FFT calculation zero-filling control device, equipment and method based on reset control
By adopting a zero-compensation control device based on reset control in the FFT computing device, resetting and writing data directly in the data storage space, the problems of complex and low efficiency of zero-compensation operations in the prior art are solved, and efficient and real-time FFT calculation zero-compensation function is realized.
Patent Information
- Application Number
- CN202311118575.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-08-31
- Publication Date
- 2025-04-25
- Estimated Expiration
- 2043-08-31
AI Technical Summary
The zero-compensation operation in existing FFT computing devices is complex, costly and inefficient in execution, making it difficult to meet real-time requirements.
The FFT calculation zero-compensation control device based on reset control is adopted. By performing reset operations in the data storage space, the entire space is zeroed, and then the real-time collected data is directly stored in the corresponding address to realize the zero-compensation function, avoiding additional zero-compensation points judgment, calculation and address generation operations.
The complexity of the zero-compensation operation is simplified, the execution efficiency of the zero-compensation operation is improved, and the real-time and robustness of FFT calculations are ensured.
Smart Images

Figure CN117216457B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of FFT (Fast Fourier Transform) computing equipment, and in particular to a reset-control-based FFT computing zero-filling control device, equipment and method. Background Art
[0002] The number of input points of the collected data supported by the FFT computing device is closely related to the FFT algorithm used. For example, if the FFT uses a radix-N algorithm (N is usually 2 or 4), the hardware device can only support N. x (x=1,2,...x max , The maximum number of points supported by the designed FFT) is the input of the data collected. In practical applications, the number of data points M is usually not equal to N. x Completely fit, but satisfy the following relationship: N x-1 <M<N x , that is, the number of data points M collected at one time cannot reach the N required for FFT calculation x Point number requirement, so it is necessary to fill the number of collected data points with zeros to expand it to N x , thus meeting the FFT operation requirements.
[0003] In an FFT computing device for real-time data collection and calculation, the collected data is sequentially written into a readable and writable RAM in the form of a data stream. After all the collected data are written, the FFT computing device reads the corresponding data sequentially for calculation. Therefore, in the prior art, the zero-filling operation in the FFT computing device is usually implemented by controlling the zero-filling at the end of the data stream, that is, firstly determining the number of points to be stored, and after storing M data in the FFT data storage RAM, determining that a zero-filling operation is required and then calculating the number of zero-filling MN. x , generate the corresponding address, and then at the end of the data stream during the data stream writing process, MN is written according to the generated address x The zero data is written into the FFT data storage RAM.
[0004] However, the above method of controlling the zero-padding of FFT calculation at the end of the data stream requires multiple operations such as zero-padding judgment, zero-padding quantity calculation, and zero-padding corresponding address generation in order to complete the zero-padding operation of an FFT calculation. This is not only complex and costly to implement, but also has low overall zero-padding execution efficiency, making it difficult to meet real-time requirements and affecting the real-time performance of the FFT calculation. Summary of the invention
[0005] The technical problem to be solved by the present invention is: in response to the technical problems existing in the prior art, the present invention provides an FFT calculation zero-padding control device, processor and method based on reset control, which has a simple implementation method, low complexity, high execution efficiency, strong real-time performance and robustness.
[0006] In order to solve the above technical problems, the technical solution proposed by the present invention is:
[0007] A reset-control-based FFT calculation zero-filling control device comprises a reset-zeroing module, a data conduction control module and an output control module. When in a reset state, the data conduction control module controls the connection between the reset-zeroing module and a data storage space for storing data required for FFT calculation, so that the reset-zeroing module controls the reset-zeroing of the entire space in the data storage space. When in a normal conduction state, the data conduction control module controls the connection between the data storage space and a real-time data acquisition module in an FFT calculation device, so as to control the writing of M FFT calculation required data collected by the real-time data acquisition module in real time into the data storage space, where M is the number of data points collected at one time, and M≤N. x , N x is the number of FFT calculation points that the FFT calculation device can support, and the output control module is used to control the data storage space to output N each time x data for FFT calculation.
[0008] Furthermore, the reset and zeroing module includes a reset start unit, an address traversal unit and an all-zero data stream generating unit connected in sequence. When the reset start unit receives a reset signal, the address traversal unit is started to traverse all addresses of the data storage space. The all-zero data stream generating unit generates an all-zero data stream and inputs it to all traversed addresses to complete the full-space reset and zeroing operation of the data storage space.
[0009] Furthermore, the reset zeroing module also includes a counter circuit arranged between the reset start unit and the address traversal unit, so as to start counting when the reset is started until all addresses of the data storage space are traversed and generated.
[0010] Furthermore, when the data conduction control module receives a reset signal, the data input end and address line of the data storage space are connected to the data output end and address line of the reset zeroing module respectively; when a normal conduction signal is received, the data input end and address line of the data storage space are connected to the data output end and address line of the real-time data acquisition module respectively.
[0011] Furthermore, it also includes a state judgment module connected to the data conduction control module, which is used to receive a state signal and perform state judgment. When a reset signal is received, it is judged that the current state is reset; when a conduction signal is received, it is judged that the current state is normal conduction.
[0012] Furthermore, the data storage space is any one of RAM memory, flash memory, cache or auxiliary memory.
[0013] Furthermore, the output control module is connected to the data reading module of the FFT computing device to control the reading of N from the data storage space each time. x data to the data reading module to perform N x Point FFT calculation.
[0014] An FFT calculation device includes a data reading module and an FFT calculation module, wherein the data reading module is used to read the N required for FFT calculation. x The FFT calculation module calculates N data according to the data reading module. x The front end of the data reading module is also provided with the above-mentioned FFT calculation zero-filling control device, which is used to collect the M data required for FFT calculation in real time and fill in zeros to expand to N data. x After the data are received, the data are output to the data reading module.
[0015] A FFT calculation zero-filling control method based on reset control, the steps comprising:
[0016] When in the reset state, the control resets the entire space in the data storage space used to store the data required for FFT calculation to zero;
[0017] When it is in the normal conduction state, the control writes the M data required for FFT calculation collected in real time by the real-time data collection module in the FFT calculation device into the data storage space, where M is the number of data points collected at one time, M≤N x , N x The number of FFT calculation points that the FFT calculation device can support;
[0018] Control the data storage space to output N each time x data for FFT calculation.
[0019] Furthermore, the control will reset the entire space in the data storage space used to store the data required for FFT calculation to zero, including: starting counting when the reset is initiated, and traversing all addresses of the data storage space, generating an all-zero data stream input to all traversed addresses to complete the full-space reset and zeroing operation of the data storage space.
[0020] Compared with the prior art, the advantages of the present invention are: the present invention realizes the zero-filling operation in the FFT calculation process from the perspective of storage space control, by first resetting the data storage space, directly setting the entire space of the data storage space to zero during the reset, and then directly storing the real-time collected data into the corresponding address of the data storage space to cover the original 0 data, because the data on the remaining space addresses remain 0, and the FFT calculation has different groups of collected data separated by N x The original space with automatic zero padding after storing the real-time collected data will not be overwritten by other groups of collected data, thus making the control output N x The data is the M data collected in real time and N x -M zero data are used to realize the zero-filling function of the data to be processed by FFT. There is no need to perform additional complex operations such as zero-filling point judgment, zero-filling point calculation or zero setting. This can greatly reduce the complexity of the zero-filling operation and improve the execution efficiency of the zero-filling operation, thereby ensuring the real-time performance of the FFT calculation. BRIEF DESCRIPTION OF THE DRAWINGS
[0021] Figure 1 It is a structural diagram of the FFT calculation zero-padding control device based on reset control in this embodiment.
[0022] Figure 2 It is a schematic diagram of the principle of implementing FFT calculation based on zero-padding operation in this embodiment.
[0023] Figure 3 It is a schematic diagram of the structural principle of realizing FFT calculation zero-padding control in a specific application embodiment of the present invention.
[0024] Figure 4 It is a detailed implementation flow diagram of FFT calculation zero padding in a specific application embodiment of the present invention.
[0025] Figure 5 It is a structural diagram of the FFT calculation device in this embodiment.
[0026] Legend: 1. FFT calculation zero-filling control device; 101. Reset zeroing module; 111. Reset start unit; 112. Address traversal unit; 113. Data stream generation unit; 114. Counter circuit; 102. Data conduction control module; 103. Output control module; 2. Data storage space; 3. Real-time data acquisition module; 4. Data reading module. DETAILED DESCRIPTION
[0027] The present invention is further described below in conjunction with the accompanying drawings and specific preferred embodiments, but the protection scope of the present invention is not limited thereby.
[0028] As shown in the disclosure of the present invention, unless the context clearly indicates an exception, the words "one", "a kind of" and / or "the" do not specifically refer to the singular, but may also include the plural. The words "first", "second" and similar words used in the disclosure of the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "include" or "comprise" mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Words such as "connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.
[0029] In the FFT computing device, a real-time data acquisition module and an FFT data storage RAM are usually provided. The real-time data acquisition module is used to collect the data required for FFT calculation in real time and store it in the FFT data storage RAM. In order to realize the zero-filling operation in the FFT computing device, the traditional operation is usually to set a zero-filling device between the real-time data acquisition module and the FFT data storage RAM. When the real-time data acquisition module starts to store data in the FFT data storage RAM, the zero-filling device obtains the number of data currently stored at one time, M, and then makes a point judgment based on this to see whether the M value is N. x , if not N x Then you need to fill in zeros and calculate M and N x (N x-1 <M<N x ) x -M is the number of zero padding; after the real-time data acquisition module stores M data in the FFT data storage RAM, the zero padding device continues to generate N zero padding at the end of the data stream of the real-time data acquisition module. x -M samples of all-zero data stream, and according to the storage address of the Mth data of the real-time data acquisition module, N x -M samples storage address, store the required zero-filled data into FFT data storage RAM, fill M data with zeros to N x The real-time data acquisition module stores the current group of M real-time data acquisition data, regardless of whether the number of data M is equal to N. x , the next set of real-time data collection will always be N times the starting storage address. x As can be seen from the above, in order to realize the zero-padding operation of FFT calculation, the traditional operations such as point number judgment, zero-padding judgment, zero-padding quantity calculation, and zero-padding address generation are required, which makes the FFT zero-padding process complex, resource-intensive and inefficient.
[0030] FFT computing devices are usually equipped with a real-time data acquisition module for real-time acquisition of data required for FFT computing. In order to meet the FFT computing requirements of the FFT computing device, the real-time data acquisition module needs to store multiple groups of computing data. When storing each group of computing data in the real-time data acquisition module of the FFT computing device, the following storage rules need to be met: 1. Data from different groups will not be overwritten; 2. The initial storage addresses between different groups will be spaced N apart. x , that is, there must be an interval of N between different groups of data x In order to avoid the complex operations of point number judgment, zero-filling judgment, zero-filling quantity calculation, zero-filling address generation, etc. in the traditional zero-filling method, and to make full use of the above storage rules of the real-time data acquisition module in the FFT computing device, the present invention, from the perspective of storage space control, first resets the data storage space, so that the entire space of the data storage space is directly reset to zero during the reset, and then the real-time acquisition data is directly stored in the corresponding address of the data storage space to cover the original 0 data. Since the data on the remaining space addresses remain 0, and the FFT calculation has different groups of acquisition data, there will be an interval of N x The original space with automatic zero padding after storing the real-time collected data will not be overwritten by other groups of collected data, thus making the control output N x The data is the M data collected in real time and N x -M zero data are used to realize the zero-filling function of the data to be processed by FFT. There is no need to perform additional complex operations such as zero-filling point judgment, zero-filling point calculation or zero setting. This can greatly reduce the complexity of the zero-filling operation and improve the execution efficiency of the zero-filling operation, thereby ensuring the real-time performance of the FFT calculation.
[0031] like Figure 1 As shown, the FFT computing device is provided with a data storage space 2 for storing the data required for the FFT computing, and a real-time data acquisition module 3 for real-time acquisition of the data required for the FFT computing and writing it into the data storage space 2. The real-time data acquisition module 3 acquires M data required for the FFT computing each time, where M is the number of data points acquired at one time, and M≤N x , N x The number of FFT calculation points that the FFT calculation device can support. Therefore, in order to complete the FFT calculation, a zero padding operation is required to expand it to N. x The FFT calculation zero-filling control device 1 of this embodiment is used to expand the M FFT calculation data collected by the real-time data collection module to N x To perform FFT calculation, the output control module 103 in the FFT calculation zero-filling control device 1 is connected to the data reading module 4 of the FFT calculation device to control the reading of N from the data storage space 2 each time. x data to the data reading module 4 for Nx Point FFT calculation, such as Figure 2 shown.
[0032] like Figure 1 As shown, the reset-controlled FFT calculation zero-filling control device 1 of this embodiment specifically includes a reset-zero module 101, a data conduction control module 102 and an output control module 103. The reset-zero module 101, the output control module 103 and the real-time data acquisition module 3 are respectively connected to the data storage space 2. The data conduction control module 102 controls the connection state between the reset-zero module 101 and the data storage space 2, and between the real-time data acquisition module 3 and the data storage space 2. When it is in the reset state, the data conduction control module 102 controls the connection between the reset-zero module 101 and the data storage space 2 for storing the data required for the FFT calculation, so that the reset-zero module 101 controls the reset-zeroing of the entire space in the data storage space 2; when it is in the normal conduction state, the data conduction control module 102 controls the connection between the data storage space 2 and the real-time data acquisition module 3 in the FFT calculation device, so as to control the writing of M FFT calculation required data collected by the real-time data acquisition module 3 into the data storage space 2. The output control module 103 is used to control the data storage space 2 to output N x Since the entire data storage space 2 has been reset to zero in the reset state, the real-time data acquisition module 3 writes M data to the data storage space 2 and overwrites the original 0 data at the corresponding position. The final output control module 103 controls the output of N x The data is the M data collected in real time and N x -M zero data are used to implement the zero-filling function of FFT calculation.
[0033] In this embodiment, the reset zeroing module 101 specifically includes a reset start unit 111, an address traversal unit 112, and an all-zero data stream generating unit 113 connected in sequence. When the reset start unit 111 receives a reset signal, the address traversal unit 112 is started to traverse all addresses of the data storage space 2, and the all-zero data stream generating unit 113 generates an all-zero data stream and inputs it to all traversed addresses, that is, in the process of traversing the address, the all-zero data stream covers the full space address of the data storage space 2 to complete the full space reset zeroing operation of the data storage space 2. By adopting the above structure, an all-zero data stream can be generated in the reset stage, and the full space scan reset of the data storage space 2 is performed. Subsequently, the automatic FFT zero filling function can be conveniently and quickly completed according to normal data writing and reading operations.
[0034] In an optional embodiment, the reset start unit 111 may specifically adopt an enable control circuit, the enable control terminal accesses the reset signal, and the enable control terminal detects in real time whether the reset signal is received. If an enable signal (such as a high level or a low level, etc.) is generated, it is given to the address traversal unit 112 to start the address traversal unit 112 to scan and traverse the entire data storage space 2. It is understandable that the reset start unit 111 may also specifically adopt other forms of start control circuits according to actual needs.
[0035] In an optional embodiment, the address traversal unit 112 can be implemented by combining an address reading circuit with a controller, and the controller controls the address reading circuit to traverse and read all addresses in the data storage space 2. The all-zero data stream generation unit 113 can be implemented by combining a signal generator with a controller, and the controller controls the signal generator to generate an all-zero data stream, and writes zero data to all read addresses, so as to achieve full-space scan reset of the data storage space 2. It can be understood that the address traversal unit 112 and the all-zero data stream generation unit 113 can also be implemented by other types of structures according to actual needs.
[0036] In this embodiment, the reset zeroing module 101 further includes a counter circuit 114 disposed between the reset start unit 111 and the address traversal unit 112, for starting counting when the reset is started until the address traversal unit 112 traverses all addresses of the data storage space 2, so as to control the full space address scanning of the data storage space 2 by using the counter circuit 114. The counter threshold of the counter circuit 114 is specifically set according to the storage space size of the data storage space 2.
[0037] In this embodiment, when the data conduction control module 102 receives a reset signal, the data input terminal and address line of the data storage space 2 are connected to the data output terminal and address line of the reset zeroing module 101 respectively, so that when in the reset state, the reset zeroing module 101 can perform a full-space scan and zeroing of the data storage space 2; when the normal conduction signal is received, the data input terminal and address line of the data storage space 2 are connected to the data output terminal and address line of the real-time data acquisition module 3 respectively, so that when in the normal conduction state, the real-time data acquisition module 3 can normally write the real-time collected data into the data storage space 2.
[0038] In this embodiment, a state judgment module connected to the data conduction control module 102 is also included to receive the state signal and perform state judgment. When a reset signal is received, it is judged that the current state is reset, and when a conduction signal is received, it is judged that the current state is normal conduction. Specifically, each time an FFT calculation is required, a reset signal is first sent to the state judgment module. The state judgment module monitors the received state signal in real time. When the reset signal is detected to be valid, the data conduction control module 102 connects the connection between the data storage space 2 and the reset-zero module 101 to perform the reset-zero operation of the entire RAM space. When the M real-time acquisition data that need to be filled with zeros enter the data storage space 2, a normal conduction signal is sent to the state judgment module. After the state judgment module detects the normal conduction signal, it controls the connection between the data storage space 2 and the real-time acquisition data module 3. The data is normally written into the first M spaces, overwriting the original zero-value data, and the last N x -M spaces are not written, and the zero value remains unchanged. At this point, M real-time collection data has completed zero filling to N x The requirement for an FFT calculation of the data.
[0039] It is understandable that the reset state can also be entered directly by sending a reset signal to the data conduction control module 102, and the state can be switched by controlling the state to enter the normal conduction state according to the reset success signal returned after the reset is completed. The specific switching implementation method can be selected and configured according to actual needs. For example, the data conduction control module 102 monitors the reset signal in real time. When it detects that a valid reset signal has been received, the connection between the data storage space 2 and the reset-zeroing module 101 is connected through the data conduction control module 102, and the reset-zeroing operation of the entire RAM space is performed. After the reset-zeroing operation of the entire RAM space is completed, a reset success signal is returned to the data conduction control module 102; when the data conduction control module 102 receives the reset success signal, it controls the connection between the data storage space 2 and the real-time data acquisition module 3, and the data is normally written into the first M spaces, overwriting the original zero-value data, and completing the zero padding to N. x The need for an FFT to calculate data.
[0040] The steps of the zero-padding control method using the FFT calculation of the above device in this embodiment include:
[0041] S01. When in the reset state, the control will be used to store the data required for FFT calculation data in the data storage space 2 full space reset to zero;
[0042] S02. When it is in the normal conduction state, the control writes the M FFT calculation data required by the real-time data acquisition module 3 in the FFT calculation device into the data storage space 2, where M is the number of data points collected at one time, M≤N x , Nx The number of FFT calculation points that the FFT calculation device can support;
[0043] S03. Control data storage space 2 to output N each time x data for FFT calculation.
[0044] In this embodiment, the control in step S01 to reset the entire space in the data storage space 2 used to store the data required for the FFT calculation specifically includes: starting counting when the reset is started, and traversing all addresses of the data storage space, generating an all-zero data stream and inputting it into all traversed addresses to complete the full-space reset and zeroing operation of the data storage space 2.
[0045] The FFT calculation zero-filling control method based on reset control in this embodiment is consistent with the principle of the FFT calculation zero-filling control device based on reset control described above, and will not be described in detail here.
[0046] In this embodiment, the data storage space may be a RAM memory, a flash memory, a cache or an auxiliary memory, etc., which may be configured according to actual needs. The following takes the use of RAM as the FFT calculation data storage space as an example to further illustrate the above method of the present invention.
[0047] like Figure 3 As shown, in a specific application embodiment of the present invention, the reset zeroing module 101 in the reset-controlled FFT calculation zero-filling control device is specifically used to control the RAM to reset and zero, and the external reset signal is used as the start signal of the module. When the reset signal is valid, the module sets the state signal to the reset state and outputs it to the data conduction control module 102. The internal counter of the module starts to start, traverses and generates all addresses of the RAM, and generates zero data, resets the output data corresponding to the address, sets the counter threshold according to the size of the RAM space, and when the counter reaches the threshold, the reset of the RAM full address space is completed, and the module sets the state signal to the conduction state, and outputs the address and data of the real-time data acquisition module 3 to the data conduction control module 102 for data conduction control. The data conduction control module 102 selects the connection state of the RAM data and address lines according to the state signal input by the RAM reset module, wherein when the state signal is in the reset state, the RAM data and address lines are connected to the data and address lines of the RAM reset module, and when the state signal is in the conduction state, the RAM data and address lines are connected to the data and address lines of the real-time data acquisition module 3.
[0048] like Figure 4 As shown, the detailed steps of using the above device of this embodiment to implement the zero-filling function of FFT calculation and then implement FFT calculation are:
[0049] Step 1) Input a valid reset signal externally and the device enters reset mode.
[0050] Step 2) The reset and zeroing module 101 is started, the state signal is set to the reset state, and is output to the data conduction control module 102. The internal counter of the module works, traverses all addresses of the generated RAM, and the corresponding output data is reset to zero.
[0051] Step 3) The data conduction control module 102 connects the RAM data and address lines to the data and address lines of the RAM reset module according to the received reset state signal.
[0052] Step 4) The internal counter of the reset zeroing module 101 reaches the set threshold, the RAM full space reset is completed, the device enters the conduction mode, and the status signal is set to the conduction state.
[0053] Step 5) The data conduction control module 102 connects the RAM data and address lines to the data and address lines of the real-time data acquisition module 3 according to the received conduction state signal, thereby restoring the normal read and write functions of the RAM.
[0054] Step 6) The data reading module 4 reads the real-time collected data and the zero-filled data, and performs N x Point FFT operation.
[0055] like Figure 5 As shown, the FFT calculation device of this embodiment specifically includes a data reading module 4 and an FFT calculation module. The data reading module 4 is used to read the N required for FFT calculation. x The FFT calculation module calculates N data according to the data reading module. x The front end of the data reading module 4 is also provided with the above-mentioned FFT calculation zero-filling control device 1, which is used to collect the M data required for FFT calculation in real time and fill in zeros to expand to N data. x After the data is output to the data reading module 4, it is then provided to the FFT calculation module for N x FFT calculation of the point.
[0056] The above is only a preferred embodiment of the present invention, and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment, it is not intended to limit the present invention. Therefore, any simple modification, equivalent change and modification made to the above embodiment according to the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the scope of protection of the technical solution of the present invention.
Claims
1. A FFT calculation zero-filling control device based on reset control, characterized in that: The invention comprises a reset zeroing module (101), a data conduction control module (102) and an output control module (103), and also comprises a state judgment module connected to the data conduction control module (102) for receiving a state signal and performing state judgment. When a reset signal is received, it is judged that the current state is a reset state. When the state is a reset state, the data conduction control module (102) controls the connection between the reset zeroing module (101) and a data storage space (2) for storing data required for FFT calculation, so that the reset zeroing module (101) controls the reset zeroing of the entire space in the data storage space (2). When a conduction signal is received, it is judged that the current state is a normal conduction state. When the state is a normal conduction state, the data conduction control module (102) controls the connection between the data storage space (2) and a real-time data acquisition module (3) in an FFT calculation device, so as to control the writing of M data required for FFT calculation collected by the real-time data acquisition module (3) into the data storage space (2), where M is the number of data points collected at one time, and M≤N. x , N x is the number of FFT calculation points that the FFT calculation device can support, and the output control module (103) is used to control the data storage space (2) to output N each time. x data for FFT calculation.
2. The FFT calculation zero-padding control device based on reset control according to claim 1, characterized in that: The reset and zeroing module (101) comprises a reset start unit (111), an address traversal unit (112) and an all-zero data stream generation unit (113) which are connected in sequence. When the reset start unit (111) receives a reset signal, the address traversal unit (112) is started to traverse all addresses of the data storage space. The all-zero data stream generation unit (113) generates an all-zero data stream which is input to all traversed addresses to complete the full-space reset and zeroing operation of the data storage space (2).
3. The FFT calculation zero-padding control device based on reset control according to claim 2, characterized in that: The reset zeroing module (101) further comprises a counter circuit (114) arranged between the reset start unit (111) and the address traversal unit (112), for starting counting when the reset is started until all addresses of the data storage space (2) are traversed.
4. The FFT calculation zero-padding control device based on reset control according to claim 1, characterized in that: When the data conduction control module (102) receives a reset signal, the data input end and address line of the data storage space (2) are connected to the data output end and address line of the reset zeroing module (101) respectively; when the normal conduction signal is received, the data input end and address line of the data storage space (2) are connected to the data output end and address line of the real-time data acquisition module (3) respectively.
5. The FFT calculation zero-padding control device based on reset control according to any one of claims 1 to 3, characterized in that: The data storage space (2) is any one of a RAM memory, a flash memory, a cache or an auxiliary memory.
6. The FFT calculation zero-padding control device based on reset control according to any one of claims 1 to 3, characterized in that: The output control module (103) is connected to the data reading module (4) of the FFT computing device to control the reading of N from the data storage space (2) each time. x data to the data reading module (4) to perform N x Point FFT calculation.
7. An FFT calculation device, comprising a data reading module and an FFT calculation module, wherein the data reading module is used to read the N required for FFT calculation x The FFT calculation module calculates N data according to the data reading module. x The FFT calculation is performed on the data, which is characterized by: The front end of the data reading module is also provided with an FFT calculation zero-filling control device as claimed in any one of claims 1 to 6, so as to collect M data required for FFT calculation in real time and fill in zeros to expand to N data. x After the data are received, the data is output to the data reading module.
8. A control method for the FFT calculation zero-filling control device based on reset control according to any one of claims 1 to 6, characterized in that the steps include: When in the reset state, the control resets the entire space in the data storage space (2) used to store the data required for FFT calculation to zero; When it is in the normal conduction state, the control writes the M data required for FFT calculation collected in real time by the real-time data collection module (3) in the FFT calculation device into the data storage space (2), where M is the number of data points collected at one time, and M≤N x , N x The number of FFT calculation points that the FFT calculation device can support; Control the data storage space (2) to output N each time x data for FFT calculation.
9. The control method according to claim 8, characterized in that: The control of resetting the entire space in the data storage space (2) for storing data required for FFT calculation includes: starting counting when the reset is started, traversing all addresses of the data storage space (2), generating an all-zero data stream and inputting it into all traversed addresses to complete the full-space reset and zeroing operation of the data storage space (2).
Citation Information
Patent Citations
DSP image matching method based on operation resource load balance technology
CN108765471A