Data protection circuit, data protection processing device and data protection processing method

By employing a feedforward data path and data segmentation and merging technology in the data protection circuit, the data processing flow is optimized, solving the problems of limited throughput and hardware resource consumption in high-speed communication, and achieving more efficient data processing.

CN123069458APending Publication Date: 2026-11-10REALTEK SINGAPORE PTE LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202610938552.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2026-02-12
Filing Date
2026-06-26
Publication Date
2026-11-10

AI Technical Summary

Technical Problem

Existing encryption methods are limited in terms of data throughput improvement in high-speed or high-bandwidth communication environments due to sequence dependencies, and they also increase hardware resource consumption, especially in high fan-out and wide bit multiplexing.

Method used

A feedforward data path is adopted, which connects multiple computing circuits in series. Each computing circuit executes an iterative round of the data protection algorithm, and optimizes data processing through data segmentation and merging, thereby reducing the need for parallel processing resources.

Benefits of technology

It increases data throughput, reduces hardware resource consumption, solves the problems of high fan-out and wide bit multiplexing, and improves processing speed and efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123069458A_ABST
    Figure CN123069458A_ABST
Patent Text Reader

Abstract

A data protection circuit, a data protection processing device, and a data protection processing method are disclosed. The data protection circuit includes a data input terminal, multiple arithmetic circuits, and a data output terminal. The arithmetic circuits are connected in series to form a feedforward data path. The input terminal of the feedforward data path is coupled to the data input terminal and configured to acquire a data block to execute a data protection algorithm. Each arithmetic circuit is configured as a logic circuit that implements one iterative round of the data protection algorithm. The data output terminal is coupled to the output terminal of the feedforward data path and configured to output the result of the data block after executing the data protection algorithm.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to a data protection technology, and more particularly to a data protection circuit, a data protection processing device, and a data protection processing method. Background Technology

[0002] Generally, communication data encryption processing relies on the round-robin nature of encryption algorithms, using the same computing circuit to perform encryption processing on a data block round by round. In this approach, the data block must undergo multiple interdependent rounds of computation, and the processing result of each round must be fed back to the next round until all rounds are completed, resulting in a sequential dependency in the data processing process.

[0003] When applied to high-speed or high-bandwidth communication environments, the aforementioned sequence-dependent encryption methods often require processing a single data block over multiple clock cycles, thus limiting the increase in data throughput. To meet the demands of higher data transmission rates, existing technologies typically require increasing computing resources to process multiple data blocks in parallel, leading to increased hardware resource consumption and system complexity. For example, if the computing circuit completes one round of encryption computation per clock cycle, completing all encryption rounds for a data block requires multiple clock cycles, necessitating the configuration of multiple parallel computing instances to maintain data processing throughput. As throughput increases and the number of parallel computing instances grows, these instances will experience high fanout at their inputs, while their outputs will face the challenge of wide muxing. Summary of the Invention

[0004] An embodiment of the present invention provides a data protection circuit. The data protection circuit includes a data input terminal, multiple arithmetic circuits, and a data output terminal. The data input terminal is configured to receive a data block. The arithmetic circuits are sequentially connected in series to form a feedforward data path. The entry terminal of the feedforward data path is coupled to the data input terminal to obtain the data block and execute a data protection algorithm. Each arithmetic circuit is configured as a logic circuit implementing an iterative round of the data protection algorithm. The data protection algorithm is an encryption algorithm or a decryption algorithm. The data output terminal is coupled to the exit terminal of the feedforward data path and is configured to output the result of the data block after executing the data protection algorithm.

[0005] An embodiment of the present invention provides a data protection processing apparatus, including a data segmenter, multiple data protection circuits, and a data merger. The data segmenter is configured to receive an input data stream and segment the input data stream into multiple data blocks. The data protection circuits are coupled to the data segmenter and configured to receive these data blocks respectively. Each data protection circuit includes multiple computational circuits, sequentially connected in series to form a feedforward data path. These data blocks are respectively input to the inlet of the feedforward data path corresponding to these data protection circuits. Each computational circuit is configured as a logic circuit implementing an iterative round of a data protection algorithm. The data protection algorithm is an encryption algorithm or a decryption algorithm. The data merger is coupled to the outlet of the feedforward data path corresponding to these data protection circuits and configured to merge the results of each data block after executing the data protection algorithm and output them as an output data stream.

[0006] An embodiment of the present invention provides a data protection processing method, comprising: receiving a data block via a data input terminal; executing a data protection algorithm on the data block along a feedforward data path formed by sequentially connecting multiple arithmetic circuits, wherein each arithmetic circuit is configured as a logic circuit implementing an iterative round of the data protection algorithm, the data protection algorithm being an encryption algorithm or a decryption algorithm; and outputting the result of the data block after executing the data protection algorithm via a data output terminal. Attached Figure Description

[0007] Figure 1 This is a schematic diagram of the architecture of a data protection circuit according to an embodiment of the present invention.

[0008] Figure 2 This is a data processing timing diagram of a data protection circuit according to an embodiment of the present invention.

[0009] Figure 3 This is a schematic diagram of the architecture of a data protection processing device according to an embodiment of the present invention.

[0010] Figure 4 This is a schematic diagram of the architecture of a data protection circuit according to another embodiment of the present invention.

[0011] Figure 5 This is a schematic diagram of the architecture of a data protection circuit according to another embodiment of the present invention.

[0012] Figure 6 This is a schematic diagram of the architecture of a data protection communication device according to an embodiment of the present invention.

[0013] Figure 7 This is a schematic diagram of the architecture of a data protection communication device according to another embodiment of the present invention. Detailed Implementation

[0014] In all the accompanying drawings, the same reference numerals will be used to denote the same or similar components. The term "comprising" as used herein is an open-ended term and should be interpreted as "including but not limited to". As used herein, "coupled" means that two or more components are in physical or electrical contact with each other "directly" or "indirectly".

[0015] Merge reference Figure 1 and Figure 2 . Figure 1 This is a schematic diagram of the architecture of a data protection circuit 10 according to an embodiment of the present invention. Figure 2 This is a data processing timing diagram of a data protection circuit 10 according to an embodiment of the present invention. The data protection circuit 10 includes a data input terminal 11, multiple arithmetic circuits 12, and a data output terminal 13.

[0016] Data input terminal 11 is configured to receive an input data stream DI. The input data stream DI is a data sequence, where the data arrives in chronological order and can be continuous or discontinuous. The input data stream DI can be divided into multiple data blocks I (e.g., ...). Figure 2 As shown, data block I is represented by I1, I2, ..., In, In+1, etc. Data block I is the data length unit for each arithmetic circuit 12. Taking the Advanced Encryption Standard (AES) algorithm as an example, the length of data block I in AES-128 is 128 bits, the length of data block I in AES-192 is 192 bits, and the length of data block I in AES-256 is 256 bits.

[0017] The arithmetic circuits 12 are connected in series to form a feedforward data path 20. The input terminal 21 of the feedforward data path 20 is coupled to the data input terminal 11 and configured to obtain a data block I from the input data stream DI, and execute a data protection algorithm on this data block I. The data protection algorithm includes multiple iterative rounds. Multiple iterative rounds are configured to gradually approach the optimal solution, refine the data, or strengthen security protection by repeatedly executing the same logical steps. Each arithmetic circuit 12 is configured as the logic circuit implementing one iterative round of the data protection algorithm. Taking the AES algorithm as an example, AES-128 includes 10 iterative rounds, AES-192 includes 12 iterative rounds, and AES-256 includes 14 iterative rounds. In some embodiments, the arithmetic circuit 12 may also be referred to as an arithmetic engine.

[0018] A logic circuit implementing a complete iteration round of an encryption algorithm (such as AES encryption) includes multiple sub-logic circuits (combinatorial logic). The sub-logic circuits in the logic circuits excluding the last round correspond to the four steps in the AES encryption algorithm: SubBytes, ShiftRows, MixColumns, and AddRoundKey. The logic circuit in the last round only includes sub-logic circuits corresponding to the three steps: SubBytes, ShiftRows, and AddRoundKey. The SubBytes sub-logic circuit is configured to perform byte substitution operations. The ShiftRows sub-logic circuit is configured to perform row shift operations. The MixColumns sub-logic circuit is configured to perform column mixing operations. The AddRoundKey sub-logic circuit is configured to add the round key to the data. These sub-logic circuits are coupled to each other to form the operational circuit structure of a complete iteration round. In some embodiments, the AES encryption and decryption algorithms operate in AES-Counting (CTR) mode. AES's CTR mode is a working mode that transforms block ciphers (such as AES) into stream ciphers. It generates a key stream by encrypting an incrementing counter, and then XORs the key stream with the plaintext to obtain the ciphertext. The decryption process is the same.

[0019] Similarly, a logic circuit implementing a decryption algorithm (such as the AES decryption algorithm) in a complete iterative round includes multiple sub-logic circuits (combinational logic). The sub-logic circuits included in the logic circuits except the last round correspond to the four steps in the AES decryption algorithm. These steps include InvShiftRows, InvSubBytes, AddRoundKey, and InvMixColumns. The logic circuit in the last round only includes sub-logic circuits corresponding to the InvShiftRows, InvSubBytes, and AddRoundKey steps. The InvShiftRows, InvSubBytes, AddRoundKey, and InvMixColumns steps are the reverse operations of the aforementioned SubBytes, ShiftRows, MixColumns, and AddRoundKey steps, respectively; their respective operations will not be repeated here. These sub-logic circuits are coupled together to form the operational circuit structure of the complete iterative round.

[0020] like Figure 2As shown, the first arithmetic circuit 12 performs the first iteration round on data block I1 in one clock cycle, and then outputs the result M1 to the second arithmetic circuit 12; the first arithmetic circuit 12 performs the same first iteration round on data block I2 in the next clock cycle. That is, the first arithmetic circuit 12 performs the first iteration round of operation on each data block I. The second arithmetic circuit 12 then performs the second iteration round of operation on the operation results M1, M2, ..., Mn, Mn+1 of the first arithmetic circuit 12. And so on, the last arithmetic circuit 12 performs the final iteration round on the operation results N1, N2, ..., Nn, Nn+1 of the previous arithmetic circuit 12, to produce the final results O1, O2, ..., On, On+1, etc. Here, although one iteration round is performed in one clock cycle as an example, the embodiments of the present invention are not limited thereto. In some embodiments, the data protection algorithm is a symmetric-key algorithm. These algorithms use the same key for both encryption and decryption, or they use two keys that can be easily deduced from each other.

[0021] Data output terminal 13 is coupled to output terminal 22 of feedforward data path 20, and is configured to output the results O1, O2, ..., On, On+1 of data block I after the data protection algorithm is executed in sequence as output datastream DO. Figure 2 As shown, since each arithmetic circuit 12 only performs one iteration round of computation, the data throughput requirement can be met by configuring the number of arithmetic circuits 12 for the corresponding iteration round, without the need to add other parallel processing computing resources.

[0022] The aforementioned embodiment transmits one data block I per clock cycle, meaning the data path width of the data input terminal 11 is equivalent to one data block I. In some embodiments, the data path width may be less than one data block I (hereinafter referred to as a narrow data path, such as half the width of data block I); ​​in some embodiments, the data path width may be greater than one data block I (hereinafter referred to as a wide data path, such as twice the width of data block I).

[0023] Reference Figure 3This is a schematic diagram of the architecture of a data protection processing device 40 according to an embodiment of the present invention, illustrating data input terminal 11 for wide data path data processing. The data protection processing device 40 includes a data splitter 41, multiple data protection circuits 10, and a data merger 42. The data splitter 41 is configured to receive an input data stream DI and split the input data stream DI into multiple data blocks I. Here, two data protection circuits 10 are used as an example. Assume the data path width of the input data stream DI is twice that of the data blocks I; for example, when the length of data block I is 128 bits, the data path width of the input data stream DI is 256 bits. The data splitter 41 divides the 256 bits into two 128-bit data blocks I, which are respectively input to the two data protection circuits 10. In some embodiments, the data splitter 41 is a one-bit split logic.

[0024] These data protection circuits 10 are coupled to the data divider 41 and configured to receive these data blocks I respectively (i.e., data blocks I are respectively input to the input terminal of the feedforward data path 20 of the corresponding data protection circuit 10), and each executes a data protection algorithm on one data block I. For the specific structure and operation of the data protection circuit 10, please refer to the foregoing. Figure 1 The relevant descriptions of the embodiments will not be repeated here.

[0025] Data merger 42 is coupled to these data protection circuits 10 (i.e., the output terminals of the feedforward data paths 20 corresponding to these data protection circuits 10) and configured to merge the results of each data block I after performing the data protection algorithm and output them as an output data stream DO. Continuing with the previous example, the results of two 128-bit data blocks I after performing the data protection algorithm will be merged into a 256-bit stream. In some embodiments, data merger 42 is a one-bit merging logic.

[0026] Although the above description uses an input data stream DI with a data path width of 256 bits, divided into two 128-bit data blocks I, the embodiments of the present invention are not limited to this. The method of dividing the data path width depends on the alignment specifications and minimum frame length of the data frames. Taking a passive optical network (PON) as an example, it uses 4-byte alignment and a minimum frame length of 16 bytes.

[0027] If the data path width of the input data stream DI is 160 bits (20 bytes), it may include 4 bytes of data in the first frame and 16 bytes of data in the second frame, thus requiring two parallel data protection circuits 10.

[0028] If the data path width of the input data stream DI is 288 bits (36 bytes), it may include 16 bytes of data for two complete frames and 4 bytes of data for a partial frame, thus requiring three parallel data protection circuits 10.

[0029] If the data path width of the input data stream DI is 128 bits (16 bytes), it may consist of two different frames. Since the amount of data in the second frame is insufficient to execute the data protection algorithm, only one data protection circuit 10 needs to be configured (i.e., Figure 1 (as shown in the architecture), without adopting Figure 3 The architecture shown.

[0030] Reference Figure 4 This is a schematic diagram of the architecture of a data protection circuit 10 according to another embodiment of the present invention, illustrating data processing with a narrow data path at the data input terminal 11. Here, we will use an example where the data path width of the data input terminal 11 is half the length of data block I, meaning that the arithmetic circuit 12 accumulates one data block I every two clock cycles. If we follow the... Figure 1 The architecture shown results in a situation where a complete data block I is available for iteration in one clock cycle, but no data is available for processing in the next clock cycle. This leads to a waste of idle hardware resources. Therefore, compared to Figure 1 ,exist Figure 4 In the illustrated architecture, each arithmetic circuit 12 includes at least one pipeline buffer 31 to divide each arithmetic circuit 12 into multiple pipeline stages 32. Here, we take an example where each arithmetic circuit 12 includes one pipeline buffer 31. This pipeline buffer 31 divides the corresponding arithmetic circuit 12 into two pipeline stages 32. The pipeline buffer 31 can latch the result of the previous stage as the input to the next stage. The pipeline buffer 31 can be implemented using a set of flip-flops. Thus, each arithmetic circuit 12 completes two iterations in two clock cycles, matching the data throughput. Similarly, when a data block I is obtained from the input data stream DI after accumulating k clock cycles (k is a positive integer), the arithmetic circuit 12 is divided into k pipeline stages 32. That is, the number of clock cycles is equal to the number of pipeline stages 32. Since the number of iteration rounds divided by k may not be an integer, one of the computation circuits 12 may need to insert one or more no-operation (NOP) cycles to fill idle cycles and match the overall round allocation. By dividing the computation circuit 12 into k pipeline stages 32, the number of computation circuits 12 can be reduced (compared to...). Figure 1 The architecture is reduced to 1 / k times, and the throughput of each arithmetic circuit 12 is to complete the processing of one data block I every k clock cycles.

[0031] Reference Figure 5 This is a schematic diagram of the architecture of a data protection circuit 10 according to another embodiment of the present invention. It differs from the aforementioned... Figure 4 The pipeline buffer 31 is placed within the arithmetic circuit 12. Figure 5 Pipeline buffers 31 are arranged between these operational circuits 12 to form a plurality of pipeline stages 32. Each pipeline stage 32 includes at least one operational circuit 12. Figure 5 This example illustrates that each pipeline stage 32 includes two operational circuits 12. By configuring pipeline stages 32, signal transmission delay can be reduced. For instance, suppose the critical timing path in the logic circuitry of each iteration round includes 50 logic gates. For the AES-256, which has 14 iteration rounds, the critical timing path of the data protection circuit 10 has approximately 700 logic gates. Assuming a delay time of 40 ps for each logic gate, the total delay time is approximately 28 ns, which translates to a maximum clock frequency of approximately 35 MHz. By placing a pipeline buffer 31 in the middle of the 14 operational circuits 12, i.e., dividing it into two pipeline stages 32 (each pipeline stage 32 includes 7 operational circuits 12), the number of logic gates can be halved, thus increasing the maximum clock frequency by approximately double. In reality, it will be slightly less than double because each flip-flop adds additional setup / hold time and clock margin to the path, narrowing the available clock pane and thus offsetting some of the theoretical frequency gain. If a pipeline buffer 31 is placed between each operational circuit 12, that is, each pipeline stage 32 includes one operational circuit 12, the maximum clock frequency can be increased by approximately 14 times, or close to 500MHz. If an even faster clock frequency is required, the aforementioned method can be used. Figure 4 The architecture includes a pipeline buffer 31 inside the arithmetic circuit 12.

[0032] According to the foregoing Figure 4 and Figure 5 The illustrated architecture explains that by setting at least one pipeline buffer 31 in the feedforward data path 20, and dividing the path into multiple pipeline stages 32 based on the at least one pipeline buffer 31, critical timing paths can be shortened to increase clock frequency and accelerate data protection algorithm processing. Although the configuration of pipeline stages 32 is based on... Figure 4 and Figure 5 This is to illustrate, however, it can also be applied to Figure 3 The data protection circuit 10 shown.

[0033] Reference Figure 6This is a schematic diagram of the architecture of a data protection communication device 60 according to an embodiment of the present invention. The data protection communication device 60 is a communication device with data protection capabilities (data encryption and decryption). The data protection communication device 60 includes two data channels, namely a data transmitting unit 61 and a data receiving unit 62.

[0034] The data transmission unit 61 is configured to process the first unencrypted data Du1 into first encrypted data De1 and transmit the first encrypted data De1. The data transmission unit 61 includes at least one data encryption circuit 51 for encryption processing. The data encryption circuit 51 is equivalent to the aforementioned data protection circuit 10 and has multiple operational circuits 12 forming a feedforward data path 20. Each operational circuit 12 is configured as a logic circuit that implements one iteration round (encryption iteration round) of the encryption algorithm. Specifically, the data encryption circuit 51 may be as follows... Figure 1 The single data protection circuit 10 shown (here, a first data protection circuit 57 is taken as an example), or as... Figure 3 The number of data protection circuits 10 shown depends on the data path width. The detailed configuration is as described above and will not be repeated here. When the data encryption circuit 51 is configured with multiple data protection circuits 10, it has the following characteristics: Figure 3 The data splitter 41 (which splits the first unencrypted data Du1 into multiple data blocks I) and the data merger 42 (which merges the data to produce the first encrypted data De1) are configured in detail as described above and will not be repeated here. Figure 4 or Figure 5 It is configured with multiple pipeline stages 32, and the detailed configuration is as described above, so it will not be repeated here.

[0035] The data receiving unit 62 is configured to receive second encrypted data De2 and process the second encrypted data De2 into second unencrypted data Du2. The data receiving unit 62 includes at least one data decryption circuit 52 for decryption processing. The data decryption circuit 52 is equivalent to the aforementioned data protection circuit 10 and has multiple operational circuits 12 forming a feedforward data path 20. Each operational circuit 12 is configured as a logic circuit that implements one iteration of the decryption algorithm (decryption iteration round). Specifically, the data decryption circuit 52 may be as follows... Figure 1 The single data protection circuit 10 shown (here, a second data protection circuit 58 is used as an example), or as... Figure 3 The number of data protection circuits 10 shown depends on the data path width. Detailed configurations are as described previously and will not be repeated here. When the data decryption circuit 52 is configured with multiple data protection circuits 10, it has the following characteristics: Figure 3The data splitter 41 (which splits the second encrypted data De2 into multiple data blocks I) and the data merger 42 (which merges the data to produce the second unencrypted data Du2) shown are configured in detail as previously described and will not be repeated here. Figure 4 or Figure 5 It is configured with multiple pipeline stages 32, and the detailed configuration is as described above, so it will not be repeated here.

[0036] like Figure 6 As shown, the data protection communication device 60 also includes a coupled key expander 53 and an asynchronous crossing circuit 54. The asynchronous crossing circuit 54 can be a clock domain crossing (CDC) circuit, which can transmit signals between different clock domains. In some embodiments, the clock domain crossing circuit is implemented using a handshake synchronizer or an asynchronous FIFO. The key expander 53 is configured to pre-generate and store a series of round keys (referred to as first expanded key K1 and second expanded key K2, respectively) required for the operation of the data encryption circuit 51 and the data decryption circuit 52 based on the initial cipher key. These first expanded keys K1 correspond to multiple encryption iteration rounds of the encryption algorithm. These second expanded keys K2 correspond to multiple decryption iteration rounds of the decryption algorithm. The key expander 53 is located in either the data transmission unit 61 or the data receiving unit 62. Here, it is described as being located in the data transmission unit 61. Since the key expander 53 is located in the data transmission unit 61... Therefore, the key expander 53 operates within the clock domain of the data transmission unit 61, and the generated expanded key is stored in the data transmission unit 61. For keys required within the same clock domain (here, the first expanded key K1), they are directly sent to the data encryption circuit 51 for its use. For keys required within different clock domains (here, the second expanded key K2), they are transmitted via the asynchronous cross-clock transmission circuit 54 to a clock domain different from the one where the key expander 53 is located (here, they are sent to the data decryption circuit 52 for its use). Through the asynchronous cross-clock transmission circuit 54, the cross-clock logic burden can be reduced, and only a single key expander 53 can be used, saving hardware.

[0037] Reference Figure 7 This is a schematic diagram of the architecture of a data protection communication device 60 according to another embodiment of the present invention. Compared to Figure 6 , Figure 7Two key expanders (a first key expander 55 and a second key expander 56) are used, located in the data transmitting unit 61 and the data receiving unit 62 respectively, to generate a first expanded key K1 and a second expanded key K2, respectively, operating in their respective clock domains. The first key expander 55 provides the first expanded key K1 to the data encryption circuit 51. The second key expander 56 provides the second expanded key K2 to the data decryption circuit 52.

[0038] In some embodiments, the aforementioned key expander includes: one or more SubWord logic, a RotWord logic, a round constant generator, and an XOR logic.

[0039] In some embodiments, the data protection communication device 60 is applied to PON network technology. By adopting the configuration of the aforementioned pipeline stage 32, it can meet high-speed requirements such as 50Gbit and 100Gbit.

[0040] In one embodiment, the data protection communication device 60 has 256-bit data channels for both transmission and reception. The data encryption circuit 51 includes two first data protection circuits 57; the data decryption circuit 52 includes two second data protection circuits 58. The asynchronous cross-clock transmission circuit 54 has a 128-bit data path width.

[0041] In this paper, the data protection processing device 40 and the data protection communication device 60 are implemented in the form of integrated circuits (ICs), and each circuit module contained within the integrated circuit is implemented in dedicated hardware circuits.

[0042] The functions and operations of the aforementioned circuits and devices are also used to implement the corresponding data protection processing method. The input data stream DI is processed into data block I, the data protection algorithm is executed on data block I along the feedforward data path 20, and the final result O is output as the output data stream DO.

[0043] The above-disclosed content is only a preferred and feasible embodiment of the present invention, and is not intended to limit the scope of the patent application of the present invention. Therefore, all equivalent technical changes made based on the content of the present invention specification and drawings are included in the scope of the patent application of the present invention.

[0044] [Symbol Explanation] 10: Data Protection Circuit 11: Data Input Terminal 12: Operational circuit 13: Data output end 20: Feedforward data path 21: Entry point 22: Export end 31: Pipeline buffer 32: Pipeline Phase 40: Data Protection Processing Device 41: Data Splitter 42: Data Merger 51: Data encryption circuit 52: Data decryption circuit 53: Key Expander 54: Asynchronous cross-clock transmission circuit 55: First Key Expander 56: Second Key Expander 57: First Data Protection Circuit 58: Second Data Protection Circuit 60: Data Protection Communication Device 61: Data Transmission Department 62: Data Receiving Unit DI: Input Data Stream DO: Output data stream De1: First encrypted data De2: Second encrypted data Du1: First Unencrypted Data Du2: Second unencrypted data I, I1, I2, In, In+1: Data blocks M1, M2, Mn, Mn+1: Results N1, N2, Nn, Nn+1: Results O1, O2, On, On+1: Result K1: First Extended Key K2: Second Extended Key

Claims

1. A data protection circuit, comprising: A data input terminal, configured to receive a data block; Multiple arithmetic circuits are sequentially connected in series to form a feedforward data path. An input terminal of this feedforward data path is electrically coupled to the data input terminal to obtain the data block and execute a data protection algorithm. Each arithmetic circuit is configured as a logic circuit implementing an iterative round of the data protection algorithm, which is either an encryption algorithm or a decryption algorithm. A data output terminal, coupled to an exit terminal of the feedforward data path, is configured to output the result of the data block after the data protection algorithm has been executed.

2. The data protection circuit as claimed in claim 1 further includes at least one pipeline buffer disposed in the feedforward data path to divide and form multiple pipeline stages according to the at least one pipeline buffer.

3. The data protection circuit of claim 2, wherein each pipeline stage includes at least one of the arithmetic circuits.

4. The data protection circuit of claim 2, wherein each of the arithmetic circuits includes at least one of the pipeline buffers to divide each of the arithmetic circuits into a plurality of pipeline stages.

5. The data protection circuit of claim 4, wherein the data block is obtained by accumulating multiple clock cycles from an input data stream, the number of such clock cycles being equal to the number of pipeline stages.

6. A data protection processing device, comprising: Multiple data protection circuits as described in claim 1; A data splitter is configured to receive an input data stream and split the input data stream into data blocks, the data blocks being respectively input to the input end of the feedforward data path corresponding to the data protection circuit; and A data merger, coupled to the output end of the feedforward data path corresponding to the data protection circuits, is configured to merge the results of each data block after the data protection algorithm is executed and output them as an output data stream.

7. The data protection processing apparatus of claim 6, wherein each of the data protection circuits further includes at least one pipeline buffer disposed in the feedforward data path to divide and form a plurality of pipeline stages according to the at least one pipeline buffer.

8. The data protection processing apparatus of claim 7, wherein each pipeline stage includes at least one of the computing circuits.

9. The data protection processing apparatus of claim 7, wherein each of the arithmetic circuits includes at least one pipeline buffer to divide each of the arithmetic circuits into a plurality of pipeline stages.

10. A data protection processing method, comprising: Receive a data block via a data input terminal; A data protection algorithm is executed on the data block along a feedforward data path formed by multiple sequentially connected arithmetic circuits, wherein each arithmetic circuit is configured as a logic circuit implementing an iterative round of the data protection algorithm, which is either an encryption algorithm or a decryption algorithm; and The result of the data block after the data protection algorithm has been executed is output via a data output terminal.

11. The data protection processing method as described in claim 10, further comprising: Based on at least one pipeline buffer in the feedforward data path, multiple pipeline stages are formed.

12. The data protection processing method of claim 11, wherein each pipeline stage includes at least one of the computing circuits.

13. The data protection processing method of claim 11, wherein each of the arithmetic circuits includes at least one pipeline buffer to divide each of the arithmetic circuits into a plurality of pipeline stages.

14. The data protection processing method of claim 13, wherein the data block is obtained by accumulating multiple clock cycles from an input data stream, the number of such clock cycles being equal to the number of pipeline stages.