multi-carrier multiplexing technology is limited, and the applicability and versatility of the existing non-orthogonal chirp multi-carrier multiplexing system structure are limited, and a flexible spectrum-efficient chirp multi-carrier signal generation method based on zero insertion and zero padding is proposed.

[0006] The technical solution adopted by the present invention to solve the above technical problems is:

[0007] According to one aspect of the present invention, a method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding, the method specifically comprises the following steps:

[0008] At the transmitter

[0009] Step S1, a signal source generates a random bit stream sequence v of length N×log2M, where N is the number of subcarriers and M is the digital modulation order;

[0010] Step S2: digitally modulate the random bit stream sequence v to obtain a digitally modulated sequence s T , where s T =[s0,s1,s2,...,s N-1 ],s0,s1,s2,...,s N-1 The sequence s T The 0th, 1st, 2nd, ..., N-1th elements in ;

[0011] Step S3: sequence s T Perform serial-to-parallel conversion to obtain s=[s0,s1,s2,...,s N-1 ] T , T represents transpose;

[0012] Step S4: Add Q / α-N zeros to the end of s to obtain the zero-filled result.

[0013] Where Q is the number of time domain samples, and α is the bandwidth compression factor;

[0014] Step S5: Multiply by Θ2 H Get s′, s′=[s′0,s1′,s2′,...,s′ Q / α-1 ] T ;

[0015] Among them, Θ2 H represents the Hermite transpose of the phase rotation factor θ2;

[0016] Step S6: insert b-1 zeros between every two adjacent data points of s′, and add cQ-b(Q / α-1)-1 zeros at the end of s′ to obtain the signal vector

[0017] Wherein, b and c are both positive integers, b<c, and α=b / c;

[0018] Step S7, perform a discrete Fourier inverse transform of cQ points on the signal vector obtained in step S6 to obtain a discrete Fourier inverse transform result, and then remove cQ-Q / α time domain data at the tail of the discrete Fourier inverse transform result, that is, only retain the first Q / α time domain data in the discrete Fourier inverse transform result, and record the vector composed of the retained time domain data as x′=[x′0,x1′,x2′,...,x′ Q / α-1 ] T ;

[0019] Step S8: Multiply x′ by θ1 H Get the signal vector [x0,x1,x2,...,x Q-1 ,..,x Q / α-1 ] T , where Θ1 H represents the Hermite transpose of the phase rotation factor Θ1;

[0020] Step S9: discard the signal vector [x0, x1, x2, ..., x Q-1 ,..,x Q / α-1 ] T The Q(1-α) / α data at the end, the vector composed of the retained data is recorded as x=[x0,x1,x2,...,x Q-1 ] T ;

[0021] Step S10: vector x = [x0, x1, x2, ..., x Q-1 ]T Perform parallel-to-serial conversion to obtain the parallel-to-serial conversion result x T =[x0,x1,x2,...,x Q-1 ], and x T Send to the additive white Gaussian noise channel;

[0022] On the receiving end

[0023] Step R1: The signal received by the receiving end is represented as y T =[y0,y1,y2,...,y Q-1 ], for the received signal y T Perform serial-to-parallel conversion to obtain the signal y=[y0,y1,y2,...,y Q-1 ] T ;

[0024] Step R2: perform zero-padding operation on signal y to obtain a vector [y0,y1,y2,...,y Q-1 ,0,...0] T , that is, add Q(1-α) / α zeros at the end of signal y;

[0025] Step R3: Substitute [y0,y1,y2,...,y Q-1 ,0,...0] T Multiplying by the phase rotation factor θ1, we get the vector y′=[y′0,y1′,y′2,...,y′ Q / α-1 ] T ;

[0026] Step R4: In the vector y′=[y′0,y1′,y′2,...,y′ Q / α-1 ] T b-1 zeros are inserted between every two adjacent data points, and cQ-b(Q / α-1)-1 zeros are added to the tail of the vector y′. The signal vector obtained after inserting and adding zeros is recorded as

[0027]

[0028] Step R5: the signal vector obtained in step R4 Perform a discrete Fourier transform of the cQ point, and record the signal vector composed of the first Q / α time domain data in the discrete Fourier transform result as r′=[r0′,r1′,r2′,...,r Q ' / α-1 ] T ;

[0029] Step R6: The vector r′ obtained in step R5 is Q ' / α-1 ] T Multiply by the phase rotation factor Θ2 to get the vector [r0,r1,r2,...,r Q-1 ,..,r Q / α-1 ] T ;

[0030] Step R7: discard the vector [r0,r1,r2,...,r Q-1 ,..,r Q / α-1 ] T The Q / α-N data at the end are used to obtain the vector to be detected r = [r0, r1, r2, ..., r N-1 ] T ;

[0031] Step R8: Use the detection algorithm to detect the vector r = [r0, r1, r2, ..., r N-1 ] T Processing to obtain the estimated value of s

[0032]

[0033] Step R9: Estimated value Perform digital demodulation to obtain an estimate of the random bit stream sequence v That is, the original bit stream data is restored.

[0034] According to another aspect of the present invention, a method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding, the working process of the method at the transmitting end is:

[0035] Step S1, a signal source generates a random bit stream sequence v of length N×log2M, where N is the number of subcarriers and M is the digital modulation order;

[0036] Step S2: digitally modulate the random bit stream sequence v to obtain a digitally modulated sequence s T , where s T =[s0,s1,s2,...,s N-1 ],s0,s1,s2,...,s N-1 The sequence s T The 0th, 1st, 2nd, ..., N-1th elements in ;

[0037] Step S3: sequence s T Perform serial-to-parallel conversion to obtain s=[s0,s1,s2,...,s N-1 ] T , T represents transpose;

[0038] Step S4: Add Q / α-N zeros to the end of s to obtain the zero-filled result.

[0039] Where Q is the number of time domain samples, and α is the bandwidth compression factor;

[0040] Step S5: Multiply by Θ2 H Get s′, s′=[s′0,s1′,s2′,...,s′ Q / α-1 ] T ;

[0041] Among them, Θ2 H represents the Hermite transpose of the phase rotation factor θ2;

[0042] Step S6: insert b-1 zeros between every two adjacent data points of s′, and add cQ-b(Q / α-1)-1 zeros at the end of s′ to obtain the signal vector

[0043] Wherein, b and c are both positive integers, b<c, and α=b / c;

[0044] Step S7, perform a discrete Fourier inverse transform of cQ points on the signal vector obtained in step S6 to obtain a discrete Fourier inverse transform result, and then remove cQ-Q / α time domain data at the tail of the discrete Fourier inverse transform result, that is, only retain the first Q / α time domain data in the discrete Fourier inverse transform result, and record the vector composed of the retained time domain data as x′=[x′0,x1′,x2′,...,x′ Q / α-1 ] T ;

[0045] Step S8: Multiply x′ by θ1 H Get the signal vector [x0,x1,x2,...,x Q-1 ,..,x Q / α-1 ] T , where Θ1 H represents the Hermite transpose of the phase rotation factor Θ1;

[0046] Step S9: discard the signal vector [x0, x1, x2, ..., x Q-1 ,..,x Q / α-1 ] T The Q(1-α) / α data at the end, the vector composed of the retained data is recorded as x=[x0,x1,x2,...,x Q-1 ] T ;

[0047] Step S10: vector x = [x0, x1, x2, ..., x Q-1 ] T Perform parallel-to-serial conversion to obtain the parallel-to-serial conversion result xT =[x0,x1,x2,...,x Q-1 ], and x T Sent to an additive white Gaussian noise channel.

[0048] The beneficial effects of the present invention are:

[0049] The present invention adds zero insertion and zero padding operations to both ends of the transmitter and receiver after the first phase rotation. By controlling the number of zero insertion and zero padding, the number of DFT and IDFT points can be adjusted to adjust the bandwidth compression factor in the chirp multi-carrier multiplexing system, thereby providing flexible and variable spectrum efficiency, solving the problem of limited spectrum efficiency range caused by the number of operation points must be an integer and a power of 2 in the chirp multi-carrier signal generation process. The structural application scope of the chirp multi-carrier multiplexing system is further expanded, which is more universal, promotes its digital realization, and provides more structural support for the field of high spectrum efficiency transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a workflow diagram of a transmitter of a method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding of the present invention;

[0051] Figure 2 It is a working flow chart of a receiving end of a method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding of the present invention;

[0052] Figure 3 It is a flow chart of a conventional non-orthogonal chirp multi-carrier multiplexing signal generation method;

[0053] Figure 4 is the bit error rate performance under AWGN channel;

[0054] Figure 5 It is the bit error rate performance under frequency selective fading channel. DETAILED DESCRIPTION

[0055] Specific implementation method 1: Combination Figure 1 and Figure 2 The present embodiment describes a method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding, and the method specifically comprises the following steps:

[0056] At the transmitter

[0057] Step S1, a signal source generates a random bit stream sequence v of length N×log2M, where N is the number of subcarriers and M is the digital modulation order;

[0058] Step S2: digitally modulate the random bit stream sequence v to obtain a digitally modulated sequence s T , where s T =[s0,s1,s2,...,s N-1 ],s0,s1,s2,...,s N-1 The sequence s T The 0th, 1st, 2nd, ..., N-1th elements in ;

[0059] Step S3: sequence s T Perform serial-to-parallel conversion to obtain s=[s0,s1,s2,...,s N-1 ] T , T represents transpose;

[0060] Step S4: Add Q / α-N zeros at the end of s (i.e., N-1 Fill the back with zeros) to get the result of filling with 0

[0061] Where Q is the number of time domain samples, and α is the bandwidth compression factor;

[0062] Step S5: Multiply by Θ2 H Get s′, s′=[s′0,s1′,s2′,...,s′ Q / α-1 ] T ;

[0063] Among them, Θ2 H represents the Hermite transpose of the phase rotation factor θ2;

[0064] Step S6: insert b-1 zeros between every two adjacent data points of s′ (i.e. insert b-1 zeros between s0′ and s1′, insert b-1 zeros between s1′ and s2′, and so on), and add cQ-b(Q / α-1)-1 zeros at the end of s′ to obtain the signal vector

[0065] Wherein, b and c are both positive integers, b<c, and α=b / c;

[0066] Step S7, perform an inverse discrete Fourier transform (IDFT) of cQ points on the signal vector obtained in step S6 to obtain an inverse discrete Fourier transform result, and then remove cQ-Q / α time domain data at the end of the inverse discrete Fourier transform result, that is, only retain the first Q / α time domain data in the inverse discrete Fourier transform result, and record the vector composed of the retained time domain data as x′=[x′0,x1′,x2′,...,x′ Q / α-1 ] T ;

[0067] Step S8: Multiply x′ by θ1 H Get the signal vector [x0,x1,x2,...,x Q-1 ,..,x Q / α-1 ] T , where Θ1 H represents the Hermite transpose of the phase rotation factor Θ1;

[0068] Step S9: discard the signal vector [x0, x1, x2, ..., x Q-1 ,..,x Q / α-1 ] T The Q(1-α) / α data at the end, the vector composed of the retained data is recorded as x=[x0,x1,x2,...,x Q-1 ] T ;

[0069] Step S10: vector x = [x0, x1, x2, ..., x Q-1 ] T Perform parallel-to-serial conversion to obtain the parallel-to-serial conversion result x T =[x0,x1,x2,...,x Q-1 ], that is, flexible spectrum effect chirp multi-carrier signal, and x T Send to the additive white Gaussian noise channel;

[0070] On the receiving end

[0071] Step R1: The signal received by the receiving end is represented as y T =[y0,y1,y2,...,y Q-1 ], for the received signal y T Perform serial-to-parallel conversion to obtain the signal y=[y0,y1,y2,...,y Q-1 ] T ;

[0072] Step R2: perform zero-padding operation on signal y to obtain a vector [y0,y1,y2,...,y Q-1 ,0,...0] T , that is, add Q(1-α) / α zeros at the end of signal y;

[0073] Step R3: Substitute [y0,y1,y2,...,y Q-1 ,0,...0] T Multiplying by the phase rotation factor θ1, we get the vector y′=[y′0,y1′,y′2,...,y′ Q / α-1 ] T ;

[0074] Step R4: In the vector y′=[y′0,y1′,y′2,...,y′ Q / α-1 ] T b-1 zeros are inserted between every two adjacent data points, and cQ-b(Q / α-1)-1 zeros are added to the tail of the vector y′. The signal vector obtained after inserting and adding zeros is recorded as

[0075]

[0076] Step R5: the signal vector obtained in step R4 Perform a discrete Fourier transform (DFT) of the cQ point, and record the signal vector composed of the first Q / α time domain data in the discrete Fourier transform result as r′=[r0′,r1′,r2′,...,r Q ' / α-1 ] T ;

[0077] Step R6: The vector r′ obtained in step R5 is Q ' / α-1 ] T Multiply by the phase rotation factor Θ2 to get the vector [r0,r1,r2,...,r Q-1 ,..,r Q / α-1 ] T ;

[0078] Step R7: discard the vector [r0,r1,r2,...,r Q-1 ,..,r Q / α-1 ] T The Q / α-N data at the end are used to obtain the vector to be detected r = [r0, r1, r2, ..., r N-1 ] T ;

[0079] Step R8: Use the detection algorithm to detect the vector r = [r0, r1, r2, ..., r N-1 ] T By processing, the interference between subcarriers can be eliminated and the estimated value of s can be obtained.

[0080] Step R9: Estimated value Perform digital demodulation to obtain an estimate of the random bit stream sequence v That is, the original bit stream data is restored.

[0081] The present invention aims at the problem that the scope of spectrum efficiency is limited due to the conditional restriction of the number of DFT points in the chirp multi-carrier multiplexing system structure, thereby limiting the applicability and versatility of the system structure. A chirp multi-carrier signal generation method with flexible spectrum efficiency is proposed. In the process of using DFT and IDFT, zero insertion and zero padding operations are added. Different bandwidth compression factors are set by independently and flexibly controlling the number of zero insertion and zero padding, thereby expanding the scope of spectrum efficiency that can be provided in the system. Compared with the traditional chirp multi-carrier signal generation method, the signal generation method proposed by the present invention is more flexible and more universal, and further promotes the digital realization of the chirp multi-carrier multiplexing system.

[0082] Specific implementation method 2: This implementation method is different from the specific implementation method 1 in that the number of time domain samples Q is:

[0083] Q=ρN

[0084] Where ρ is the oversampling factor.

[0085] The other steps and parameters are the same as those in the first embodiment.

[0086] Specific implementation method three: This implementation method is different from specific implementation methods one or two in that the bandwidth compression factor α is:

[0087] α=Δf / Δf′

[0088] Wherein, Δf is the subcarrier spacing of the chirp multi-carrier when it is non-orthogonal, and Δf′ is the subcarrier spacing of the chirp multi-carrier when it is orthogonal.

[0089] The other steps and parameters are the same as those in the first or second embodiment.

[0090] Specific implementation mode 4: This implementation mode is different from any one of the specific implementation modes 1 to 3 in that the phase rotation factor θ2 is a diagonal matrix having non-zero elements only on the diagonal line, and the elements on the diagonal line are respectively:

[0091]

[0092] Where n = 0, 1, ... Q / α-1, j is the imaginary unit, e is the base of the natural logarithm, and Θ2(n,n) is the nth element on the diagonal.

[0093] The other steps and parameters are the same as those in Specific Embodiments 1 to 3.

[0094] Specific implementation mode 5: This implementation mode is different from any one of specific implementation modes 1 to 4 in that the phase rotation factor θ1 is a diagonal matrix having non-zero values ​​only on the diagonal line, and the values ​​of the elements on the diagonal line are:

[0095]

[0096] Where m=0,1,...Q / α-1, Θ1(m,m) is the mth element on the diagonal.

[0097] The other steps and parameters are the same as those in Specific Embodiments 1 to 4.

[0098] Specific implementation method 6: This implementation method is different from any one of the specific implementation methods 1 to 5 in that the signal y received by the receiving end T for:

[0099] y T =x T +n T

[0100] Among them, n T is the noise sequence introduced by the signal through the channel.

[0101] The other steps and parameters are the same as those in Specific Implementations 1 to 5.

[0102] Specific implementation method seven: This implementation method is different from any one of specific implementation methods one to six in that the detection algorithm is a maximum likelihood detection method.

[0103] The other steps and parameters are the same as those in Specific Embodiments 1 to 6.

[0104] The transmission process of the Chirp multi-carrier multiplexing system is similar to that of the SEFDM system, both of which are block transmission forms, and a cyclic prefix (CP) is inserted at the beginning of each symbol before entering the fading channel to reduce the impact of multipath interference. After removing the CP at the receiving end, the received signal sampling can be expressed as:

[0105] y=ΗΘ1 H F H Θ2 H s+n=HΦs+n

[0106] In particular, when H is the identity matrix, the channel is an additive Gaussian white noise channel. Then according to Figure 2 By following the process shown in , we can get:

[0107] r=Θ2FΘ1ΗΘ1 H F H Θ2 H s+n′=Φ H ΗΦs+n′

[0108] If the detection algorithm selects maximum likelihood detection, the detection is performed according to the following formula to obtain In this way, the originally transmitted bit stream sequence is restored;

[0109]

[0110] Specific implementation method eight: Combination Figure 1 The present embodiment describes a method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding, and the working process of the method at the transmitting end is as follows:

[0111] Step S1, a signal source generates a random bit stream sequence v of length N×log2M, where N is the number of subcarriers and M is the digital modulation order;

[0112] Step S2: digitally modulate the random bit stream sequence v to obtain a digitally modulated sequence s T , where s T =[s0,s1,s2,...,s N-1 ],s0,s1,s2,...,s N-1 The sequence s T The 0th, 1st, 2nd, ..., N-1th elements in ;

[0113] Step S3: sequence s T Perform serial-to-parallel conversion to obtain s=[s0,s1,s2,...,s N-1 ] T , T represents transpose;

[0114] Step S4: Add Q / α-N zeros at the end of s (i.e., N-1 Fill the back with zeros) to get the result of filling with 0

[0115] Where Q is the number of time domain samples, and α is the bandwidth compression factor;

[0116] Step S5: Multiply by Θ2 H Get s′, s′=[s′0,s1′,s2′,...,s′ Q / α-1 ] T ;

[0117] Among them, Θ2 H represents the Hermite transpose of the phase rotation factor θ2;

[0118] Step S6: insert b-1 zeros between every two adjacent data points of s′ (i.e. insert b-1 zeros between s0′ and s1′, insert b-1 zeros between s1′ and s2′, and so on), and add cQ-b(Q / α-1)-1 zeros at the end of s′ to obtain the signal vector

[0119] Wherein, b and c are both positive integers, b<c, and α=b / c;

[0120] Step S7, perform an inverse discrete Fourier transform (IDFT) of cQ points on the signal vector obtained in step S6 to obtain an inverse discrete Fourier transform result, and then remove cQ-Q / α time domain data at the end of the inverse discrete Fourier transform result, that is, only retain the first Q / α time domain data in the inverse discrete Fourier transform result, and record the vector composed of the retained time domain data as x′=[x′0,x1′,x2′,...,x′ Q / α-1 ] T ;

[0121] Step S8: Multiply x′ by θ1 H Get the signal vector [x0,x1,x2,...,x Q-1 ,..,x Q / α-1 ] T , where Θ1 H represents the Hermite transpose of the phase rotation factor Θ1;

[0122] Step S9: discard the signal vector [x0, x1, x2, ..., x Q-1 ,..,x Q / α-1 ] T The Q(1-α) / α data at the end, the vector composed of the retained data is recorded as x=[x0,x1,x2,...,x Q-1 ] T ;

[0123] Step S10: vector x = [x0, x1, x2, ..., x Q-1 ] T Perform parallel-to-serial conversion to obtain the parallel-to-serial conversion result x T =[x0,x1,x2,...,x Q-1 ], that is, flexible spectrum effect chirp multi-carrier signal, and x T Sent to an additive white Gaussian noise channel.

[0124] The numerical selection process of b and c is independent. By setting b and c, the transmitter expands the selectable range of the bandwidth compression factor α in the chirp multi-carrier multiplexing system in two different dimensions, which is convenient for providing flexible and variable spectrum efficiency in the chirp multi-carrier multiplexing system and expanding the scope of application of the chirp multi-carrier multiplexing system structure.

[0125] Specific implementation method 9: This implementation method is different from specific implementation method 8 in that the number of time domain samples Q is:

[0126] Q=ρN

[0127] Where ρ is the oversampling factor;

[0128] The bandwidth compression factor α is:

[0129] α=Δf / Δf′

[0130] Wherein, Δf is the subcarrier spacing of the chirp multi-carrier when it is non-orthogonal, and Δf′ is the subcarrier spacing of the chirp multi-carrier when it is orthogonal.

[0131] The other steps and parameters are the same as those in the eighth embodiment.

[0132] Specific implementation method ten: This implementation method is different from specific implementation methods eight or nine in that the phase rotation factor θ2 is a diagonal matrix having non-zero elements only on the diagonal line, and the elements on the diagonal line are respectively:

[0133]

[0134] Where n = 0, 1, ... Q / α-1, j is the imaginary unit, e is the base of the natural logarithm, Θ2(n,n) is the nth element on the diagonal;

[0135] The phase rotation factor θ1 is a diagonal matrix with non-zero values ​​only on the diagonal, and the values ​​of the elements on the diagonal are:

[0136]

[0137] Where m=0,1,...Q / α-1, Θ1(m,m) is the mth element on the diagonal.

[0138] The other steps and parameters are the same as those in the eighth or ninth embodiment.

[0139] Equivalence Proof

[0140] The general mathematical expression of a multi-carrier signal with a chirp signal as the carrier is:

[0141]

[0142] Where T is the duration of the symbol, N is the number of subcarriers, and s n,lis the modulation symbol modulated on the nth subcarrier of the lth multicarrier symbol, g(t) is the pulse shaping function, and the rectangular window function is selected here. α=ΔfT≤1, α is the bandwidth compression factor, which measures the spectral efficiency provided by the chirp multicarrier multiplexing system; Δf is the distance between two adjacent subcarriers, that is, the smaller Δf is, the smaller α is, and the higher the spectral efficiency is. In particular, when α=1, the system degenerates into a traditional OCDM system, and when α<1, it is a non-orthogonal chirp multicarrier multiplexing system.

[0143] For convenience, only l = 0 is considered. The above continuous signal is sampled at intervals of T / Q, that is, t = kT / Q, where Q = ρN, and ρ is an oversampling factor.

[0144] The discrete form of this signal is:

[0145]

[0146] The signal form of formula (2) can be completed by discrete Fourier transform and two phase rotation operations. The transmitter structure of the existing chirp multi-carrier multiplexing system is as follows: Figure 3 As shown, the specific process is: multiply by Θ2 H , perform Q / α point IDFT, multiply by Θ1 H .

[0147] Figure 3 In the process shown, the DFT input Q / α must be a positive integer. If the fast DFT algorithm FFT is used, the number of points must generally be a power of 2. This limits the selection range of α, that is, the selection range of spectrum efficiency, and also reduces the practicality of the system structure.

[0148] The operation of the IDFT part can be expressed as:

[0149]

[0150] Among them, U=[u′0,u1′,...u′ Q-1 ], any multi-carrier signal form similar to equation (3) can be expressed as:

[0151]

[0152] Among them, U′=[u0′,u1′,...u′ Q-1 ,...u′ L-1 ], is the normalization factor, Ω is the L×L IDFT matrix, whose elements can be expressed as:

[0153]

[0154] According to the transmitter structure proposed in the present invention, the input signal of IDFT can be expressed as:

[0155]

[0156] Among them, S is Figure 3 The input signal of IDFT is 0≤i<L, L=cQ, α=b / c, b, c are both positive integers and b<c. According to the signal form, U′=[u′0,u1′,...u′ Q-1 ,...u′ L-1 ] can also be expressed as:

[0157]

[0158] Let l = bl′, Then we get:

[0159]

[0160] According to the above proof process, the process of the chirp multi-carrier multiplexing signal generation method can be expressed as:

[0161]

[0162] So far, the equivalence of a flexible spectral effect chirp multi-carrier signal generation method based on zero insertion and zero padding proposed in the present invention and a traditional chirp multi-carrier multiplexing signal generation method is explained, and they can be flexibly selected according to different requirements.

[0163] Experimental Section

[0164] The bit error rate performance of the transmitter and receiver models of the present invention is simulated and analyzed in AWGN channels and frequency selective fading channels, and compared with SEFDM systems and traditional non-orthogonal chirp multi-carrier multiplexing systems. During the simulation, the number of subcarriers is set to 8, 4QAM modulation is adopted, the maximum likelihood detection method is adopted at the receiving end, and different b and c are set to complete the setting of different bandwidth compression factors α.

[0165] Depend on Figure 4 and Figure 5It can be seen that under different selections of bandwidth compression factors, the bit error rate performance of the chirp multi-carrier signal generation method proposed in the present invention is consistent with that of the traditional non-orthogonal chirp multi-carrier multiplexing system, and is superior to the SEFDM system. It can provide higher spectrum efficiency without other performance losses and complex calculation costs. Moreover, the method proposed in the present invention can control the bandwidth compression factor α by independently setting the numerical values ​​of b and c, so as to provide flexible spectrum efficiency and enhance the practicality of the system structure. When providing the same spectrum efficiency, the above results reflect the performance advantages of the multi-carrier system with the chirp signal as the carrier. At the same time, simulation verifies the feasibility of the method proposed in the present invention.

[0166] The structure of the non-orthogonal chirp multi-carrier multiplexing system of the present invention can be well compatible with the SEFDM system, and the smooth switching and mutual compatibility of the two systems can be achieved, and only the phase rotation factor needs to be added. Moreover, the compatibility of the system structure of the chirp multi-carrier multiplexing system in the orthogonal and non-orthogonal cases is achieved, and the switching of the orthogonal and non-orthogonal modes can be completed by setting different parameters.

[0167] The above calculation examples of the present invention are only used to explain the calculation model and calculation process of the present invention in detail, and are not intended to limit the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding, characterized in that: The method specifically comprises the following steps: At the transmitter Step S1, a signal source generates a random bit stream sequence v of length N×log2M, where N is the number of subcarriers and M is the digital modulation order; Step S2: digitally modulate the random bit stream sequence v to obtain a digitally modulated sequence s T , where s T =[s0,s1,s2,...,s N-1 ],s0,s1,s2,...,s N-1 The sequence s T The 0th, 1st, 2nd, ..., N-1th elements in ; Step S3: sequence s T Perform serial-to-parallel conversion to obtain s=[s0,s1,s2,...,s N-1 ] T , T represents transpose; Step S4: Add Q / α-N zeros to the end of s to obtain the zero-filled result. Where Q is the number of time domain samples, and α is the bandwidth compression factor; Step S5: Multiply by Θ2 H Get s′, s′=[s′0,s1′,s2′,...,s′ Q / α-1 ] T ; Among them, Θ2 H represents the Hermite transpose of the phase rotation factor θ2; Step S6: insert b-1 zeros between every two adjacent data points of s′, and add cQ-b(Q / α-1)-1 zeros at the end of s′ to obtain the signal vector Wherein, b and c are both positive integers, b<c, and α=b / c; Step S7, perform a discrete Fourier inverse transform of cQ points on the signal vector obtained in step S6 to obtain a discrete Fourier inverse transform result, and then remove cQ-Q / α time domain data at the tail of the discrete Fourier inverse transform result, that is, only retain the first Q / α time domain data in the discrete Fourier inverse transform result, and record the vector composed of the retained time domain data as x′=[x′0,x1′,x2′,...,x′ Q / α-1 ] T ; Step S8: Multiply x′ by θ1 H Get the signal vector [x0,x1,x2,...,x Q-1 ,..,x Q / α-1 ] T , where Θ1 H represents the Hermite transpose of the phase rotation factor Θ1; Step S9: discard the signal vector [x0, x1, x2, ..., x Q-1 ,..,x Q / α-1 ] T The Q(1-α) / α data at the end, the vector composed of the retained data is recorded as x=[x0,x1,x2,...,x Q-1 ] T ; Step S10: vector x = [x0, x1, x2, ..., x Q-1 ] T Perform parallel-to-serial conversion to obtain the parallel-to-serial conversion result x T =[x0,x1,x2,...,x Q-1 ], and x T Send to the additive white Gaussian noise channel; On the receiving end Step R1: The signal received by the receiving end is represented as y T =[y0,y1,y2,...,y Q-1 ], for the received signal y T Perform serial-to-parallel conversion to obtain the signal y=[y0,y1,y2,...,y Q-1 ] T ; Step R2: perform zero-padding operation on signal y to obtain a vector [y0,y1,y2,...,y Q-1 ,0,...0] T , that is, add Q(1-α) / α zeros at the end of signal y; Step R3: Substitute [y0,y1,y2,...,y Q-1 ,0,...0] T Multiplying by the phase rotation factor θ1, we get the vector y′=[y′0,y1′,y′2,...,y′ Q / α-1 ] T ; Step R4: In the vector y′=[y′0,y1′,y′2,...,y′ Q / α-1 ] T b-1 zeros are inserted between every two adjacent data points, and cQ-b(Q / α-1)-1 zeros are added to the tail of the vector y′. The signal vector obtained after inserting and adding zeros is recorded as Step R5: the signal vector obtained in step R4 Perform a discrete Fourier transform of the cQ point, and record the signal vector composed of the first Q / α time domain data in the discrete Fourier transform result as r′=[r0′,r1′,r2′,...,r Q ' / α-1 ] T ; Step R6: The vector r′ obtained in step R5 is Q ' / α-1 ] T Multiply by the phase rotation factor Θ2 to get the vector [r0,r1,r2,...,r Q-1 ,..,r Q / α-1 ] T ; Step R7: discard the vector [r0,r1,r2,...,r Q-1 ,..,r Q / α-1 ] T The Q / α-N data at the end are used to obtain the vector to be detected r = [r0, r1, r2, ..., r N-1 ] T ; Step R8: Use the detection algorithm to detect the vector r = [r0, r1, r2, ..., r N-1 ] T Processing to obtain the estimated value of s Step R9: Estimated value Perform digital demodulation to obtain an estimate of the random bit stream sequence v That is, the original bit stream data is restored.

2. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 1, characterized in that: The number of time domain samples Q is: Q=ρN Where ρ is the oversampling factor.

3. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 2, characterized in that: The bandwidth compression factor α is: α=Δf / Δf′ Wherein, Δf is the subcarrier spacing of the chirp multi-carrier when it is non-orthogonal, and Δf′ is the subcarrier spacing of the chirp multi-carrier when it is orthogonal.

4. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 3, characterized in that: The phase rotation factor θ2 is a diagonal matrix with non-zero elements only on the diagonal, and the elements on the diagonal are: Where n = 0, 1, ... Q / α-1, j is the imaginary unit, e is the base of the natural logarithm, and Θ2(n,n) is the nth element on the diagonal.

5. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 4, characterized in that: The phase rotation factor θ1 is a diagonal matrix with non-zero values ​​only on the diagonal, and the values ​​of the elements on the diagonal are: Where m=0,1,...Q / α-1, Θ1(m,m) is the mth element on the diagonal.

6. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 5, characterized in that: The signal y received by the receiving end T for: and T =x T +n T Among them, n T is the noise sequence introduced by the signal through the channel.

7. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 6, characterized in that: The detection algorithm is a maximum likelihood detection method.

8. A method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding, characterized in that: The working process of the method at the sending end is: Step S1, a signal source generates a random bit stream sequence v of length N×log2M, where N is the number of subcarriers and M is the digital modulation order; Step S2: digitally modulate the random bit stream sequence v to obtain a digitally modulated sequence s T , where s T =[s0,s1,s2,...,s N-1 ],s0,s1,s2,...,s N-1 The sequence s T The 0th, 1st, 2nd, ..., N-1th elements in ; Step S3: sequence s T Perform serial-to-parallel conversion to obtain s=[s0,s1,s2,...,s N-1 ] T , T represents transpose; Step S4: Add Q / α-N zeros to the end of s to obtain the zero-filled result. Where Q is the number of time domain samples, and α is the bandwidth compression factor; Step S5: Multiply by Θ2 H Get s′, s′=[s′0,s1′,s2′,...,s′ Q / α-1 ] T ; Among them, Θ2 H represents the Hermite transpose of the phase rotation factor θ2; Step S6: insert b-1 zeros between every two adjacent data points of s′, and add cQ-b(Q / α-1)-1 zeros at the end of s′ to obtain the signal vector Wherein, b and c are both positive integers, b<c, and α=b / c; Step S7, perform a discrete Fourier inverse transform of cQ points on the signal vector obtained in step S6 to obtain a discrete Fourier inverse transform result, and then remove cQ-Q / α time domain data at the tail of the discrete Fourier inverse transform result, that is, only retain the first Q / α time domain data in the discrete Fourier inverse transform result, and record the vector composed of the retained time domain data as x′=[x′0,x1′,x2′,...,x′ Q / α-1 ] T ; Step S8: Multiply x′ by θ1 H Get the signal vector [x0,x1,x2,...,x Q-1 ,..,x Q / α-1 ] T , where Θ1 H represents the Hermite transpose of the phase rotation factor Θ1; Step S9: discard the signal vector [x0, x1, x2, ..., x Q-1 ,..,x Q / α-1 ] T The Q(1-α) / α data at the end, the vector composed of the retained data is recorded as x=[x0,x1,x2,...,x Q-1 ] T ; Step S10: vector x = [x0, x1, x2, ..., x Q-1 ] T Perform parallel-to-serial conversion to obtain the parallel-to-serial conversion result x T =[x0,x1,x2,...,x Q-1 ], and x T Sent to an additive white Gaussian noise channel.

9. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 8, characterized in that: The number of time domain samples Q is: Q=ρN Where ρ is the oversampling factor; The bandwidth compression factor α is: α=Δf / Δf′ Wherein, Δf is the subcarrier spacing of the chirp multi-carrier when it is non-orthogonal, and Δf′ is the subcarrier spacing of the chirp multi-carrier when it is orthogonal.

10. The method for generating a flexible spectral effect chirp multi-carrier signal based on zero insertion and zero padding according to claim 9, characterized in that: The phase rotation factor θ2 is a diagonal matrix with non-zero elements only on the diagonal, and the elements on the diagonal are: Where n = 0, 1, ... Q / α-1, j is the imaginary unit, e is the base of the natural logarithm, Θ2(n,n) is the nth element on the diagonal; The phase rotation factor θ1 is a diagonal matrix with non-zero values ​​only on the diagonal, and the values ​​of the elements on the diagonal are: Where m=0,1,...Q / α-1, Θ1(m,m) is the mth element on the diagonal.

Images