CN106878222A

CN106878222A - The generation method and device of a kind of multi-carrier signal

Info

Publication number: CN106878222A
Application number: CN201510927332.9A
Authority: CN
Inventors: 黄琛; 胡留军; 辛雨
Original assignee: ZTE Corp
Current assignee: ZTE Corp
Priority date: 2015-12-14
Filing date: 2015-12-14
Publication date: 2017-06-20
Anticipated expiration: 2035-12-14
Also published as: WO2017101462A1; CN106878222B

Abstract

The generation method and generating means of a kind of multi-carrier signal；The generation method includes：Single-carrier signal is segmented, and is mapped to each subband position, obtain one or more subband signals；The multi-carrier modulation with Cyclic Prefix and/or cyclic suffix, sub-band filter and windowing process are carried out to the subband signal.The present invention not only can improve system effectiveness with the overhead of larger reduction CP-UFMC, and still have less band outward leakage characteristic.

Description

Method and device for generating multi-carrier signal

Technical Field

The present invention relates to the field of wireless communications, and in particular, to a method and an apparatus for generating a multi-carrier signal.

Background

In an LTE (Long Term Evolution) system and an LTE-a (Long Term Evolution-Advanced, subsequent Evolution of Long Term Evolution), OFDM (orthogonal frequency Division Multiplexing) is a basic technology for physical layer signal transmission. However, the sideband leakage of the OFDM signal is large, especially when there is a time offset and a frequency offset. This results in a larger guard bandwidth between OFDM and other systems, and in order to reduce interference to other user equipments, signals of each user equipment of the OFDM system need to be more strictly time and frequency synchronized. To reduce the sideband leakage of the OFDM system, several techniques are proposed: FBMC (Filter-Bank Multi-Carrier, Filter Bank Multi-Carrier), UFMC (Universal Filtered Multi-Carrier, full-Filtered Multi-Carrier), GFDM (Generalized Frequency Division multiplex), and the like. Among them, UFMC is also called f-OFDM (filtered OFDM, filtered orthogonal frequency division multiplexing).

The UFMC filters each subband based on the OFDM signal, i.e. the same filter within the same subband (within the resources of the same user equipment). The UFMC technology filters the sub-bands, thereby greatly reducing the side-band leakage, greatly reducing the interference power between the sub-bands, and flexibly configuring different parameters for different sub-bands to better adapt to the service and channel requirements. This feature of UFMC is particularly suitable for the situation where services with different requirements and features coexist in the future wireless communication system.

The UFMC signals, as generally discussed in the industry, are without CP (cyclic Prefix) and/or cyclic suffix. When there is multipath delay of channel, there is ISI (inter-symbol interference), and if the multipath delay of channel is small, the interference can be ignored; if the channel multipath delay is large and the interference is not negligible, the technique of UFMC for ISI cancellation needs to be considered.

The common technique for eliminating ISI is to add CP, and the UFMC can also adopt the method for adding CP to resist the influence of channel multipath delay and propagation delay, thereby eliminating ISI. However, since the UFMC signal needs to undergo filtering, in order to ensure no interference between symbols and existence of multipath time delay, the time domain signals are circularly arranged, and the increased CP length of the UFMC should be L_CP＝L_CD+L_FilterWherein L is_CDIs a measure against the length of the channel multipath delay and propagation delay, L_FilterIs the length of the Filter. Fig. 1(a) is a signal diagram of UFMC after adding CP. It can be seen that the signal length of the UFMC after adding the CP (hereinafter referred to as CP-UFMC) is the original signal length + the length of the CP + the length of the Tx (transmit end) Filter. FIG. 1(b) shows the signal of FIG. 1(a) after the signal has passed through Tx Finiter, that is, the UFMC signal after adding CP has an overhead of L_CD+2*L_Filter. Such overhead is large and has a large impact on system efficiency. In addition, both the UFMC with normal and the UFMC with CP added generally require the same filter length for each sub-band. When the filtering requirements of different sub-bands are different and the filter lengths are different, the method also leads toInterference between incoming subbands.

Disclosure of Invention

The invention provides a method and a device for generating a multi-carrier signal, which can greatly reduce the extra overhead of CP-UFMC, improve the system efficiency and have smaller out-of-band leakage characteristics.

In order to solve the above problems, the following technical solutions are adopted.

A method of generating a multi-carrier signal, comprising:

segmenting the single carrier signal, and mapping to each subband position to obtain one or more subband signals;

and carrying out multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering and windowing on the sub-band signals.

Optionally, the performing multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering and windowing on the sub-band signals comprises the following operations arranged in any order:

performing signal transformation to transform the frequency domain signal into a time domain signal;

adding a cyclic prefix and/or cyclic suffix to the signal;

circularly filtering the subband signals;

superposing the sub-band signals;

windowing is performed on the signal.

Optionally, the cyclically filtering the signal comprises:

performing cyclic filtering on each sub-band signal by using a cyclic filter;

or,

filtering each sub-band signal respectively; for each sub-band signal after filtering, truncating the tail signal or the head signal, and copying and superposing the truncated tail signal on the head of the corresponding signal, or superposing the truncated head signal on the tail of the corresponding signal;

or,

filtering each sub-band signal respectively; and after the sub-band signals are superposed, for the superposed signals, cutting off the tail signals or the head signals, and copying and superposing the cut-off tail signals at the heads of the corresponding signals or superposing the cut-off head signals at the tails of the corresponding signals.

Optionally, the copying and superimposing the truncated tail signal on the head of the corresponding signal includes:

when the length L of the truncated tail signal is equal to L_filter-1, sequentially superimposing the 1 st to the last signal point of the truncated tail signal on the 1 st to the Lth signal points of the respective signal_filter-1 signal point; l is_filterIs the time domain total order of the sub-band filter;

when the length of the truncated tail signal is greater than L ═ L_filterAt time-1, the L-L of the truncated tail signal_filterThe +2 signal points to the last signal point are sequentially superposed on the 1 st signal point to the Lth signal point of the corresponding signal_filter-1 signal point; the 1 st signal point of the truncated tail signal is connected to the L-L_filterThe +1 signal point is placed before the 1 st signal point of the corresponding signal;

the copying and superimposing the truncated head signal on the tail of the corresponding signal comprises:

when the length L of the truncated header signal is equal to L_filter1, sequentially superimposing the 1 st signal point to the last signal point of the truncated header signal on the last-to-last L of the corresponding signal_filter-1 signal point to the last signal point;

when cutting offThe length of the broken head signal is greater than L ═ L_filter1, from the 1 st signal point to the L-th signal point of the truncated header signal_filter-1 signal point superimposed in succession on the last but one L of the corresponding signal_filter-1 signal point to the last signal point; l-th of the truncated header signal_filterThe signal points to the last signal point are positioned after the last signal point of the corresponding signal.

Optionally, L of different sub-bands_filterThe same or different.

Optionally, the performing multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering, and windowing on the sub-band signal further includes:

performing back-end processing on the signal; the back-end processing includes any one or any combination of: multiplying by a predetermined complex vector, clipping.

An apparatus for generating a multicarrier signal, comprising:

the segment mapping module is used for segmenting the single carrier signal and mapping the single carrier signal to each subband position to obtain one or more subband signals;

and the modulation filtering module is used for carrying out multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering and windowing on the sub-band signals.

Optionally, the modulation filtering module includes the following sub-modules connected in any order:

the signal transformation submodule is used for carrying out signal transformation and transforming the frequency domain signal into a time domain signal;

an adding submodule for adding a cyclic prefix and/or a cyclic suffix to the signal;

the circulating type filtering submodule is used for circularly filtering the subband signals;

the signal superposition submodule is used for superposing the subband signals;

and the windowing processing submodule is used for carrying out windowing processing on the signal.

Optionally, the circular filtering submodule includes:

the circulating filter is used for respectively and circularly filtering each subband signal;

or comprises the following steps:

a subband signal filtering unit for filtering each subband signal respectively;

a truncation duplication and superposition unit, configured to truncate the tail signal or the head signal for each filtered subband signal or a signal obtained by superimposing the subband signal, and duplicate and superimpose the truncated tail signal at the head of the corresponding signal, or superimpose the truncated head signal at the tail of the corresponding signal;

the sub-band signal filtering unit and the truncation, copying and overlapping unit are connected through the signal overlapping sub-module; or, the truncation, duplication and superposition unit is connected between the subband signal filtering unit and the signal superposition submodule.

Optionally, the copying, copying and superimposing unit copies and superimposes the truncated tail signal on the head of the corresponding signal:

the truncation copy superposition unit truncates the tail signal when the length L of the truncated tail signal is equal to L_filter-1, sequentially superimposing the 1 st to the last signal point of the truncated tail signal on the 1 st to the Lth signal points of the respective signal_filter-1 signal point; l is_filterIs the time domain total order of the sub-band filter;

when the length of the truncated tail signal is greater than L ═ L_filterAt time-1, the L-L of the truncated tail signal_filterThe +2 signal points to the last signal point are sequentially superposed on the 1 st signal point to the Lth signal point of the corresponding signal_filter-1 signal point; the first of the truncated tail signals1 signal point to the L-L_filterThe +1 signal point is placed before the 1 st signal point of the corresponding signal;

the method for copying and superposing the truncated head signal on the tail part of the corresponding signal by the truncation-copying-superposing unit comprises the following steps:

the truncated duplicate superposition unit is configured to perform a duplication when the length L of the truncated header signal is equal to L_filter1, sequentially superimposing the 1 st signal point to the last signal point of the truncated header signal on the last-to-last L of the corresponding signal_filter-1 signal point to the last signal point;

when the length of the truncated head signal is greater than L ═ L_filter1, from the 1 st signal point to the L-th signal point of the truncated header signal_filter-1 signal point superimposed in succession on the last but one L of the corresponding signal_filter-1 signal point to the last signal point; l-th of the truncated header signal_filterThe signal points to the last signal point are positioned after the last signal point of the corresponding signal.

Optionally, L of different sub-bands_filterThe same or different.

Optionally, the apparatus further comprises:

the back-end processing module is used for carrying out back-end processing on the signal output by the modulation filtering module; the back-end processing includes any one or any combination of: multiplying by a predetermined complex vector, clipping.

The scheme used by the invention is to obtain the improved CP-UFMC signal by generating the UFMC signal with the cyclic prefix and/or the cyclic suffix and performing time domain windowing on the signal, so that the additional overhead of the CP-UFMC signal is greatly reduced, the system efficiency is improved, and the signal still has smaller out-of-band leakage characteristic. Alternatively, filters of different lengths may be used between different subbands to accommodate the filtering requirements of the different subbands. Alternatively, the circular filtering may be implemented differently.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the example serve to explain the principles of the invention and not to limit the invention.

FIG. 1(a) is a signal diagram of UFMC with CP added;

FIG. 1(b) is the signal of FIG. 1(a) after it has passed through Tx Filter;

fig. 2 is a flowchart illustrating a method for generating a multi-carrier signal according to a first embodiment;

FIGS. 3(a) and (b) are schematic diagrams of a truncated tail signal replica superimposed on a signal header in the first embodiment;

fig. 4 is a schematic diagram of a multicarrier signal generation apparatus according to the second embodiment;

fig. 5 is a schematic diagram of a generation method of a multicarrier signal of example 1;

fig. 6 is a schematic diagram of a generation apparatus of a multicarrier signal in example 1;

FIG. 7 is a schematic diagram of a first combination of signal transformation and filtering in example 1;

FIG. 8 is a schematic diagram of a second combination of signal transformation and filtering in example 1;

FIG. 9 is a schematic representation of a third combination of signal transformation and filtering in example 1;

FIG. 10 is a schematic representation of a fourth combination of signal transformation and filtering in example 1;

fig. 11 is a schematic diagram of a generation method of a multicarrier signal in example 2;

fig. 12 is a schematic diagram of a generation apparatus of a multicarrier signal in example 2;

fig. 13 is a schematic diagram of a generation method of a multicarrier signal in example 3;

fig. 14 is a schematic diagram of a generation apparatus of a multicarrier signal in example 3;

fig. 15 is a schematic diagram of a generation method of a multicarrier signal in example 4;

fig. 16 is a schematic diagram of a generation apparatus of a multicarrier signal in example 4;

fig. 17 is a schematic diagram of a generation method of a multicarrier signal in example 5;

fig. 18 is a schematic diagram of a generation apparatus of a multicarrier signal in example 5;

fig. 19 is a schematic diagram of a generation method of a multicarrier signal in example 6;

fig. 20 is a schematic diagram of a generation apparatus of a multicarrier signal in example 6;

fig. 21 is a schematic diagram of a generation method of a multicarrier signal in example 7;

fig. 22 is a schematic diagram of a generation apparatus of a multicarrier signal in example 7.

Detailed Description

The technical solution of the present invention will be described in more detail with reference to the accompanying drawings and examples.

It should be noted that, if not conflicting, the embodiments of the present invention and the features of the embodiments may be combined with each other within the scope of protection of the present invention. Additionally, while a logical order is shown in the flow diagrams, in some cases, the steps shown or described may be performed in an order different than here.

In an embodiment, a method for generating a multi-carrier signal, as shown in fig. 2, includes S110 and S120:

s110, segmenting the single carrier signal, and mapping to each subband position to obtain one or more subband signals.

In step S110, after the single carrier signal is segmented and before it is mapped to each sub-band position, the single carrier signal may be further transformed from the time domain to the frequency domain in the manner adopted

And S120, performing multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering and windowing on the sub-band signals.

In step S120, the purpose of multicarrier modulation and filtering is to output a multicarrier signal, and the multicarrier signal includes a plurality of subband signals subjected to sideband leakage suppression. The step S120 has various implementation manners, and several implementation manners of the step S120 are given in the example; the practical application is not limited to the way given in the examples.

The step S120 may include sub-steps 21 to 25, and those skilled in the art will understand that the sequence between these five sub-steps can be arbitrarily adjusted:

substep 21: and performing signal transformation to transform the frequency domain signal into a time domain signal.

Optionally, in sub-step 21, the signal may be transformed in a variety of ways, including but not limited to IDFT (Inverse Discrete Fourier Transform) or IFFT (Inverse fast Fourier Transform); corresponding to the transformation used when transforming the signal from the time domain to the frequency domain, for example, the Discrete Fourier Transform (DFT) is used when transforming the signal from the time domain to the frequency domain, and then IDFT is used in sub-step 21.

Substep 22: a cyclic prefix and/or cyclic suffix is added to the signal.

Substep 23: and circularly filtering the sub-band signals. Optionally, in the sub-step 23, the filtering may be performed in the time domain, filtering the time domain signal, or performing equivalent filtering in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

Alternatively, the combination of signal transformation and filtering may take a variety of forms in order to output a multi-carrier signal containing a plurality of sideband signals with sideband leakage suppression. Four alternative combinations are given in example 1.

Substep 24: superposing the sub-band signals;

substep 25: windowing is performed on the signal.

Optionally, in sub-step 25, the windowing process may be performed in the time domain, or may be performed in the transform domain as an equivalent of time domain windowing. The transform domain includes, but is not limited to, the frequency domain.

Optionally, substep 23 may comprise:

performing cyclic filtering on each sub-band signal by using a cyclic filter;

or,

filtering each sub-band signal respectively; after performing said sub-step 24 (superimposing the sub-band signals), for the superimposed signal, the tail signal or the head signal is truncated, and the truncated tail signal is copied, superimposed on the head of the corresponding signal, or the truncated head signal is superimposed on the tail of the corresponding signal.

It can be seen that the head signal of the filtered (or superimposed subband signal) signal may be truncated and copied and superimposed on the tail signal of the filtered (or superimposed subband signal) signal, or the tail signal of the filtered or superimposed signal may be truncated and copied and superimposed on the head signal of the filtered (or superimposed subband signal) signal. These two operations are similar and easy to be deduced by analogy, and the following gives a detailed description of the truncation of the tail signal of the filtered signal and the duplication and superposition on the head signal of the filtered signal, and the truncation of the tail signal of the signal after the superposition (i.e.: sub-step 24) of the subband signal and the duplication and superposition on the head signal of the signal after the superposition of the subband signal are similar, with the difference that: when the above operation is performed on the signals after the subband signals are superposed, the tail signals of all the subband signals are copied and superposed to the head at the same time, and when the above operation is performed on the signals after filtering and before the subband signals are superposed, the tail signals of each subband signal are respectively copied and superposed to the head.

In sub-step 23, for the convenience of the following description, let the truncated tail signal have a length of L (i.e. the signal from the last L signal point to the last 1 signal point in the filtered signal is truncated as the truncated tail signal), and assume the total time-domain order L of the subband filter_filter. If the length L of the truncated tail signal is equal to (L)_filter-1), then when signals are superimposed, as shown in fig. 3(a), the 1 st signal point to the last signal point of the truncated tail signal (diagonal filled portion in the figure) is added (i.e.: the lth signal point) is sequentially superimposed on the header signal of the conventional filtered time-domain signal, i.e., superimposed on the 1 st signal point to the (L) th signal point of the conventional filtered time-domain signal_filter-1) signal points.

In sub-step 23, if the length of the truncated tail signal is greater than (L)_filter-1), then when signals are superimposed, as shown in fig. 3(b), the truncated tail signal is added(L-L) of number (diagonal filled portion in the figure)_filter+2) signal points to the last signal point (i.e.: the lth signal point) is sequentially superimposed on the header signal of the conventional filtered time-domain signal, i.e., superimposed on the 1 st signal point to the (L) th signal point of the conventional filtered time-domain signal_filter-1) signal points; dropping the 1 st signal point of the truncated tail signal to the (L-L) th signal point_filter+1) signal points are placed before the 1 st signal point of the filtered time domain signal.

The conventional filtering refers to a common filtering mode of non-cyclic filtering, such as band-pass filtering, low-pass filtering, and the like.

The operation of truncating the head signal of the filtered (or superimposed subband signal) signal and copying and superimposing it on the tail signal of the filtered (or superimposed subband signal) signal comprises:

Similarly, when the above operation is performed on the signals after the subband signals are superimposed, it is equivalent to copy and superimpose the head signals of all the subband signals to the tail simultaneously, and when the above operation is performed on the signals after filtering and before the subband signals are superimposed, the head signals of each subband signal are respectively copied and superimposed to the tail.

Optionally, time of different sub-band filtersL of the total order of the field_filterMay be different or the same.

When the sub-step 24 is performed first and then the operations of truncation, duplication and superposition are performed, if L is_filterOtherwise, the operations of truncation, copying and superposition of each sub-band should be processed separately, similar to the operations when the filtered signals and the signals before the sub-band signals are superposed are truncated, copied and superposed.

In an alternative embodiment, it is possible to set sub-step 21 (signal transformation) at the top and sub-step 25 (windowing) after sub-steps 22 (cyclic prefix and/or cyclic suffix addition) and 23 (cyclic filtering); substep 26 is set to the end.

In other embodiments, the order of sub-steps 21-25 may be self-arranging.

Optionally, after the step S120, the method may further include:

s130: the signal is back-end processed.

In S130, the purpose of the back-end processing is to reduce a peak-to-average ratio of the final time domain signal, enhance an ability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce energy consumption of the final time domain signal, reduce out-of-band energy leakage of the signal, and the like. The back-end processing may include, but is not limited to, any one or any few of the following: multiplication by a predetermined complex vector, peak clipping, etc.

The step S130 may be performed before or after any of the substeps 21 to 25.

In a second embodiment, a multicarrier signal generating apparatus, as shown in fig. 4, includes:

a segment mapping module 41, configured to segment the single carrier signal and map the single carrier signal to each subband position to obtain one or more subband signals;

a modulation filtering module 42, configured to perform multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering, and windowing on the sub-band signal.

The modulation filtering module 42 may include the following sub-modules connected in any order:

optionally, the signal transformation employed by the signal transformation sub-module may be implemented in various ways, including but not limited to IDFT or IFFT; corresponding to the transformation used when transforming the signal from the time domain to the frequency domain.

the signal superposition submodule is used for superposing the subband signals;

The sub-modules are connected in a front-to-back order, the input signal of the sub-module in the first order is the output signal of the segment mapping module, and the input signal of each of the other sub-modules is the output signal of the previous sub-module connected with the sub-module. Through the joint serial work of the modules/sub-modules, the extra cost of the cyclic prefix can be reduced, the system efficiency is improved, and the low out-of-band leakage characteristic is still achieved.

Optionally, the filtering submodule may perform filtering in a time domain, that is: the time domain signal may be filtered, or the equivalent filtering may be performed in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

Optionally, the windowing processing sub-module may perform windowing in the time domain, or perform equivalent operation of time domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

Optionally, the circular filtering submodule includes:

or comprises the following steps:

a subband signal filtering unit for filtering each subband signal respectively;

when the length of the truncated tail signal is greater than L ═ L_filterAt time-1, the L-L of the truncated tail signal_filterThe +2 signal points to the last signal point are sequentially superposed on the 1 st signal point to the Lth signal point of the corresponding signal_filter-1 signal point; the 1 st signal point of the truncated tail signal is connected to the L-L_filterThe 1 st signal point of the corresponding signal is placedBefore the signal point;

Optionally, L of different sub-bands_filterThe same or different.

Optionally, the apparatus may further include:

and the back-end processing module is used for performing back-end processing on the signal output by the modulation filtering module.

The purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal against intersymbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, reduce the out-of-band energy leakage of the signal, and the like. The back-end processing may include, but is not limited to, any one or any few of the following: multiplication by a predetermined complex vector, peak clipping, etc.

The back-end processing module may also be connected before or after any of the sub-modules comprised in the modulation filtering module.

The above embodiments are further illustrated below by 7 examples.

Example 1

The present example provides a method and apparatus for generating an improved multi-carrier signal with a cyclic prefix and/or cyclic suffix such that the CP signal overhead is greatly reduced, system efficiency is improved, and still has a smaller out-of-band leakage characteristic.

The specific method adopted in the present example is shown in fig. 5, and includes the following steps 101 to 108:

101. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

102. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

103. And filtering each subband signal through a filter to obtain a signal filtered by each subband.

In 103, the filtering may be performed in the time domain, the filtering may be performed on the time domain signal, or the filtering may be performed in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

103, for the convenience of the following description, assume that the total time domain order of the filter is L_filter。

102 and 103, and four implementations of several combinations of 102 and 103 are given in this example, and will be described later in this example.

104. The resulting filtered signals for each subband at 103 are superimposed.

In 104, the signal superposition may be performed in the time domain or in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

105. The head/tail signals of the signal are truncated and the truncated head/tail signals are copied and superimposed on the tail/head signals of the resulting signal at 104.

In the step 105, all the header/trailer signals are copied and superimposed at the same time, and in actual application, the header/trailer signals of each sub-band may be copied and superimposed (that is, the header/trailer signals are first copied and superimposed 105 and then 104 before the superimposition operation of 104).

In 105, the head signal of the signal generated in 104 may be cut off, copied, and superimposed on the tail signal of the signal generated in 104, or the tail signal of the signal generated in 104 may be cut off, copied, and superimposed on the head signal of the signal generated in 104. The two operations are similar and can be easily derived from each other, and a specific description is given below of the operations of truncating the tail signal of the signal generated at 104, copying and superimposing the tail signal of the signal generated at 104, and the operations of truncating the head signal of the signal generated at 104, copying and superimposing the head signal of the signal generated at 104 are similar and will not be described again.

In 105, if all the tail signals are copied and superimposed at the same time, the tail signal of the signal generated in 104 is cut off, the cut-off tail signal is copied and superimposed, and for convenience of subsequent description, the length of the cut-off tail signal is set to be L, and the specific operations of signal superimposition are as follows:

if the length L of the truncated tail signal is equal to (L)_filter-1), then the 1 st signal point to the last signal point of the tail signal are superimposed on the head signal of the 104 generated time domain signal in sequence when the signals are superimposed, i.e. superimposed on the 1 st signal point to the (L) th signal point of the 104 generated time domain signal_filter-1) signal points.

If the length of the truncated tail signal is greater than (L)_filter-1), then the (L-L) th of the tail signal when the signals are superimposed_filter+2) signal points to the last signal point are sequentially superimposed on the header signal of the time-domain signal generated at 104, i.e. superimposed on the 1 st signal point to the (L) th signal point of the time-domain signal generated at 104_filter-1) signal points; the 1 st signal point to the (L-L) th signal point of the tail signal_filter+1) signal points are placed in the 104 generated time domain signalNumber 1 signal point.

If the tail signals of each sub-band are respectively copied and superposed, the tail part of the signal generated by 103 is taken not to be cut off for each sub-band by 104, the cut-off tail signals are copied and superposed, and then 104 is carried out. For the convenience of the following description, let the length of the tail signal truncated by the sub-band be L, and the total order of the time domain of the sub-band filter be L_filterThe specific operation of signal superposition is as follows:

for each subband, if the length of the truncated tail signal is equal to (L)_filter-1), then the 1 st signal point to the last signal point of the tail signal are superimposed on the head signal of the 103 generated time domain signal in sequence when the signals are superimposed, i.e. superimposed on the 1 st signal point to the (L) th signal point of the 103 generated time domain signal_filter-1) signal points.

For each subband, if the length of the truncated tail signal is greater than (L)_filter-1), then the (L-L) th of the tail signal when the signals are superimposed_filter+2) signal points to the last signal point are sequentially superimposed on the head signal of the time-domain signal generated at 103, i.e. superimposed on the 1 st signal point to the (L) th signal point of the time-domain signal generated at 103_filter-1) signal points; the 1 st signal point to the (L-L) th signal point of the tail signal_filter+1) signal points are placed before the 1 st signal point of the time domain signal generated at 103.

106. A cyclic prefix and/or cyclic suffix is added to the signal generated by 105.

107. The signal generated by 106 is windowed.

In 107, the windowing process may be performed in the time domain, or may be equivalent to performing time-domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

108. The signal generated by 107 is back-end processed.

In 108, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and so on. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 6, and comprises, connected in sequence: a segmentation mapping module 811, a signal transformation submodule 812, a filtering unit 813 corresponding to the subbands one to one, a signal superposition submodule 814, a truncated copy superposition unit 815, an adding submodule 816, a windowing processing submodule 817, a back-end processing submodule 818, and the like.

In this example, the following four implementations of steps 102 and 103 are provided, and the practical application is not limited to the implementations described below.

In a first implementation, the inverse transform of 102 is inverse discrete fourier transform, and the filtering of 103 is performed by using BPF (Band-Pass Filter).

This implementation is shown in fig. 7 and includes: the inverse discrete Fourier transform units 5-1, 5-2, … … and 5-M; the filtering unit 813 comprises band pass filters 7-1, 7-2, … …, 7-M. The inverse discrete Fourier transform units are connected with the band-pass filters in a one-to-one correspondence mode.

The sub-band signals are respectively input into the inverse discrete Fourier transform units 5-1, 5-2, … … and 5-M; the band pass filters 7-1, 7-2, … …, 7-M output filtered subband signals, respectively.

The difference between the second implementation manner and the first implementation manner is that the filtering of the sub-unit 103 is implemented by using an LPF (Low pass filter) and a multiplier.

This implementation is shown in fig. 8 and includes: the inverse discrete Fourier transform units 5-1, 5-2, … … and 5-M; low pass filters 8-1, 8-2, … …, 8-M and M multipliers. The inverse discrete Fourier transform units are connected with the LPFs in a one-to-one corresponding manner; LPF is connected with the multipliers in one-to-one correspondence and is respectively connected with the reference frequency f_1-1、f_1-2、……、f_1-MMultiplying; is connected to oneA low-pass filter and a multiplier form a filter unit.

The sub-band signals are respectively input into the inverse discrete Fourier transform units 5-1, 5-2, … … and 5-M; the M multipliers respectively output the filtered subband signals.

The third implementation mode is different from the second implementation mode in that the sub-band 102 uses inverse fast fourier transform, all sub-band signals are inverse transformed by using one IFFT unit, and in 103, the inverse transformed signals are multiplied by a reference frequency, and then sent to the LPF, and the output of the LPF is up-sampled and then input to the multiplier.

This implementation is shown in fig. 9 and includes: an inverse fast Fourier transform unit 10-0; m first multipliers; low pass filters 8-1, 8-2, … …, 8-M; upsampling units 9-1, 9-2, … …, 9-M, and M second multipliers. The inverse fast Fourier transform unit is connected with the M multipliers and respectively connected with the reference frequency f_2-1、f_2-2、……、f_2-MAfter multiplication, inputting the multiplied signal into an LPF; the LPFs are connected with the up-sampling units in a one-to-one corresponding mode; the up-sampling units are correspondingly connected with the second multipliers one by one and respectively connected with the reference frequency f_1-1、f_1-2、……、f_1-MMultiplying; a first multiplier, a low-pass filter, an up-sampling unit and a second multiplier which are connected form a filtering unit.

All sub-band signals are input into an inverse fast Fourier transform unit 10-0; and M second multipliers respectively output the sub-band signals after modulation and filtration.

The difference between the fourth implementation and the second implementation is that 102 uses inverse fast fourier transform, and 103, the output of the LPF is up-sampled and then input to the multiplier.

This implementation is shown in fig. 10 and includes: inverse fast Fourier transform units 10-1, 10-2, … …, 10-M; low pass filters 8-1, 8-2, … …, 8-M; upsampling units 9-1, 9-2, … …, 9-M, and M multipliers. The inverse fast Fourier transform units are connected with the LPFs in a one-to-one corresponding manner; LPF andthe upper sampling units are connected in a one-to-one correspondence manner; the up-sampling units are correspondingly connected with the multipliers one by one and respectively connected with the reference frequency f_1-1、f_1-2、……、f_1-MMultiplying; a low-pass filter, an up-sampling unit and a multiplier which are connected form a filtering module.

The sub-band signals are respectively input into inverse fast Fourier transform units 10-1, 10-2, … … and 10-M; m multipliers respectively output the sub-band signals after modulation and filtering.

In the implementation mode, middle processing can be added between signal conversion and filtering; the middle treatment comprises any one or more of the following steps: multiply by a predetermined complex vector, add a cyclic prefix, add a cyclic suffix, etc. This middle processing may be another step, the above-described substep 22, and the like.

Signal transformation/filtering in other examples may refer to the implementations described above.

Example 2

The specific method adopted in the present example is shown in fig. 11, and includes the following steps 201 to 207:

201. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

202. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

203. A cyclic prefix and/or cyclic suffix is added to the 202 signal.

204. And performing cyclic filtering on each subband signal to obtain a signal after each subband is filtered.

In 204, the cyclic filtering may be performed in the time domain, filtering the time domain signal, or performing equivalent filtering in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

204, for the convenience of the following description, assume that the total time domain order of the subband filter is L_filter. The total number of time-domain orders of the filters used for different sub-bands may be different.

204, the circular filtering can be implemented by two specific embodiments:

the first method comprises the following steps: and obtaining the circularly filtered signal through a circulating filter.

The second method comprises the following steps: firstly, carrying out conventional filtering through a filter; then, the head/tail signal of the conventional filtered signal is truncated, and the truncated head/tail signal is copied and superimposed on the tail/head signal of the conventional filtered signal to obtain a cyclic filtered signal.

In the second method, the header signal of the normal filtered signal may be truncated and copied and superimposed on the tail signal of the normal filtered signal, or the tail signal of the normal filtered signal may be truncated and copied and superimposed on the header signal of the normal filtered signal. The two operations are similar and easy to be deduced by analogy, and a specific description is given below for truncating the tail signal of the conventionally filtered signal, and copying and superimposing the tail signal of the conventionally filtered signal, and the operations for truncating the head signal of the conventionally filtered signal, and copying and superimposing the head signal of the conventionally filtered signal on the tail signal of the conventionally filtered signal are similar and will not be described again.

In the second method, for convenience of description, the length of the truncated tail signal is L. If the length L of the truncated tail signal is equal to (L)_filter-1), then the 1 st signal point to the last signal point of the truncated tail signal are superimposed in sequence on the header signal of the conventional filtered time domain signal, i.e. superimposed, when the signals are superimposedIn the 1 st signal point to the (L) th signal point of the conventional filtered time domain signal_filter-1) signal points.

In the second method, if the length of the truncated tail signal is greater than (L)_filter-1), then the (L-L) th of the truncated tail signal when the signals are superimposed_filter+2) signal points to the last signal point are sequentially superimposed on the header signal of the time-domain signal after the conventional filtering, i.e. superimposed on the 1 st signal point to the (L) th signal point of the time-domain signal after the conventional filtering_filter-1) signal points; the 1 st signal point to the (L-L) th signal point of the tail signal_filter+1) signal points are placed before the 1 st signal point of the conventional filtered time domain signal.

205. For each sub-band, the signal generated by 204 is windowed.

In 205, the windowing process may be performed in the time domain, or may be equivalent to performing time domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

206. The resulting filtered signals for each subband are superimposed 205.

206. The signal superposition may be performed in the time domain or in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

207. The signal generated by 206 is back-end processed.

In 207, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and so on. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 12, and comprises, connected in series: a segmentation mapping module 821, a signal transformation sub-module 822, an adding sub-module 823, a circular filtering sub-module 824 corresponding to the sub-bands one by one, a windowing sub-module 825, a signal superposition sub-module 826, a back-end processing sub-module 827, and the like. Through the joint serial work of the modules/sub-modules, the additional expenditure of the CP can be reduced, the system efficiency is improved, and the low out-of-band leakage characteristic is still achieved.

Example 3

The specific method adopted in the present example is shown in fig. 13, and includes the following steps 301 to 307:

301. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

302. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

303. And performing cyclic filtering on each subband signal to obtain a signal after each subband is filtered.

303, the loop filtering may be performed in the time domain, filtering a time domain signal, or performing equivalent filtering in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

303, for the convenience of the following description, assume that the total time domain order of the subband filter is L_filterThen for that subband, the resulting time domain signal length of 302 is (time domain signal length of 301 + L)_filter-1). The total number of time-domain orders of the filters used for different sub-bands may be different.

303, the circular filtering can be implemented by two specific embodiments:

In the second method, for convenience of description, the length of the truncated tail signal is L. If the length L of the truncated tail signal is equal to (L)_filter-1), then the signal overlap time, the 1 st signal point to the last signal point of the truncated tail signal are superimposed in sequence on the header signal of the conventional filtered time domain signal, i.e. on the 1 st signal point to the (L) th signal point of the conventional filtered time domain signal_filter-1) signal points.

304: a cyclic prefix and/or cyclic suffix is added to the signal obtained 303.

305. The resulting 304 individual subband filtered signals are superimposed.

In 305, the signal superposition may be performed in the time domain or in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

306. For each subband, the signal generated by 305 is windowed.

In 306, the windowing process may be performed in the time domain, or may be equivalent to performing time domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

307. The resulting signal 305 is post-processed.

In 307, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and so on. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 14, and comprises, connected in series: a segment mapping module 831, a signal transformation submodule 832, a cyclic filtering submodule 833 corresponding to the sub-bands one by one, an adding submodule 834, a signal superposition submodule 835, a windowing processing submodule 836, a back-end processing submodule 837 and the like.

Example 4

The specific method adopted in the present example is shown in fig. 15, and includes the following steps 401 to 408:

401. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

402. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

403. A cyclic prefix and/or cyclic suffix is added to the signal generated by 402.

404. And filtering each subband signal through a filter to obtain a signal filtered by each subband.

In 404, the filtering may be performed in the time domain, the time domain signal may be filtered, or the equivalent filtering may be performed in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

In 404, for convenience of the following description, it is assumed that the total time domain order of the filter is L_filter。

405. The resulting 405 filtered signals for each subband are superimposed.

In 405, the signal superposition may be performed in the time domain or in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

406. The head/tail signal of the signal is truncated and the truncated head/tail signal is copied and superimposed on the tail/head signal of the signal generated at 405.

In 406, when the head/tail signals are copied and superimposed, the head/tail signals may be copied and superimposed at the same time, or the head/tail signals of the respective subbands may be copied and superimposed, respectively.

In 406, the head signal of the signal generated in 405 may be cut off, copied, and superimposed on the tail signal of the signal generated in 405, or the tail signal of the signal generated in 405 may be cut off, copied, and superimposed on the head signal of the signal generated in 405. The two operations are similar and can be easily derived from each other, and a specific description is given below of the operations of truncating the tail signal of the signal generated by 405, copying and superimposing the tail signal of the signal generated by 405, and the operations of truncating the head signal of the signal generated by 405, copying and superimposing the head signal of the signal generated by 405 are similar and will not be described again.

In 406, if all the tail signals are copied and superimposed at the same time, the tail signal of the signal generated in 405 is cut off, and the cut-off tail signal is copied and superimposed, and for convenience of the following description, the length of the cut-off tail signal is set to L, and the specific operations of signal superimposition are as follows:

if the length L of the truncated tail signal is equal to (L)_filter-1), then the signal is superimposed, the 1 st signal point to the last signal point of the tail signal are superimposed on the head signal of the time domain signal generated at 405 in sequence, i.e. the 1 st signal point to the (L) th signal point of the time domain signal generated at 405 are superimposed_filter-1) signal points.

If the length of the truncated tail signal is greater than (L)_filter-1), then the (L-L) th of the tail signal when the signals are superimposed_filter+2) signal points to the last signal point are superimposed in sequence on the header signal of the time-domain signal generated at 405, i.e. superimposed on the 1 st signal point to the (L) th signal point of the time-domain signal generated at 405_filter-1) signal points; the 1 st signal point to the (L-L) th signal point of the tail signal_filter+1) signal points are placed before the 1 st signal point of the time domain signal generated at 405.

In 406, if the tail signals of the respective subbands are copied and superimposed, the tail of the signal generated in 404 is truncated for each subband, and the truncated tail signals are copied and superimposed. For the convenience of the following description, let the length of the tail signal truncated by the sub-band be L, and the total order of the time domain of the sub-band filter be L_filterThe specific operation of signal superposition is as follows:

for each subband, if the length of the truncated tail signal is equal to (L)_filter-1), then the 1 st signal point to the last signal point of the tail signal are superimposed on the head signal of the time domain signal generated at 404 in sequence during signal superposition, i.e. superimposed on the time domain signal generated at 404Number 1 signal point to (L)_filter-1) signal points.

For each subband, if the length of the truncated tail signal is greater than (L)_filter-1), then the (L-L) th of the tail signal when the signals are superimposed_filter+2) signal points to the last signal point are sequentially superimposed on the header signal of the time-domain signal generated at 404, i.e. superimposed on the 1 st signal point to the (L) th signal point of the time-domain signal generated at 404_filter-1) signal points; the 1 st signal point to the (L-L) th signal point of the tail signal_filter+1) signal points are placed before the 1 st signal point of the time domain signal generated at 405.

407. The signal generated by 406 is windowed.

In 407, the windowing process may be performed in the time domain, or may be equivalent to performing time domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

408. The signal generated by 407 is post-processed.

In 408, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and so on. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 16 and comprises, in succession: a segment mapping module 841, a signal transformation sub-module 842, an addition sub-module 843, a filtering unit 844 corresponding to the sub-bands one by one, a signal superposition sub-module 845, a truncation and duplication superposition unit 846, a windowing sub-module 847, a back-end processing sub-module 848, and the like.

Example 5

The specific method adopted in the present example is shown in fig. 17, and includes the following steps 501 to 507:

501. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

502. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

503. And performing cyclic filtering on each subband signal to obtain a signal after each subband is filtered.

In 503, the cyclic filtering may be performed in the time domain, filtering the time domain signal, or performing equivalent filtering in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

503, for the convenience of the following description, assume that the total time domain order of the subband filter is L_filterThen for that subband, 502 results in a time domain signal length of (501 results in a time domain signal length + L)_filter-1). The total number of time-domain orders of the filters used for different sub-bands may be different.

503, the circular filtering can be implemented by two specific embodiments:

504. For each sub-band, a cyclic prefix and/or cyclic suffix is added to the signal generated by 503.

505. For each subband, the signal generated at 504 is windowed.

505, the windowing process may be performed in the time domain, or may be equivalent to performing time domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

506. The resulting 505 filtered signals for each sub-band are superimposed.

In 506, the signal superposition may be performed in the time domain or in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

507. The signal generated by 506 is back-end processed.

In 507, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and the like. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 18 and comprises, in succession: a segment mapping module 851, a signal transformation sub-module 852, a cyclic filtering sub-module 853 corresponding to the sub-bands one by one, an adding sub-module 854, a windowing sub-module 855, a signal superposition sub-module 856, a back-end processing sub-module 857, and the like.

Example 6

The specific method adopted in the present example is shown in fig. 19, and includes the following steps 601 to 607:

601. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

602. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

603. A cyclic prefix and/or cyclic suffix is added to the signal generated at 602.

604. And performing cyclic filtering on each subband signal to obtain a signal after each subband is filtered.

In 604, the cyclic filtering may be performed in the time domain, filtering the time domain signal, or performing equivalent filtering in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

604, for the convenience of the following description, assume that the total time domain order of the subband filter is L_filter. The total number of time-domain orders of the filters used for different sub-bands may be different.

604, the circular filtering can be implemented by two specific embodiments:

In the second method, for convenience of description, the length of the truncated tail signal is L. If the length L of the truncated tail signal is equal to (L)_filter-1), then the 1 st signal point to the last signal point of the truncated tail signal are superimposed in sequence on the header signal of the conventional filtered time domain signal when the signals are superimposedOn the sign, i.e. superimposed on the 1 st to (L) th signal points of the conventionally filtered time-domain signal_filter-1) signal points.

605. The resulting filtered signals for each subband are superimposed 604.

606. The signal generated by 605 is windowed.

At 606, the windowing may be performed in the time domain, or may be equivalent to performing time-domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

607. The signal generated 607 is post-processed.

In 607, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist intersymbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and so on. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 20 and comprises, connected in series: a segmentation mapping module 861, a signal transformation submodule 862, an adding submodule 863, a cyclic filtering submodule 864 corresponding to the sub-bands one by one, a signal superposition submodule 865, a windowing processing submodule 866, a back-end processing submodule 867, and the like.

Example 7

The specific method adopted in this example is shown in fig. 20, and includes the following steps 701 to 707:

701. and segmenting the single-carrier signal, and mapping to each subband position to obtain a plurality of subband signals.

702. And generating a time domain signal by respectively adopting IDFT or IFFT transformation on each sub-band signal.

703. And performing cyclic filtering on each subband signal to obtain a signal after each subband is filtered.

703, the cyclic filtering may be performed in the time domain, filtering a time domain signal, or performing equivalent filtering in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

703, for the convenience of the following description, assume that the total time domain order of the subband filter is L_filter. The total number of time-domain orders of the filters used for different sub-bands may be different.

703, the circular filtering can be implemented by two specific embodiments:

704. The resulting 703 filtered signals for each subband are superimposed.

705. A cyclic prefix and/or cyclic suffix is added to the signal generated at 704.

706. The signal generated by 705 is windowed.

In 706, the windowing may be performed in the time domain, or may be equivalent to performing time-domain windowing in the transform domain. The transform domain includes, but is not limited to, the frequency domain.

707. The generated signal is back-end processed 706.

In 707, the purpose of the back-end processing is to reduce the peak-to-average ratio of the final time domain signal, enhance the capability of the final time domain signal to resist inter-symbol interference or frequency deviation, reduce the energy consumption of the final time domain signal, and so on. The back-end processing includes, but is not limited to, the following operations: multiplication by a predetermined complex vector, peak clipping, etc.

The apparatus used in this example is shown in fig. 22, and comprises, connected in series: a segment mapping module 871, a signal transformation sub-module 872, a cyclic filtering sub-module 873 corresponding to the sub-bands one by one, a signal superposition sub-module 874, an adding sub-module 875, a windowing sub-module 876, a back-end processing sub-module 877, and the like.

It will be understood by those skilled in the art that all or part of the steps of the above methods may be implemented by instructing the relevant hardware through a program, and the program may be stored in a computer readable storage medium, such as a read-only memory, a magnetic or optical disk, and the like. Alternatively, all or part of the steps of the above embodiments may be implemented using one or more integrated circuits. Accordingly, each module/unit in the above embodiments may be implemented in the form of hardware, and may also be implemented in the form of a software functional module. The present invention is not limited to any specific form of combination of hardware and software.

Although the embodiments of the present invention have been described above, the above description is only for the convenience of understanding the present invention, and is not intended to limit the present invention. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims

1. A method of generating a multi-carrier signal, comprising:

2. The method of claim 1, wherein said multi-carrier modulation with cyclic prefix and/or cyclic suffix, sub-band filtering and windowing of sub-band signals comprises the following operations in any order:

adding a cyclic prefix and/or cyclic suffix to the signal;

circularly filtering the subband signals;

superposing the sub-band signals;

windowing is performed on the signal.

3. The method of claim 2, wherein the cyclically filtering the signal comprises:

performing cyclic filtering on each sub-band signal by using a cyclic filter;

or,

4. The method of claim 3, wherein the copying and superimposing the truncated tail signal on the head of the corresponding signal comprises:

when cutting offThe length of the broken tail signal is greater than L ═ L_filterAt time-1, the L-L of the truncated tail signal_filterThe +2 signal points to the last signal point are sequentially superposed on the 1 st signal point to the Lth signal point of the corresponding signal_filter-1 signal point; the 1 st signal point of the truncated tail signal is connected to the L-L_filterThe +1 signal point is placed before the 1 st signal point of the corresponding signal;

5. The method of claim 4, wherein:

l of different sub-bands_filterThe same or different.

6. The method of claim 1, wherein said processing said subband signals for multicarrier modulation with a cyclic prefix and/or cyclic suffix, subband filtering, and windowing further comprises:

7. An apparatus for generating a multicarrier signal, comprising:

8. The apparatus of claim 7, wherein the modulation filtering module comprises the following sub-modules connected in any order:

the signal superposition submodule is used for superposing the subband signals;

9. The apparatus of claim 8, wherein the circular filtering submodule comprises:

or comprises the following steps:

a subband signal filtering unit for filtering each subband signal respectively;

10. The apparatus of claim 9, wherein the truncated replica superimposing unit that duplicates and superimposes the truncated tail signal on the head of the corresponding signal means that:

11. The apparatus of claim 10, wherein:

l of different sub-bands_filterThe same or different.

12. The apparatus of claim 7, further comprising: