CN107483378B - FTN block transmission method, transmitter, receiver and system based on DFT - Google Patents

Abstract

The present invention provides a DFT-based FTN block transmission method, a transmitter, a receiver and a system. Based on discrete Fourier transform and frequency domain windowing technology, the traditional waveform generation and reception matching filtering method based on time domain convolution is equivalently realized. Then, through simple frequency domain equalization, channel multipath interference and self-ISI are eliminated in steps, thereby simplifying the implementation structure of baseband processing at the transmitter and receiver. In addition, by further utilizing the high efficiency of DFT calculation, the present invention can achieve the effect of effectively reducing the implementation complexity of the transmitter and receiver of the system.

Description

DFT-based FTN block transmission method, transmitter, receiver and system

technical field

The present invention relates to the field of communication technologies, and in particular, to a DFT-based FTN block transmission method, transmitter, receiver and system.

Background technique

With the explosive growth of wireless devices in recent years, especially the rapid popularization of smart devices capable of transmitting high-speed multimedia streams, wireless data traffic has shown a trend of exponential growth. Under the realistic dilemma of limited total amount of spectrum resources, finding a transmission technology with higher spectral efficiency is a crucial link in the design of future-oriented wireless communication systems. One of the potential technologies is Faster than Nyquist (FTN), which allows data transmission over the same bandwidth at a code rate higher than the Nyquist rate to improve band utilization rate, thus triggering a research boom.

The Nyquist criterion requires signals to be orthogonal to each other to avoid a decrease in the accuracy of the decision at the receiving end caused by inter-symbol interference (English: Inter-symbol interference, ISI for short), and then the cost of ensuring service quality is sacrificing spectral efficiency. FTN achieves higher data transmission rates by introducing non-orthogonal signals from ISI, but after the concept of FTN was proposed by Mazo in the 1970s, the main reason why traditional communication system design still follows Nyquist's first criterion is that The hardware circuit cannot implement the high-complexity algorithm that the receiver needs to eliminate ISI. Thanks to the rapid development of semiconductor and integrated circuit technology, hardware circuits can implement algorithms of higher complexity, and research on FTN has become a hot spot again in recent years.

At present, the research on FTN transmission technology is mainly limited to simple binary modulation. Although the FTN transmission technology has verified its advantages in non-binary and high-order modulation, the simulation assumes that the channel is an additive white Gaussian noise channel, and does not consider the influence of channel fading, and its practicability is still limited. The design of the multi-carrier FTN transceiver based on spectrally efficient frequency division multiplexing (English: spectrally efficient frequency division multiplexing, SEFDM for short) has a situation where the peak-to-average ratio is too high, which will cause waste of transmission energy. In addition, in order to eliminate the intersymbol interference in the received signal, in terms of detection algorithm, the maximum likelihood sequence detection (English: Maximum Likelihood Sequence Detection, referred to as: MLSD) algorithm can obtain the best performance, but its too high to achieve The complexity makes it difficult to apply in practice. If the received symbol is regarded as the result of convolutional coding, it can be detected by the Viterbi algorithm or the BCJR algorithm, but the increase in the number of states caused by the shortening of the interval between symbols will greatly improve the above two. The complexity of the algorithm. In addition, due to the introduction of ISI, the super-Nyquist system will increase the implementation complexity compared with the Nyquist system in terms of synchronization, channel estimation, and equalization.

To sum up, in view of the problems existing in the existing FTN transmission technology in terms of modulation mode, applicable channel model and ISI elimination algorithm complexity, how to reduce the overall implementation complexity of the super-Nyquist transmission system suitable for multipath fading channels? It is a technical problem that those skilled in the art need to solve urgently.

SUMMARY OF THE INVENTION

In view of the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a DFT-based FTN block transmission method, transmitter, receiver and system, which are used to solve the high complexity of super-Nyquist transmission in the prior art. The problem.

In order to achieve the above object and other related objects, according to the first aspect of the present invention, an embodiment of the present invention provides a DFT-based FTN block transmission method, which is applied to a transmitter, and the method includes:

dividing the modulation symbol sequence into a plurality of symbol data blocks;

Perform discrete Fourier transform on each symbol data block to obtain a first signal sequence;

cyclically extending the first signal sequence into a second signal sequence;

Generate a window function according to a preset shaping filter;

using the window function to perform a windowing operation on the second signal sequence;

Perform inverse discrete Fourier transform on the second signal sequence after the windowing operation to obtain a cyclic signal sequence;

A guard interval is added to the cyclic signal sequence to obtain an output signal sequence, which is sent to the receiver.

Optionally, the lengths of the multiple symbol data blocks are all the same.

Optionally, the length of the second signal sequence is greater than the length of the first signal sequence.

Optionally, cyclically extending the first signal sequence into a second signal sequence includes:

Make the n×D+kth data symbol of the second signal sequence equal to the k data symbol of the first signal sequence, where D is the length of the symbol data block, k is a natural number less than or equal to D-1, n is a natural number less than or equal to N _FTN -1, N _FTN is the time shift interval adopted by the super-Nyquist shaping filtering of the symbol data block; and, the length of the second signal sequence is the symbol data block The product of the length of the time shift interval.

Optionally, the time shift interval is smaller than the upsampling rate of the shaping filter.

Optionally, generate a window function according to a preset shaping filter, including:

According to the length of the second signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the window function.

Optionally, use the window function to perform a windowing operation on the second signal sequence, including:

A multiplication factor operation is performed on the second signal sequence and the window function.

Optionally, a guard interval is added to the cyclic signal sequence to obtain an output signal sequence, including:

The guard interval is added at the head or tail of the cyclic signal sequence, and the length of the guard interval is greater than or equal to the maximum delay extension length of the channel.

Optionally, when adding the guard interval to the header of the cyclic signal sequence, copy and add the data block symbol corresponding to the length of the guard interval at the end of the cyclic signal sequence to the header of the cyclic signal sequence .

According to a second aspect of the present invention, an embodiment of the present invention provides a DFT-based FTN block transmission method, which is applied to a receiver, and the method includes the following steps:

receiving the output signal sequence sent by the transmitter, and removing the guard interval in the output signal sequence to obtain the input signal sequence;

performing discrete Fourier transform on the input signal sequence to obtain a third signal sequence;

performing a first frequency domain equalization operation on the third signal sequence to obtain a first equalized signal sequence;

generate filter coefficients;

performing a matched filtering operation on the first equalized signal sequence according to the filter coefficient to obtain a matched filtered signal sequence;

A cyclic accumulation operation is performed on the matched filter signal sequence to obtain a fourth signal sequence;

generate equalization coefficients;

performing a second frequency domain equalization operation on the fourth signal sequence according to the equalization coefficient to obtain a second equalized signal sequence;

Inverse discrete Fourier transform is performed on the second equalized signal sequence to obtain a transformed output signal sequence.

Optionally, receive the output signal sequence sent by the transmitter, and remove the guard interval in the output signal sequence to obtain the input signal sequence, including:

acquiring guard interval configuration information sent by the transmitter, where the guard interval configuration information at least carries the setting position and length of the guard interval and the length of the cyclic signal sequence;

Remove the guard interval in the output signal sequence according to the setting position and length of the guard interval;

The length of the input signal sequence is equal to the length of the cyclic signal sequence.

Optionally, the generating filter coefficients includes:

acquiring shaping filter configuration information sent by the transmitter, where the shaping filter configuration information at least carries the type of shaping filter;

According to the length of the input signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the filter coefficients.

Optionally, a cyclic accumulation operation is performed on the matched filtered signal sequence to obtain a fourth signal sequence, including:

Obtaining cyclic accumulation configuration information, the cyclic accumulation configuration information carries at least the length of the symbol data block and the time shift interval;

Accumulate the N _FTN data symbols of the k+q×D-th signal sequence as the k-th data symbol of the fourth signal sequence, where Q is the length of the input signal sequence, and D is the length of the symbol data block , k is a natural number less than or equal to D-1, q is a natural number less than or equal to N _FTN -1.

Optionally, the length of the fourth signal sequence is equal to the length of the symbol data block.

Optionally, the generating an equalization coefficient includes:

obtaining equalization coefficient configuration information, where the equalization coefficient configuration information at least carries the type of the shaping filter and the time shift interval;

According to the shaping filter corresponding to the type and the time shift interval, through a cyclic autocorrelation operation, a filter cyclic autocorrelation function is generated;

According to the length of the symbol data block, discrete Fourier transform is performed on the filter cyclic autocorrelation function to obtain the equalization coefficient;

The length of the equalization coefficients is equal to the length of the symbol data block.

Optionally, the second equalization module is used for,

A vector division operation of the fourth signal sequence and the equalization coefficient is calculated to obtain a second equalized signal sequence.

According to a third aspect of the present invention, an embodiment of the present invention provides a transmitter, the transmitter comprising:

a data block division module, which is used to divide the modulation symbol sequence into a plurality of symbol data blocks;

The discrete Fourier transform module is used to perform discrete Fourier transform on each symbol data block to obtain the first signal sequence;

a cyclic extension module, configured to cyclically extend the first signal sequence into a second signal sequence;

a window function generation module for generating a window function according to a preset shaping filter;

a windowing module for performing a windowing operation on the second signal sequence using the window function;

an inverse discrete Fourier transform module, configured to perform an inverse discrete Fourier transform on the second signal sequence after the windowing operation to obtain a cyclic signal sequence;

A guard interval module, configured to add a guard interval to the cyclic signal sequence, obtain an output signal sequence, and send it to the receiver.

Optionally, the lengths of the plurality of symbol data blocks obtained by dividing by the data block dividing module are all the same.

Optionally, the cyclic expansion module is used for,

Make the n×D+kth data symbol of the second signal sequence equal to the k data symbol of the first signal sequence, where D is the length of the symbol data block, k is a natural number less than or equal to D-1, n is a natural number less than or equal to N _FTN -1, N _FTN is the time shift interval adopted by the super-Nyquist shaping filtering of the symbol data block; and, the length of the second signal sequence is the symbol data block The product of the length of the time shift interval.

Optionally, the time shift interval is smaller than the upsampling rate of the shaping filter.

Optionally, the window function generation module is used for,

According to the length of the second signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the window function.

Optionally, the windowing module is used to,

A multiplication factor operation is performed on the second signal sequence and the window function.

Optionally, the guard interval module is configured to add the guard interval to the head or tail of the cyclic signal sequence, and the length of the guard interval is greater than or equal to the maximum delay extension length of the channel.

Optionally, the guard interval module is configured to, when adding the guard interval to the header of the cyclic signal sequence, copy the data block symbol located at the tail of the cyclic signal sequence and corresponding to the length of the guard interval and added to the header of the cyclic signal sequence.

According to a fourth aspect of the present invention, an embodiment of the present invention provides a receiver, the receiver comprising:

a guard interval removing module, configured to receive the output signal sequence sent by the transmitter, and remove the guard interval in the output signal sequence to obtain the input signal sequence;

a discrete Fourier transform module, configured to perform discrete Fourier transform on the input signal sequence to obtain a third signal sequence;

a first equalization module, configured to perform a first frequency domain equalization operation on the third signal sequence to obtain a first equalized signal sequence;

A filter coefficient generation module for generating filter coefficients;

a matched filtering module, configured to perform a matched filtering operation on the first equalized signal sequence according to the filter coefficient to obtain a matched filtered signal sequence;

The cyclic accumulation module is used to perform cyclic accumulation operation on the matched filter signal sequence to obtain the fourth signal sequence;

an equalization coefficient generation module for generating equalization coefficients;

a second equalization module, configured to perform a second frequency domain equalization operation on the fourth signal sequence according to the equalization coefficient to obtain a second equalized signal sequence;

an inverse discrete Fourier transform module, configured to perform an inverse discrete Fourier transform on the second equalized signal sequence to obtain a transformed output signal sequence.

Optionally, the guard interval removal module is used for,

acquiring guard interval configuration information sent by the transmitter, where the guard interval configuration information at least carries the setting position and length of the guard interval and the length of the cyclic signal sequence;

Remove the guard interval in the output signal sequence according to the setting position and length of the guard interval;

The length of the input signal sequence is equal to the length of the cyclic signal sequence.

Optionally, the filter coefficient generation module is used for,

acquiring shaping filter configuration information sent by the transmitter, where the shaping filter configuration information at least carries the type of shaping filter;

According to the length of the input signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the filter coefficients.

Optionally, the cyclic accumulation module is used for,

Obtaining cyclic accumulation configuration information, the cyclic accumulation configuration information carries at least the length of the symbol data block and the time shift interval;

Accumulate the N _FTN data symbols of the k+q×D-th signal sequence as the k-th data symbol of the fourth signal sequence, where Q is the length of the input signal sequence, and D is the length of the symbol data block , k is a natural number less than or equal to D-1, q is a natural number less than or equal to N _FTN -1.

Optionally, the length of the fourth signal sequence is equal to the length of the symbol data block.

Optionally, the equalization coefficient generation module is used to:

obtaining equalization coefficient configuration information, where the equalization coefficient configuration information at least carries the type of the shaping filter and the time shift interval;

According to the shaping filter corresponding to the type and the time shift interval, through a cyclic autocorrelation operation, a filter cyclic autocorrelation function is generated;

According to the length of the symbol data block, discrete Fourier transform is performed on the filter cyclic autocorrelation function to obtain the equalization coefficient;

The length of the equalization coefficients is equal to the length of the symbol data block.

Optionally, the second equalization module is used for,

A vector division operation of the fourth signal sequence and the equalization coefficient is calculated to obtain a second equalized signal sequence.

According to a fifth aspect of the present invention, an embodiment of the present invention provides a transmission system, including the transmitter described in the foregoing embodiment and the receiver described in the foregoing embodiment.

As mentioned above, a DFT-based FTN block transmission method, transmitter, receiver and system of the present invention have the following beneficial effects: based on discrete Fourier transform and frequency-domain windowing technology, it is equivalent to realize traditional time-domain convolution-based The waveform generation and reception matched filtering method is adopted, and then through simple frequency domain equalization, the channel multipath interference and self-ISI are eliminated step by step, thereby simplifying the implementation structure of baseband processing at the transmitter and receiver. In addition, by further utilizing the high efficiency of DFT calculation, the present invention can effectively reduce the realization complexity of the system.

Description of drawings

FIG. 1 shows a schematic flowchart of a method for transmitting a DFT-based FTN block at the transmitter side according to an embodiment of the present invention.

FIG. 2 shows a schematic flowchart of a DFT-based FTN block transmission method at the receiver side according to an embodiment of the present invention.

FIG. 3 shows a schematic flowchart of a method for removing a guard interval according to an embodiment of the present invention.

FIG. 4 shows a schematic flowchart of a method for generating a filter coefficient according to an embodiment of the present invention.

FIG. 5 is a schematic flowchart of a cyclic accumulation operation according to an embodiment of the present invention.

FIG. 6 shows a schematic flowchart of a method for generating an equalization coefficient according to an embodiment of the present invention.

FIG. 7 is a schematic structural diagram of a transmitter according to an embodiment of the present invention.

FIG. 8 is a schematic structural diagram of a receiver according to an embodiment of the present invention.

FIG. 9 shows a schematic structural diagram of a DFT-based FTN block transmission system according to an embodiment of the present invention.

FIG. 10 is a schematic diagram showing the results of the structure and performance of a simplified implementation of the frequency domain of an FTN-BT system according to an embodiment of the present invention.

Detailed ways

The embodiments of the present invention are described below through specific specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention.

See Figures 1 through 10. It should be noted that the diagrams provided in this embodiment are only to illustrate the basic concept of the present invention in a schematic way, so the diagrams only show the components related to the present invention rather than the number, shape and the number of components in actual implementation. For dimension drawing, the type, quantity and proportion of each component can be arbitrarily changed in actual implementation, and the component layout may also be more complicated.

Referring to FIG. 1, it is a schematic flowchart of a DFT-based FTN block transmission method on the transmitter side provided by an embodiment of the present invention. As shown in FIG. 1, the embodiment of the present invention shows the process of implementing the transmission method on the transmitter side:

Step S101: The modulation symbol sequence is divided into a plurality of symbol data blocks.

In an exemplary embodiment, the modulation symbol sequence may be represented as {x(n), n=0, 1, 2....}. In the embodiment of the present invention, the modulation symbol sequence is divided into a plurality of symbol data blocks, and the length of each symbol data block may be equal to D. In this way, a plurality of symbol data blocks can form a symbol data block sequence. A symbol data block can be exemplarily represented as {a(d), d=0,1,2....D-1}. Since the operation of the transmitter on each symbol data block is the same and independent, the subsequent steps will be described in detail by taking the processing process of one symbol data block as an example.

Of course, it should be noted that the lengths of the symbol data blocks may also be different, and the division rules of the symbol data blocks may be preset between the receiver and the transmitter, or, during the communication process between the receiver and the transmitter, The transmitter can send the division rule of the symbol data block to the receiver, so that the receiver and the transmitter have a unified symbol data block.

Step S102: Perform discrete Fourier transform on each symbol data block to obtain a first signal sequence.

According to the symbol data block {a(d), d=0, 1, 2, L, D-1} obtained in step S101, perform discrete Fourier transform at point D on the symbol data block, and output the first signal sequence {b(k),k=0,1,2,L,D-1}. The calculation formula of the first signal sequence is as follows:

Among them, D is the length of the symbol data block.

Step S103: Cyclic extending the first signal sequence into a second signal sequence.

In this embodiment of the present invention, a specific process of the transmitter cyclically extending the first signal sequence into the second signal sequence may be: making the n×D+k th data symbol of the second signal sequence equal to the first signal A sequence of k data symbols, where D is the length of the symbol data block, k is a natural number less than or equal to D-1, n is a natural number less than or equal to N _FTN -1, and N _FTN is the symbol data block. and the length of the second signal sequence is the product of the length of the symbol data block and the time shift interval.

Specifically, the first signal sequence {b(k), k=0, 1, 2, . =0,1,2,L,Q-1}, that is, the second signal sequence. The expression of the second signal sequence {c(k),k=0,1,2,L,Q-1} is as follows:

{c(nD+k)=b(k),k=0,L,D-1,n=0,L,N _FTN -1},

Among them, N _FTN is the time shift interval adopted by the super-Nyquist shaping filtering of the data block {a(d), d=0,1,2....D-1} at point D, and at a given This value determines the time compression ratio of the transmission at the super-Nyquist rate, ie the spectral efficiency of the transmission, given the upsampling rate of the shaping filter. If the discrete impulse response of the shaping filter is f _p (t), its length is L, and the upsampling rate is N, then for super Nyquist transmission, N _FTN <N, that is, the time shift interval is less than The upsampling rate of the shaping filter. Q is the length of the second signal sequence. In this embodiment of the present invention, the length of the second signal sequence is greater than the length of the first signal sequence; and the length Q of the second signal sequence is the length of the symbol data block. The product of the length D and the time shift interval N _FTN , ie Q=D×N _FTN . In an exemplary embodiment, the length D of the symbol data block may be 16, 20, etc., the time shift interval N _FTN may be 16, 18, 20, etc., the shaping filter The upsampling rate N can be set to 20, etc., in the specific implementation, those skilled in the art can set the length and time shift interval of any symbol data block according to the transmission needs of the FTN block, and select the corresponding shaping filter, It is not limited in this embodiment of the present invention.

Step S104: Generate a window function according to a preset shaping filter.

The shaping filter may be any one of a root raised cosine filter, a Gaussian filter and an isotropic orthogonal transform algorithm (English: Isotropic orthogonal transform algorithm, IOTA for short) filter. In order to generate the window function, in this embodiment of the present invention, the window function {w(k), k=0, 1, 2, L, Q-1 may be obtained by performing discrete Fourier transform on the shaping filter }, that is, the window function can be expressed as the discrete Fourier transform of the shaping filter f _p (t), the specific formula is as follows:

Wherein, L is the length of the discrete impulse response of the shaping filter, and Q is the length of the second signal sequence.

Step S105: Use the window function to perform a windowing operation on the second signal sequence.

A windowing operation is performed on the cyclically extended second signal sequence {c(k), k=0, 1, 2, . . . , Q-1}.

In the embodiment of the present invention, according to the window function {w(k), k=0, 1, 2, L, Q-1} obtained in step S104, for the second signal sequence {c(k), k= 0, 1, 2, .

{g(k)=c(k)·w(k), k=0, 1, 2, ..., Q-1}

Step S106: Perform inverse discrete Fourier transform on the second signal sequence after the windowing operation to obtain a cyclic signal sequence.

According to the second signal sequence g(k) after the windowing operation obtained in step S105, perform Q-point inverse discrete Fourier transform on g(k) to obtain an equivalent time-domain cyclic data block {q(t), t=0, 1, 2, ..., Q-1} as the cyclic signal sequence. The expression of the cyclic signal sequence is as follows:

Step S107 : adding a guard interval to the cyclic signal sequence to obtain an output signal sequence, which is sent to the receiver.

In order to reduce inter-channel interference, in an exemplary embodiment, a guard interval may be added to the head or tail of the cyclic signal sequence q(t); and, preferably, the length of the guard interval is greater than or equal to the maximum channel delay Extended length.

In a specific embodiment, the guard interval may be added to the head of the cyclic signal sequence by means of a cyclic prefix: copy and add the data block symbol located at the tail of the cyclic signal sequence and corresponding to the length of the guard interval to The header of the cyclic signal sequence, thereby forming the output signal sequence to be cyclic prefixed.

By increasing the guard interval, the cyclic signal sequence {q(t), t=0, 1, 2, ..., Q-1} is transformed into a complete output signal sequence {s(t), t=0, 1 ,2....Q+C-1}, where C is the length of the guard interval. The generated sequence of output signals is sent to the receiver, or broadcast to the surrounding environment, to enable the receiver to receive the sequence of output signals. Of course, it should be noted that, in specific implementation, the process of sending the output signal sequence may also include steps such as channel coding, digital modulation, radio frequency conversion, and transmission, which will not be repeated in the implementation of the present invention.

Referring to FIG. 2 , it is a schematic flowchart of a DFT-based FTN block transmission method on the receiver side provided by an embodiment of the present invention. As shown in FIG. 2 , an embodiment of the present invention shows a process in which the receiver implements the transmission method:

Step S201: Receive the output signal sequence sent by the transmitter, and remove the guard interval in the output signal sequence to obtain the input signal sequence.

First of all, it should be noted that, taking a digital communication system as an example, the receiver may also need to perform steps such as radio frequency conversion, synchronization, channel estimation, and digital demodulation when receiving signals, which will not be repeated in the implementation of the present invention. In this embodiment of the present invention, the DFT-based FTN block is implemented on the transmitter side by taking the output signal sequence {r(t), t=0, 1, ..., Q+C-1} received as an example. The transmission method is described in detail; where Q is the length of the output signal sequence, and C is the length of the guard interval.

According to the transmitter guard interval addition rule, the receiver discards C in the data block as the sample value of the guard interval, thereby forming an input signal sequence of length Q {y(t), t=0, 1, 2, . ..,Q-1}.

In the first implementation, the addition rule of the guard interval can be preset, and the addition rule specifies parameters such as the addition position of the guard interval and the length of the guard interval; in this way, the transmitter can increase the guard interval according to the preset addition rule, The output signal sequence is sent to the receiver, and the receiver removes the guard interval added by the transmitter according to the preset addition rule, thereby obtaining the input signal sequence.

In the second implementation case, since the transmitter may need to change the addition rule of the guard interval, in order to improve the flexibility and efficiency of removing the guard interval, referring to FIG. 3 , it is a flowchart of a method for removing the guard interval provided by an embodiment of the present invention The schematic diagram, as shown in Figure 3, the method includes:

Step S2011: Obtain the guard interval configuration information sent by the transmitter, where the guard interval configuration information at least carries the setting position and length of the guard interval and the length of the cyclic signal sequence.

When the transmitter sends the output signal sequence, it can also send guard interval configuration information; wherein the guard interval configuration information can be integrated into a field in the output signal sequence, or can be sent to the receiver in the form of an independent field Sending out the implementation guard interval configuration information is not limited in this embodiment of the present invention. Moreover, the guard interval configuration information carries the setting position of the guard interval added by the transmitter and the length of the guard interval. Further, the receiver obtains the guard interval configuration information sent by the transmitter.

Step S2022: Remove the guard interval in the output signal sequence according to the setting position and length of the guard interval.

According to the setting position and length of the guard interval obtained from the guard interval configuration information, the receiver starts from the setting position in the output signal sequence, and removes the data symbol in the output signal sequence corresponding to the length, thereby obtaining the removal protection. The input signal sequence after the interval. The length of the input signal sequence is equal to the length of the cyclic signal sequence.

Step S202: Perform discrete Fourier transform on the input signal sequence to obtain a third signal sequence.

Perform Q-point discrete Fourier transform on the input signal sequence {y(t), t=0, 1, 2, ..., Q-1} obtained after step S201 to obtain a third signal sequence {Y(k) , k=0, 1, 2, ..., Q-1}, the expression of the third signal sequence is as follows:

Step S203: Perform a first frequency domain equalization operation on the third signal sequence to obtain a first equalized signal sequence.

The first frequency domain equalization operation is performed on the data sequence {Y(k), k=0, 1, 2, ..., Q-1} after DFT, and the corresponding first equalized signal sequence is obtained as {E(k), k =0,1,2,...,Q-1}. In the embodiment of the present invention, the first frequency domain equalization operation is a commonly used frequency domain equalization, which is not repeated in the embodiment of the present invention.

Step S204: Generate filter coefficients.

The filter coefficients are expressed as {F(k), k=0, 1, 2, L, Q-1}, which are used to implement frequency-domain matched filtering in subsequent steps.

In the first implementation, the shaping filter configuration information may be preset, and the shaping filter configuration information specifies parameters such as the selected shaping filter type, the discrete impulse response length of the shaping filter, etc.; in this way, the transmitter can The symbol data block is processed according to the preset shape filter configuration information, and an output signal sequence is sent to the receiver, and the receiver performs corresponding operations on the received output signal sequence according to the preset shape filter configuration information.

In the second implementation case, since the transmitter may need to change and use different shaping filters, in order to improve the flexibility and efficiency of FTN block transmission, referring to FIG. The schematic flow chart, as shown in Figure 4, the method includes:

Step S2041: Acquire the shaping filter configuration information sent by the transmitter, where the shaping filter configuration information at least carries the type of the shaping filter.

By analyzing the shaping filter configuration information, the type of shaping filter is obtained, the discrete impulse response f _p (t) of the shaping filter to be used is determined by the type of the shaping filter, and the discrete impulse response of the shaping filter length L.

Step S2042: According to the length of the input signal sequence, perform discrete Fourier transform on the shaping filter to obtain the filter coefficient.

Discrete Fourier transform is performed on the discrete impulse response f _p (t) of the shaping filter corresponding to the type to obtain the filter coefficient {F(k),k=0,1,2,L,Q-1}, the formula as follows:

where f _p (t) is the discrete impulse response of the shaping filter, and L is the discrete impulse response length of the shaping filter.

Step S205: Perform a matched filtering operation on the first equalized signal sequence according to the filtering coefficient to obtain a matched filtered signal sequence.

Perform matched filtering operation on the first equalized signal sequence {E(k), k=0, 1, 2, ..., Q-1} after channel equalization, and output the matched filtered signal sequence of data and expressed as {M(k) , k=0, 1, 2, ..., Q-1}, the formula is as follows:

M(k)=E(k)·F ^* (k),

where (g) ^* represents a conjugation operation.

Step S206: Perform a cyclic accumulation operation on the matched filtered signal sequence to obtain a fourth signal sequence.

Perform a cyclic accumulation operation on the matched filter signal sequence {M(k), k=0, 1, 2, ..., Q-1} to convert the original Q point data into D point data, thereby obtaining the fourth signal sequence {o (k), k=0, 1, 2, ..., D-1}.

In the first implementation case, the same cyclic accumulation configuration information can be preset for the transmitter and the receiver, and the cyclic accumulation configuration information carries the length of the symbol data block, so that the transmitter can separate symbols according to the cyclic accumulation configuration information For the data block, the receiver can directly parse the length of the symbol data block from the cyclic accumulation configuration information without communicating with the receiver, and perform the subsequent cyclic accumulation operation.

In the second implementation case, since the length of the symbol data block may need to be flexibly adjusted during the FTN block transmission process, in order to improve the flexibility of the FTN block transmission, see FIG. 5 , which is a loop provided by an embodiment of the present invention A schematic flowchart of the accumulation operation, as shown in Figure 5, the method includes:

Step S2061: Acquire cyclic accumulation configuration information, where the cyclic accumulation configuration information carries at least the length of the symbol data block and the time shift interval.

The receiver acquires the cyclic accumulation configuration information sent by the transmitter, and the transmitter may send the cyclic accumulation configuration information independently, or may carry the cyclic accumulation configuration information in the sent output signal sequence.

Step S2062: Accumulate the N _FTN data symbols whose serial number is k+q×D in the matched filtered signal sequence, as the kth data symbol of the fourth signal sequence, wherein Q is the length of the input signal sequence, and D is the symbol data The length of the block, k is a natural number less than or equal to D-1, q is a natural number less than or equal to N _FTN -1.

Specifically, the fourth signal sequence {o(k), k=0, 1, 2, ..., D-1} can be obtained by the following formula:

Among them, Q is the length of the input signal sequence, and D is the length of the symbol number block.

In this embodiment of the present invention, the length of the fourth signal sequence is equal to the length of the symbol data block.

Step S207: Generate equalization coefficients.

The equalization coefficients are frequency domain equalization coefficients required for self-intersymbol interference (ISI) equalization, and the equalization coefficients may be expressed as {cf(k), k=0, . . . , D-1}.

In the first implementation, the transmitter and the receiver may agree to use the same shaping filter and time shift interval, so that the receiver can generate equalization coefficients according to the agreed shaping filter and time shift interval.

In the second implementation case, the transmitter may adjust parameters such as the shaping filter and the time shift interval at any time. In order to improve the flexibility of FTN block transmission, see FIG. 6 , which is an equalization coefficient provided by an embodiment of the present invention. A schematic flowchart of the generation method, as shown in Figure 6, the method includes:

Step S2071: Obtain equalization coefficient configuration information, where the equalization coefficient configuration information at least carries the type of the shaping filter and the time shift interval.

Obtain the equalization coefficient configuration information sent by the transmitter, where the equalization coefficient configuration information at least carries the type of the shaping filter and the time shift interval. In a specific implementation, the transmitter may send the equalization coefficient configuration information to the receiver alone, or the transmitter may carry the equalization coefficient configuration information in the output signal sequence sent out, so that the receiver receives the equalization coefficients After the configuration information, the type of shaping filter to be used in the subsequent steps can be obtained by analysis, so that the type of shaping filter to be used can be determined according to the type, and parameters such as time shift interval can be obtained by analysis.

Step S2072: According to the shaping filter corresponding to the type and the time shift interval, through the cyclic autocorrelation operation, generate a filter cyclic autocorrelation function; according to the length of the symbol data block, the filter cyclic autocorrelation function is The correlation function is subjected to discrete Fourier transform to obtain the equalization coefficients.

According to the shaping filter determined in the above two implementations and the time shift interval, the calculation formula of the equalization coefficient is as follows:

where ((·)) _Q represents the modulo Q operation. the length of the equalization coefficients is equal to the length of the symbol data block

Step S208: Perform a second frequency equalization operation on the fourth signal sequence according to the equalization coefficient to obtain a second equalized signal sequence.

Obtain equalization coefficients {cf(k), k=0,..., D-1} according to step S207, for the fourth signal sequence {o(k), k=0, 1, 2,..., D-1} Perform frequency domain equalization to obtain a second equalized signal sequence {q(k), k=0, 1, 2, ..., D-1}.

In an exemplary embodiment, a second equalized signal sequence is obtained by calculating the vector division operation of the fourth signal sequence and the equalization coefficient, and the expression of the second equalized signal sequence is as follows:

q(k)=o(k)/cf(k)

Step S209: Perform inverse discrete Fourier transform on the second equalized signal sequence to obtain a transformed output signal sequence.

Perform D-point inverse discrete Fourier transform on the second equalized signal sequence {q(k), k=0, 1, 2, ..., D-1} obtained in step S208, thereby obtaining the transformed output signal sequence {z(d ), d=0, 1, 2, L, D-1}, the expression of the transformed output signal sequence is as follows:

Corresponding to the embodiment of the method for transmitting a block with a super Nyquist rate provided by the embodiment of the present invention, the embodiment of the present invention also provides an embodiment of the apparatus for the transmission method.

Referring to FIG. 7, it is a schematic structural diagram of a transmitter according to an embodiment of the present invention. As shown in FIG. 7, the transmitter includes:

a data block dividing module 11, for dividing the modulation symbol sequence into a plurality of symbol data blocks;

The discrete Fourier transform module 12 is used to perform discrete Fourier transform on each symbol data block to obtain a first signal sequence;

a cyclic extension module 13, configured to cyclically extend the first signal sequence into a second signal sequence;

The window function generation module 14 is used to generate a window function according to a preset shaping filter;

A windowing module 15 is used to perform a windowing operation on the second signal sequence using the window function;

an inverse discrete Fourier transform module 16, configured to perform an inverse discrete Fourier transform on the second signal sequence after the windowing operation to obtain a cyclic signal sequence;

The guard interval module 17 is configured to add a guard interval to the cyclic signal sequence, obtain an output signal sequence, and send it to the receiver.

Optionally, the lengths of the plurality of symbol data blocks obtained by the data block dividing module 11 are all the same.

Optionally, the cyclic extension module 13 is configured to make the n×D+k th data symbol of the second signal sequence equal to the k data symbol of the first signal sequence, where D is the length of the symbol data block , k is a natural number less than or equal to D-1, n is a natural number less than or equal to N _FTN -1, N _FTN is the time shift interval adopted by the super-Nyquist shaping filtering of the symbol data block; and , the length of the second signal sequence is the product of the length of the symbol data block and the time shift interval.

Optionally, the time shift interval is smaller than the upsampling rate of the shaping filter.

Optionally, the window function generating module 14 is configured to, according to the length of the second signal sequence, perform discrete Fourier transform on the shaping filter to obtain the window function.

Optionally, the windowing module 15 is configured to perform a coefficient multiplication operation on the second signal sequence and the window function.

Optionally, the guard interval module 17 is configured to add the guard interval to the head or tail of the cyclic signal sequence, and the length of the guard interval is greater than or equal to the maximum delay extension length of the channel.

Optionally, the guard interval module 17 is configured to, when adding the guard interval to the header of the cyclic signal sequence, place the data block symbol at the tail of the cyclic signal sequence and corresponding to the length of the guard interval. Copied and added to the head of the cyclic signal sequence.

Referring to FIG. 8, it is a schematic structural diagram of a receiver according to an embodiment of the present invention. As shown in FIG. 8, the receiver includes:

The guard interval removal module 21 is used for receiving the output signal sequence sent by the transmitter, and removing the guard interval in the output signal sequence to obtain the input signal sequence;

A discrete Fourier transform module 22, configured to perform discrete Fourier transform on the input signal sequence to obtain a third signal sequence;

a first equalization module 23, configured to perform a first frequency domain equalization operation on the third signal sequence to obtain a first equalized signal sequence;

a filter coefficient generation module 24 for generating filter coefficients;

a matched filtering module 25, configured to perform a matched filtering operation on the first equalized signal sequence according to the filter coefficient;

The cyclic accumulation module 26 is used to perform cyclic accumulation operation on the matched filter signal sequence to obtain the fourth signal sequence;

an equalization coefficient generation module 27, configured to generate an equalization coefficient;

A second equalization module 28, configured to perform a second frequency domain equalization operation on the fourth signal sequence according to the equalization coefficient to obtain a second equalized signal sequence;

The inverse discrete Fourier transform module 29 is configured to perform inverse discrete Fourier transform on the second equalized signal sequence to obtain a transformed output signal sequence.

Optionally, the guard interval removal module 21 is configured to acquire guard interval configuration information sent by the transmitter, where the guard interval configuration information at least carries the setting position and length of the guard interval, and the cyclic signal sequence. length; according to the setting position and length of the guard interval, remove the guard interval in the output signal sequence; the length of the input signal sequence is equal to the length of the cyclic signal sequence.

Optionally, the filter coefficient generation module 24 is configured to obtain the shaping filter configuration information sent by the transmitter, where the shaping filter configuration information at least carries the type of shaping filter; according to the length of the input signal sequence, The filter coefficients are obtained by performing discrete Fourier transform on the shaping filter.

Optionally, the cyclic accumulation module 26 is used to obtain cyclic accumulation configuration information, where the cyclic accumulation configuration information at least carries the length of the symbol data block and the time shift interval; the sequence number in the accumulated matched filtered signal sequence is the kth The N _FTN data symbols of +q× D are used as the kth data symbol of the fourth signal sequence, where Q is the length of the input signal sequence, D is the length of the symbol data block, and k is a natural number less than or equal to D-1 , q is a natural number less than or equal to N _FTN -1.

Optionally, the length of the fourth signal sequence is equal to the length of the symbol data block.

Optionally, the equalization coefficient generation module 27 is configured to obtain equalization coefficient configuration information, where the equalization coefficient configuration information at least carries the type and time shift interval of the shaping filter; The time shift interval, through a cyclic autocorrelation operation, generates a filter cyclic autocorrelation function; according to the length of the symbol data block, discrete Fourier transform is performed on the filter cyclic autocorrelation function to obtain the equalization coefficient. The length of the equalization coefficients is equal to the length of the symbol data block.

Optionally, the second equalization module is configured to calculate a vector division operation of the fourth signal sequence and an equalization coefficient to obtain a second equalized signal sequence.

Referring to FIG. 9 , it is a schematic structural diagram of a DFT-based FTN block transmission system provided by an embodiment of the present invention, where the transmission system includes the transmitter 31 described in the foregoing embodiment and the receiver 32 described in the foregoing embodiment.

In order to illustrate the effects of a DFT-based FTN transmission method, transmitter, receiver, and system provided by the embodiment of the present invention, a system simulation is performed in the embodiment of the present invention, and the specific system simulation parameter settings are shown in Table 1 below:

Table I:

Nyquist rate system FTN-BT system Modulation QPSK QPSK Shaping filter type root raised cosine root raised cosine Shaping filter roll-off factor 0.3 0.3 Shaping filter upsampling rate (N) 20 20 Shaping filter displacement 20 16, 18 Number of symbols transmitted per data block (D) 16 20 Data block length (number of sampling points) (Q) 320 320, 360 time compression ratio 1 0.8, 0.9 Spectral efficiency (bps/Hz) 2 2.5, 2.22

The embodiment of the present invention intends to equivalently implement a block transmission-based super Nyquist rate communication system with lower complexity in the frequency domain. Referring to FIG. 10 , it is a schematic diagram of a structure and performance result of a simplified implementation of the frequency domain of an FTN-BT system according to an embodiment of the present invention. Figure 10 shows the comparison of the reconstructed SNR performance between the FTN-FBT system based on DFT and the FTN system based on time domain convolution. In terms of reconstructed signal-to-noise ratio performance, the simulation results show that the reconstructed signal-to-noise ratio-channel signal-to-noise ratio curves realized in the time domain and frequency domain are coincident, that is, the effect realized in the frequency domain is equivalent to that in the time domain, and the simplified The frequency domain implementation structure does not degrade system performance.

因此，对于复序列a(d)，和长度为L的实系数成形滤波器f_p(t)，生成该信号所需计算量(实乘数)为2L×D。当D＝32，成形滤波器f_p(t)的长度L＝241，所需实乘数为15424。In terms of complexity performance, the baseband output signal (one data block) of the transmitter of the FTN system based on time domain convolution can be expressed as {g(t), t=0,1,2....Q-1},

Therefore, for a complex sequence a( _d ), and a real coefficient shaping filter fp(t) of length L, the amount of computation (real multipliers) required to generate this signal is 2L×D. When D=32, the length L ₌ 241 of the shaping filter fp(t), and the required real multiplier is 15424.

For the FTN-FBT system based on the DFT implementation, the transmitting end generates a data block with a large amount of computation, including D-point DFT, Q-point windowing in the frequency domain, and Q-point IDFT. The computational complexity of the D-point DFT is negligible, so the amount of computation (real multipliers) required to generate this signal is about 4Q+2Q log ₂ (Q). When D=32, the length L=241 of the shaping filter f _p (t) corresponds to Q=512, and the required real multiplier is 11264. Compared with the time domain implementation, the complexity is reduced by about 27%. And, as D increases, the complexity decreases more obviously. Based on the same mechanism, a similar conclusion can also be drawn for the processing at the receiving end.

Therefore, the DFT-based FTN block transmission system proposed by the present invention ensures system performance on the basis of significantly reducing the implementation complexity.

To sum up, a DFT-based FTN block transmission method, transmitter, receiver and system provided by the present invention are based on discrete Fourier transform and frequency domain windowing technology, and equivalently realize the traditional waveform generation based on time domain convolution And receive matched filtering method, and then through simple frequency domain equalization, step by step to eliminate channel multipath interference and self-ISI, thus simplifying the implementation structure of the transmitter and receiver baseband processing. In addition, by further utilizing the high efficiency of DFT calculation, the present invention can effectively reduce the realization complexity of the system. Therefore, the present invention effectively overcomes various shortcomings in the prior art and has high industrial utilization value.

The above-mentioned embodiments merely illustrate the principles and effects of the present invention, but are not intended to limit the present invention. Anyone skilled in the art can modify or change the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or changes made by those with ordinary knowledge in the technical field without departing from the spirit and technical idea disclosed in the present invention should still be covered by the claims of the present invention.

Claims (28)

1. a DFT-based FTN block transmission method, applied to a transmitter, is characterized in that, the transmission method comprises the following steps: dividing the modulation symbol sequence into a plurality of symbol data blocks; Perform discrete Fourier transform on each symbol data block to obtain a first signal sequence; Cyclically extending the first signal sequence into a second signal sequence, including making the n×D+kth data symbol of the second signal sequence equal to the k data symbols of the first signal sequence, where D is symbol data The length of the block, k is a natural number less than or equal to D-1, n is a natural number less than or equal to N _FTN -1, and N _FTN is the time shift used by the super-Nyquist shaping filter for the symbol data block interval; and, the length of the second signal sequence is the product of the length of the symbol data block and the time shift interval; Generate a window function according to a preset shaping filter; using the window function to perform a windowing operation on the second signal sequence; Perform inverse discrete Fourier transform on the second signal sequence after the windowing operation to obtain a cyclic signal sequence; A guard interval is added to the cyclic signal sequence to obtain an output signal sequence, which is sent to the receiver. 2 . The DFT-based FTN block transmission method according to claim 1 , wherein the lengths of the multiple symbol data blocks are all the same. 3 . 3 . The DFT-based FTN block transmission method according to claim 1 , wherein the length of the second signal sequence is greater than the length of the first signal sequence. 4 . 4. The DFT-based FTN block transmission method according to claim 1, wherein the time shift interval is smaller than the upsampling rate of the shaping filter. 5. the FTN block transmission method based on DFT according to claim 1, is characterized in that, according to preset shaping filter, generates window function, comprising: According to the length of the second signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the window function. 6. the FTN block transmission method based on DFT according to claim 1, is characterized in that, using described window function to carry out windowing operation to described second signal sequence, comprising: A multiplication factor operation is performed on the second signal sequence and the window function. 7. The DFT-based FTN block transmission method according to claim 1, wherein a guard interval is added to the cyclic signal sequence to obtain an output signal sequence, comprising: The guard interval is added at the head or tail of the cyclic signal sequence, and the length of the guard interval is greater than or equal to the maximum delay extension length of the channel. 8. The DFT-based FTN block transmission method according to claim 7, wherein when the guard interval is added at the head of the cyclic signal sequence, the cyclic signal sequence tail and the guard interval The length of the corresponding data block symbol is copied and added to the header of the cyclic signal sequence. 9. A DFT-based FTN block transmission method, applied to a receiver, is characterized in that, comprising the following steps: receiving the output signal sequence sent by the transmitter, and removing the guard interval in the output signal sequence to obtain the input signal sequence; performing discrete Fourier transform on the input signal sequence to obtain a third signal sequence; performing a first frequency domain equalization operation on the third signal sequence to obtain a first equalized signal sequence; generate filter coefficients; performing a matched filtering operation on the first equalized signal sequence according to the filter coefficient to obtain a matched filtered signal sequence; Performing a cyclic accumulation operation on the matched filtered signal sequence to obtain a fourth signal sequence, including: acquiring cyclic accumulation configuration information, where the cyclic accumulation configuration information at least carries the length of the symbol data block and the time shift interval; accumulating the matched filtered signal sequence in The N _FTN data symbols whose serial numbers are k+q×D are used as the kth data symbol of the fourth signal sequence, where Q is the length of the input signal sequence, D is the length of the symbol data block, and k is less than or equal to D A natural number of -1, q is a natural number less than or equal to N _FTN -1; generate equalization coefficients; performing a second frequency domain equalization operation on the fourth signal sequence according to the equalization coefficient to obtain a second equalized signal sequence; Inverse discrete Fourier transform is performed on the second equalized signal sequence to obtain a transformed output signal sequence. 10. The DFT-based FTN block transmission method according to claim 9, wherein receiving the output signal sequence sent by the transmitter and removing the guard interval in the output signal sequence to obtain the input signal sequence, comprising: acquiring guard interval configuration information sent by the transmitter, where the guard interval configuration information at least carries the setting position and length of the guard interval and the length of the cyclic signal sequence; Remove the guard interval in the output signal sequence according to the setting position and length of the guard interval; The length of the input signal sequence is equal to the length of the cyclic signal sequence. 11. The DFT-based FTN block transmission method according to claim 9, wherein the generating filter coefficients comprises: acquiring shaping filter configuration information sent by the transmitter, where the shaping filter configuration information at least carries the type of shaping filter; According to the length of the input signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the filter coefficients. 12. The DFT-based FTN block transmission method according to claim 9, wherein the length of the fourth signal sequence is equal to the length of the symbol data block. 13. The DFT-based FTN block transmission method according to claim 9, wherein the generating equalization coefficients comprises: obtaining equalization coefficient configuration information, where the equalization coefficient configuration information at least carries the type of the shaping filter and the time shift interval; According to the shaping filter corresponding to the type and the time shift interval, through a cyclic autocorrelation operation, a filter cyclic autocorrelation function is generated; According to the length of the symbol data block, discrete Fourier transform is performed on the cyclic autocorrelation function of the filter to obtain the equalization coefficient; The length of the equalization coefficients is equal to the length of the symbol data block. 14. The DFT-based FTN block transmission method according to claim 9, wherein the second frequency domain equalization operation to obtain a second equalized signal sequence comprises: A vector division operation of the fourth signal sequence and the equalization coefficient is calculated to obtain a second equalized signal sequence. 15. A transmitter, characterized in that it comprises: a data block division module, which is used to divide the modulation symbol sequence into a plurality of symbol data blocks; The discrete Fourier transform module is used to perform discrete Fourier transform on each symbol data block to obtain the first signal sequence; a cyclic extension module, configured to cyclically extend the first signal sequence into a second signal sequence, so that the n×D+k th data symbol of the second signal sequence is equal to the k data symbols of the first signal sequence, Among them, D is the length of the symbol data block, k is a natural number less than or equal to D-1, n is a natural number less than or equal to N _FTN -1, and N _FTN is the super Nyquist shaping of the symbol data block. the time shift interval used for filtering; and, the length of the second signal sequence is the product of the length of the symbol data block and the time shift interval; a window function generation module for generating a window function according to a preset shaping filter; a windowing module for performing a windowing operation on the second signal sequence using the window function; an inverse discrete Fourier transform module, configured to perform an inverse discrete Fourier transform on the second signal sequence after the windowing operation to obtain a cyclic signal sequence; A guard interval module, configured to add a guard interval to the cyclic signal sequence, obtain an output signal sequence, and send it to the receiver. 16 . The transmitter according to claim 15 , wherein the lengths of the plurality of symbol data blocks obtained by dividing by the data block dividing module are all the same. 17 . 17. The transmitter of claim 15, wherein the time shift interval is less than an upsampling rate of the shaping filter. 18. transmitter according to claim 15, is characterized in that, described window function generation module is used for, According to the length of the second signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the window function. 19. The transmitter according to claim 15, wherein the windowing module is used for, A multiplication factor operation is performed on the second signal sequence and the window function. 20 . The transmitter according to claim 15 , wherein the guard interval module is configured to add the guard interval to the head or tail of the cyclic signal sequence, and the length of the guard interval is greater than or equal to the length of the guard interval. 21 . The maximum delay spread length of the channel. 21. The transmitter according to claim 20, wherein the guard interval module is configured to, when adding the guard interval to the head of the cyclic signal sequence, remove the Data block symbols corresponding to the length of the guard interval are copied and added to the header of the cyclic signal sequence. 22. A receiver, characterized in that, comprising: a guard interval removing module, configured to receive the output signal sequence sent by the transmitter, and remove the guard interval in the output signal sequence to obtain the input signal sequence; a discrete Fourier transform module, configured to perform discrete Fourier transform on the input signal sequence to obtain a third signal sequence; a first equalization module, configured to perform a first frequency domain equalization operation on the third signal sequence to obtain a first equalized signal sequence; A filter coefficient generation module for generating filter coefficients; a matched filtering module, configured to perform a matched filtering operation on the first equalized signal sequence according to the filter coefficient to obtain a matched filtered signal sequence; The cyclic accumulation module is used to obtain the cyclic accumulation configuration information, the cyclic accumulation configuration information carries at least the length of the symbol data block and the time shift interval; the sequence number in the accumulated matched filter signal sequence is the k+q×D N _FTN data symbols as the kth data symbol of the fourth signal sequence, where Q is the length of the input signal sequence, D is the length of the symbol data block, k is a natural number less than or equal to D-1, and q is less than or equal to N A natural number of _FTN -1 to obtain the fourth signal sequence; an equalization coefficient generation module for generating equalization coefficients; a second equalization module, configured to perform a second frequency domain equalization operation on the fourth signal sequence according to the equalization coefficient to obtain a second equalized signal sequence; an inverse discrete Fourier transform module, configured to perform an inverse discrete Fourier transform on the second equalized signal sequence to obtain a transformed output signal sequence. 23. The receiver of claim 22, wherein the guard interval removal module is used to: acquiring guard interval configuration information sent by the transmitter, where the guard interval configuration information at least carries the setting position and length of the guard interval and the length of the cyclic signal sequence; Remove the guard interval in the output signal sequence according to the setting position and length of the guard interval; The length of the input signal sequence is equal to the length of the cyclic signal sequence. 24. The receiver of claim 22, wherein the filter coefficient generation module is used to: acquiring shaping filter configuration information sent by the transmitter, where the shaping filter configuration information at least carries the type of shaping filter; According to the length of the input signal sequence, discrete Fourier transform is performed on the shaping filter to obtain the filter coefficients. 25. The receiver of claim 22, wherein the length of the fourth signal sequence is equal to the length of the symbol data block. 26. The receiver of claim 22, wherein the equalization coefficient generation module is configured to: obtaining equalization coefficient configuration information, where the equalization coefficient configuration information at least carries the type of the shaping filter and the time shift interval; According to the shaping filter corresponding to the type and the time shift interval, through a cyclic autocorrelation operation, a filter cyclic autocorrelation function is generated; According to the length of the symbol data block, discrete Fourier transform is performed on the cyclic autocorrelation function of the filter to obtain the equalization coefficient; The length of the equalization coefficients is equal to the length of the symbol data block. 27. The receiver of claim 22, wherein the second equalization module is configured to: A vector division operation of the fourth signal sequence and the equalization coefficient is calculated to obtain a second equalized signal sequence. 28. A transmission system, comprising a transmitter as claimed in any one of claims 15 to 21, and a receiver as claimed in any one of claims 22 to 27.

Images