Method and apparatus for forward error correction encoding of magnetic resonance data and magnetic resonance imaging system

Abstract

The embodiment of the application discloses a forward error correction encoding method and device of magnetic resonance data and a magnetic resonance imaging system. The method comprises the following steps: sequentially reading each original MR signal sample group from the digital MR signal collected by the MR wireless coil end; for each original MR signal sample group, finding the maximum effective bit width in the effective bit width of each original MR signal sample in the group as the effective bit width of the group; determining the puncturing mode used for the FEC encoding of each original MR signal sample in the group according to the effective bit width of the group, wherein the smaller the effective bit width of the group is, the less the corresponding puncturing number of the group is; for any original MR signal sample in any group, determining the corresponding effective MR signal sample of the original MR signal sample according to the effective bit width of the group, and performing the FEC encoding on the effective MR signal sample according to the corresponding puncturing mode of the group. The embodiment of the application improves the utilization rate of the channel bandwidth resource when the MR data is wirelessly transmitted, and improves the reliability of the MR data transmission.

Description

Technical Field

[0001] This invention relates to the field of MRI (Magnetic Resonance Imaging) technology, and in particular to FEC (Forward Error Correction) encoding methods, devices and MRI systems for MR (Magnetic Resonance) data. Background Technology

[0002] In MR (Magnetic Resonance) systems, wireless coil systems typically employ MIMO (Multiple-Input Multiple-Output) communication technology with 4 transmitters and 8 receivers. Because the four transmitters share spectrum resources and transmit simultaneously, mutual interference is inevitable. This interference stems not only from signal superposition and aliasing in space but is also affected by factors such as differences in antenna layout, channel fading, and multipath effects. Simultaneously, due to transient environmental interference and uneven antenna distribution, bit errors are inevitable in wireless transmission systems.

[0003] When the signals emitted by each transmitter of the wireless coil undergo parallel-to-serial conversion, the bit error location may occur at the highest bit of the signal sample. In MR imaging applications, this type of bit error directly introduces artifact interference, severely affecting the image quality and diagnostic accuracy of the MR imaging area.

[0004] Currently, wireless coil communication systems in MR systems generally employ fixed FEC encoding / decoding schemes, such as the classic (9,12,7) convolutional code encoder, where 9 represents the bit width of the original data, 12 represents the bit width of the FEC-encoded data, and 7 represents the bit width of the encoding register. However, due to the characteristics of MR sampled data, this traditional scheme has significant limitations: the sampling bit width of MR systems is typically set to 18 bits, but in actual scanning, the effective data bit width in most imaging scenarios only occupies 11-13 bits. This results in redundant transmission of the remaining high-order bits, wasting channel bandwidth resources and failing to effectively improve data reliability. Summary of the Invention

[0005] In view of this, embodiments of the present invention propose, on the one hand, an FEC encoding method and apparatus for MR data to improve the utilization of channel bandwidth resources during wireless transmission of MR data and to improve the reliability of MR data transmission; on the other hand, an MRI system is proposed to improve the utilization of channel bandwidth resources during wireless transmission of MR data and to improve the reliability of MR data transmission.

[0006] A forward error correction (FEC) coding method for magnetic resonance (MR) data, characterized in that the method includes:

[0007] The MR wireless coil terminal performs analog-to-digital conversion on the acquired analog MR signal to obtain a digital MR signal;

[0008] Based on the number of coded MR signal samples contained in a training frame and the length of an original MR signal sample, each original MR signal sample group is sequentially read from the digital MR signal, wherein the number of original MR signal samples contained in each original MR signal sample group is equal to the number of coded MR signal samples contained in a training frame.

[0009] For each group of original MR signal samples read, the maximum effective bit width is found among the effective bit widths of the original MR signal samples in the group, and the maximum effective bit width is taken as the effective bit width of the group. Based on the effective bit width of the group, the puncturing method used for FEC encoding of each original MR signal sample in the group is determined. The smaller the effective bit width of the group, the fewer the number of punctures in the corresponding puncturing method of the group.

[0010] For any original MR signal sample within any original MR signal sample group, the effective MR signal sample corresponding to the original MR signal sample is determined according to the effective bit width of the group, wherein the bit width of the effective MR signal sample is equal to the effective bit width of the group; the effective MR signal sample is then FEC encoded according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0011] Finding the maximum effective bit width among the effective bit widths of each original MR signal sample point within the group includes:

[0012] Among all the original MR signal samples in the group, find the largest sample and use the effective bit width of the largest sample as the maximum effective bit width.

[0013] The step of determining the puncturing method for FEC encoding of each original MR signal sample point within the group based on the effective bit width of the group includes:

[0014] In the pre-defined correspondence between each effective bit width and the punching method, find the punching method corresponding to the effective bit width of this group;

[0015] The correspondence between each effective bit width and the punching method is determined as follows:

[0016] For any valid bit width, the total input bit width of the puncturing process corresponding to the valid bit width is obtained based on the valid bit width and the number of convolutional codes used in the pre-set FEC encoding; the number of punctures corresponding to the valid bit width is obtained based on the total input bit width of the puncturing process and the output bit width of the FEC encoding; and each puncture code corresponding to the valid bit width is determined based on the number of punctures and the number of puncture codes used in the pre-set FEC encoding, wherein the total number of 0 bits contained in all puncture codes is equal to the number of punctures corresponding to the valid bit width.

[0017] The step of performing FEC encoding on the effective MR signal samples according to the corresponding punching method includes:

[0018] The valid MR signal sample is encoded using each of the pre-defined convolutional codes used in FEC encoding.

[0019] For each encoding result, select one punch code sequentially from the punch codes corresponding to the effective bit width of the group, and punch the encoding result to obtain a punch result.

[0020] The punching results are combined according to a predefined punching result combination method to obtain the coded MR signal sample corresponding to the original MR signal sample.

[0021] After obtaining the encoded MR signal sample corresponding to the original MR signal sample, the process further includes:

[0022] For each original MR signal sample group, a training frame is constructed based on the coded MR signal sample corresponding to each original MR signal sample in the group. At the same time, the training frame carries the FEC coding information used by the group. The FEC coding information is used to indicate the effective bit width and punching method of the group. The training frame is then transmitted through the transmit antenna.

[0023] The MR system receives a training frame from the receiving antenna, parses each coded MR signal sample from the training frame, determines the puncturing method indicated by the FEC encoding information carried in the training frame, performs FEC decoding on each coded MR signal sample according to the puncturing method, obtains each decoded MR signal sample, and fills the missing high bits of each decoded MR signal sample with zeros according to the effective bit width indicated by the FEC encoding information carried in the training frame, to obtain the recovered original MR signal sample.

[0024] The step of performing FEC decoding on each coded MR signal sample according to the puncturing method to obtain each decoded MR signal sample includes:

[0025] Based on the punching method, determine the corresponding punch codes;

[0026] For each coded MR signal sample in the training frame, multiple puncture results are recovered from the coded MR signal sample according to the determined puncture codes and the predefined combination of puncture results; each puncture result is decoded using the pre-set convolutional codes used for FEC encoding to obtain the decoded MR signal sample of the coded MR signal sample. The bit width of the decoded MR signal sample is equal to the effective bit width indicated by the FEC encoding information carried in the training frame.

[0027] A forward error correction (FEC) encoding device for magnetic resonance (MR) data, the device being located at the MR wireless coil end, the device comprising:

[0028] Raw MR signal buffer module: buffers the digital MR signal obtained by analog-to-digital conversion of the acquired analog MR signal at the MR wireless coil end;

[0029] Maximum effective bit width detection module: Sequentially reads each original MR signal sample in an original MR signal sample group, wherein the number of original MR signal samples in an original MR signal sample group is equal to the number of encoded MR signal samples in a training frame; for each original MR signal sample group read, finds the maximum effective bit width among the effective bit widths of each original MR signal sample in the group, and uses the maximum effective bit width as the effective bit width of the group; based on the effective bit width of the group, determines the puncturing method used for FEC encoding of each original MR signal sample in the group, wherein the smaller the effective bit width of the group, the fewer punctures are used in the corresponding puncturing method of the group;

[0030] Dynamic FEC encoding module: For any original MR signal sample in a group of original MR signal samples, determine the effective MR signal sample corresponding to the original MR signal sample based on the effective bit width of the group, wherein the bit width of the effective MR signal sample is equal to the effective bit width of the group; perform FEC encoding on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0031] The maximum effective bit width detection module searches for the maximum effective bit width among the effective bit widths of each original MR signal sample point in the group, including:

[0032] Among all the original MR signal samples in the group, find the largest sample and use the effective bit width of the largest sample as the maximum effective bit width.

[0033] The maximum effective bit width detection module determines the punching method used for FEC encoding of each original MR signal sample point in the group based on the effective bit width of the group, including:

[0034] In the pre-defined correspondence between each effective bit width and the punching method, find the punching method corresponding to the effective bit width of this group;

[0035] The correspondence between each effective bit width and the punching method is determined as follows:

[0036] For any valid bit width, the total input bit width of the puncturing process corresponding to the valid bit width is obtained based on the valid bit width and the number of convolutional codes used in the pre-set FEC encoding; the number of punctures corresponding to the valid bit width is obtained based on the total input bit width of the puncturing process and the output bit width of the FEC encoding; and each puncture code corresponding to the valid bit width is determined based on the number of punctures and the number of puncture codes used in the pre-set FEC encoding, wherein the total number of 0 bits contained in all puncture codes is equal to the number of punctures corresponding to the valid bit width.

[0037] The dynamic FEC encoding module performs FEC encoding on the effective MR signal sample points according to the corresponding punching method, including:

[0038] The valid MR signal sample is encoded using each of the pre-defined convolutional codes used in FEC encoding.

[0039] For each encoding result, select one punch code sequentially from the punch codes corresponding to the effective bit width of the group, and punch the encoding result to obtain a punch result.

[0040] The punching results are combined according to a predefined punching result combination method to obtain the coded MR signal sample corresponding to the original MR signal sample.

[0041] The device further includes a transmission mapping module, used for:

[0042] For each original MR signal sample group, a training frame is constructed based on the coded MR signal sample corresponding to each original MR signal sample in the group. At the same time, the training frame carries the FEC coding information used in the group. The FEC coding information is used to indicate the effective bit width and punching method of the group. The training frame is then transmitted through the transmit antenna.

[0043] A magnetic resonance imaging system, the system comprising:

[0044] At the wireless coil end: the acquired analog magnetic resonance (MR) signal is converted from analog to digital to obtain a digital MR signal; based on the number of coded MR signal samples in a training frame and the length of an original MR signal sample, each original MR signal sample group is sequentially read from the digital MR signal, wherein the number of original MR signal samples in each original MR signal sample group is equal to the number of coded MR signal samples in a training frame; for each read original MR signal sample group, the maximum effective bit width is found among the effective bit widths of the original MR signal samples in that group, and this maximum effective bit width is taken as the effective bit width of that group; Based on the effective bit width of the group, the puncturing method used for forward error correction (FEC) encoding of each original MR signal sample within the group is determined. The smaller the effective bit width of the group, the fewer punctures are used in the corresponding puncturing method. For any original MR signal sample within any original MR signal sample group, the effective MR signal sample corresponding to that original MR signal sample is determined based on the effective bit width of the group. The bit width of the effective MR signal sample is equal to the effective bit width of the group. FEC encoding is then performed on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0045] On the system side: The system receives a training frame from the receiving antenna, parses each coded MR signal sample from the training frame, determines the puncturing method indicated by the FEC coding information carried in the training frame, performs FEC decoding on each coded MR signal sample according to the puncturing method, obtains each decoded MR signal sample, and fills the missing high bits of each decoded MR signal sample with zeros according to the effective bit width indicated by the FEC coding information carried in the training frame, to obtain the recovered original MR signal sample.

[0046] The wireless coil terminal searches for the maximum effective bit width among the effective bit widths of each original MR signal sample point in the group, including:

[0047] Among all the original MR signal samples in the group, find the largest sample and use the effective bit width of the largest sample as the maximum effective bit width.

[0048] The wireless coil terminal determines the punching method used for FEC encoding of each original MR signal sample point within the group based on the effective bit width of the group, including:

[0049] In the pre-defined correspondence between each effective bit width and the punching method, find the punching method corresponding to the effective bit width of this group;

[0050] The correspondence between each effective bit width and the punching method is determined as follows:

[0051] For any valid bit width, the total input bit width of the puncturing process corresponding to the valid bit width is obtained based on the valid bit width and the number of convolutional codes used in the pre-set FEC encoding; the number of punctures corresponding to the valid bit width is obtained based on the total input bit width of the puncturing process and the output bit width of the FEC encoding; and each puncture code corresponding to the valid bit width is determined based on the number of punctures and the number of puncture codes used in the pre-set FEC encoding, wherein the total number of 0 bits contained in all puncture codes is equal to the number of punctures corresponding to the valid bit width.

[0052] The wireless coil terminal performs FEC encoding on the effective MR signal sample points according to the corresponding punching method, including:

[0053] The valid MR signal sample is encoded using each of the pre-defined convolutional codes used in FEC encoding.

[0054] For each encoding result, select one punch code sequentially from the punch codes corresponding to the effective bit width of the group, and punch the encoding result to obtain a punch result.

[0055] The punching results are combined according to a predefined punching result combination method to obtain the coded MR signal sample corresponding to the original MR signal sample.

[0056] After the wireless coil terminal obtains the coded MR signal sample corresponding to the original MR signal sample, it further includes:

[0057] For each original MR signal sample group, a training frame is constructed based on the coded MR signal sample corresponding to each original MR signal sample in the group. At the same time, the FEC coding information used in the group is carried in the training frame. The FEC coding information is used to indicate the effective bit width and punching method of the group. The training frame is then transmitted through the transmit antenna.

[0058] The system performs FEC decoding on each coded MR signal sample according to the puncturing method to obtain each decoded MR signal sample, including:

[0059] Based on the puncturing method, determine the corresponding puncturing codes; for each coded MR signal sample in the training frame, recover multiple puncturing results from the coded MR signal sample according to the determined puncturing codes and the predefined puncturing result combination method; decode each puncturing result using each convolutional code used in the pre-set FEC encoding to obtain the decoded MR signal sample of the coded MR signal sample, the bit width of the decoded MR signal sample is equal to the effective bit width indicated by the FEC encoding information carried in the training frame.

[0060] In this embodiment of the invention, when the MR wireless coil acquires raw MR signal samples, the effective bit width of the raw MR signal sample group corresponding to each training frame is detected based on the number of encoded MR signal samples contained in each training frame and the length of a raw MR signal sample. The puncturing method of the FEC encoding for each group is determined based on the effective bit width of each group. The corresponding puncturing method is used to puncture each raw MR signal sample group. The smaller the effective bit width of a raw MR signal sample group, the fewer punctures are required in the corresponding puncturing method. Therefore, the puncturing method of the FEC encoding is dynamically adjusted according to the different effective bit widths of the raw MR signal sample groups. This ensures that the smaller the effective bit width of the raw MR signal sample group, the more effective information is contained in the encoded MR signal sample, resulting in a higher decoding accuracy at the MR system end. This improves the utilization rate of channel bandwidth resources during wireless transmission of MR data and enhances the reliability of MR data transmission. Attached Figure Description

[0061] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which will make the above and other features and advantages of the present invention more apparent to those skilled in the art. In the drawings:

[0062] Figure 1 A flowchart of an FEC encoding method for MR data provided in an embodiment of the present invention;

[0063] Figure 2 This is a schematic diagram of FEC encoding of a valid MR signal sample in an application example of the present invention;

[0064] Figure 3 This is a schematic diagram illustrating an example of wireless transmission of MR signals using the present invention;

[0065] Figure 4 This is a schematic diagram of the structure of an FEC encoding device for MR data provided in an embodiment of the present invention;

[0066] Figure 5 This is a schematic diagram of the structure of an MRI system provided in an embodiment of the present invention.

[0067] The reference numerals and their meanings in the attached figures are as follows:

[0068] 101~104: Steps;

[0069] X: Effective MR signal sample point;

[0070] X1, X2: Encoding results;

[0071] 21: Encoding register;

[0072] D0~D6: 0th~6th bit;

[0073] 22: Drilling results for X1;

[0074] 23: Drilling results for X2;

[0075] R0~R13: bit 0~13;

[0076] 24: Output register;

[0077] D0~D23: Bits 0 to 23;

[0078] 31: MR wireless coil end;

[0079] 32: MR system end;

[0080] 311: ADC interface;

[0081] 312: Original MR signal buffer module;

[0082] 313: Maximum effective bit width detection module;

[0083] 314: Dynamic FEC encoding module;

[0084] 315: Transmission mapping module;

[0085] 321: Receive demapping module;

[0086] 322: Training frame receiving buffer module;

[0087] 323: FEC decoding module;

[0088] 324: ADC sampling recovery module;

[0089] 40: FEC encoding device for MR data;

[0090] 41: Original MR signal buffer module;

[0091] 42: Maximum effective bit width detection module;

[0092] 43: Dynamic FEC encoding module;

[0093] 50: MRI system;

[0094] 51: Wireless coil end;

[0095] 52: System side. Detailed Implementation

[0096] To make the objectives, technical solutions, and advantages of the present invention clearer, the following embodiments are provided to further illustrate the present invention in detail.

[0097] The inventors observed and analyzed the MR signals acquired by the MR wireless coil and discovered that the ADC (Analog-to-Digital Converter) at the MR wireless coil sampled the analog MR signals acquired by the wireless coil, with a sampling bit width typically of 18 bits. Specifically, when the MR system is not scanning, the signal sampled by the ADC is usually low noise, and due to the shielding effect between MR scans, the value of the digital signal sampled by the ADC is usually small, no more than 3 bits. When the MR system is scanning, the effective bit width of the digital MR signal sampled by the ADC ranges from 11 to 18 bits depending on the selected scan sequence, but cases with an effective bit width greater than 15 bits are rare. Existing FEC encoding schemes do not consider the effective bit width of the digital MR signal output by the ADC, applying the same FEC encoding method to all digital MR signals based on the principle that the bit width of all digital MR signals is the effective bit width. This wastes channel bandwidth resources and fails to improve data reliability. Based on this observation and analysis, the inventors proposed the following solution of this invention to fully utilize channel bandwidth resources and improve data reliability.

[0098] Figure 1 A flowchart illustrating an embodiment of the FEC encoding method for MR data provided by the present invention. Figure 1 As shown, the specific steps are as follows:

[0099] Step 101: The MR wireless coil terminal performs analog-to-digital conversion on the acquired analog MR signal to obtain a digital MR signal.

[0100] In practical applications, the ADC at the MR wireless coil end samples the analog MR signal collected by the wireless coil and outputs a digital MR signal. The sampling bit width of the ADC is usually 18 bits, that is, the bit width of a sample point of the original MR signal is usually 18 bits.

[0101] Step 102: Based on the number of coded MR signal samples in a training frame and the length of an original MR signal sample, sequentially read each original MR signal sample group from the digital MR signal in Step 101. The number of original MR signal samples in each original MR signal sample group is equal to the number of coded MR signal samples in a training frame.

[0102] In practical applications, MR signals are sent to the MR system via training frames. The number of coded MR signal samples carried in a training frame is predetermined based on factors such as channel bandwidth. The coded MR signal samples are those obtained by FEC encoding the original MR signal samples. For example, in one application scenario, the bit width (i.e., length) of an original MR signal sample is 18 bits, and the bit width (i.e., length) of the coded MR signal sample obtained after FEC encoding of the original MR signal sample is 24 bits. Assuming that in this scenario, a training frame can carry 128 coded MR signal samples, then in step 102, each original MR signal sample (with a bit width of 18 bits) is sequentially read from the digital MR signal in step 101, and every 128 original MR signal samples read constitute a group of original MR signal samples.

[0103] Step 103: For each group of original MR signal samples read, find the maximum effective bit width among the effective bit widths of all original MR signal samples in the group, and use this maximum effective bit width as the effective bit width of the group; based on the effective bit width of the group, determine the puncturing method used for FEC encoding of each original MR signal sample in the group. The smaller the effective bit width of the group, the fewer punctures are required in the corresponding puncturing method.

[0104] In one optional embodiment, step 103, finding the maximum effective bit width among the effective bit widths of each original MR signal sample point in the group, includes: finding the maximum sample point among all original MR signal samples in the group, and using the effective bit width of the maximum sample point as the maximum effective bit width.

[0105] When determining the effective bit width of the maximum sample point, the 0 bit preceding the first 1 bit starting from the highest bit needs to be taken into account, as shown in the following example:

[0106] Suppose the bit width of the original MR signal sample is 18 bits. The largest sample in a group of original MR signal samples is: 000011101000110101. The first four bits of the high-order bits of this sample are all 0, so the effective bit width of this sample is 15 bits. The reason why the effective bit width of this sample is 18-4+1=15 bits instead of 18-4=14 bits is as follows: If the effective bit width were 14 bits, the sample decoded during FEC decoding at the MR system would be: 11101000110101. However, for the MR system, since the highest bit of 11101000110101 is "1", when supplementing the four missing high-order bits, it defaults to using the highest bit of 11101000110101 as "1", meaning the final recovered original MR signal sample is 1. The result 11111101000110101 is clearly incorrect. However, if the effective bit width is set to 15 bits, the sample point decoded during FEC decoding at the MR system end will be 011101000110101. Since the highest bit of 011101000110101 is "0", when supplementing the missing 3 high bits, it will default to using the highest bit "0" of 011101000110101 to fill in the missing bits. Therefore, the final recovered original MR signal sample point is 000011101000110101, which is correct.

[0107] In an optional embodiment, step 103, determining the puncturing method used for FEC encoding of each original MR signal sample in the group based on the effective bit width of the group, includes: searching for the puncturing method corresponding to the effective bit width of the group in a pre-defined correspondence between effective bit widths and puncturing methods. The correspondence between effective bit widths and puncturing methods is determined as follows: for any effective bit width, the total input bit width for puncturing processing corresponding to that effective bit width is obtained based on the effective bit width and the pre-defined number of convolutional codes used for FEC encoding; the number of punctures corresponding to that effective bit width is obtained based on the total input bit width for puncturing processing and the output bit width of FEC encoding; and each puncturing code corresponding to that effective bit width is determined based on the number of punctures and the pre-defined number of puncturing codes used for FEC encoding, wherein the total number of 0 bits contained in all puncturing codes is equal to the number of punctures corresponding to that effective bit width.

[0108] Step 104: For any original MR signal sample in any original MR signal sample group, determine the effective MR signal sample corresponding to the original MR signal sample according to the effective bit width of the group, wherein the bit width of the effective MR signal sample is equal to the effective bit width of the group; perform FEC encoding on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0109] In one optional embodiment, step 104 involves FEC encoding of the effective MR signal sample according to the corresponding puncturing method of the group, including: encoding the effective MR signal sample using each of the pre-defined convolutional codes used for FEC encoding; for each encoding result, sequentially selecting a puncturing code from the puncturing codes corresponding to the effective bit width of the group, and performing puncturing processing on the encoding result to obtain a puncturing result; and combining each puncturing result according to a predefined puncturing result combination method to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0110] Figure 2 This is a schematic diagram illustrating FEC encoding of a valid MR signal sample in an application example of the present invention. In this example, the bit width of the original MR signal sample is 18 bits, and the effective bit width of the current original MR signal sample group is 14 bits. The current original MR signal sample in this group is: 000000101011011111. Since the effective bit width of this group is 14 bits, the valid MR signal sample X corresponding to the current original MR signal sample is: 00101011011111. Assuming the output bit width of FEC encoding is 24 bits, the bit width of the encoding register is 7 bits, and FEC encoding uses two sets of convolutional codes and two sets of punctured codes, the process of FEC encoding the valid MR signal sample: 00101011011111 is as follows:

[0111] Step 01: First, based on the effective bit width of the current original MR signal sample group of 14 bits, find the corresponding two punch codes, set as: Punch code 0: 11111101111011, Punch code 1: 11011111111110.

[0112] The design principle of the two punch codes corresponding to the current original MR signal sample group is as follows:

[0113] The effective bit width of the original MR signal sample group is 14 bits. FEC encoding uses two puncture codes, so the bit width of the input data for puncturing processing is 14 × 2 = 28 bits, while the bit width of the output data for FEC encoding is 24 bits. Therefore, the total number of punctures can be calculated as: 28 - 24 = 4 bits. The two puncture codes designed satisfy the following conditions: the length of each puncture code is 14 bits, and the number of puncture bits is 2 bits. In this example, the two puncture codes designed to meet the above conditions are: Punch code 0: 11111101111011, Punch code 1: 110111111111110.

[0114] Step 02: Encode the effective MR signal sample X: 0010101101111 using the two convolutional codes used in FEC encoding: convolutional code 0: 1101101 and convolutional code 1: 1001111, to obtain two encoded results: X1 and X2. The bit width of both X1 and X2 is the same as the bit width of the effective MR signal sample, both being 14 bits. Here, 21 is the encoding register; in this example, the encoding register has a bit width of 7 bits, and D0~D6 represent bits 0~6.

[0115] It should be noted that in this invention, the two convolutional codes used in FEC encoding remain unchanged and are applicable to all valid MR signal samples corresponding to all original MR signal sample groups. The only difference in the FEC encoding method in this invention is that the puncturing method (i.e., puncturing code) in the corresponding FEC encoding differs for original MR signal sample groups with different effective bit widths.

[0116] Step 03: Punch the encoded result X1 using punch code 0. That is, if the value of a certain bit in punch code 0 is 0, then add a punch mark to the bit at the same position in X1; punch the encoded result X2 using punch code 1. Wherein, 22 is the punch result of X1, 23 is the punch result of X2, and R0~R13 represent bits 0~13.

[0117] For example, the punch code 0 is 11111101111011, which has two 0 bits, namely the 2nd bit and the 7th bit. Then, punch marks are made on the 2nd bit and the 7th bit of X1.

[0118] Step 04: Combine and sort the punching results of X1 and X2 one by one in order from the lowest bit to the highest bit, and then put them into the output register. Here, 24 is the output register, and D0~D23 represent bits 0~23.

[0119] like Figure 2As shown, specifically, first check the lowest bit of X1, i.e., bit 0. This bit is not marked with a punch mark, so it is placed into bit 0 of the output register. Then check the lowest bit of X2, i.e., bit 0. This bit is marked with a punch mark, so it is not placed into the output register. Then check the second lowest bit of X1, i.e., bit 1. This bit is not marked with a punch mark, so it is placed into bit 1 of the output register. Then check the second lowest bit of X2, i.e., bit 1. This bit is not marked with a punch mark, so it is placed into bit 2 of the output register. Then check the second bit of X1. This bit is marked with a punch mark, so it is not placed into the output register. Then check the second bit of X2. This bit is not marked with a punch mark, so it is placed into bit 3 of the output register. Then check the third bit of X1, ...; and so on, until finally all 24 bits of X1 and X2 that are not marked with punch marks are placed into the 24-bit output register.

[0120] In the above embodiments, when the MR wireless coil acquires raw MR signal samples, the effective bit width of the raw MR signal sample group corresponding to each training frame is detected based on the number of coded MR signal samples contained in each training frame and the length of a raw MR signal sample. The puncturing method of FEC encoding for each group is determined based on the effective bit width of each group, and the corresponding puncturing method is used to puncture each raw MR signal sample group. The smaller the effective bit width of a raw MR signal sample group, the fewer punctures are in the corresponding puncturing method. Thus, the puncturing method of FEC encoding is dynamically adjusted according to the different effective bit widths of the raw MR signal sample groups. This ensures that the smaller the effective bit width of the raw MR signal sample group, the more effective information is contained in the coded MR signal sample, thereby increasing the decoding accuracy of the MR system, improving the utilization rate of channel bandwidth resources when wirelessly transmitting MR data, and improving the reliability of MR data transmission.

[0121] Once the MR wireless coil obtains the coded MR signal samples corresponding to all the original MR signal samples in a set of original MR signal samples, a training frame can be constructed.

[0122] In an optional embodiment, after step 104, the method further includes: for each original MR signal sample group, constructing a training frame based on the coded MR signal samples corresponding to each original MR signal sample in the group, and carrying the FEC encoding information used by the group in the training frame. The FEC encoding information is used to indicate the effective bit width and puncturing method of the group. The training frame is then transmitted through the transmit antenna. The MR system receives the training frame from the receive antenna, parses each coded MR signal sample from the training frame, determines the puncturing method indicated by the FEC encoding information based on the FEC encoding information carried in the training frame, performs FEC decoding on each coded MR signal sample according to the puncturing method, obtains each decoded MR signal sample, and pads the missing high bits of each decoded MR signal sample with zeros according to the effective bit width indicated by the FEC encoding information carried in the training frame, thereby obtaining the recovered original MR signal sample.

[0123] In practical applications, when constructing training frames based on the coded MR signal samples corresponding to each original MR signal sample in the group, a training sequence containing multiple training samples is inserted into each coded MR signal sample. The training sequence is used for frame synchronization and / or clock synchronization between the MR wireless coil and the MR system. For example, if a training frame contains 128 coded MR signal samples, and each coded MR signal sample is 24 bits; and a training sequence contains 128 training samples, each training sample being 1 bit, then a training sample can be inserted after each coded MR signal sample, meaning the payload length of a training frame is 128 × (24 + 1) = 3200 bits.

[0124] Since the FEC encoding method (i.e., puncturing method) corresponding to the coded MR signal samples in different training frames is dynamically changing, the MR wireless coil needs to inform the MR system of the FEC encoding method (mainly including: the effective bit width of the original MR signal sample group and the puncturing method used for FEC encoding) corresponding to the coded MR signal samples carried in each training frame so that the MR system can correctly perform FEC decoding. Therefore, the FEC encoding information (mainly including: the effective bit width of the original MR signal sample group and the puncturing method used for FEC encoding) can be carried in the reporting information bits of the training frame. Specifically, since the effective bit width and puncturing method are in one-to-one correspondence, and since the bit width of the original MR signal sample is fixed, the number of effective bit width types is also fixed. Therefore, for simplicity, different effective bit widths can be represented by different serial numbers. At the same time, the correspondence between serial numbers and effective bit widths, and the correspondence between effective bit widths and puncturing methods are stored at both the MR wireless coil and the MR system. The puncturing method can be specifically represented by the puncturing codes used for that puncturing method.

[0125] For example, when the bit width of the original MR signal sample is 18 bits, it is known from experience that the effective bit width of the original MR signal sample is mostly between 12 and 18 bits. Therefore, we can use numbers 1-7 to represent the effective bit widths of 12-18 bits respectively, and pre-set the corresponding puncturing method (i.e., puncturing code) for the effective bit widths of 12-18 bits. That is, the correspondence between the number, effective bit width, and puncturing method (i.e., puncturing code) is as follows:

[0126] Serial number 1, effective bit width 12 bits, punching method 1;

[0127] Serial number 2, effective bit width 13 bits, punching method 2;

[0128] Serial number 3, effective bit width 14 bits, punching method 3;

[0129] Serial number 4, effective bit width 15 bits, punching method 4;

[0130] Serial number 5, effective bit width 16 bits, punching method 5;

[0131] Serial number 6, effective bit width 17 bits, punching method 6;

[0132] Number 7, effective bit width 18 bits, punching method 7.

[0133] In this way, the MR wireless coil end can represent the FEC encoded information carried in the training frame using numbers 1 to 7. After receiving the training frame, the MR system end can determine the effective bit width and punching method corresponding to the number in the training frame based on the above correspondence between the numbers, effective bit width, and punching method.

[0134] It should be noted that since the convolutional code is the same for different effective bit widths, meaning the convolutional code remains unchanged during FEC encoding, the MR system will pre-store the convolutional code for use during FEC decoding.

[0135] In one optional embodiment, FEC decoding is performed on each coded MR signal sample according to the puncturing method to obtain each decoded MR signal sample, including: determining each puncturing code corresponding to the puncturing method according to the puncturing method; for each coded MR signal sample in the training frame, recovering multiple puncturing results from the coded MR signal sample according to the determined puncturing codes and a predefined combination method of puncturing results; and decoding each puncturing result using each convolutional code used in the pre-set FEC encoding to obtain the decoded MR signal sample of the coded MR signal sample, wherein the bit width of the decoded MR signal sample is equal to the effective bit width indicated by the FEC encoding information carried in the training frame.

[0136] by Figure 2Taking the FEC encoding process shown as an example, the decoding process is as follows:

[0137] First, for a coded MR signal sample parsed from the training frame, the puncture results corresponding to the original two coded results X1 and X2 are recovered according to the two puncture codes and the pre-set combination method of puncture results. Then, two convolutional codes, convolutional code 0 and convolutional code 1, are used to decode puncture result 1 and puncture result 2 respectively to obtain the estimated effective MR signal sample. After that, according to the effective bit width, the missing high bits of the effective MR signal sample are padded with zeros to obtain the recovered original MR signal sample.

[0138] Figure 3 This is a schematic diagram illustrating an example of wireless transmission of MR signals using the present invention. Wherein:

[0139] 31 is the MR wireless coil end, specifically, 31 is the control module of the MR wireless coil end; 32 is the MR system end, specifically, 32 is the control module of the MR system end.

[0140] 311 is the ADC interface, 312 is the raw MR signal buffer module, 313 is the maximum effective bit width detection module, 314 is the dynamic FEC encoding module, and 315 is the transmit mapping module;

[0141] 321 is the receiver demapping module, 322 is the training frame receiver buffer module, 323 is the FEC decoding module, and 324 is the ADC sampling recovery module.

[0142] The wireless transmission process of MR signals is as follows:

[0143] Step 01: ADC interface 311 receives the digital MR signal sampled by ADC. In this example, the sampling bit width of ADC is 18 bits, that is, the bit width of the original MR signal sample point is 18 bits.

[0144] Step 02: The digital MR signal received by the ADC interface 311 is sent to the original MR signal buffer module 312 and the maximum effective bit width detection module 313.

[0145] Assuming that one training frame can transmit 128 coded MR signal samples, the length of the training frame receiving buffer module 312 is 128 × 18 = 2304 bits.

[0146] Step 03: Whenever the maximum effective bit width detection module 313 finishes receiving 128 original MR signal samples corresponding to a training frame, the 128 original MR signal samples form an original MR signal sample group. The maximum effective bit width detection module 313 searches for the largest sample in the 128 original MR signal samples in the group, takes the effective bit width of the largest sample as the effective bit width of the group, determines the corresponding puncturing method according to the effective bit width of the group, and sends the effective bit width and puncturing method of the group to the dynamic FEC encoding module 314 and the transmission mapping module 315.

[0147] Step 04: The dynamic FEC encoding module 314 automatically reads data from the original MR signal buffer module 312, reading 128 original MR signal samples each time, i.e., reading a group of original MR signal samples. Then, for each original MR signal sample in the group, the dynamic FEC encoding module 314 determines the effective MR signal sample corresponding to the original MR signal sample according to the effective bit width of the group, and encodes the effective MR signal sample using pre-set convolution codes (two convolution codes are set in this example) to obtain two convolution results X1 and X2. Then, it uses puncturing codes 0 and 1 in the corresponding puncturing method of the group to puncture X1 and X2 respectively to obtain puncturing result 1 and puncturing result 2. Then, puncturing result 1 and puncturing result 2 are merged according to the pre-set puncturing result combination method to obtain the encoded MR signal sample, and the encoded MR signal sample is placed in the output register.

[0148] Step 05: The transmission mapping module 315 reads each coded MR signal sample from the output register of the dynamic FEC coding module 314, and evenly inserts each training sample in the predefined training sequence into the coded MR signal sample. The 128 coded MR signal samples with the inserted training sequence are mapped into the payload of the training frame. At the same time, the effective bit width and FEC coding information corresponding to the original MR signal sample group corresponding to the 128 coded MR signal samples are put into the reporting information bit of the training frame, and the training frame is transmitted through the transmission antenna.

[0149] As mentioned earlier, since the effective bit width and the puncturing method are in one-to-one correspondence, a sequence number can be assigned to different effective bit widths and puncturing methods, and this sequence number can be used as FEC encoded information and placed in the reporting information bit of the training frame.

[0150] Step 06: The receiving and demapping module 321 receives the training frame from the receiving antenna and performs correlation processing with the locally stored training sequence and the training frame to determine the position of each training sample in the training frame. This allows the training sequence, each coded MR signal sample, and FEC coding information to be parsed from the training frame. The correlation processing results can be used for clock synchronization between the MR wireless coil and the MR system. Each coded MR signal sample is placed in the training frame receiving buffer module 322. The puncturing method indicated by the FEC coding information is sent to the FEC decoding module 323, and the effective bit width indicated by the FEC coding information is sent to the ADC sampling recovery module 324.

[0151] Step 07: The FEC decoding module 323 reads each encoded MR signal sample from the training frame receiving buffer module 322. For any encoded MR signal sample, it recovers the original two encoded results X1 and X2 corresponding to the puncturing result 1 and puncturing result 2 according to the puncturing method (i.e., puncturing code 0 and puncturing code 1) and the pre-set combination of puncturing results. Then, it uses two convolutional codes: convolutional code 0 and convolutional code 1 to decode puncturing result 1 and puncturing result 2 respectively to obtain the estimated effective MR signal sample. The effective MR signal sample is then sent to the ADC sampling recovery module 324.

[0152] Step 08: The ADC sampling recovery module 324 receives the valid MR signal sample and fills the missing high bits of the valid MR signal sample with zeros according to the valid bit width to obtain the recovered original MR signal sample.

[0153] As can be seen from the above process, for the MR wireless coil end, the dynamic FEC encoding process adopted in this invention, compared with the existing fixed FEC encoding process, only needs to save a few more sets of correspondences between effective bit width and punching method, and add an operation to query the punching method based on the effective bit width. It can be seen that the power consumption added to the MR wireless coil end is extremely small and can be ignored.

[0154] Figure 4 This is a schematic diagram of the structure of an FEC encoding device 40 for MR data provided in an embodiment of the present invention. The device 40 is located at the MR wireless coil end. Figure 4 As shown, the device 40 mainly includes: a raw MR signal buffer module 41, a maximum effective bit width detection module 42, and a dynamic FEC encoding module 43, wherein:

[0155] Original MR signal buffer module 41: buffers the digital MR signal obtained by analog-to-digital conversion of the collected analog MR signal at the MR wireless coil end.

[0156] Maximum effective bit width detection module 42: sequentially reads each original MR signal sample contained in an original MR signal sample group, wherein the number of original MR signal samples contained in an original MR signal sample group is equal to the number of encoded MR signal samples contained in a training frame; for a read original MR signal sample group, finds the maximum effective bit width among the effective bit widths of each original MR signal sample in the group, and uses the maximum effective bit width as the effective bit width of the group; based on the effective bit width of the group, determines the puncturing method used for FEC encoding of each original MR signal sample in the group, wherein the smaller the effective bit width of the group, the fewer punctures are in the corresponding puncturing method of the group.

[0157] Dynamic FEC encoding module 43: For any original MR signal sample in a group of original MR signal samples, determine the effective MR signal sample corresponding to the original MR signal sample according to the effective bit width of the group, wherein the bit width of the effective MR signal sample is equal to the effective bit width of the group; perform FEC encoding on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0158] In one optional embodiment, the maximum effective bit width detection module 42 searches for the maximum effective bit width among the effective bit widths of each original MR signal sample point in the group, including: searching for the largest sample point among all original MR signal samples in the group, and taking the effective bit width of the largest sample point as the maximum effective bit width.

[0159] In one optional embodiment, the maximum effective bit width detection module 42 determines the puncturing method used for FEC encoding of each original MR signal sample in the group based on the effective bit width of the group. This includes: searching for the puncturing method corresponding to the effective bit width of the group in a pre-defined correspondence between effective bit widths and puncturing methods. The correspondence between effective bit widths and puncturing methods is determined as follows: for any effective bit width, the total input bit width of the puncturing process corresponding to that effective bit width is obtained based on the effective bit width and the pre-defined number of convolutional codes used for FEC encoding; the number of punctures corresponding to that effective bit width is obtained based on the total input bit width of the puncturing process and the output bit width of FEC encoding; and each puncturing code corresponding to that effective bit width is determined based on the number of punctures and the pre-defined number of puncturing codes used for FEC encoding, wherein the total number of 0 bits contained in all puncturing codes is equal to the number of punctures corresponding to that effective bit width.

[0160] In one optional embodiment, the dynamic FEC encoding module 43 performs FEC encoding on the effective MR signal sample according to the corresponding puncturing method of the group, including: encoding the effective MR signal sample using each of the pre-set convolutional codes used for FEC encoding; for each encoding result, sequentially selecting a puncturing code from the puncturing codes corresponding to the effective bit width of the group, performing puncturing processing on the encoding result to obtain a puncturing result; and combining each puncturing result according to a predefined puncturing result combination method to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0161] In an optional embodiment, the above-mentioned device 40 further includes a transmission mapping module 44, which is used to: construct a training frame for each original MR signal sample group according to the coded MR signal sample corresponding to each original MR signal sample in the group, and carry the FEC encoding information used by the group in the training frame, the FEC encoding information being used to indicate the effective bit width and punching method of the group, and transmit the training frame through the transmission antenna.

[0162] Figure 5 This is a schematic diagram of the structure of an MRI system 50 provided in an embodiment of the present invention. Figure 5 As shown, the system mainly includes: a wireless coil terminal 51 and a system terminal 52, wherein:

[0163] Wireless coil terminal 51: Performs analog-to-digital conversion on the acquired analog MR signal to obtain a digital MR signal; according to the number of coded MR signal samples contained in a training frame and the length of an original MR signal sample, sequentially reads each original MR signal sample group from the digital MR signal, wherein the number of original MR signal samples contained in each original MR signal sample group is equal to the number of coded MR signal samples contained in a training frame; for each read original MR signal sample group, finds the maximum effective bit width among the effective bit widths of each original MR signal sample in the group, and uses the maximum effective bit width as the effective bit width of the group. Based on the effective bit width of the group, the puncturing method used for FEC encoding of each original MR signal sample in the group is determined. The smaller the effective bit width of the group, the fewer punctures are used in the corresponding puncturing method. For any original MR signal sample in any original MR signal sample group, the effective MR signal sample corresponding to the original MR signal sample is determined based on the effective bit width of the group. The bit width of the effective MR signal sample is equal to the effective bit width of the group. FEC encoding is performed on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0164] System terminal 52: Receives a training frame from the receiving antenna, parses each coded MR signal sample from the training frame, determines the puncturing method indicated by the FEC coding information carried in the training frame, performs FEC decoding on each coded MR signal sample according to the puncturing method, obtains each decoded MR signal sample, and fills the missing high bits of each decoded MR signal sample with zeros according to the effective bit width indicated by the FEC coding information carried in the training frame, to obtain the recovered original MR signal sample.

[0165] In one optional embodiment, the wireless coil terminal 51 searches for the maximum effective bit width among the effective bit widths of each original MR signal sample in the group, including: searching for the maximum sample among all original MR signal samples in the group, and using the effective bit width of the maximum sample as the maximum effective bit width.

[0166] In one optional embodiment, the wireless coil terminal 51 determines the puncturing method used for FEC encoding of each original MR signal sample in the group based on the effective bit width of the group. This includes: searching for the puncturing method corresponding to the effective bit width of the group in a pre-defined correspondence between effective bit widths and puncturing methods. The correspondence between effective bit widths and puncturing methods is determined as follows: for any effective bit width, the total input bit width for puncturing processing corresponding to that effective bit width is obtained based on the effective bit width and the pre-defined number of convolutional codes used for FEC encoding; the number of punctures corresponding to that effective bit width is obtained based on the total input bit width for puncturing processing and the output bit width of FEC encoding; and each puncturing code corresponding to that effective bit width is determined based on the number of punctures and the pre-defined number of puncturing codes used for FEC encoding, wherein the total number of 0 bits contained in all puncturing codes is equal to the number of punctures corresponding to that effective bit width.

[0167] In one optional embodiment, the wireless coil terminal 51 performs FEC encoding on the effective MR signal sample according to the corresponding punching method of the group, including: encoding the effective MR signal sample using each convolutional code used for FEC encoding in a pre-defined manner; for each encoding result, sequentially selecting a punching code from the punching codes corresponding to the effective bit width of the group, punching the encoding result to obtain a punching result; and combining the punching results according to a predefined punching result combination method to obtain the encoded MR signal sample corresponding to the original MR signal sample.

[0168] In one optional embodiment, after the wireless coil terminal 51 obtains the coded MR signal sample corresponding to the original MR signal sample, it further includes: for each original MR signal sample group, constructing a training frame according to the coded MR signal sample corresponding to each original MR signal sample in the group, and carrying the FEC encoding information used by the group in the training frame. The FEC encoding information is used to indicate the effective bit width and punching method of the group, and transmitting the training frame through the transmitting antenna.

[0169] In one optional embodiment, the system terminal 52 performs FEC decoding on each coded MR signal sample according to the puncturing method to obtain each decoded MR signal sample, including: determining each puncturing code corresponding to the puncturing method according to the puncturing method; for each coded MR signal sample in the training frame, recovering multiple puncturing results from the coded MR signal sample according to the determined puncturing codes and a predefined combination method of puncturing results; and decoding each puncturing result using each convolutional code used in the pre-set FEC encoding to obtain the decoded MR signal sample of the coded MR signal sample, wherein the bit width of the decoded MR signal sample is equal to the effective bit width indicated by the FEC encoding information carried in the training frame.

[0170] Those skilled in the art will understand that the features described in the various embodiments and / or claims disclosed in this application can be combined and / or combined in various ways, even if such combinations or combinations are not explicitly described in this application. In particular, without departing from the spirit and teachings of this application, the features described in the various embodiments and / or claims of this application can be combined and / or combined in various ways, and all such combinations and / or combinations fall within the scope of this application.

[0171] This document uses specific embodiments to illustrate the principles and implementation methods of this application. The descriptions of the embodiments above are only for the purpose of helping to understand the methods and core ideas of this application, and are not intended to limit this application. For those skilled in the art, changes can be made to the specific implementation methods and application scope based on the ideas, spirit and principles of this application. Any modifications, equivalent substitutions, improvements, etc., made should be included within the scope of protection of this application.

Claims

1. A forward error correction (FEC) coding method for magnetic resonance (MR) data, characterized in that, The method includes: The MR wireless coil terminal performs analog-to-digital conversion on the acquired analog MR signal to obtain a digital MR signal; Based on the number of coded MR signal samples contained in a training frame and the length of an original MR signal sample, each original MR signal sample group is sequentially read from the digital MR signal, wherein the number of original MR signal samples contained in each original MR signal sample group is equal to the number of coded MR signal samples contained in a training frame. For each group of original MR signal samples read, the maximum effective bit width is found among the effective bit widths of the original MR signal samples in the group, and the maximum effective bit width is taken as the effective bit width of the group. Based on the effective bit width of the group, the puncturing method used for FEC encoding of each original MR signal sample in the group is determined. The smaller the effective bit width of the group, the fewer the number of punctures in the corresponding puncturing method of the group. For any original MR signal sample within any original MR signal sample group, the effective MR signal sample corresponding to the original MR signal sample is determined according to the effective bit width of the group, wherein the bit width of the effective MR signal sample is equal to the effective bit width of the group; the effective MR signal sample is then FEC encoded according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

2. The method according to claim 1, characterized in that, Finding the maximum effective bit width among the effective bit widths of each original MR signal sample point within the group includes: Among all the original MR signal samples in the group, find the largest sample and use the effective bit width of the largest sample as the maximum effective bit width.

3. The method according to claim 1, characterized in that, The step of determining the puncturing method for FEC encoding of each original MR signal sample point within the group based on the effective bit width of the group includes: In the pre-defined correspondence between each effective bit width and the punching method, find the punching method corresponding to the effective bit width of this group; The correspondence between each effective bit width and the punching method is determined as follows: For any valid bit width, the total input bit width of the puncturing process corresponding to the valid bit width is obtained based on the valid bit width and the number of convolutional codes used in the pre-set FEC encoding; the number of punctures corresponding to the valid bit width is obtained based on the total input bit width of the puncturing process and the output bit width of the FEC encoding; and each puncture code corresponding to the valid bit width is determined based on the number of punctures and the number of puncture codes used in the pre-set FEC encoding, wherein the total number of 0 bits contained in all puncture codes is equal to the number of punctures corresponding to the valid bit width.

4. The method according to claim 3, characterized in that, The step of performing FEC encoding on the effective MR signal samples according to the corresponding punching method includes: The valid MR signal sample is encoded using each of the pre-defined convolutional codes used in FEC encoding. For each encoding result, select one punch code sequentially from the punch codes corresponding to the effective bit width of the group, and punch the encoding result to obtain a punch result. The punching results are combined according to a predefined punching result combination method to obtain the coded MR signal sample corresponding to the original MR signal sample.

5. The method according to claim 1, characterized in that, After obtaining the encoded MR signal sample corresponding to the original MR signal sample, the process further includes: For each original MR signal sample group, a training frame is constructed based on the coded MR signal sample corresponding to each original MR signal sample in the group. At the same time, the training frame carries the FEC coding information used by the group. The FEC coding information is used to indicate the effective bit width and punching method of the group. The training frame is then transmitted through the transmit antenna. The MR system receives a training frame from the receiving antenna, parses each coded MR signal sample from the training frame, determines the puncturing method indicated by the FEC encoding information carried in the training frame, performs FEC decoding on each coded MR signal sample according to the puncturing method, obtains each decoded MR signal sample, and fills the missing high bits of each decoded MR signal sample with zeros according to the effective bit width indicated by the FEC encoding information carried in the training frame, to obtain the recovered original MR signal sample.

6. The method according to claim 5, characterized in that, The step of performing FEC decoding on each coded MR signal sample according to the puncturing method to obtain each decoded MR signal sample includes: Based on the punching method, determine the corresponding punch codes; For each coded MR signal sample in the training frame, multiple puncture results are recovered from the coded MR signal sample according to the determined puncture codes and the predefined combination of puncture results; each puncture result is decoded using the pre-set convolutional codes used for FEC encoding to obtain the decoded MR signal sample of the coded MR signal sample. The bit width of the decoded MR signal sample is equal to the effective bit width indicated by the FEC encoding information carried in the training frame.

7. A forward error correction (FEC) encoding device for magnetic resonance (MR) data, characterized in that, The device is located at the MR wireless coil end and includes: Raw MR signal buffer module: buffers the digital MR signal obtained by analog-to-digital conversion of the acquired analog MR signal at the MR wireless coil end; Maximum effective bit width detection module: Sequentially reads each original MR signal sample in an original MR signal sample group, wherein the number of original MR signal samples in an original MR signal sample group is equal to the number of encoded MR signal samples in a training frame; for each original MR signal sample group read, finds the maximum effective bit width among the effective bit widths of each original MR signal sample in the group, and uses the maximum effective bit width as the effective bit width of the group; based on the effective bit width of the group, determines the puncturing method used for FEC encoding of each original MR signal sample in the group, wherein the smaller the effective bit width of the group, the fewer punctures are used in the corresponding puncturing method of the group; Dynamic FEC encoding module: For any original MR signal sample in a group of original MR signal samples, determine the effective MR signal sample corresponding to the original MR signal sample based on the effective bit width of the group, wherein the bit width of the effective MR signal sample is equal to the effective bit width of the group; perform FEC encoding on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample.

8. The apparatus according to claim 7, characterized in that, The maximum effective bit width detection module searches for the maximum effective bit width among the effective bit widths of each original MR signal sample point in the group, including: Among all the original MR signal samples in the group, find the largest sample and use the effective bit width of the largest sample as the maximum effective bit width.

9. The apparatus according to claim 7, characterized in that, The maximum effective bit width detection module determines the punching method used for FEC encoding of each original MR signal sample point in the group based on the effective bit width of the group, including: In the pre-defined correspondence between each effective bit width and the punching method, find the punching method corresponding to the effective bit width of this group; The correspondence between each effective bit width and the punching method is determined as follows: For any valid bit width, the total input bit width of the puncturing process corresponding to the valid bit width is obtained based on the valid bit width and the number of convolutional codes used in the pre-set FEC encoding; the number of punctures corresponding to the valid bit width is obtained based on the total input bit width of the puncturing process and the output bit width of the FEC encoding; and each puncture code corresponding to the valid bit width is determined based on the number of punctures and the number of puncture codes used in the pre-set FEC encoding, wherein the total number of 0 bits contained in all puncture codes is equal to the number of punctures corresponding to the valid bit width.

10. The apparatus according to claim 9, characterized in that, The dynamic FEC encoding module performs FEC encoding on the effective MR signal sample points according to the corresponding punching method, including: The valid MR signal sample is encoded using each of the pre-defined convolutional codes used in FEC encoding. For each encoding result, select one punch code sequentially from the punch codes corresponding to the effective bit width of the group, and punch the encoding result to obtain a punch result. The punching results are combined according to a predefined punching result combination method to obtain the coded MR signal sample corresponding to the original MR signal sample.

11. The apparatus according to claim 7, characterized in that, The device further includes a transmission mapping module, used for: For each original MR signal sample group, a training frame is constructed based on the coded MR signal sample corresponding to each original MR signal sample in the group. At the same time, the training frame carries the FEC coding information used in the group. The FEC coding information is used to indicate the effective bit width and punching method of the group. The training frame is then transmitted through the transmit antenna.

12. A magnetic resonance imaging system, characterized in that, The system includes: At the wireless coil end: the acquired analog magnetic resonance (MR) signal is converted from analog to digital to obtain a digital MR signal; based on the number of coded MR signal samples in a training frame and the length of an original MR signal sample, each original MR signal sample group is sequentially read from the digital MR signal, wherein the number of original MR signal samples in each original MR signal sample group is equal to the number of coded MR signal samples in a training frame; for each read original MR signal sample group, the maximum effective bit width is found among the effective bit widths of the original MR signal samples in that group, and this maximum effective bit width is taken as the effective bit width of that group; Based on the effective bit width of the group, the puncturing method used for forward error correction (FEC) encoding of each original MR signal sample in the group is determined. The smaller the effective bit width of the group, the fewer punctures are used in the corresponding puncturing method. For any original MR signal sample in any original MR signal sample group, the effective MR signal sample corresponding to the original MR signal sample is determined based on the effective bit width of the group. The bit width of the effective MR signal sample is equal to the effective bit width of the group. FEC encoding is performed on the effective MR signal sample according to the puncturing method corresponding to the group to obtain the encoded MR signal sample corresponding to the original MR signal sample. On the system side: The system receives a training frame from the receiving antenna, parses each coded MR signal sample from the training frame, determines the puncturing method indicated by the FEC coding information carried in the training frame, performs FEC decoding on each coded MR signal sample according to the puncturing method, obtains each decoded MR signal sample, and fills the missing high bits of each decoded MR signal sample with zeros according to the effective bit width indicated by the FEC coding information carried in the training frame, to obtain the recovered original MR signal sample.

13. The system according to claim 12, characterized in that, The wireless coil terminal searches for the maximum effective bit width among the effective bit widths of each original MR signal sample point in the group, including: Among all the original MR signal samples in the group, find the largest sample and use the effective bit width of the largest sample as the maximum effective bit width.

14. The system according to claim 12, characterized in that, The wireless coil terminal determines the punching method used for FEC encoding of each original MR signal sample point within the group based on the effective bit width of the group, including: In the pre-defined correspondence between each effective bit width and the punching method, find the punching method corresponding to the effective bit width of this group; The correspondence between each effective bit width and the punching method is determined as follows: For any valid bit width, the total input bit width of the puncturing process corresponding to the valid bit width is obtained based on the valid bit width and the number of convolutional codes used in the pre-set FEC encoding; the number of punctures corresponding to the valid bit width is obtained based on the total input bit width of the puncturing process and the output bit width of the FEC encoding; and each puncture code corresponding to the valid bit width is determined based on the number of punctures and the number of puncture codes used in the pre-set FEC encoding, wherein the total number of 0 bits contained in all puncture codes is equal to the number of punctures corresponding to the valid bit width.

15. The system according to claim 14, characterized in that, The wireless coil terminal performs FEC encoding on the effective MR signal sample points according to the corresponding punching method, including: The valid MR signal sample is encoded using each of the pre-defined convolutional codes used in FEC encoding. For each encoding result, select one punch code sequentially from the punch codes corresponding to the effective bit width of the group, and punch the encoding result to obtain a punch result. The punching results are combined according to a predefined punching result combination method to obtain the coded MR signal sample corresponding to the original MR signal sample.

16. The system according to claim 12, characterized in that, After the wireless coil terminal obtains the coded MR signal sample corresponding to the original MR signal sample, it further includes: For each original MR signal sample group, a training frame is constructed based on the coded MR signal sample corresponding to each original MR signal sample in the group. At the same time, the FEC coding information used in the group is carried in the training frame. The FEC coding information is used to indicate the effective bit width and punching method of the group. The training frame is then transmitted through the transmit antenna.

17. The system according to claim 12, characterized in that, The system performs FEC decoding on each coded MR signal sample according to the puncturing method to obtain each decoded MR signal sample, including: Based on the puncturing method, determine the corresponding puncturing codes; for each coded MR signal sample in the training frame, recover multiple puncturing results from the coded MR signal sample according to the determined puncturing codes and the predefined puncturing result combination method; decode each puncturing result using each convolutional code used in the pre-set FEC encoding to obtain the decoded MR signal sample of the coded MR signal sample, the bit width of the decoded MR signal sample is equal to the effective bit width indicated by the FEC encoding information carried in the training frame.

Images