ENCODING AND DECODING DATA

Abstract

A method for encoding data, a method for decoding data, an encoder, and a decoder are shown. The decoder is configured to determine a convolutionally encoded message consisting of two or more symbol strings, where each of the symbol strings is assigned to different bit sequences according to an assignment rule, and the convolutionally encoded message corresponds to a bit sequence assigned to all symbol strings according to the common assignment rule, and to decode the convolutionally encoded message.

Description

AREA

[0001] The present invention relates to an encoder, a decoder, and methods for encoding and decoding data. In particular, the present invention relates to the encoder-side derivation of symbol sequences (from the data), which are transmitted via different transmission channels or stored in different memory locations, and the decoder-side determination of a convolutionally encoded message from the symbol sequences. BACKGROUND

[0002] Cryptographic encoding and encoding to increase error tolerance (by adding systematic redundancy) are typically performed separately. For example, cryptographic encoding and channel encoding are usually carried out in separate steps on the sender's side. In the first step of the encoding process, the data to be transmitted (the plaintext) is cryptographically encoded using a secret key to increase information security. In the second step, channel encoding adds systematic redundancy, enabling error detection and / or correction on the receiver's side. The symbols thus generated are transmitted via transmission channels and can be converted back into the original data stream in a decoding process by reversing the order.The same applies if the symbols are not transmitted via different transmission channels, but stored in different memories.

[0003] Convolutional coding can be used to increase error tolerance (e.g., for channel coding). This method distributes the input data stream to be encoded smoothly and evenly across multiple positions in the encoded output data stream by introducing redundancy. The redundancy added during encoding increases the number of symbols that can be transmitted / stored. A key feature is that the decoding process can still restore the input data stream even if individual symbols in the output data stream are modified or deleted to a limited extent. This property is used to enable error detection and / or error correction on the receiver / readout side during data transmission / storage. PRESENTATION OF THE INVENTION

[0004] The idea underlying the present invention is based on a combination of (channel) coding using convolutional encoders for receiver-side / reader-side error detection and / or error correction with cryptographic coding to increase information security.

[0005] A method according to the invention for encoding data comprises deriving a first symbol sequence from the data, wherein the first symbol sequence is intended for transmission on a first transmission channel or for storage in a first memory, and deriving a second symbol sequence from the data, wherein the second symbol sequence is intended for transmission on a second transmission channel or for storage in a second memory. The derivation of the first symbol sequence is based on a first masking of the data encoding by a convolutional encoder. The derivation of the second symbol sequence is based on a second masking of the data encoding by the convolutional encoder.

[0006] The first masking assigns a first group of state transitions of the convolutional encoder to each symbol in the first symbol sequence. The second masking assigns a second group of state transitions of the convolutional encoder to each symbol in the second symbol sequence. The first group has at least one first state transition that the second group does not have. The second group has at least one second state transition that the first group does not have. The first and second groups have at least one common third state transition. The first, second, and third state transitions all originate from the same state and lead to different states. One of the state transitions can lead to the state from which the state transitions originate.

[0007] Furthermore, the term "transmission channel," as used in the description and claims, is to be understood in particular as a wired medium. For example, the first transmission channel can be a first wired medium and the second transmission channel a second wired medium. In particular, the first wired medium can be a first electrically conductive medium and the second wired medium a second electrically conductive medium. Furthermore, the first wired medium (on the transmitter and / or receiver side) can be connected to a first interface and the second wired medium (on the transmitter and / or receiver side) to a second interface, wherein the second wired medium cannot be addressed via the first interface and the first wired medium cannot be addressed via the second interface. For example,The wired media can be connected to different ports of a transmitter designed as a microcontroller. Furthermore, the term "memory," as used in the description and claims, refers specifically to (electronic) data storage. For example, the first memory can be a primary data storage device and the second memory a secondary data storage device, whereby access to the first memory does not automatically grant access to the second memory, and vice versa.

[0008] Furthermore, the term "convolutional encoder," as used in the description and claims, refers in particular to an automaton with a finite number of states, wherein a current state of the automaton results from a previous state of the automaton and one or more bits of a bit sequence supplied to the automaton, by determining the state elements representing the current state from the state elements representing the previous state and the one or more bits. In this context, the term "state transition," as used in the description and claims, refers in particular to the transition between two successive states (e.g.,The transition from the previous state to the current state is understood as the change triggered by the processing of one or more bits, thus allowing (especially if the previous or current state is known) a deduction about that one or more bits. It should also be noted that a state transition can lead back to the state from which it originates.

[0009] Furthermore, the term "encoding," as used in the description and claims, refers in particular to the mapping of a bit sequence representing plaintext to several symbol sequences according to a predetermined mapping rule. In this context, the term "symbol," as used in the description and claims, refers in particular to an element of a group of elements that contains information about a desired state transition of the convolutional encoder, triggered by the processing of one or more bits of the bit sequence. This allows a subset of state transitions containing the desired state transition to be determined from the set of all state transitions by knowing a symbol.Furthermore, the term "symbol sequence," as used in the description and claims, refers in particular to a sequence of symbols containing information about the state transitions of the convolutional encoder triggered by the processing of the bits of the bit sequence. Furthermore, the term "masking," as used in the description and claims, refers in particular to a mapping rule that is ambiguous in the reverse direction (i.e., from the symbol sequence to the bit sequence). A single mask can thus obscure a bit sequence, even though the totality of all masks allows for a unique mapping and thus the reconstruction of the bit sequence.

[0010] By using masking, a convolutional encoding is created, in which channel encoding is combined with cryptographic encoding, and whose output data, in the form of multiple symbol streams, is suitable for transmission over a multi-channel communication system. The convolutional encoding can exhibit the following properties: • From the portion of the output data transmitted on any single transmission channel (i.e., from a symbol string), little or no information about the input data (i.e., the plaintext) can be gleaned. A complete reconstruction of the input data is therefore impossible for an attacker using only a single symbol string, thus increasing the information security of the transmission. • From the totality of the parts of the output data transmitted via all transmission channels (i.e., from the totality of the symbol sequences), the input data (i.e., the plaintext) can be reconstructed by combination. • When reconstructing the input data (i.e., the plaintext), the decoder has the possibility to detect and / or correct symbol errors to a limited extent. • No secret is required for the encoding and decoding processes. Information security arises solely from the type of encoding and the distribution of the output data across the transmission channels, which themselves are not secret. • The output data can optionally be generated in such a way that a redundant symbol sequence can be transmitted on an additional transmission channel, whereby the aforementioned properties remain valid for any of the transmission channels. The use of an additional transmission channel also provides tolerance to the failure of any transmission channel, since the symbol sequences transmitted via the remaining transmission channels are sufficient for decoding by the receiver.

[0011] The state transitions included in the first group can lead to first states, whereby the first masking assigns a first symbol to at least two state transitions that originate from a first of the first states, and assigns the first symbol to at least two state transitions that originate from a second of the first states.

[0012] Therefore, knowledge of the first symbol does not allow conclusions to be drawn as to whether the convolutional encoder was in the first of the first states or in the second of the first states, nor which of the state transitions the convolutional encoder went through.

[0013] The state transitions included in the second group can lead to second states, whereby the second masking assigns a second symbol to at least two state transitions that originate from a first of the second states, and assigns the second symbol to at least two state transitions that originate from a second of the second states.

[0014] Therefore, the knowledge of the second symbol does not allow conclusions to be drawn as to whether the convolutional encoder was in the first of the two states or in the second of the two states, nor which of the state transitions the convolutional encoder went through.

[0015] The method can further include deriving a third symbol sequence from the data, wherein the third symbol sequence is intended for transmission on a third transmission channel. The derivation of the third symbol sequence can be based on a third masking of the data encoding by the convolutional encoder. The third masking can assign a third group of state transitions of the convolutional encoder to each symbol of the third symbol sequence, wherein the first group has at least a fourth state transition that the third group does not have, the second group has at least a fifth state transition that the third group does not have, the third group has at least a sixth state transition that neither the first nor the second group has, and the third group has at least the common third state transition.

[0016] The third, fourth, fifth, and sixth state transitions can originate from the same state and lead to different states. Furthermore, one of the third, fourth, fifth, and sixth state transitions can lead to the state from which the third, fourth, fifth, and sixth state transitions originate.

[0017] The state transitions included in the third group can lead to third states, whereby the third masking assigns a third symbol to at least two state transitions that originate from a first of the third states, and assigns the third symbol to at least two state transitions that originate from a second of the third states.

[0018] Therefore, the knowledge of the third symbol does not allow conclusions to be drawn as to whether the convolutional encoder was in the first or second of the third states, nor which of the state transitions the convolutional encoder went through.

[0019] A method according to the invention for decoding data comprises determining a convolutionally encoded message from two or more symbol sequences, wherein each of the symbol sequences is assigned to different bit sequences according to a (common) assignment rule and the convolutionally encoded message corresponds to a bit sequence that is assigned to all symbol sequences according to the (common) assignment rule, and decoding the convolutionally encoded message.

[0020] The term "common assignment rule," as used in the description and claims, refers in particular to an assignment rule that, based on the masking, assigns symbols to be transmitted via the transmission channels to each state transition of the convolutional encoder. The common assignment rule can, for example, be represented in the form of a state diagram in which each state transition is assigned the symbols to be transmitted via the various transmission channels, so that the symbol sequences can be determined by following a path through the state diagram.

[0021] An encoder according to the invention is configured to derive a first symbol sequence from data to be encoded and to transmit it via a first transmission channel or to store it in a first memory, and to derive a second symbol sequence from the data to be encoded and to transmit it via a second transmission channel or to store it in a second memory, wherein the derivation of the first symbol sequence is based on a first masking of an encoding of the data by a convolutional encoder, the derivation of the second symbol sequence is based on a second masking of the encoding of the data by the convolutional encoder, the first masking assigns a first symbol to a first group of state transitions of the convolutional encoder, and the second masking assigns a second symbol to a second group of state transitions of the convolutional encoder.

[0022] The first group and the second group each have at least one state transition that is included in both groups, at least one state transition that is included only in the first group, and at least one state transition that is included only in the second group.

[0023] The state transitions included in the first group can lead to first states, whereby the first masking assigns a first symbol to at least two state transitions that originate from a first of the first states, and assigns the first symbol to at least two state transitions that originate from a second of the first states.

[0024] The state transitions included in the second group can lead to second states, whereby the second masking assigns a second symbol to at least two state transitions that originate from a first of the second states, and assigns the second symbol to at least two state transitions that originate from a second of the second states.

[0025] The encoder can also be configured to determine a third symbol sequence based on the first and second symbol sequences and to transmit it via a third transmission channel or to store it in a third memory.

[0026] Each symbol in the third symbol sequence can be assigned to a third group of state transitions of the convolutional encoder. The first group can have at least one fourth state transition that the third group does not have. The second group can have at least one fifth state transition that the third group does not have. The third group can have at least one sixth state transition that neither the first nor the second group has. The third group can have at least the common third state transition. The third, fourth, fifth, and sixth state transitions can originate from the same state and lead to different states.

[0027] The state transitions included in the third group can lead to third states. At least two state transitions that originate from a first of the third states can be assigned a third symbol, and at least two state transitions that originate from a second of the third states can be assigned the third symbol.

[0028] A decoder according to the invention is configured to determine a convolutionally encoded message from two or more symbol sequences, wherein each of the symbol sequences is assigned to different bit sequences according to a (common) assignment rule, and the convolutionally encoded message corresponds to a bit sequence that is assigned to all symbol sequences according to the (common) assignment rule, and to decode the convolutionally encoded message.

[0029] Furthermore, it is understood that the features described in connection with the encoder and the decoder can also be features of the procedures and vice versa. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The invention is explained below in detail by means of exemplary embodiments, with reference to drawings in which: Fig. 1 a convolution encoder; Fig. 2 one the in Fig. 1 showing a convolution encoder representing a finite automaton; Fig. 3 a communication system; Fig. 4 a state diagram of a finite automaton representing a convolution code according to the present invention; Fig. 5a an illustration of data encoding; Fig. 5b an illustration of the first, second and third masking; Fig. 5c a further illustration of the first, second and third masking; Fig. 5d an illustration of the decoding; Fig. 6 symbol pairs of the state transitions of the in Fig. 4 finite automata shown; Fig. 7 state types and incoming and outgoing state transitions; and Fig. 8 a flowchart of a procedure for encoding data; and Fig. Figure 9 shows a flowchart of a procedure for decoding data.

[0031] In the drawings, identical or functionally similar elements are identified by the same reference symbols. WAYS TO IMPLEMENT THE INVENTION

[0032] Fig. Figure 1 shows a shift register representation of a binary convolutional encoder. This representation comprises two bit positions s0 and s1, into which the bits u of the data to be encoded are shifted sequentially. Each new input bit u is logically combined with the stored values ​​s0 and s1 according to the arrangement shown, generating two new output bits v0 and v1. These output bits v0 and v1, and the subsequent bit pairs, are interdependent by the influence length of the convolutional encoder, which is determined by the number of internal bit positions, thus enabling the distribution of the information content across multiple output bits.

[0033] Fig. Figure 2 shows a representation corresponding to the shift register representation in the form of a finite automaton. In this representation, the states of the automaton are labeled with the values ​​of the bits s1 and s0, and the state transitions between the states are labeled with the value of the input bit u, which triggers the respective state transition. Furthermore, each state transition is assigned the output bits v0 and v1, which are generated by the input bit u. The code generated by the convolutional encoder has a free distance (Hamming distance) of d. f -5. This means that for any two paths from one state to another or the same state, the symbol sequences of the two paths differ in at least 5 symbols.

[0034] Fig. Figure 3 shows a system 10 with a transmitter 12, which has an encoder 14, and a receiver 16, which has a decoder 18. Transmitter 12 and receiver 16 are connected to each other via three transmission channels 20, 22, and 24. If the encoder 14 were the one referred to in Fig. 1 and Fig. If the two described convolutional encoders were implemented, the binary symbols v0 and v1 could not (easily) be divided between two of the transmission channels 20, 22 and 24, so that the path through the states of the automaton could not be traced by observing the symbols v0 and v1 of only one of the transmission channels 20, 22 and 24.

[0035] In particular, the information security of system 10 cannot be increased (solely) by alternating the assignment of symbols v0 and v1 to transmission channels 20 and 22, since it is irrelevant which of the symbols v0 and v1 is known. Starting from a known state, the (additional) knowledge of just one of the two symbols v0 and v1 is sufficient to determine the subsequent state transition, thus revealing the subsequent state. Even alternating the pairwise assignment of symbol pairs to transmission channels 20 and 22 would not (easily) increase information security, because the path is revealed by observing multiple state transitions. A pseudorandom distribution generated using a secret key would indeed increase information security, but at the cost of using a secret key.

[0036] However, the information security of system 10 can be increased by transmitting symbols over two of the transmission channels 20, 22 and 24, if the encoder 14 is configured according to the specifications in Fig. The quaternary convolutional encoder 100 shown in section 4 is configured as follows. The quaternary convolutional encoder 100 has two symbol positions for recording the bitstream of the input data (memory depth m=2), resulting in a total of q when using a symbol alphabet of size q=4. mThis results in 16 states that the convolutional encoder 100 can assume. Furthermore, four state transitions originate from each state, and four state transitions lead into each state, one for each input symbol u={0,1,2,3}. To increase the information security of system 10 during the transmission of the output symbols w2={0,1,2,3}, w1={0,1,2,3}, and w0={0,1,2,3}, the output symbols w2, w1, or w0 are assigned to the state transitions such that in each state, two of the four state transitions use the same output symbol (0, 1, 2, or 3) on one of the transmission channels 20, 22, and 24. Therefore, observing the output symbols w2, w1, or w0 on one of the transmission channels 20, 22, and 24 does not reveal unambiguous information about the state transition. Rather, only half of the subsequent states can be ruled out.In order for the state transitions to be unique for the decoder 28, which receives the symbols of two transmission channels 20, 22 and 24, the pairs of state transitions that use the same output symbol w2, w1 or w0 must be different on both transmission channels 20, 22 and 24.

[0037] For example, in encoder 14, as in Fig. 5a illustrates that the data 26, consisting of the symbol sequence “1-3-3”, is entered, and the result is derived according to the in Fig. 4 quaternary convolutional encoders shown, 100 trained encoders, 14 of which are in Fig. The symbol sequences 28, 30, and 32 shown in Figure 5b are transmitted, with the first symbol sequence 28 "0-0-3" being transmitted via the first transmission channel 20, the second symbol sequence 30 "3-2-3" being transmitted via the second transmission channel 22, and the third symbol sequence 32 "3-2-0" being transmitted via the third transmission channel 24 to the receiver 16. This is illustrated by the example of the first symbols 34, 36, and 38 of the symbol sequences 28, 30, and 32 in Figure 5b. Fig. As illustrated in Figure 5b, the first symbols 34, 36 and 38 are assigned to groups 40, 42 and 44 at state transitions 46, 48, 50 and 52, with each of the groups 40, 42 and 44 having the common state transition 48 and another state transition 46, 50 and 52 which the others of the groups 40, 42 and 44 do not include.

[0038] To increase the information security of system 10, it is further advantageous if an observer cannot retrospectively exclude state transitions from a sequence of several output symbols w2, w1, or w0 on one of the transmission channels 20, 22, and 24 to such an extent that only one possibility remains for a state transition, thereby revealing the underlying input symbol u. Therefore, it is advantageous if, for an output symbol w2, w1, or w0 that marks a transition from one state to possible subsequent states, the output symbols w2, w1, or w0 that are possible from the subsequent states are identical, because then the preceding state transition cannot be inferred from the observation of the subsequent transition.

[0039] This is exemplified in Fig. Figure 5c illustrates the state transitions 60 and 62, which begin with state S. 00depart, and the state transitions 64 and 66, which are from state S 10 The symbol 54 is assigned to the transitions from state S. State transitions 68 and 66, which originate from state S, are also assigned to state S. 10 depart, and the state transitions 70 and 72, which are from state S 30 The symbol 56 is assigned to the transitions from state S. State transitions 74 and 66, which originate from state S, are also included. 10 depart, and the state transitions 76 and 78, which are from state S 20 The symbol 58 is assigned to the exit. In decoder 18, as in Fig. Figure 5d illustrates that the convolutionally coded message 80 can be derived from the symbol sequences 28, 30 and 32, and that by decoding the convolutionally coded message 80 the data 26 can be determined.

[0040] For such a construction, the output symbols w2, w1, or w0 can be calculated from the quaternary code symbols v0 and v1. Since these are transmitted on the individual transmission channels 20, 22, and 24, they are subsequently also referred to as channel symbols w2, w1, or w0. The channel symbols w1 or w0 for transmission channels 20 and 22 are calculated by interchanging the code bits as follows: w0={v1(0),v0(0)} and w1={v1(1),v0(1)}.

[0041] For the (optional) third transmission channel 24, the channel symbol w2 can be used. w2=w0⊕w1 The code symbols v0 and v1 can be calculated. This allows the code symbols v0 and v1 to be reconstructed from any two of the three channel symbols. Furthermore, the conversion ensures that when decoding using the symbol w2, a single-bit error (for example, caused during data transmission) only destroys one of the quaternary code symbols v0 and v1, and not both.

[0042] It can now be arbitrarily defined that the transitions exiting a state use either the channel symbol pair w={0,3} or w={1,2} for a transmission channel. Two of the four transitions thus receive the same channel symbol. Furthermore, it can be defined that transitions with the channel symbols w={0,2} lead to a state from which the symbol pair 0 / 3 is then used. Similarly, transitions with the channel symbols w={1,3} lead to a state from which the symbol pair 1 / 2 is used. This relationship is in Fig. Figure 6 is schematically illustrated, although the depiction of the fact that several transitions with the same symbol originate from each state and each lead to two different target states has been omitted for the sake of clarity.

[0043] To achieve these conditions, the exclusive construction of the convolutional encoder 100 via generator polynomials can be abandoned. However, the structure consisting of states, transitions, and their associated code symbols v0 and v1 can be retained. The code symbols v0 and v1 can, however, be modified as described in Fig. Figure 7 shows that the transitions can be directly assigned. In particular, the 16 combinations of code and channel symbols (v1v0=00,01,02...32,33) resulting from Formula 1 and Formula 2 can be sorted such that they each have two identical channel symbols per transmission channel. The pairs can also be offset by four combinations on the transmission channels. Each set of four combinations is assigned to one of four state types.

[0044] This results in the following: Fig. The 7 depicted state types A to D show the following properties: • For each state type and each transmission channel, only two of the four channel symbols are used to identify the outgoing transition. • The two channel symbols used are either from the pair 0 / 3 or 1 / 2. • The pairs are arranged offset across the three transmission channels, so that each transition is unique with any two of the three channel symbols. • The transitions lead to a state type as shown in the illustration in Fig. 6. This allows the channel symbol pair of the subsequent transition to be predicted. This applies to all three transmission channels, even though the pairs are staggered. • All 16 possible combinations of code symbols v1v0 are used. • Each state type, with its four transitions, leads to all four state types, and each state type is reachable from all four types. The states, or types, are therefore all interconnectable.

[0045] The q m The state types must now be assigned to the 16 states. Due to the structure in Fig. 7 requires that the same type be assigned to each of the 4 states. This results, for example, in the convolutional encoder 100. Fig. 4, whose states are labeled with their respective state type. The start state can be, for example, state S. 00 The code has a free distance (Hamming distance) of d, based on the code symbols that can be calculated from the channel symbols. f =3 and is therefore suitable for error correction or error detection.

[0046] Furthermore, the use of convolutional encoder 100 provides (additional) cryptographic protection because an attacker who can only observe one of the transmission channels 20, 22, or 24 cannot uniquely determine the state of convolutional encoder 100 (after the first transition). Specifically, there are always multiple transitions, and thus data symbols, to choose from, which could be the cause of the observed channel symbol. Therefore, the data symbol can no longer be uniquely determined. Since the channel symbols do not allow any prediction of the expected channel symbol pair of the subsequent transition, no relationship between the individual data symbols can be established.

[0047] Fig. Figure 8 shows a flowchart of a procedure for encoding the data 26. This includes step 82 of deriving the symbol sequence 28 from the data 26 and step 84 of deriving the symbol sequence 30 from the data 26. Subsequently, the symbol sequences 28 and 30 are sent via transmission channels 20 and 22 to the receiver 16, where the symbol sequences 28 and 30 are decoded.

[0048] Fig. Figure 9 shows a flowchart of a procedure for decoding the symbol sequences 28 and 30. This includes step 86 of determining the convolutionally encoded message 80 from the symbol sequences 28 and 30 and step 88 of decoding the convolutionally encoded message 80. REFERENCE MARK LIST 10 System 12 channels 14 encoders 16 recipients 18 decoders 20 transmission channels 22 transmission channels 24 transmission channels 26 data 28 symbol sequence 30 symbol sequence 32 symbol sequence 34 Symbol 36 Symbol 38 Symbol Group 40 Group 42 Group 44 46 State transition 48 State transition 50 State transition 52 State transition 54 Symbol 56 Symbol 58 Symbol 60 State transition 62 State transition 64 State transition 66 State transition 68 State transition 70 State transition 72 State transition 74 State transition 76 State transition 78 State transition 80 messages 82 steps 84 steps 86 steps 88 steps 100 convolutional encoders

Claims

[1] Methods for encoding data (26), comprising: Deriving (82) a first symbol sequence (28) from the data (26), wherein the first symbol sequence (28) is intended for transmission on a first transmission channel (20) or for storage in a first memory; and Deriving (84) a second symbol sequence (30) from the data (26), wherein the second symbol sequence (30) is intended for transmission on a second transmission channel (22) or for storage in a second memory; wherein the derivation of the first symbol sequence (28) is based on a first masking of an encoding of the data (26) by a convolutional encoder (100); wherein the derivation of the second symbol sequence (30) is based on a second masking of the encoding of the data (26) by the convolutional encoder (100); wherein the first masking assigns to each symbol (34, 54) of the first symbol sequence (28) a first group (40) of state transitions (46, 48) of the convolutional encoder (100); wherein the second masking assigns to each symbol (36, 56) of the second symbol sequence (30) a second group (42) of state transitions (46, 50) of the convolutional encoder (100); wherein the first group (40) has at least one first state transition (46) which the second group (42) does not have; wherein the second group (42) has at least one second state transition (50) which the first group (40) does not have; wherein the first group (40) and the second group (42) have at least one common third state transition (48); where the first, second and third state transitions (46, 48, 50) from an identical state (S 00 ) depart and enter different states (S 00 , S 10 , S 30 ) flow into. [2] Method for encoding data (26) according to claim 1, wherein the state transitions (46, 48) included in the first group (40) into first states (S 00 , S 10 ) flow into; where the first masking involves at least two state transitions (60, 62), which are from a first (S 00 ) of the first states (S 00 , S 10 ) depart, assigns a first symbol (54) and at least two state transitions (64, 66), which are from a second (S 10 ) of the first states (S 00 , S 10 ) depart, assigning the first symbol (54). [3] Method for encoding data according to claim 1 or 2, wherein the state transitions (48, 50) to second states (S) included in the second group (42) 10 , S 30 ) flow into; where the second masking involves at least two state transitions (66, 68), which are from a first (S 10 ) of the second states (S 10 , S30 ) depart, assigns a second symbol (56), and at least two state transitions (70, 72), which are from a second (S 30 ) of the second states (S 10 , S 30 ) depart, assigning the second symbol (56). [4] Method for encoding data according to any one of claims 1 to 3, further comprising: Deriving a third symbol sequence (32) from the data (26), wherein the third symbol sequence (32) is intended for transmission on a third transmission channel (24) or for storage in a third memory; wherein the derivation of the third symbol sequence (32) is based on a third masking of the encoding of the data (26) by the convolutional encoder (100); wherein the third masking assigns to each symbol (38, 58) of the third symbol sequence (32) a third group (44) of state transitions (48, 52) of the convolutional encoder (100); where the first group (40) has at least one fourth state transition (46) which the third group (44) does not have; where the second group (42) has at least one fifth state transition (50) which the third group (44) does not have; wherein the third group (44) has at least one sixth state transition (52) which the first group (40) and the second group (42) do not have; wherein the third group (44) has at least the common third state transition (48); where the third, fourth, fifth and sixth state transitions (46, 48, 50, 52) from an identical state (S 00 ) depart and enter different states (S 00 , S 10 , S 20 , S 30 ) flow into. [5] Method for encoding data according to claim 4, wherein the state transitions (48, 52) included in the third group (44) into third states (S 10 , S 20) flow into; where the third masking involves at least two state transitions (66, 74), which are from a first (S 10 ) of the third states (S 10 , S 20 ) depart, assigns a third symbol (58), and at least two state transitions (76, 78), which are from a second (S 20 ) of the third states (S 10 , S 20 ) depart, assigning the third symbol (58). [6] Methods for decoding data, comprising: Determine (86) a convolutionally coded message (80) from two or more symbol sequences (28, 30, 32), wherein each of the symbol sequences (28, 30, 32) is assigned to different bit sequences according to an assignment rule and the convolutionally coded message corresponds to a bit sequence that is assigned to all symbol sequences according to the common assignment rule; and Decoding (88) the convolutionally encoded message (80). [7] Encoder (14) which is set up: to derive a first symbol sequence (28) from the data to be encoded (26) and to transmit it via a first transmission channel (20) or to store it in a first memory; and to derive a second symbol sequence (30) from the data (26) to be encoded and to transmit it via a second transmission channel (22) or to store it in a second memory; wherein the derivation of the first symbol sequence (28) is based on a first masking of an encoding of the data (26) by a convolutional encoder (100); wherein the derivation of the second symbol sequence (30) is based on a second masking of the encoding of the data (26) by the convolutional encoder (100); wherein the first masking of a first group (40) of state transitions (46, 48) of the convolutional encoder (100) assigns a first symbol (34); wherein the second masking of a second group (42) of state transitions (48, 50) of the convolutional encoder assigns a second symbol (36); where the first group (40) and the second group (42): exhibit at least one state transition (48) that is included in both groups (40, 42), exhibit at least one state transition (46) that is only included in the first group (40), and exhibit at least one state transition (50) that is only included in the second group (42). [8] Encoder according to claim 7, wherein the state transitions (46, 48) included in the first group (40) into first states (S 00 , S 10 ) flow into; where the first masking involves at least two state transitions (60, 62), which are from a first (S 00 ) of the first states (S 00 , S 10) depart, assigns a first symbol (54) and at least two state transitions (64, 66), which are from a second (S 10 ) of the first states (S 00 , S 10 ) depart, assigning the first symbol (54). [9] Encoder according to claim 7 or 8, wherein the state transitions (48, 50) to second states (S) included in the second group (42) 10 , S 30 ) flow into; where the second masking involves at least two state transitions (66, 68), which are from a first (S 10 ) of the second states (S 10 , S 30 ) departs a second symbol (56), and at least two state transitions (70, 72), which are assigned a second (S 30 ) of the second states (S 10 , S 30 ) depart, assigning the second symbol (56). [10] Encoder according to any one of claims 7 to 9, wherein the encoder is further configured as follows: to determine a third symbol sequence (32) based on the first and second symbol sequences (28, 30) and to transmit it via a third transmission channel (24) or to store it in a third memory. [11] Decoder (18) which is set up: to determine a convolutionally encoded message (80) from two or more symbol sequences (28, 30, 32), wherein each of the symbol sequences (28, 30, 32) is assigned to different bit sequences according to an assignment rule, and the convolutionally encoded message (80) corresponds to a bit sequence that is assigned to all symbol sequences (28, 30, 32) according to the common assignment rule; and to decode the convolutionally encoded message (80).

Images