Subscriber station for a bus system and method for improving the reception quality of messages at a subscriber station of a bus system

Abstract

Subscriber station (10; 30; 50; 60) for a bus system (1; 2), comprising an estimating device (12) for estimating a channel impulse response (120) when and / or after only one other subscriber station (10; 20; 30; 50; 60) of the bus system (1; 2) sends a message (41; 42; 43) to the bus system (1; 2), and a correction device (13) for correcting a signal received by the subscriber station (10; 50; 60) on the basis of the channel impulse response (120) estimated by the estimating device (12), wherein the estimating device (12) is configured such that, in estimating the channel impulse response (120), it uses parameters and / or coefficients of previously received messages (41, 42, 43) identified by the message identifier (411, 421, 431) are assigned to a subscriber station which sent the received message (41, 42, 43).

Description

Technical field The present invention relates to a subscriber station for a bus system and a method for improving the reception quality of messages at a subscriber station of a bus system, in which, in particular, the reception quality during information transmission on a CAN bus system is improved even under unfavorable conditions, such as those that may be present due to branches, faulty terminations, mismatches, or pinched cable routing in the bus system. State of the art Patent application US 2002 / 0181567 A1 discloses a method for the rapid identification of transmission channel characteristics. A known training sequence is used to calculate an estimate of the channel's impulse response. This impulse response is then used to set the initial values ​​of filter coefficients, for example, for an echo canceller and an equalizer. US patent 5,870,433 A discloses a method for digital telecommunications that uses a Viterbi decoder based on the DDFSE (Delayed Decision Feedback Sequence Estimation) algorithm. To determine the branch metrics in the trellis diagram, separate first feedback filters are used only for a specific, short number of symbol intervals. For all other, more distant symbol intervals, a common feedback filter is used for all states. Document US 2003 / 0016770 A1 discloses a comprehensive system and method for communication systems, describing a novel network architecture of packet / circuit communication processors. This system aims to increase transmission speeds across various communication channels and topologies. To achieve this, techniques such as multi-channel signal coding, advanced adaptive equalization, and precision synchronization are employed to improve channel capacity and seamlessly integrate services. Document US 2003 / 0081668 A1 discloses a method for the rapid calculation of the coefficients of a decision feedback equalizer (DFE). Starting with a channel estimate, the problem is formulated as a recursive least-squares (RLS) problem. A fast algorithm, such as a fast transverse filter (FTF), is used to calculate the forward filter coefficients (FFE). The backward filter coefficients (FBE) are then determined by convolving the FFE coefficients with the channel impulse response, which can be efficiently performed in the frequency domain using a fast Fourier transform (FFT). The CAN bus system has become widely used for communication between sensors and control units. With the CAN bus system, messages are transmitted using the CAN protocol, as described in the CAN specification in ISO 11898. More recently, techniques such as CAN-FD have been proposed, in which messages are transmitted according to the specification "CAN with Flexible Data-Rate, Specification Version 1.0" (source: http: / / www.semiconductors.bosch.de), etc. These techniques increase the maximum possible data rate beyond 1 Mbit / s by using a higher clock frequency in the data fields. This generally comes at the expense of transmission quality, for example, in the form of a higher bit error rate, assuming actual bus topologies are used. Actual bus topologies typically deviate from the theoretical model in that reflections occur on the bus line at points where the line exhibits a characteristic impedance that differs from the theoretical value. Such points include, for example, branch connections, incorrect terminations, impedance mismatches, and pinched cable routing, which are often found in practical implementations such as spur lines, passive star points, etc. The resulting reflections lead to temporal crosstalk of states on the bus line, such that a transmitted symbol or bit can interfere with subsequent symbols, potentially distorting their detection. According to the CAN specification in ISO 11898, the bus line should be terminated at both ends with the line impedance, so that transients decay within a transmitted symbol for the specified maximum cable length, resulting in a clear state at the end of the symbol interval. In reality, however, crosstalk between two or more CAN symbols is often unavoidable. The receiver of a CAN bus system consists of a communication processor, usually integrated into a microcontroller, and a transmitter / receiver, also called a transceiver, which is typically a separate chip with a direct connection to the bus line. In such a transceiver, the receive path usually comprises only a comparator with upstream voltage dividers for bias adjustment of the bus levels. The comparator directly evaluates the bus levels of dominant and recessive bit states and makes a decision at its output. However, directly generating and outputting signal decisions has the disadvantage that reflections on the bus line can negatively influence the decision and lead to incorrect decisions in signal transmission. This is particularly true at higher clock rates for data fields exceeding 1 Mbit / s, as is the case with CAN-FD, for example. Here, due to the shortened bit duration, reflections that would otherwise contribute constructively to the decision in a conventional CAN bus system with a lower clock rate have a negative impact. Reflections at cable transitions towards higher impedances generally result in positive reflections. Conversely, reflections at cable transitions towards lower impedances result in negative reflections. Due to two reflections at different distances, time shifts occur. The reflections considered severely restrict the application area, for example with regard to possible topologies, cable lengths, etc., of currently considered techniques with higher data transmission rates, such as CAN-FD, etc. Although equalization methods for improving receiver quality are generally known in the field of communication technology, their application to CAN communication systems is not yet known. Furthermore, special measures are required to use known equalization methods for CAN communication systems, as these were not considered in the system design. Disclosure of the invention Therefore, the object of the present invention is to provide a participant station for a bus system and a method that solve the aforementioned problems. In particular, a participant station for a bus system and a method are to be provided that improve the reception quality during information transmission on a bus system, which is in particular a CAN bus system, even under the aforementioned unfavorable conditions of a real bus line. The problem is solved by a participant station for a bus system with the features of claim 1. The participant station comprises an estimation device for estimating a channel impulse response when and / or after only one other participant station of the bus system sends a message to the bus system, or for determining required filter functions directly from a signal received by the participant station, and a correction device for correcting a signal received by the participant station based on the channel impulse response estimated by the estimation device. The receiving station is also suitable for use in higher-frequency systems, such as CAN-FD, etc. The receiving station's functionality with respect to the received signal can also be implemented, in particular, as preprocessing in a transmitter / receiver or transceiver, a CAN transceiver, or a transceiver chipset. Specifically, the functionality under consideration can be embedded either in the transceiver as a separate electronic component (chip) or in an integrated solution containing only one electronic component (chip). The receiving station is capable of improved reception in a CAN communication system and therefore offers higher reliability. Here, the bus signal is not decided directly by a comparator or similar device, as before, but is (first) corrected, for example, by additional filtering and feedback of previously determined received values. It is also possible to represent the decisions using hypotheses and to further process the determined received values. Advantageous further embodiments of the participant station are specified in the dependent patent claims. The estimation device may be designed to perform the estimation of the channel impulse response based on an LMS algorithm or an RLS algorithm, or to directly determine the channel impulse response. The estimation device can be designed in such a way that, when estimating the channel impulse response, it uses parameters and / or coefficients of previously received messages, which are assigned to the corresponding transmitting station by the message identifier. The estimation device can also be designed to display a result of its estimation based on hypotheses. The correction device can be designed for further processing received values ​​determined according to, for example, the DDFSE algorithm. It is possible that the correction device is also designed to correct the signal received from the receiving station according to a DFE algorithm, a BCJR algorithm, or a DDFSE algorithm. The correction device can also include a filter for correcting the signal received from the receiving station, wherein, for example, the signal-to-noise ratio at the respective decision point is maximized or optimized at the filter's input. A further filter can be used to filter received values ​​previously decided according to the DDFSE algorithm, and the correction device can be designed to provide feedback to the filter's output. Preferably, the subscriber station also comprises a transmit / receive device for direct connection to a bus line of the bus system, and a communication control device for processing the signals received by the transmit / receive device and for providing the messages to be sent by it in the form of signals, wherein the estimation device and / or the correction device are part of the transmit / receive device or the communication control device. The previously described participant station can be part of a bus system that also includes a bus line and at least two participant stations connected to each other via the bus line in such a way that they can communicate with each other. At least one of the at least two participant stations is a participant station as described above. The aforementioned problem is further solved by a method for improving the reception quality of messages at a subscriber station of a bus system according to claim 9. The method comprises the following steps: Estimating, using an estimating device, a channel impulse response when and / or after only one other subscriber station of the bus system transmits a message to the bus system; or determining, using the estimating device, required filter functions directly from a signal received from the subscriber station; and correcting, using a correction device, a signal received from the subscriber station based on the channel impulse response estimated by the estimating device. This method is particularly suitable for use in higher-frequency systems, such as CAN-FD, etc., and can be implemented in one embodiment, in particular, as preprocessing in the receiving station. Furthermore, the method is also applicable to other systems with a similar structure, such as FlexRay, etc. The procedure offers the same advantages as previously mentioned in relation to the participant station. Other possible implementations of the invention also include combinations of features or embodiments described previously or subsequently with regard to the exemplary embodiments, even if not explicitly mentioned. In such cases, the person skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention. Drawings The invention is described in more detail below with reference to the accompanying drawing and exemplary embodiments. The figures show: Fig. 1 a simplified block diagram of a bus system according to a first embodiment; Fig. 2 an electrical circuit diagram of a transmit / receive device of the bus system according to the first embodiment; Fig. 3 a signal waveform of a transmit signal transmitted via the bus system according to the first embodiment; Fig. 4 a first example of a channel impulse response to the signal waveform shown in Fig. 3; Fig. 5 a signal waveform of a receive signal received by the transmit / receive device of Fig. 2 with a channel impulse response according to Fig. 4; Fig. 6 a second example of a channel impulse response to the signal waveform shown in Fig. 3; Fig. 7 a signal waveform of a receive signal received by the transmit / receive device of Fig. 2 with a channel impulse response according to Fig. 4.6 is received; Fig. 8 a flowchart of a method according to the first embodiment; Fig. 9 a simplified block diagram of a bus system according to a second embodiment; and Fig. 10 a block diagram to illustrate the function of an estimation device and a correction device of the bus system according to the second embodiment. In the figures, identical or functionally equivalent elements are provided with the same reference symbols unless otherwise specified. Description of the exemplary implementations Fig. 1 shows a bus system 1, which can be, for example, a CAN bus system, a CAN FD bus system, etc. The bus system 1 can be used in a vehicle, in particular a motor vehicle, an aircraft, etc., or in a hospital, etc. In Fig. 1, the bus system 1 has a plurality of participant stations 10, 20, 30, each connected to a bus line 40. Messages 41, 42, 43 in the form of signals can be transmitted between the individual participant stations 10, 20, 30 via the bus line 40. The messages 41, 42, 43 each have one of the message identifiers 411, 421, 431. The participant stations 10, 20, 30 can be, for example, control units or display devices of a motor vehicle. As shown in Fig. 1, subscriber station 10 has a communication control unit 11, an estimating unit 12, a correction unit 13, and a transmit / receive unit 14. Subscriber station 20, on the other hand, has a communication control unit 11 and a transmit / receive unit 14. Subscriber station 30, like subscriber station 10, has a communication control unit 11, an estimating unit 12, a correction unit 13, and a transmit / receive unit 14. The transmit / receive units 14 of subscriber stations 10, 20, and 30 are each directly connected to the bus line 40, even though this is not shown in Fig. 1. The communication control unit 11 serves to control communication between each participant station 10, 20, 30 via bus line 40 and another participant station 10, 20, 30 connected to bus line 40. The estimation unit 12 and the correction unit 13 serve to improve the reception quality of the message 41, 42, 43 received by the transmit / receive unit 14, as will be described in more detail later. The communication control unit 11 can be implemented like a conventional CAN controller. The transmit / receive unit 14 can be implemented like a conventional CAN transceiver with respect to its transmission functionality. Consequently, the quality of the signal received by the transmit / receive unit 14, which is based on one of the messages 41, 42, 43, can be improved by the two participant stations 10, 30.In contrast, participant station 20 corresponds to a conventional CAN participant station in terms of both its transmitting and receiving functionality. Fig. 2 shows the structure of a transmit / receive device 14 of the subscriber station 20 in more detail as an example. The transmit / receive device 14 has a receive path 140 for receiving the signal based on one of the messages 41, 42, 43. The receive path includes a first and second input terminal 141, 142, two resistors 143, 144, a comparator 145, a processing element 146, and an output terminal 147. Resistor 143 is connected between the first input of comparator 145, which is at positive potential, and the first input terminal 141. Resistor 144 is connected between the second input of comparator 143, which is at negative potential, and the second input terminal 142. Figures 3, 4, 5, 6 to 7 show examples of the transmit / receive signals of the transmit / receive devices 14. In these examples, one of the messages 41, 42, 43 is fed onto the bus line 40 in the form of a signal from one of the transmit / receive devices 14 of the subscriber stations 10, 20, 30, as shown in Figure 3. The vertical axis in Figures 3, 4, 5, 6 to 7 represents the level difference of the message, which is plotted against time (horizontal axis) in µs. If the channel impulse response is as shown in Fig. 4 according to a first example, the communication control unit 11 of the subscriber station 20 receives a signal as shown in Fig. 5. If, on the other hand, the channel impulse response is as shown in Fig. 6 according to a second example, the communication control unit 11 of the subscriber station 20 receives a signal as shown in Fig. 7. In contrast, the communication control devices 11 of the subscriber stations 10, 30 in the first example shown in Fig. 3, Fig. 4, Fig. 5, Fig. 6 to Fig. 7 receive, due to their estimation device 12 and correction device 13, a signal that is very similar to or almost identical to the transmitted signal of Fig. 3. Fig. 8 shows a method for improving the reception quality of a signal based on one of the messages 41, 42, 43 at a subscriber station 10, 30 of the bus system 1. The method is explained using an example in which subscriber station 10 sends message 41 to subscriber station 30. Accordingly, at step S1 in Fig. 8, the estimating device 12 of the participant station 30 estimates the channel impulse response 120, which is generally expected for the transmission channel (channel impulse response) of both participants in the case of the actually installed bus line 40. For this purpose, the estimating device 12 formulates a hypothesis based on experience gained during a transmission between participant station 10 as sender and participant station 30 as receiver. The hypothesis of the estimating device 12 can thus also be described as the result of a learning process. In a CAN bus system, several participating stations 10, 20, 30 are active, transmitting their messages 41, 42, 43 with message identifiers 411, 421, 431. These message identifiers 411, 421, 431 are used for arbitration during the arbitration phase. After arbitration, only one participating station (10, 20, 30) transmits signals in the form of one or more messages 41, 42, 43 onto the bus line 40. From this point on, each listening participating station (10, 20, 30) can observe the bus signals or messages 41, 42, 43 and, based on this, estimate the channel impulse response 120 using its estimating device 12. This can also be done adaptively and during detection or acquisition. In this case, a learning system is present. The estimating unit 12 of the participant station 30 then provides the estimated channel impulse response 120 to the correction unit 13. Subsequently, in step S2, the correction device 13 corrects the signal received from the subscriber station 30, which is based on the message 41, on the basis of the channel impulse response 120 estimated by the estimating device 12. The procedure is then completed. A delay in signal or message 41, 42, 43 is not inherently present, as the delay falls within the time frame tolerated by the CAN protocol. However, it can be utilized through optimization measures, e.g., in procedures with hypotheses and delayed decision-making, provided this is acceptable to the application. In such cases, a delay of 1-2 symbol clock cycles may be sufficient. The estimating unit 12 of subscriber station 30 operates similarly when subscriber station 20 sends one of the messages 41, 42, or 43 to subscriber station 30. Furthermore, the estimating unit 12 of subscriber station 10 operates similarly when it receives one of the messages 41, 42, or 43 from subscriber stations 20 or 30. This allows the reception quality for information transmission on bus system 1 to be improved even under unfavorable conditions, such as branches, faulty terminations, mismatches, and cramped cable routing in bus system 1. Fig. 9 shows a bus system 2 according to a second embodiment. In addition to at least one participant station 10, which is configured as in the first embodiment, the bus system 2 comprises at least one participant station 50 and at least one participant station 60. The participant stations 10, 50, and 60 are each connected to the bus line 40, as in the first embodiment. Messages 41, 42, and 43 in the form of signals can be transmitted between the individual participant stations 10, 50, and 60 via the bus line 40, as in the first embodiment. Here, too, the messages 41, 42, and 43 each have one of the message identifiers 411, 421, and 431, respectively. The participant stations 50 and 60 can also be, for example, control units or display devices of a motor vehicle, etc. As shown in Fig. 9, each of the subscriber stations 10, 50, 60 has an estimating device 12 and a correction device 13. At subscriber station 50, however, the estimating device 12 and the correction device 13 are part of a transmit / receive device 54. At subscriber station 60, however, the estimating device 12 and the correction device 13 are part of a communication control device 61. The communication control device 11 of subscriber station 50 is otherwise identical to the communication control device 11 of subscriber station 10. Furthermore, the transmit / receive device 54 of subscriber station 50 is otherwise identical to the transmit / receive device 14 of subscriber station 10. Fig. 10 shows the structure of the estimation device 12 and the correction device 13 of the transmit / receive device 54, which are used in place of the comparator 145 (Fig. 2). The correction device 13 forms, for example, a Decision-Feedback Equalization structure, which is also referred to as the DFE structure below, as shown in Fig. 10. The estimation device 12 provides the channel impulse response 120 required for correction, which can be represented by a signal b(k). Fig. 10 shows the basic structure of a DFE structure with a preceding channel model. The disturbance on the channel or bus line 40 is represented by a signal n(k) and a filter 541 with transfer function G(z), which corresponds to the z-transform of the discrete-time sequence g(k). The received signal at the subscriber stations 10, 50, 60 results from a transmitted signal a(k) after passing through a channel, which is modeled by a filter 542 with transfer function H(z). At point 543, the resulting signals are added and then fed to a filter 544. Thus, the filter 544 receives the signal received from subscriber station 50 at its input.Filter 544 is a possible additional filter in the transmit / receive device 54 with a transfer function F(z), which can be used to concentrate the signal energy, but for the sake of simplicity, it can be assumed to have a constant transfer function, such as F(z) = 1. The modeled disturbance is then subtracted from the signal output by filter 544 at point 545. The normalized channel impulse response 120 is the resulting total channel, composed of the channel modulated by filter 542 and the output of filter 544. The time-compensated total channel is represented by B(z), the z-transform of b(k), in Fig. 10. The value at k=0 represents the main tap considered for the decision, and the subsequent values ​​at k>0 correspond to the taps of preceding symbols.For the coefficients b(k) of the channel impulse response 120, b(k) = 0, so that the main tap of the filter 547 is eliminated, which is represented by the term -1 of the filter 547 given by B(z)-1. The signal â(k-k0) resulting from the compensation by the correction signal after the addition stage 545 is then fed to a decision point 546, which corresponds to the comparator 145 of the first embodiment. Subsequently, the received signal corrected by the correction device 13 is available for further processing by the communication control device 11. In Fig. 10, the term k0 represents a decision delay, with delay k0 of the filtered signal after the filter 544, which can be optimized with the filter 544 with the transfer function F(z) and is assumed here with k0 = 0 for the simplified consideration. For example: - Example channel impulse response h(k) = δ(k) + 0.5 δ (k-1) + 0.2 δ (k-2) - Pre-filter impulse response f(k) = δ(k) => F(z) = constant - Decision delay k0 = 0 In this case, b(k) = h(k) is a reasonable solution, and B(z)-1 corresponds to the channel impulse response b(k) - δ(k) = h(k) - δ(k) = 0.5 δ(k-1) + 0.2 δ(k-2). It is easy to see from this that only previously decided and known symbols are used for feedback, since the current decision time corresponds to k and this impulse response only considers terms related to received values ​​(k-1, k-2) from the past. Although the estimating device 12 and the correction device 13 can be implemented very easily in the transmitting / receiving device 54, their implementation in the communication control device 11 is also feasible, as shown by the communication control device 61. The method according to this embodiment monitors the CAN bus signal during transmission and corrects the signal using the DFE structure shown above. Knowledge of the channel impulse response is advantageous here, as it depends on the transmitter-receiver pairing, as described in the first embodiment. The channel impulse response can be determined as described in the first embodiment. All previously described configurations of bus system 1, participant stations 10, 20, 30, and the procedure can be used individually or in any possible combination. In addition, the following modifications are particularly conceivable. The bus system 1, 2 described above, according to the first and second embodiments, is based on a CAN protocol-based bus system. However, the bus system 1 according to the first and second embodiments can also be a different type of communication network. It is advantageous, but not a necessary requirement, that the bus system 1, 2 ensures exclusive, collision-free access to a common channel for at least certain periods of time for each participant station 10, 20, 30, 50, 60. The bus system 1, 2 according to the first and second embodiments is in particular a CAN network or a TTCAN network or a CAN FD network. The number and arrangement of the participant stations 10, 20, 30, 50, 60 in the bus systems 1, 2 of the first and second embodiments is arbitrary. In particular, there can also be only participant stations 10, 50, or 60 in the bus systems 1, 2 of the first and second embodiments. Instead of the Decision-Feedback Equalization structure (DFE structure) described in the first embodiment for the estimation device 12 and the correction device 13, any trellis-based equalization methods can also be used, such as the BCJR algorithm (in BCJR the individual letters stand for the initials of the developers: B for L. Bahl, C for J. Cocke, J for F. Jelinek, R for J. Raviv,) and / or a DDFSE algorithm (DDFSE = Delayed Decision-Feedback Sequence Estimation). Several applications for the receiving station and the process it implements are conceivable. In particular, in addition to its use with CAN and CAN-FD, an application in FlexRay is also possible. These systems currently do not use any equalization methods in the receiver, the receiving station. The process can be optimized separately for each specific application, such as CAN-FD, FlexRay, etc. For multiple application areas, the application can also be automatically detected and adapted accordingly. For example, the process is faster for CAN-FD than for CAN, so that no delay occurs that exceeds the tolerance of the respective protocol. A basis for the procedure is the determination of the channel impulse response 120, which is used for equalization. This can be obtained by estimating the channel impulse response 120 of the transmission channel of bus line 40. Alternatively, the channel impulse response 120 can also be directly determined or optimized in the form of the signal b(k). The estimation of the channel impulse responses 120 can be carried out using any suitable method. Since no training sequence is available, adaptive methods are particularly suitable, such as an LMS algorithm (LMS = least mean squares) and / or an RLS algorithm (RLS = recursive least squares filter). To stabilize the estimation and increase the accuracy of the channel impulse response 120, the parameters and coefficients of previously received messages 41, 42, 43 or packets of messages 41, 42, 43, which can also be called receive bursts, can be assigned to the message identifiers or the sending subscriber stations and used for subsequently received messages 41, 42, 43 or packets of messages 41, 42, 43. In addition to or as an alternative to estimating the channel or its channel impulse response 120, the filters 541, 542, 544 in the transmit / receive device 14, 54, 61 can be optimized. For this purpose, filters 544 (transfer function F(z)) and 547 (transfer function B(z)-1) as well as the decision delay k0 are considered. Various criteria can be used for optimization, such as maximizing the signal-to-noise power ratio at the input of the decision controller 546. Alternatively, the required filter coefficients 544, 547 are set by the estimating device without explicitly estimating the channel impulse response 120, but are determined directly from the signal received by the subscriber station 10, 50, 60. Participant stations 10, 30, 50 offer a particular opportunity for CAN-FD to increase the reception quality of CAN-FD to the range of typical CAN transmissions while using a significantly higher data rate. The method can be implemented, with regard to the functionality of the received signal, for example in a transceiver or a transmit / receive device 14, 54, in a communication control device 61, etc. Additionally or alternatively, it can be integrated into existing products.

Claims

Subscriber station (10; 30; 50; 60) for a bus system (1; 2), comprising an estimating device (12) for estimating a channel impulse response (120) when and / or after only one other subscriber station (10; 20; 30; 50; 60) of the bus system (1; 2) sends a message (41; 42; 43) to the bus system (1; 2), and a correction device (13) for correcting a signal received by the subscriber station (10; 50; 60) on the basis of the channel impulse response (120) estimated by the estimating device (12), wherein the estimating device (12) is configured such that, in estimating the channel impulse response (120), it uses parameters and / or coefficients of previously received messages (41, 42, 43) identified by the message identifier (411, 421, 431) are assigned to a subscriber station which sent the received message (41, 42, 43). Subscriber station (10; 30; 50; 60) according to claim 1, wherein the estimating device (12) is configured to perform the estimation of the channel impulse response (120) based on an LMS algorithm or an RLS algorithm, or to determine the channel impulse response (120) directly from a transmit signal. Participant station (10; 30; 50; 60) according to one of the preceding claims, wherein the estimation device (12) is configured such that it previously represents a result of its estimation by hypotheses and the correction device (13) is configured for further processing of received values ​​decided according to the DDFSE algorithm Subscriber station (10; 30; 50; 60) according to one of the preceding claims, wherein the correction device (13) is further configured to correct the signal received by the subscriber station (10; 30; 50; 60) according to a DFE algorithm or a BCJR algorithm or a DDFSE algorithm. Subscriber station (10; 30; 50; 60) according to one of the preceding claims, wherein the correction device (13) comprises a filter (546) for correcting the signal received by the subscriber station (10; 30; 50; 60), and wherein the signal-to-noise power ratio is maximized at the input of the filter (546) at the respective time of reception. Subscriber station (10; 30; 50; 60) according to claim 5, wherein a further filter can be used to filter received values ​​previously determined according to the DDFSE algorithm, and wherein the correction device (13) is designed to provide feedback of an output of the filter (546). Subscriber station (10; 30; 50; 60) according to one of the preceding claims, furthermore comprising a transmitting / receiving device (14) for direct connection to a bus line (4) of the bus system (1; 2), and a communication control device (11) for processing the signals received by the transmitting / receiving device (14) and for providing the messages (41; 42; 43) to be sent by it in the form of signals, wherein the estimating device (12) and / or the correction device (13) are part of the transmitting / receiving device (14) or the communication control device (11). Bus system (1; 2), comprising a bus line (4), and at least two subscriber stations (10; 29; 30; 50; 60), which are connected to each other via the bus line (4) in such a way that they can communicate with each other, wherein at least one of the at least two subscriber stations (10; 20; 30; 50; 60) is a subscriber station (10; 30; 50; 60) according to one of the preceding claims. Method for improving the reception quality of messages (41; 42; 43) at a subscriber station (10; 30; 50; 60) of a bus system (1; 2), comprising the steps of estimating, using an estimating device (12), a channel impulse response (120), when and / or after only one other subscriber station (10; 20; 30; 50; 60) of the bus system (1; 2) sends a message (41; 42; 43) to the bus system (1; 2), and correcting, using a correction device (13), a signal received by the subscriber station (10; 30; 50; 60) based on the channel impulse response (120) estimated by the estimating device (12), wherein the estimating device (12) is configured such that, when estimating the channel impulse response (120), it considers parameters and / or coefficients of previously received messages. (41, 42, 43) are used, which are assigned by the message identifier (411, 421, 431) to a subscriber station that sent the received message (41, 42, 43).

Images