SENDER AND RECEIVER AND CORRESPONDING PROCEDURES

Description

[0001] The invention relates to a transmitter and a receiver and corresponding methods for transmitting signals or for synchronizing a receiver with a transmitter.

[0002] In many data transmission systems, pilot sequences (also called training sequences or synchronization sequences) are inserted into the data streams to be transmitted for signal detection or parameter estimation. This can involve the transmission of an uninterrupted data stream into which pilot sequences are interspersed at certain intervals, or a packet-oriented transmission, in which each packet (also called a telegram) usually contains exactly one pilot sequence. The pilot sequence is also called a preamble or midamble if it is located at the beginning or in the middle of the packet, respectively. A pilot sequence can also be distributed within the packet in the form of two or more subsequences.

[0003] In some radio-based systems with asynchronous packet transmission, the transmission pauses are considerably longer than the respective packet duration. Some systems employ telegram splitting (see [4] and [1], DE 10 2011 082 098 A1), in which each packet is divided into a number of subpackets – called fragments or telegram fragments. Each fragment typically contains its own pilot sequence. Telegram splitting proves particularly robust in interference-limited systems in which a large number of transmitters transmit uncoordinated telegrams that are to be received and decoded by a single receiver. This occurs, for example, in telemetry systems, sensor networks, or applications under the heading Internet of Things (IoT).

[0004] The demodulation of data from a radio-based data transmission system requires receiver synchronization consisting of: Time synchronization: Estimation of the precise temporal position of the packet and the optimal sampling point. Frequency synchronization: Estimation and correction of the frequency offset between the transmit and receive oscillators, i.e., the frequency difference between the carrier frequency of the transmit signal and the center frequency of the receive filter. Phase synchronization: Estimation of the phase after frequency correction (only required for coherent demodulation).

[0005] In asynchronous packet transmission, these three types of synchronization must be performed for each packet prior to demodulation, independently of the previously received packets. For this purpose, each packet contains a pilot sequence that allows for initial synchronization (so-called "initial acquisition"). Furthermore, parameter tracking may be required during data demodulation. The problem of initial synchronization will be considered below.

[0006] Telegram splitting offers the advantage of making systems robust against packet collisions, especially with a large number of uncoordinated transmitters. However, no approaches are known for telegram splitting in the area of ​​synchronization, and especially frequency synchronization, that lead to satisfactory results.

[0007] EP 2914039 A1 describes a data transmission arrangement, wherein the data transmission arrangement comprises a power supply device for supplying the data transmission arrangement with electrical energy; a device for determining payload data provided by a sensor element connected to the data transmission arrangement; a device for channel-coding the payload data to obtain channel-coded data; a device for dividing the channel-coded data into a plurality of data packets with a code rate of less than one; and a device for transmitting data packets, which is designed to transmit the plurality of data packets at a time interval via a communication channel, provided that an amount of electrical energy that can be provided by the power supply device is sufficient to transmit the respective data packets with a standard transmission power.The device for transmitting data packets is designed to transmit a data packet of the plurality of data packets that is pending for transmission at a reduced transmission power if an amount of electrical energy that can be provided by the energy supply device is sufficient to transmit the respective data packet at the reduced transmission power, wherein the reduced transmission power is reduced by a maximum of 40 dB compared to the standard transmission power; and / or not to transmit a data packet of the plurality of data packets that is pending for transmission or to transmit it with a delay if an amount of electrical energy that can be provided by the energy supply device is insufficient to transmit the respective data packet.

[0008] A transmitter is configured to transmit at least one signal with a pilot sequence or partial pilot sequence comprising multiple pilot symbols. The transmitter comprises a signal generator that provides the pilot sequence or partial pilot sequence.

[0009] In one embodiment, the transmitter is designed to transmit at least one signal with a pilot sequence comprising a plurality of pilot symbols. The signal generator provides the pilot sequence such that the pilot sequence has at least two symbol groups, each with at least two pilot symbols. When evaluating a signal received by a receiver as a result of the transmission of the signal, the symbol groups generate phase errors with respect to a phase, which are reduced by a time offset between a reference time of transmission; and / or not to transmit a data packet of the plurality of data packets that is due to be transmitted, or to transmit it with a delay, if the amount of electrical energy that can be provided by the energy supply device is insufficient to transmit the respective data packet.

[0010] The object of the invention is to propose a receiver, a receiving method and a corresponding computer program which in particular use telegram splitting and which are improved with regard to synchronization compared to the prior art.

[0011] This problem is solved by the independent patent claims.

[0012] Advantageous further developments can be found in the dependent patent claims.

[0013] One embodiment of the receiver provides that the receiver is configured to receive multiple telegram fragments transmitted by the transmitter, wherein the telegram fragments each have a partial pilot sequence. The telegram fragments complement each other to form a telegram containing data transmitted by the transmitter, and wherein the telegram fragments are shorter than a single telegram containing the transmitted data. Furthermore, the partial pilot sequences complement each other to form a pilot sequence. Based on the partial pilot sequences and applying the DFT method, the signal evaluation device determines a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver.This determination is performed by the signal evaluation device determining values ​​of decision variables for each partial pilot sequence, combining the determined values ​​of the decision variables across the partial pilot sequences, and using the combined values ​​to determine a maximum value. Optionally, the maximum value is also compared with a decision threshold.

[0014] In one embodiment, it is provided that the signal evaluation device determines the values ​​of the decision variables for each partial pilot sequence by the signal evaluation device forming, for a plurality of frequencies of a complex exponential oscillation, referred to as frequency hypotheses, multiplication values ​​of a multiplication of sample values ​​of the respective partial pilot sequence with conjugate complex values ​​of an equal number of reference symbols of a known reference sequence and with sample values ​​of the complex exponential oscillation and summing the multiplication values.

[0015] According to one embodiment, the signal evaluation device combines the values ​​of the decision variables determined for the partial pilot sequences together (i.e. jointly across the partial pilot sequences) for the partial pilot sequences by the signal evaluation device adding the absolute values ​​of the decision variables.

[0016] In one embodiment, it is provided that the signal evaluation device combines the values ​​of the decision variables determined for the partial pilot sequences together for the partial pilot sequences by the signal evaluation device forming a sum of the amounts of a real part and an imaginary part of the values ​​of the decision variables.

[0017] According to one embodiment, the signal evaluation device combines the values ​​of the decision variables determined for the partial pilot sequences together for the partial pilot sequences by the signal evaluation device summing the absolute value squares of the values ​​of the decision variables.

[0018] In one embodiment, it is provided that the signal evaluation device combines the values ​​of the decision variables determined for the partial pilot sequences together for the partial pilot sequences coherently, taking into account the magnitude and phase of the values ​​of the decision variables.

[0019] According to one embodiment, the signal evaluation device combines the determined values ​​of the decision variables with each other, taking into account weighting factors.

[0020] In one embodiment, it is provided that the signal evaluation device combines the determined values ​​of the decision variables with each other, taking into account weighting factors relating to the partial pilot sequences.

[0021] According to one embodiment, the signal evaluation device determines the weighting factors based on a signal-to-noise ratio of the respective telegram fragment.

[0022] In one embodiment, it is provided that the signal evaluation device determines the weighting factors as proportional to a root of the signal-to-noise ratio.

[0023] According to one embodiment, the signal evaluation device determines the weighting factors as proportional to a root of a quotient of the signal-to-noise ratio and a noise power of the respective telegram fragment.

[0024] In one embodiment, it is provided that the signal evaluation device determines the weighting factors as a function of a computing power available to the signal evaluation device or as a function of a ratio of the signal-to-noise ratio relative to a predeterminable relation value or as a function of an interference power either as proportional to a root of the signal-to-noise ratio or as proportional to a root of a quotient of the signal-to-noise ratio and a noise power of the respective telegram fragment.

[0025] According to one embodiment, the receiver is designed to receive a plurality of telegram fragments transmitted by the transmitter, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram containing data transmitted by the transmitter, wherein the telegram fragments are shorter than a single telegram containing the transmitted data and wherein the partial pilot sequences complement each other to form a pilot sequence.Based on the partial pilot sequences and using the DFT method, the signal evaluation device determines a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver by determining values ​​of decision variables for each partial pilot sequence, combining the determined values ​​of the decision variables across the partial pilot sequences, and using the combined values ​​to determine a maximum value.The signal evaluation device carries out a time estimate for a time offset between a reference time of transmission of the telegram fragment and a value assumed and / or estimated for the evaluation of the telegram fragment for the reference time, in that the signal evaluation device determines a time estimate for a time offset based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

[0026] In one embodiment, the receiver is configured to receive at least one telegram transmitted by the transmitter, wherein the telegram has a pilot sequence. Based on the pilot sequence and using the DFT method, the signal evaluation device determines a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver by determining values ​​of decision variables for the pilot sequence and determining a maximum value.The signal evaluation device carries out a time estimate for a time offset between a reference time of transmission of the telegram and a value assumed and / or estimated for the evaluation of the telegram for the reference time, in that the signal evaluation device determines a time estimate for a time offset based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

[0027] According to one embodiment, the signal evaluation device obtains the maximum value based on the determination of the maximum value and a positive comparison of the maximum value with a decision threshold.

[0028] In one embodiment, it is provided that the signal evaluation device generates the time estimate from the maximum value and two adjacent values, and that the two adjacent values ​​precede or follow the determined maximum value with regard to the time variable.

[0029] According to one embodiment, the signal evaluation device determines a polynomial for the maximum value and the two adjacent values. Furthermore, the signal evaluation device determines the time estimate from an extreme value associated with the polynomial.

[0030] One embodiment consists in that the signal evaluation device performs the interpolation with a second-degree polynomial.

[0031] According to one embodiment, the signal evaluation device (12) carries out the interpolation with a polynomial of the following form: y(x) = y 0 - c(xx 0 ) 2,< where the free parameters y 0 , c and x 0 are determined on the basis of the maximum value and the neighboring values.

[0032] The transformation device determines a maximum value of the interpolation curve using the following function: x 0 = 1 2 y 1 − y − 1 2 y 0 − y 1 − y − 1 , where x 0 is an abscissa value of a maximum of the polynomial, y(0) is the maximum value and y(-1) and y(1) are the neighboring values

[0033] In one embodiment, a second degree polynomial of the form y(x) = y 0 - c(xx 0 ) 2< is used as the interpolation function.

[0034] For example, the abscissa value x 0 of the polynomial maximum represents the improved time estimate (normalized to the sampling interval T / N).

[0035] From the abscissa value x 0 of the polynomial maximum, the improved frequency estimate can be calculated using ν ^ T = i 0 + x 0 N FFT − 0 für i 0 < N FFT / 2 1 für i 0 ≥ N FFT / 2 calculate.

[0036] In one embodiment, it is provided that the receiver is designed to receive a plurality of telegram fragments transmitted by the transmitter, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram containing data transmitted by the transmitter, wherein the telegram fragments are shorter than a single telegram containing the transmitted data and wherein the partial pilot sequences complement each other to form a pilot sequence.

[0037] Based on the partial pilot sequences and using the DFT method, the signal evaluation device determines a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver. This is done by determining decision variable values ​​for each partial pilot sequence, combining the determined decision variable values ​​across the partial pilot sequences, and using the combined values ​​to determine a maximum value.

[0038] The signal evaluation device carries out a frequency estimation for the frequency difference in that the signal evaluation device determines a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a frequency variable.

[0039] According to one embodiment, the receiver is configured to receive at least one telegram transmitted by the transmitter, the telegram having a pilot sequence. Based on the pilot sequence and using the DFT method, the signal evaluation device determines a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver. The signal evaluation device determines values ​​of decision variables for the pilot sequence and carries out a determination of a maximum value. The signal evaluation device determines a frequency estimate for the frequency difference. The signal evaluation device determines a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a frequency variable.

[0040] In one embodiment, it is provided that the signal evaluation device obtains the maximum value based on the maximum value and a positive comparison of the maximum value with a decision threshold.

[0041] According to one embodiment, the signal evaluation device generates the frequency estimate from the maximum value and two adjacent values, wherein the two adjacent values ​​precede or follow the determined maximum value with respect to the frequency variable.

[0042] In one embodiment, it is provided that the signal evaluation device determines a polynomial for the maximum value and the two adjacent values, and that the signal evaluation device determines the frequency estimate from an extreme value associated with the polynomial.

[0043] One embodiment consists in that the signal evaluation device performs the interpolation with a second-degree polynomial.

[0044] According to one embodiment, the signal evaluation device (12) carries out the interpolation with a polynomial of the following form: y(x) = y 0 - c(xx 0 ) 2,< where the free parameters y 0 , c and x 0 are determined on the basis of the maximum value and the neighboring values.

[0045] The transformation device determines a maximum value of the interpolation curve using the following function: x 0 = 1 2 y 1 − y − 1 2 y 0 − y 1 − y − 1 , where x 0 is an abscissa value of a maximum of the polynomial, y(0) is the maximum value and y(-1) and y(1) are the neighboring values

[0046] In one embodiment, a second degree polynomial of the form y(x) = y 0 - c(xx 0 ) 2< is used as the interpolation function.

[0047] For example, the abscissa value x 0 of the polynomial maximum represents the improved time estimate (normalized to the sampling interval T / N).

[0048] From the abscissa value x 0 of the polynomial maximum, the improved frequency estimate can be calculated using ν ^ T = i 0 + x 0 N FFT − 0 für i 0 < N FFT / 2 1 für i 0 ≥ N FFT / 2 calculate.

[0049] According to one embodiment, the signal evaluation device determines the time estimate and uses it for a renewed determination of the frequency difference by using samples of the respective partial pilot sequences shifted by the time estimate when re-determining the values ​​of the decision variables or by shifting reference symbols of a reference sequence in time by the time estimate.

[0050] In one embodiment, it is provided that the signal evaluation device carries out an interpolation of the sample values ​​of the respective partial pilot sequences in order to re-determine the frequency difference.

[0051] According to one embodiment, after the renewed determination of the frequency difference, the signal evaluation device carries out a frequency estimation for the frequency difference in that the signal evaluation device determines a frequency estimate based on a maximum value obtained by determining the maximum value and a positive comparison with the decision threshold and at least one value adjacent to a frequency variable.

[0052] The aforementioned embodiments of the receiver can also be implemented in the synchronization method - possibly as part of a method for receiving signals.

[0053] A variant of the procedure, which is complementary or alternative to the aforementioned variants, provides that several telegram fragments transmitted by the transmitter are received by the receiver, ∘ wherein the telegram fragments each have a partial pilot sequence, ∘ wherein the telegram fragments add up to form a telegram that contains data transmitted by the transmitter, ∘ wherein the telegram fragments are shorter than the telegram, ∘ wherein the partial pilot sequences add up to form a pilot sequence, that based on the partial pilot sequences, the receiver is synchronized with the transmitter, o based on the partial pilot sequences and using the DFT method, a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver is determined, ▪ by determining values ​​of decision variables for each partial pilot sequence,▪ by combining the determined values ​​of the decision variables across all partial pilot sequences and ▪ by using the combined values ​​to determine a maximum value.

[0054] One embodiment provides that the receiver receives a plurality of telegram fragments transmitted by the transmitter, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram that contains data transmitted by the transmitter, wherein the telegram fragments are shorter than the telegram, wherein the partial pilot sequences complement each other to form a pilot sequence.

[0055] Based on the partial pilot sequences, the receiver is synchronized with the transmitter. Based on the partial pilot sequences and using the DFT method, a value for a frequency difference between a transmitter's transmit frequency and a receiver's receive frequency is determined by determining decision variable values ​​for each partial pilot sequence, combining the determined decision variable values ​​across all partial pilot sequences, and using the combined values ​​to determine a maximum value.A time estimate for a time offset between a reference time of transmission of the telegram fragment and a value assumed and / or estimated for the evaluation of the telegram fragment for the reference time is carried out by the signal evaluation device determining a time estimate for a time offset based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

[0056] One embodiment consists in that at least one telegram transmitted by the transmitter is received by the receiver, the telegram having a pilot sequence, the receiver being synchronized with the transmitter on the basis of the pilot sequence, a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver being determined on the basis of the pilot sequence and using the DFT method, by determining values ​​of decision variables for the pilot sequence, by determining a maximum value using the determined values, a time estimate being made for a time offset between a reference time of transmission of the telegram fragment and a value assumed and / or estimated for the evaluation of the telegram fragment for the reference time,in that the signal evaluation device determines a time estimate for a time offset based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

[0057] One embodiment provides that the receiver receives a plurality of telegram fragments transmitted by the transmitter, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram that contains data transmitted by the transmitter, wherein the telegram fragments are shorter than the telegram, wherein the partial pilot sequences complement each other to form a pilot sequence, wherein the receiver is synchronized with the transmitter based on the partial pilot sequences.Based on the partial pilot sequences and applying the DFT method, a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver is determined by determining values ​​of decision variables for each partial pilot sequence, by combining the determined values ​​of the decision variables across all partial pilot sequences, and by determining a maximum value using the combined values. A frequency estimate for the frequency difference is performed by the signal evaluation device determining a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a frequency variable.

[0058] One embodiment consists in that the receiver receives at least one telegram transmitted by the transmitter, wherein the telegram has a pilot sequence. Based on the pilot sequence, the receiver is synchronized with the transmitter. Based on the pilot sequence and using the DFT method, a value for a frequency difference between a transmission frequency of the transmitter and a reception frequency of the receiver is determined by determining values ​​of decision variables for the pilot sequence and by determining a maximum value using the determined values. A frequency estimate is carried out for the frequency difference by the signal evaluation device determining a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with regard to a frequency variable.

[0059] The above embodiments of the receiver can be realized by steps of corresponding embodiments of the method, so that a repetition of the explanations is omitted here.

[0060] Depending on the design, the receiver uses a method for frequency estimation in the case of fragmented transmission, i.e. when the telegrams are fragmented during "telegram splitting".

[0061] In one embodiment, the DFT method for frequency estimation is extended to telegram splitting.

[0062] In one embodiment, a combination is performed after the DFT, which is applied to the individual telegram fragments, and before the maximum value search, which thus refers to all telegram fragments.

[0063] In one embodiment, the signal processing of the DFT method up to and including the DFT is applied to each telegram fragment to combine the results before the maximum value search using the frequency hypotheses. Further processing is performed based on the combined values.

[0064] The advantage of merging is that it results in significant noise suppression, which can significantly improve the estimation results for frequency and time.

[0065] In one embodiment, a combination takes place through weighted addition of individual results, e.g., the amounts or squares of the amounts.

[0066] In different configurations, a weighted addition of the individually determined values ​​of the decision variables is carried out by an addition the amounts, the sum of the amounts of the real and imaginary parts or the squares of the amounts.

[0067] Mathematically, the configurations can be formulated as follows: Let di,n [ k ] the decision variables (for i = 0, 1, ..., N DFT -1) of the k-th sample time after the DFT of the n-th fragment.

[0068] Then the following applies: Addition of the amounts: d i k = ∑ n = 0 N frag − 1 c n d i , n k Addition of the amounts of Re and Im: d i k = ∑ n = 0 N frag − 1 c n Re d i , n k + Im d i , n k Addition of the squares of the absolute values: d i k 2 = ∑ n = 0 N frag − 1 c n 2 d i , n k 2

[0069] In one embodiment, the symbol grid remains constant throughout the duration of all transmitted fragments, and the distance between the fragments (e.g., in the number of symbol intervals) is known to the receiver. In this case, the kth samples from each telegram fragment (counted from the beginning of the fragment) correspond to one another. Furthermore, in one embodiment, the frequency offset does not change significantly throughout the duration of all fragments.

[0070] In one embodiment, a coherent combination is performed, i.e., a combination based on the magnitude and phase of the respectively determined values ​​of the decision variables. This is particularly important in the case where the carrier phase remains the same from telegram fragment to telegram fragment, i.e., coherence must be maintained across the telegram fragments.

[0071] The choice of weighting factors for adding the values ​​of the decision variables is specified in more detail in some embodiments.

[0072] In one embodiment, the aforementioned weighting factors cn are chosen proportional to the square root of the estimated signal-to-noise ratio (SNRn) of the nth telegram fragment, whereby the estimated SNR also includes the power of any interference signals present. This corresponds to maximum ratio combining (MRC) in antenna diversity.

[0073] In the case of interference powers that may be different for the individual fragments, one embodiment provides for setting the weighting factors proportional to the square root of the ratio SNR n / P Nn of the estimated SNR to the estimated noise power PN of the n-th telegram fragment.

[0074] By choosing the weight factors, the SNR in the decision variable is maximized after the combination.

[0075] Alternatively, the above weight factors cn are set equal to 1. This corresponds to the well-known Equal Gain Combining (EGC) in antenna diversity.

[0076] One refinement represents an improvement in the parameter estimates.

[0077] In one embodiment, the time estimation is improved by means of interpolation.

[0078] In order to increase the accuracy of the time estimation, one embodiment provides for improving the time estimation after determining the time index k 0 in the DFT method by an interpolation between the maximum value (i.e. |di [k 0 ]|), the value before (i.e. |di [k 0 -1]|) and the value after (i.e. |di [k 0 +1]|).

[0079] The two neighboring values ​​must have the same frequency index as the maximum value, ie i = i 0 [k 0 ].

[0080] In one embodiment, a second degree polynomial of the form y(x) = y 0 - c(xx 0 ) 2< is used as the interpolation function.

[0081] In one embodiment, the free parameters y 0 , c and x 0 are determined such that the polynomial passes through the three values ​​y(-1) = |di [k 0 -1]|, y(0) = |di [k 0 ]| and y(1) = |di [k 0 +1]|.

[0082] The abscissa value x 0 of the polynomial maximum represents the improved time estimate (normalized to the sampling interval T / N). It can be calculated from the three given y-values ​​as follows: x 0 = 1 2 y 1 − y − 1 2 y 0 − y 1 − y − 1 .

[0083] In an alternative or supplementary embodiment, the frequency estimation is improved by means of interpolation.

[0084] To increase the accuracy of the frequency estimation, a design provides that, after determining the time index k 0 , the frequency estimation is carried out by interpolation between the maximum value at the position i 0 (i.e. d i 0 [ k 0 ]), the value for i 0 -1 (i.e. d i 0-1 [ k 0 ]) and the value for i 0 +1 (i.e. d i 0+1 [ k 0 ]).

[0085] The two neighboring values ​​must come from the same DFT from which the maximum value was determined. This is the DFT at time k 0 .

[0086] In one embodiment, a second-degree polynomial of the form y(x) = y 0 - c(xx 0 ) 2< is used as the interpolation function. This corresponds to the previously mentioned polynomial as described for improving time estimation.

[0087] The free parameters y 0 , c and x 0 are determined in such a way that the polynomial is exactly defined by the three values ​​y(-1) = | d i0-1 [ k 0 ]| , y(0) = | d i 0 [ k 0 ]| and y(1) = | d i 0+1 [ k 0 ]| runs.

[0088] From the abscissa value x 0 of the polynomial maximum, the improved frequency estimate can be calculated using ν ^ T = i 0 + x 0 N FFT − 0 für i 0 < N FFT / 2 1 für i 0 ≥ N FFT / 2 calculate.

[0089] The value x 0 can be calculated from the three given y-values ​​according to formula (8).

[0090] In one embodiment, the following sequence of measures to improve the estimated value is provided.

[0091] To optimize time and frequency estimation, it is planned to work with an oversampling factor of N = 2 in one embodiment and to perform the following operations in the given order: 1. Interpolation in the time domain as described above. One result is the time estimate τ̂ 2. Approximate calculation of L new samples of the signal x(t) in symbol time, where the sampling times are different from the original ones by τ̂are shifted. This is conveniently done by interpolation filtering of the existing sample values ​​x[k-LN-1], x[k-LN], ..., x[k+2]. In one embodiment, an FIR filter with four coefficients is used as the interpolation filter. Experience has shown that this provides sufficient accuracy with 2-fold oversampling. To calculate the filter coefficients, one embodiment provides for the use of the method of minimizing the mean square error (MMSE), whereby the convolution of the modulation pulse with the impulse response of the receive filter serves as the calculation basis. With a matched filter as the receive filter, this is the ACF of the modulation pulse. The filter coefficients can be calculated and stored a priori for a sufficiently fine grid of time estimation hypotheses (e.g., in a grid of 0.01T). The filter whose time hypothesis corresponds to the current time estimate is used in each case. τ̂3. Recalculate the decision variable di [k 0 ] based on the samples interpolated in the previous step. A different DFT length can be chosen. Since only three output values ​​of the DFT are of interest, this can be replaced by multiplying three times with samples of three complex exponential oscillations with the frequencies ν 0 T − 1 N DFT , v0 T and ν 0 T + 1 N DFT , maintaining the DFT length. 4. Polynomial interpolation in the frequency domain as described previously. One result is the frequency estimate v 5. To further improve the estimates, a second polynomial interpolation is subsequently performed in a further embodiment, using values ​​that lie in a denser frequency grid. For this purpose, the DFT input values ​​must be multiplied three times with samples of three complex exponential oscillations with the frequencies v̂T - Δ vT , v̂T and v̂T + Δ vTmultiplied, where ΔγT<<1 / N DFT is to be chosen. In one embodiment, the following applies: Δ νT = 1 8 N DFT .

[0092] As an alternative to steps 3 to 5, one embodiment uses a method for directly estimating the remaining frequency offset. This is, for example, the method according to Louise and Reggiannini.

[0093] All you need to do is: 3. Calculation of L values ​​according to z l = x k − L − 1 − l N a ∗ l e j 2 πlν 0 T für l = 0 , 1 , 2 , … , L − 1 , where the sample values ​​x[.] are the interpolated values ​​from step 2. Multiplication with a complex exponential oscillation with the frequency v 0 T corresponds to a frequency correction by the frequency v 0 T. 4. Carry out the calculations according to formulas (2) and (4) using the values ​​from formula (9). The estimated value for the total frequency deviation is then the sum of v 0 T and the estimated value from formula (4).

[0094] Furthermore, the invention relates to a system for signal transmission which has at least one transmitter according to one of the embodiments discussed above and at least one receiver according to one of the above embodiments.

[0095] Finally, the invention relates to a computer program with a program code for carrying out the aforementioned method according to one of the above embodiments.

[0096] In detail, there are numerous possibilities for designing and developing the transmitter, the receiver, the system, and the corresponding methods. Reference is made to the patent claims and the following description of exemplary embodiments in conjunction with the drawings. They show: Fig. 1 a schematic representation of a signal processing for time and frequency estimation, Fig. 2 a representation of an MSK baseband signal for three positively mapped symbols, Fig. 3 a representation of an MSK baseband signal with ideal time offset, Fig. 4 a representation of an MSK baseband signal with negative time offset, Fig. 5 a curve of the symbol amplitude after matched filtering over the time offset, Fig. 6 a curve of the phase offset of the Fig. 5 considered symbol over the time offset after matched filtering and multiplication with the conjugate complex of the reference symbol, Fig. 7 a representation of an MSK baseband signal for three symbols, where the last symbol is negatively mapped, Fig. 8 a course of the symbol amplitude from Fig. 7 after matched filtering over the time offset, Fig. 9 a course of the phase offset of the Fig. 8considered symbol over the time offset after matched filtering and multiplication with the conjugate complex of the reference symbol, Fig. 10 a schematic representation of a communication system with a transmitter and a receiver, Fig. 11 a schematic representation of an example of a modulation used and Fig. 12 an illustration of an interpolation by a second degree polynomial.

[0097] With telegram splitting (see, for example, DE 10 2011 082 098 A1), the data to be transmitted is not sent in a single data packet, but rather multiple data packets are generated, which may be transmitted at different frequencies and at different times. Thus, multiple telegram fragments are generated from one telegram. Each telegram fragment has a pilot sequence that serves to synchronize the sender and receiver. The pilot sequences are identical in one embodiment and different from each other in another.

[0098] A pilot sequence generally consists of a number L of modulation symbols (hereinafter also referred to as pilot symbols). It is usually transmitted compactly either at the beginning (preamble) or in the middle (midamble) of the telegram or telegram fragment. However, it can also be divided into two or more segments, between which data symbols are transmitted. It is common practice to take the pilot symbols from the same modulation alphabet as the data symbols (e.g., multiple phase-shift keying, M-PSK, or M-ary Quadrature Amplitude Modulation, M-QAM). The pilot symbols are known in advance to the receiver or, for example, are stored appropriately.

[0099] In receivers, it is common to bandpass filter the received signal, then downconvert it to baseband, and then sample and quantize it equidistantly using an analog-to-digital converter (ADC). Each sampled value is complex-valued and consists of a real and an imaginary part. Sampling occurs at least at the symbol rate or a multiple thereof (so-called oversampling).

[0100] Several methods are known for the time and frequency estimation required for synchronization, which are described below. DFT method:

[0101] In [9], a method is described that is also suitable for large frequency offsets. The frequency offset is the frequency difference between the carrier frequency of the transmitted signal and the center frequency of the receive filter. The essential signal processing steps are described in Fig. 1 illustrated.

[0102] Where T is the symbol spacing or 1 / T is the symbol rate. N is the oversampling factor, and k*T / N denotes the time of the k-th sample. The pilot symbols are denoted by a[0], a[1], ..., a[L-1], where a[0] is transmitted first and a[L-1] last. The superscript * indicates that the complex conjugate is to be used. The delay element is denoted by z -N<, where z is the delay of N samples. Furthermore, W is the window length for the maximum value search in number of samples.

[0103] Time variables in parentheses are always continuous-time; for example, r(t) denotes the continuous-time received signal. In contrast, time variables in square brackets are discrete-time and usually represent a consecutive numbering of sample values. For example, x[k] denotes the kth value of the (continuous-time) signal x(t) after the receive filter.

[0104] After the receive filter, which generates the signal x(t) from the signal r(t), the signal is sampled and quantized equidistantly in time with N values ​​per symbol. At each time point k, L samples are taken at symbol spacing (i.e., every Nth sample) and first multiplied by the L conjugate complex pilot symbols. These are then multiplied by the samples of a complex exponential oscillation and then summed. The last process is performed several times for different frequencies of this oscillation, which are referred to as frequency hypotheses. For equidistant frequency hypotheses, this corresponds to the discrete Fourier transform (DFT) of the L products x[k]a*[L-1], x[kN]a*[L-2] to x[k-(L-1)N]a*[0].

[0105] The DFT can be performed particularly efficiently using the well-known fast Fourier transform (FFT) algorithm, provided L is a power of two. If L is not a power of two, the DFT length is rounded up to the next higher power of two and the L values ​​are supplemented by the corresponding number of zeros, allowing an FFT to be applied. To increase the accuracy of the frequency estimation, the L FFT input values ​​can be supplemented by any number of additional zeros.

[0106] Each frequency hypothesis yields a complex value, denoted by di [k] and referred to as the decision variable. The index i refers to the i-th frequency. For each time step k, the maximum of the magnitude |di [k]| is determined over all frequency indices i. The frequency index corresponding to the maximum is denoted by i 0 . If this maximum lies above a decision threshold d thr , a pilot sequence (alternatively also referred to as a training sequence) is considered to have been recognized. The corresponding time index is denoted by k'. In the positive case that the maximum lies above the decision threshold, k' is set equal to k. In the negative case, k is increased (k becomes k + 1, Fig. 1 ). The comparison with the decision threshold is optional in each case.

[0107] It is known that further analysis of the received signal is necessary for a more accurate timing estimation. Further processing is as follows:

[0108] The decision variables are analyzed at time k' and for each time point following k', and the maximum magnitude within a time window of given duration is determined. The time index corresponding to this maximum is denoted by k 0 .

[0109] The frequency belonging to the index i 0 [k 0 ] already represents a rough estimate of the frequency offset. Let it be v 0 and is given by ν 0 T = i 0 N DFT − 0 für i 0 < N DFT / 2 1 für i 0 ≥ N DFT / 2

[0110] For additive Gaussian interference, the DFT method is optimal in terms of maximum likelihood estimation. It is applicable for frequency offsets up to almost half the symbol rate. The estimation accuracy is essentially limited by the oversampling factor N and the DFT length NDFT. For large N and large NDFT, the mean square estimation error for both time and frequency approaches the theoretically achievable limit (Cramer-Rao bound). The method is applicable for any pilot sequence. Frequency estimation according to Luise and Reggiannini:

[0111] The Luise and Reggiannini estimator is based on the following values: R λ = 1 L − λ ∑ l = 0 L − λ − 1 z l + λ z ∗ l für λ = 0 , 1 , … , Λ with z l = x k − L − 1 − l N a ∗ l

[0112] The frequency estimate is given by: ν ^ T = 1 π Λ + 1 arg ∑ λ = 1 Λ R λ

[0113] The estimator has the following properties: Within the usable estimation range, it is unbiased as a spectral noise power density even for low ES / N 0 – i.e., energy per symbol as a measure of the signal-to-noise ratio. For Λ = L / 2 and ES / N 0 > 0 dB, the estimator lies close to the Cramer-Rao bound and for a frequency offset of zero, it lies exactly on the Cramer-Rao bound. It can be described as optimal. The estimation range is limited to: νT ≤ 0 , 85 Λ + 1 .

[0114] One disadvantage is that sufficiently accurate time synchronization is required prior to application. Another disadvantage is that the method is not suitable for large frequency deviations due to the limited estimation range.

[0115] The phase estimation can be improved as described below.

[0116] For data-based phase shift estimation in coherent receivers, the synchronization symbols present in a transmitted packet are typically used. This involves using the pilot symbols of the pilot sequences. The phases of the received symbols are compared with the expected phases (reference symbols). The difference between the two results in the phase shift. For noise suppression, for example, averaging over several symbols is performed.

[0117] An example of such a phase estimation is the maximum likelihood estimator (see [6]).

[0118] This estimator produces useful results if the exact time of transmission is known. However, this is usually determined in a prior estimate, and deviations can occur due to noise and other effects. Furthermore, due to limitations in computing power, it is not always possible to make a sufficiently accurate time estimate.

[0119] Thus, the received symbols contain a residual time offset that is incorporated into the phase estimation and degrades it. This residual time offset, or more generally, time offset, results in a phase error that is added to the phase of interest between the transmitter's transmission frequency and the receiver's reception frequency, thus impairing the evaluation.

[0120] The following describes embodiments that result in improved phase estimation. These embodiments also apply to improved frequency estimation, in the case where the frequency estimation is based on phase differences between neighboring symbols.

[0121] During symbol recovery (e.g., through matched filtering followed by subsampling), a time offset also introduces a portion of at least one neighboring symbol into the result. This phenomenon is described below using MSK modulation. The mapping serves, for example, to map the pilot sequence bits to the actual symbols of the pilot sequences.

[0122] To simplify the presentation, the following graphics contain only three symbols each. The middle symbol represents the symbol under investigation, and the other two represent the preceding and following symbols.

[0123] Fig. 2shows an MSK baseband signal with three symbols. Here, both the symbol of interest and the preceding and following symbols have been mapped with positive amplitude. This represents a central symbol and two flanking symbols. Curve a shows the real part, and curves b show the imaginary part of the signal. The amplitude is plotted against time on the y-axis, with sampling times on the x-axis.

[0124] The following section examines the influence of phase on the middle symbol, which is transmitted on the real part. The two other flanking symbols (the preceding and following symbols) are transmitted on the imaginary part according to the MSK specification as an example of the mapping used to generate the transmitted symbols.

[0125] Symbol recovery is performed in the receiver using a matched filter, which is multiplied by the signal and then summed over the entire symbol. Fig. 3 shows the length of the matched filter (stronger curve c). The filter extends over the entire middle symbol. In Fig. 3 This represents the ideal case in which no time offset occurs.

[0126] Due to the special MSK modulation, a so-called ISI (intersymbol interference, see [7]) is introduced during matched filtering to reconstruct the symbols. The symbol crosstalk is known and can be taken into account when calculating the phase differences.

[0127] For symbol recovery, the data within the displayed range is extracted and multiplied by the matched filter (optimal filter) or an approximation. The symbol is obtained by integration (in the continuous-time case) or summation (in the discrete-time case).

[0128] If the reconstructed symbol is Fig. 3 compared with the transmitted symbol (or a reference symbol available to the receiver), it can be seen that there is no phase shift between the two symbols.

[0129] However, if a time offset occurs, the window for the matched filter is shifted. Fig. 4 This is shown for a negative time offset, where the correct time is later than the assumed time. The symbol start is thus assumed to be too early. For phase determination, the time offset results in a phase error as an additional error in the actual phase determination.

[0130] Due to the time offset, more energy from the preceding flanking symbol and less from the central symbol of interest is incorporated into the reconstruction of the central symbol. The value thus deviates from the expected target value. However, if the received value deviates from the expected value, this deviation results in an error in the phase estimation.

[0131] This also applies to a positive time offset, where the actual symbol begins before the assumed time. However, this increases the influence of the flanking symbol following the middle symbol.

[0132] A progression of the results of the symbol reconstruction at different time offsets is shown Fig. 5 The center represents the ideal time. The amplitude is plotted on the y-axis against the time offset (or alternatively, the timing error) in symbol lengths. The real part R and the imaginary part I are plotted.

[0133] Fig. 5 shows that at the ideal time the influence of the neighboring symbols (i.e. the flanking symbols in the example of Fig. 2 ) is smallest. With increasing time offset, the influence of neighboring symbols increases, ie, in this example with MSK modulation, the amplitude of the imaginary part increases.

[0134] If the received symbol (here the middle symbol) is multiplied after the matched filtering with the conjugate complex of the transmitted symbol (i.e. the reference symbol or the corresponding symbol of the reference sequence), the phase offset between the reference symbol and the considered middle symbol of the combination of three symbols of the pilot sequence of the received signal is obtained.

[0135] The phase offset is shown Fig. 6 for the Fig. 5corresponding symbol over the time offset. In this symbol constellation, the phase offset—that is, the resulting phase error due to the time offset—is always positive. However, the magnitude depends on the magnitude of the time offset.

[0136] The previous explanations show that the result of the matched filtering and thus also the phase error in the case of time offsets depends on the previous and following symbols - i.e. the symbols flanking the middle symbol considered in the example.

[0137] To fully investigate this, the result of the matched filtering with a time offset was recalculated for a different symbol constellation. The symbol under consideration was also mapped with a positive amplitude, but the two flanking symbols have different amplitudes. The complex baseband signal and the result of the matched filtering with different time offsets are shown in the Fig. 7 and the Fig. 8 shown.

[0138] The Fig. 7 shows the amplitude of the three symbols b, a, b versus time in sampling rates. The two flanking symbols b encompass the middle symbol a. For this MSK signal in the selected constellation, at the ideal sampling time, no symbol crosstalk (ISI) occurs between the flanking symbols and the middle symbol.

[0139] The Fig. 8 shows the amplitude of the middle symbol after matched filtering as a function of the time offset.

[0140] If the reception symbol is processed after matched filtering in the same way as for the Fig. 6 multiplied by the complex conjugate value of the transmitted symbol, one obtains the phase offset between the reference symbol and the received symbol under consideration.

[0141] The phase offset is shown Fig. 9 for the Fig. 8 considered mean symbol over the time offset. Compared to Fig. 6The phase offset curve is no longer positive for all possible time offsets. For negative time offsets, the phase offset is positive, and for positive time offsets, it changes its sign and becomes negative.

[0142] Overall, the phase offset depends on the symbols that flank the symbol being evaluated.

[0143] Therefore, the symbols are selected in such a way that the phase errors resulting from a time offset when evaluating the individual symbols compensate each other when averaging over the affected symbols - whether from one signal or over several signals.

[0144] The Fig. 10 shows a system 50 with a transmitter 1 and a receiver 10.

[0145] The transmitter 1 has a signal generator 2 and a signal output device 3. The signal generator 2 generates the signals to be transmitted based on data that the transmitter 1 has to transmit. The data is, for example, sensor data or status data of the transmitter 1 itself.

[0146] In the illustrated embodiment, telegram splitting is used so that the data to be output is distributed across at least two signals, which can also be referred to as telegram fragments. The individual telegram fragments each have a pilot sequence provided by the signal generator 2. In one embodiment, the provision for the individual signals to be transmitted consists in the signal generator 2 accessing a data memory 4 in which the data for the appropriate pilot sequence is stored. The signal output device 3 transmits the individual telegram fragments.

[0147] The individual telegram fragments are received and processed by receiver 10. Synchronization between transmitter 1 and receiver 10 is required for processing, for which purpose synchronization device 11 is provided. The pilot sequences with their pilot symbols are used for synchronization.

[0148] Synchronization is required with respect to time. This refers to the estimation of the temporal position of the received signal and the associated optimal time for sampling the signal. For this synchronization, the synchronization device 11 is connected to a sampling device 13. The sampling times cannot generally be influenced. Rather, the time is necessary to reconstruct the symbols from the usually oversampled signal or, if necessary, to perform interpolation.

[0149] In the example shown, further synchronization relates to frequency. Relevant factors here are the carrier frequency at which the signals are transmitted by transmitter 1 and the center frequency of filter device 14, which is used to filter the received signals. Alternatively, the signals are frequency-shifted before the filter after reception. The difference between the two frequencies is referred to as the frequency offset. For frequency synchronization, filter device 14 is connected to synchronization device 11.

[0150] In the case of coherent demodulation or decoding, frequency synchronization is followed by phase synchronization. For this synchronization, the signal evaluation device 12, which transmits the corresponding synchronization data to the synchronization device 11, is described in more detail below.

[0151] To determine the phase, each pilot symbol of the pilot sequence of the received signal is compared with the corresponding reference symbol of a reference sequence. The reference sequence is identical to the pilot sequence used by transmitter 1 to generate the signal or inserted into the signal. Once a phase value has been determined for each pilot symbol, a total value is subsequently generated, e.g., by averaging the individual phase values. The reference sequences are stored in a data memory 15.

[0152] However, as previously explained, a timing error between the window of the filter device 14 and the actual start of the symbol has a detrimental effect on the determination of the phase.

[0153] By averaging the evaluation of multiple symbols for noise suppression, the errors in the phase value are also averaged. Therefore, the average phase error for all pilot symbols is used for synchronization. To ensure that the average error when averaging the individual phase values ​​ideally results in zero or at least becomes very small, the dependence of the phase error of a symbol on the preceding and following symbols is taken into account, based on the above investigations. The symbols are therefore selected in such a way that the phase errors just compensate for each other due to time offsets when averaging the individual phase values ​​generated for each symbol.

[0154] Since several pilot symbols are transmitted one after the other, the previous and the following symbol (i.e. the symbols used in the example of the Fig. 2directly flanking a middle symbol or symbol considered for evaluation), or these are known to the receiver. To compensate for the overall phase error caused by the individual pilot symbols, for each symbol in which a phase error can occur, a second symbol is transmitted that has the opposite error (with the same time offset). This applies here to the symbols of a signal carrying a pilot sequence.

[0155] In an alternative embodiment, the phase errors are averaged over the partial pilot sequences of several signals, which are in particular telegram fragments.

[0156] In one embodiment, it is therefore provided that the length of the pilot sequence is a multiple of the length of two, so that an even number of additions is created for the averaging.

[0157] The effect will be explained using an example.

[0158] Is the sequence from Fig. 2a part of the pilot sequence, a positive phase offset is generated in the event of a time offset, which applies to both positive and negative time offsets. To correct this offset, a symbol must also be transmitted in which the error at the same level indicates a negative phase error.

[0159] The Fig. 11 shows the mapping rule for MSK modulation. It shows the possible constellation points of MSK with precoding (also known as non-diff MSK in MATLAB).

[0160] The symbols to be transmitted are divided into groups of four symbols, with the first symbol being transmitted at time T0. Accordingly, at time T0, the constellation point +1 + 0j is chosen for a binary zero, and the constellation point -1 + 0j is chosen for a binary one. For the following symbol, time T0 + ΔT is chosen. The constellation points are thus 0 + 1j (binary one) and 0 - 1j (binary zero). The constellation points are calculated in the same way for the two following times. After four symbols have been mapped to the constellation points, the process starts again at time T0.

[0161] For a pilot sequence length of eight symbols and the MSK mapping rule described above, the following sequences result, which exhibit both good autocorrelation and a phase error close to zero due to a time offset. Consecutive number Sequence of symbols note 0 [0, 1, 0, 0, 0, 0, 1, 0] 1 [1, 0, 1, 1, 1, 1, 0, 1] Inverted or mirrored sequence 0. 2 [0, 0, 0, 1, 0, 1, 1, 1] 3 [1, 1, 1, 0, 1, 0, 0, 0] Inverted or mirrored sequence 2.

[0162] Therefore, the pilot sequence with which transmitter 1 transmits the signals is defined in such a way that the sum of the phase offsets is minimized in the case of a time shift.

[0163] For a pilot sequence length of twelve symbols and the previously described MSK mapping rule, the following sequences were found, which exhibit both good autocorrelation and a phase error close to zero under time offset. Consecutive number Sequence of symbols in bits | ε v | max 0 [0, 0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 1] 0.0132892511307 1 [0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1] 0.00604862976372 2 [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 1] 0.00669831471134 3 [0, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1] 0.00214046117234 4 [0, 0, 0, 1, 1, 0, 1, 0, 1, 1, 1, 1] 0.00228803319793 5 [0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0] 0.00214191618191 6 [0, 1, 0, 0, 1, 1, 1, 1, 1, 0, 1, 0] 0.00228803319793 7 [0, 1, 0, 1, 0, 0, 1, 1, 1, 1, 1, 0] 0.0132920410548 8 [0, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0] 0.00228488288132 9 [0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0] 0.00214482669223 10 [0, 1, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0] 0.00604466832967 11 [1, 0, 0, 0, 0, 1, 0, 1, 1, 1, 0, 1] 0.00604862889634 12 [1, 0, 1, 0, 0, 0, 0, 0, 1, 1, 0, 1] 0.00229275981363 13 [1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1] 0.0089182616364 14 [1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 0, 1] 0.0132892511307 15 [1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1] 0.00229275939358 16 [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1] 0.0060426888597 17 [1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 0, 1] 0.00670003795849 18 [1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 0] 0.00604466832967 19 [1, 1, 0, 1, 1, 1, 1, 0, 1, 0, 0, 0] 0.00214046162904 20 [1, 1, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0] 0.00228645779938 21 [1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0] 0.00214191618191 22 [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0] 0.00229295513208 23 [1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0] 0.00669917633491 24 [1, 1, 1, 0, 1, 1, 1, 1, 0, 1, 0, 0] 0.0060426888597 25 [1, 1, 1, 1, 1, 0, 0, 1, 0, 1, 0, 0] 0.0132808813584 26 [1, 1, 1, 1, 1, 0, 1, 0, 1, 1, 0, 0] 0.00855923745514

[0164] The zeros and ones are the pilot sequence bits, which, through appropriate mapping, result in the symbols of the pilot sequence or the partial pilot sequence.

[0165] The value | ε v | max is the maximum systematic frequency estimate deviation using a correlation, taking into account a sampling time error of -0.25T to +0.25T and a frequency deviation of vT of the received signal between -0.05 to +0.05.

[0166] The Fig. 12 shows a second degree polynomial of the form y(x) = y 0 - c(xx 0 ) 2< v, as described in embodiments for improving time and / or frequency estimation by means of interpolation.

[0167] In the following, further embodiments of the present invention are described, which can be used in combination with the embodiments described above or on their own.

[0168] According to a first aspect, a transmitter 1 is designed to transmit at least one signal with a pilot sequence comprising a plurality of pilot symbols, wherein the transmitter 1 has a signal generator 2, wherein the signal generator 2 provides the pilot sequence, wherein the signal generator 2 provides the pilot sequence in such a way that the pilot sequence has at least two symbol groups, each with at least two pilot symbols, and wherein the symbol groups, when evaluating a signal received by a receiver 10 as a result of the transmission of the signal, generate phase errors with regard to a phase that are dependent on a time offset between a reference time of the transmission of the signal by the transmitter 1 and a value assumed and / or estimated for the evaluation for the reference time, which phase errors essentially compensate for one another during the evaluation across the symbol groups.

[0169] According to a second aspect with reference to the first aspect, the symbol groups partially overlap.

[0170] According to a third aspect with reference to at least one of the first to second aspects, at least one symbol group consists of a central symbol and two flanking symbols.

[0171] According to a fourth aspect with reference to at least one of the first to third aspects, the transmitter 1 is designed to transmit signals each having a pilot sequence comprising at least four pilot symbols.

[0172] According to a fifth aspect with reference to at least one of the first to fourth aspects, the transmitter 1 transmits signals having pilot sequences resulting from MSK modulation or GMSK modulation.

[0173] According to a sixth aspect with reference to at least one of the first to fifth aspects, the transmitter 1 outputs data to be output by at least two telegram fragments which the transmitter 1 transmits as individual signals and which are shorter than a single telegram comprising the data as a whole, and at least one telegram fragment comprises the pilot sequence generated by the signal generator 2.

[0174] According to a seventh aspect, referring to at least one of the first to sixth aspects, the signal generator 2 provides the pilot sequence with a length of at least eight pilot symbols such that the pilot sequence has one of the following forms: [0, 1, 0, 0, 0, 1, 0], [1, 0, 1, 1, 1, 1, 0, 1], [0, 0, 1, 0, 1, 1, 1, 1] or [1, 1, 1, 0, 1, 0, 0, 0], where the zeros and ones are pilot sequence bits before modulation.

[0175] According to an eighth aspect, with reference to at least one of the first to sixth aspects, the signal generator 2 provides the pilot sequence with a length of at least eight pilot symbols such that a portion of the pilot sequence which is a single or multiple part of the pilot sequence has one of the following forms: [0, 1, 0, 0, 0, 1, 0], [1, 0, 1, 1, 1, 1, 0, 1], [0, 0, 1, 0, 1, 1, 1, 1] or [1, 1, 1, 0, 1, 0, 0, 0], where the zeros and ones are pilot sequence bits before modulation.

[0176] According to a ninth aspect with reference to at least one of the first to sixth aspects, the signal generator 2 provides the pilot sequence with a length of at least twelve pilot symbols such that the pilot sequence has one of the following forms: [0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 1], [0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 1], [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1], [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 1], [0, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1], [0, 0, 0, 1, 1, 0, 1, 1, 1, 1], [0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0], [0, 1, 0, 1, 0, 0, 1, 1, 1, 1, 0], [0, 1, 0, 1, 1, 1, 1, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0], [1, 0, 0, 0, 1, 0, 1, 1, 0, 1], [1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 1], [1, 0, 1, 0, 0, 1, 0, 0, 0, 1], [1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 1], [1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 1], [1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0], [1, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0] or [1, 1, 1,1, 1, 0, 1, 0, 1, 1, 0, 0], where the zeros and ones are pilot sequence bits before modulation.

[0177] According to a tenth aspect with reference to at least one of the first to sixth aspects, the signal generator 2 provides the pilot sequence with a length of at least twelve pilot symbols such that a portion of the pilot sequence which is a single or multiple part of the pilot sequence has one of the following forms: [0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 1], [0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 1], [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1], [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 1], [0, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1], [0, 0, 0, 1, 1, 0, 1, 1, 1, 1], [0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0], [0, 1, 0, 1, 0, 0, 1, 1, 1, 1, 0], [0, 1, 0, 1, 1, 1, 1, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0], [1, 0, 0, 0, 1, 0, 1, 1, 0, 1], [1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 1], [1, 0, 1, 0, 0, 1, 0, 0, 0, 1], [1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 1], [1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 1], [1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0], [1, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0] or [1, 1, 1,1, 1, 0, 1, 0, 1, 1, 0, 0], where the zeros and ones are pilot sequence bits before modulation.

[0178] According to an eleventh aspect, a transmitter 1 is designed to transmit at least two signals, each having a partial pilot sequence comprising a plurality of pilot symbols, wherein the transmitter 1 has a signal generator 2, wherein the signal generator 2 provides the partial pilot sequence, wherein the signal generator 2 provides the partial pilot sequences of the signals in such a way that the partial pilot sequences each have at least one symbol group with at least two pilot symbols, and wherein the symbol groups of the signals, when evaluating signals received by a receiver 10 as a result of the transmission of the signals, generate phase errors with respect to a phase dependent on a time offset between a reference time of the respective transmission of a signal and a value assumed and / or estimated for the evaluation for the reference time, which phase errors essentially compensate for one another during the evaluation together across the signals.

[0179] According to a twelfth aspect with reference to the eleventh aspect, at least one partial pilot sequence comprises at least two symbol groups and the symbol groups partially overlap.

[0180] According to a thirteenth aspect with reference to at least one of the eleventh to twelfth aspects, at least one symbol group consists of a central symbol and two flanking symbols.

[0181] According to a fourteenth aspect with reference to at least one of the eleventh to thirteenth aspects, the transmitter 1 transmits signals having partial pilot sequences resulting from MSK modulation or GMSK modulation.

[0182] According to a fifteenth aspect, referring to at least one of the eleventh to fourteenth aspects, the signal generator 2 provides the partial pilot sequences such that the partial pilot sequences taken together comprise a portion which is present singly or multiple times and which has the following forms: [0, 1, 0, 0, 0, 1, 0], [1, 0, 1, 1, 1, 1, 0, 1], [0, 0, 1, 0, 1, 1, 1, 1] or [1, 1, 1, 0, 1, 0, 0, 0], where the zeros and ones are pilot sequence bits before modulation.

[0183] According to a sixteenth aspect, referring to at least one of the eleventh to fourteenth aspects, the signal generator 2 provides the partial pilot sequences such that the partial pilot sequences taken together comprise a section which is present singly or multiple times and which has the following forms: [0, 0, 0, 0, 1, 1, 0, 1, 0, 1, 1], [0, 0, 0, 1, 0, 0, 0, 1, 0, 1, 1], [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1], [0, 0, 0, 1, 0, 1, 0, 0, 1, 1, 1, 1], [0, 0, 0, 1, 0, 1, 1, 1, 1, 0, 1, 1], [0, 0, 0, 1, 1, 0, 1, 1, 1, 1], [0, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1, 0], [0, 1, 0, 1, 0, 0, 1, 1, 1, 1, 0], [0, 1, 0, 1, 1, 1, 1, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0], [0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0], [1, 0, 0, 0, 1, 0, 1, 1, 0, 1], [1, 0, 1, 0, 0, 0, 0, 1, 1, 0, 1], [1, 0, 1, 0, 0, 1, 0, 0, 0, 1], [1, 0, 1, 0, 1, 1, 0, 0, 0, 0, 1], [1, 0, 1, 1, 0, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 1], [1, 0, 1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 1], [1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 0, 1, 0, 1, 0, 0, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 0, 1, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0], [1, 1, 1, 0, 1, 0, 1, 1, 0, 0, 0, 0], [1, 1, 1, 0, 1, 1, 0, 1, 0, 0, 1, 0, 0] or [1, 1, 1,1, 1, 0, 1, 0, 1, 1, 0, 0], where the zeros and ones are pilot sequence bits before modulation.

[0184] A seventeenth aspect relates to a method for transmitting signals, wherein the signals are transmitted with a pilot sequence each having a plurality of pilot sequence symbols, wherein the pilot sequences are provided in such a way that the pilot sequences each have at least two symbol groups each having at least two pilot symbols, and wherein the symbol groups, when evaluating a signal received by a receiver 10 as a result of the transmission of the signal with regard to a phase, generate phase errors which are dependent on a time offset between a reference time of transmission of the signal and a value assumed and / or estimated for the evaluation for the reference time, which phase errors essentially compensate for one another during the evaluation across the symbol groups.

[0185] An eighteenth aspect relates to a method for transmitting signals, wherein at least two signals are transmitted, each having a partial pilot sequence comprising a plurality of pilot sequence symbols, wherein the partial pilot sequences are provided in such a way that the partial pilot sequences each have at least one symbol group each having at least two pilot symbols, and wherein the symbol groups of the signals, when evaluating signals received by a receiver 10 as a result of the transmission of the signals, generate phase errors with respect to a phase dependent on a time offset between a reference time of the respective transmission of a signal and a value assumed and / or estimated for the evaluation for the reference time, which phase errors essentially compensate for one another during the evaluation together across the signals.

[0186] A nineteenth aspect relates to a computer program having program code for carrying out the method according to the seventeenth or eighteenth aspect.

[0187] A twentieth aspect relates to a receiver 10, wherein the receiver 10 is designed to receive at least one signal transmitted by a transmitter 1, wherein the receiver 10 has a synchronization device 11, wherein the synchronization device 11 is designed to synchronize the receiver 10 with the transmitter 1 based on the received signal, wherein the receiver 10 has a signal evaluation device 12, wherein the signal evaluation device 12 determines data for the synchronization from the received signal and transfers it to the synchronization device 11 for the synchronization,wherein the signal evaluation device 12, when evaluating a pilot sequence or a partial pilot sequence of a received signal, determines a time offset between a reference time of transmission of the signal by the transmitter 1 and a value assumed and / or estimated for the evaluation for the reference time, and wherein the signal evaluation device 12 accesses symbols of a reference sequence adapted to the time offset for the further evaluation of the pilot sequence, or wherein the signal evaluation device 12 accesses symbols of a part of a reference sequence adapted to the time offset or of a part of a reference sequence adapted to the time offset for the further evaluation of the partial pilot sequence.

[0188] According to a twenty-first aspect with reference to the twentieth aspect, the signal evaluation device 12 carries out the adaptation of the reference sequence or the part of the reference sequence in that the signal evaluation device 12 shifts a known reference sequence or a part of a known reference sequence in time by the determined time offset.

[0189] According to a twenty-second aspect with reference to the twentieth aspect, the signal evaluation device 12 carries out the adaptation of the reference sequence or the part of the reference sequence by the signal evaluation device 12 accessing a data memory 15 with correspondingly stored reference sequences or parts of reference sequences.

[0190] According to a twenty-third aspect with reference to at least one of the twentieth to twenty-second aspects, the receiver 10 comprises a filter device 14 and a sampling device 13.

[0191] A twenty-fourth aspect relates to a method for synchronizing a receiver 10 with a transmitter 1, wherein at least one signal transmitted by the transmitter 1 is received by the receiver 10, wherein a synchronization of the receiver 10 with the transmitter 1 is carried out starting from a pilot sequence or a partial pilot sequence of the received signal, wherein a time offset between a reference time of a transmission of the signal by the transmitter 1 and an assumed and / or estimated value for the reference time is determined, and wherein a pilot sequence of the received signal is evaluated with symbols of a reference sequence adapted to the time offset, or wherein a partial pilot sequence of the received signal is evaluated with symbols of a part of a reference sequence adapted to the time offset or with symbols of a part of a reference sequence adapted to the time offset.

[0192] A twenty-fifth aspect relates to a receiver 10, wherein the receiver 10 is configured to receive at least one signal transmitted by a transmitter 1, wherein the receiver 10 has a synchronization device 11, wherein the synchronization device 11 is configured to synchronize the receiver 10 with the transmitter 1 based on the received signal, wherein the receiver 10 has a signal evaluation device 12, wherein the signal evaluation device 12 determines data for the synchronization from the received signal and transfers it to the synchronization device 11 for synchronization, wherein the receiver 10 is configured to receive a plurality of telegram fragments transmitted by the transmitter 1, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram containing data transmitted by the transmitter 1,wherein the telegram fragments are shorter than a single telegram containing the transmitted data, wherein the partial pilot sequences complement each other to form a pilot sequence, and wherein the signal evaluation device 12, based on the partial pilot sequences and using the DFT method, determines a value for a frequency difference between a transmission frequency of the transmitter 1 and a reception frequency of the receiver 10 by the signal evaluation device 12 determining values ​​of decision variables di [k] for each partial pilot sequence, combining the determined values ​​of the decision variables di [k] across the partial pilot sequences, and using the combined values ​​to determine a maximum value.

[0193] According to a twenty-sixth aspect with reference to the twenty-fifth aspect, the signal evaluation device 12 determines the values ​​of the decision variables di [k] for each partial pilot sequence by the signal evaluation device 12 forming, for a plurality of frequencies of a complex exponential oscillation, referred to as frequency hypotheses, multiplication values ​​of a multiplication of sample values ​​of the respective partial pilot sequence with conjugate complex values ​​of an equal number of reference symbols of a known reference sequence and with sample values ​​of the complex exponential oscillation and summing the multiplication values.

[0194] According to a twenty-seventh aspect with reference to at least one of the twenty-fifth to twenty-sixth aspects, the signal evaluation device 12 combines the values ​​of the decision variables di [k] determined for the partial pilot sequences together for the partial pilot sequences by the signal evaluation device 12 adding the absolute values ​​of the decision variables di [k].

[0195] According to a twenty-eighth aspect with reference to at least one of the twenty-fifth to twenty-sixth aspects, the signal evaluation device 12 combines the values ​​of the decision variables di [k] determined for the partial pilot sequences together for the partial pilot sequences by the signal evaluation device 12 forming a sum of the absolute values ​​of a real part and an imaginary part of the values ​​of the decision variables di [k].

[0196] According to a twenty-ninth aspect, with reference to at least one of the twenty-fifth to twenty-sixth aspects, the signal evaluation device 12 combines the values ​​of the decision variables di [k] determined for the partial pilot sequences together for the partial pilot sequences by the signal evaluation device 12 summing the absolute value squares of the values ​​of the decision variables di [k]. According to a thirtieth aspect, with reference to at least one of the twenty-fifth to twenty-sixth aspects, the signal evaluation device 12 combines the values ​​of the decision variables di [k] determined for the partial pilot sequences together for the partial pilot sequences coherently, taking into account the absolute value and phase of the values ​​of the decision variables di [k].

[0197] According to a thirty-first aspect with reference to at least one of the twenty-fifth to thirtieth aspects, the signal evaluation device 12 combines the determined values ​​of the decision variables di [k] with one another, taking into account weighting factors.

[0198] According to a thirty-second aspect with reference to at least one of the twenty-fifth to thirtieth aspects, the signal evaluation device 12 combines the determined values ​​of the decision variables di [k] with one another, taking into account weighting factors relating to the partial pilot sequences.

[0199] According to a thirty-third aspect with reference to at least one of the thirty-first to thirty-second aspects, the signal evaluation device 12 determines the weighting factors based on a signal-to-noise ratio of the respective telegram fragment.

[0200] According to a thirty-fourth aspect with reference to the thirty-third aspect, the signal evaluation device 12 determines the weighting factors as proportional to a square root of the signal-to-noise ratio.

[0201] According to a thirty-fifth aspect with reference to the thirty-third aspect, the signal evaluation device 12 determines the weighting factors as proportional to a root of a quotient of the signal-to-noise ratio and a noise power of the respective telegram fragment.

[0202] According to a thirty-sixth aspect with reference to at least the thirty-third aspect, the signal evaluation device 12 determines the weighting factors as a function of a computing power available to the signal evaluation device or as a function of a ratio of the signal-to-noise ratio relative to a predeterminable relational value or as a function of an interference power either as proportional to a root of the signal-to-noise ratio or as proportional to a root of a quotient of the signal-to-noise ratio and a noise power of the respective telegram fragment.

[0203] According to a thirty-seventh aspect, with reference to at least one of the twentieth to twenty-third aspects or the twenty-fifth to thirty-sixth aspects, the receiver 10 is configured to receive a plurality of telegram fragments transmitted by the transmitter 1, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram containing data transmitted by the transmitter 1, wherein the telegram fragments are shorter than a single telegram containing the transmitted data, wherein the partial pilot sequences complement each other to form a pilot sequence, and wherein the signal evaluation device 12, based on the partial pilot sequences and using the DFT method, determines a value for a frequency difference between a transmission frequency of the transmitter 1 and a reception frequency of the receiver 10,in that the signal evaluation device 12 determines values ​​of decision variables di [k] for each partial pilot sequence, combines the determined values ​​of the decision variables di [k] across the partial pilot sequences, and uses the combined values ​​to determine a maximum value, wherein the signal evaluation device 12 carries out a time estimate for a time offset between a reference time of transmission of the telegram fragment and a value assumed and / or estimated for the evaluation of the telegram fragment for the reference time, in that the signal evaluation device 12 determines a time estimate for a time offset based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

[0204] According to a thirty-eighth aspect with reference to at least one of the twentieth to twenty-third aspects or the twenty-fifth to thirty-sixth aspects, the receiver 10 is configured to receive at least one telegram transmitted by the transmitter 1, wherein the telegram has a pilot sequence, wherein the signal evaluation device 12, based on the pilot sequence and using the DFT method, determines a value for a frequency difference between a transmission frequency of the transmitter 1 and a reception frequency of the receiver 10, by the signal evaluation device 12 determining values ​​of decision variables di [k] for the pilot sequence, determining a maximum value,wherein the signal evaluation device 12 carries out a time estimate for a time offset between a reference time of transmission of the telegram and a value assumed and / or estimated for the evaluation of the telegram for the reference time, in that the signal evaluation device 12 determines a time estimate for a time offset based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

[0205] According to a thirty-ninth aspect with reference to at least one of the thirty-seventh to thirty-eighth aspects, the signal evaluation device 12 obtains the maximum value based on the determination of the maximum value and a positive comparison of the maximum value with a decision threshold.

[0206] According to a fortieth aspect with reference to at least one of the thirty-seventh to thirty-ninth aspects, the signal evaluation device 12 generates the time estimate from the maximum value and two adjacent values, and wherein the two adjacent values ​​precede or follow the determined maximum value with respect to the time variable.

[0207] According to a forty-first aspect with reference to the fortieth aspect, the signal evaluation device 12 determines a polynomial for the maximum value and the two adjacent values, and the signal evaluation device 12 determines the time estimate from an extreme value associated with the polynomial.

[0208] According to a forty-second aspect with reference to the forty-first aspect, the signal evaluation device 12 performs the interpolation with a second-degree polynomial.

[0209] According to a forty-third aspect with reference to the forty-second aspect, the signal evaluation device 12 performs the interpolation with a polynomial of the following form: y(x) = y 0 - c(xx 0 ) 2< , and wherein the free parameters y 0 , c and x 0 are determined from the maximum and the neighboring points.

[0210] According to a forty-fourth aspect, with reference to at least one of the twentieth to twenty-third aspects or the twenty-fifth to forty-third aspects, the receiver 10 is configured to receive a plurality of telegram fragments transmitted by the transmitter 1, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram containing data transmitted by the transmitter 1, wherein the telegram fragments are shorter than a single telegram containing the transmitted data, wherein the partial pilot sequences complement each other to form a pilot sequence, and wherein the signal evaluation device 12, based on the partial pilot sequences and using the DFT method, determines a value for a frequency difference between a transmission frequency of the transmitter 1 and a reception frequency of the receiver 10,in that the signal evaluation device 12 determines values ​​of decision variables di [k] for each partial pilot sequence, combines the determined values ​​of the decision variables di [k] across the partial pilot sequences, and uses the combined values ​​to determine a maximum value, wherein the signal evaluation device 12 performs a frequency estimation for the frequency difference by determining a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a frequency variable.

[0211] According to a forty-fifth aspect, with reference to at least one of the twentieth to twenty-third aspects or the twenty-fifth to forty-third aspects, the receiver 10 is configured to receive at least one telegram transmitted by the transmitter 1, wherein the telegram has a pilot sequence, wherein the signal evaluation device 12, based on the pilot sequence and using the DFT method, determines a value for a frequency difference between a transmission frequency of the transmitter 1 and a reception frequency of the receiver 10, by the signal evaluation device 12 determining values ​​of decision variables di [k] for the pilot sequence, determining a maximum value, wherein the signal evaluation device 12 performs a frequency estimation for the frequency difference,in that the signal evaluation device 12 determines a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a frequency variable.

[0212] According to a forty-sixth aspect with reference to at least one of the forty-fourth to forty-fifth aspects, the signal evaluation device 12 obtains the maximum value based on the maximum value and a positive comparison of the maximum value with a decision threshold.

[0213] According to a forty-seventh aspect with reference to at least one of the forty-fourth to forty-sixth aspects, the signal evaluation device 12 generates the frequency estimate from the maximum value and two adjacent values, wherein the two adjacent values ​​precede or follow the determined maximum value with respect to the frequency variable.

[0214] According to a forty-eighth aspect with reference to the forty-seventh aspect, the signal evaluation device 12 determines a polynomial for the maximum value and the two adjacent values, and the signal evaluation device 12 determines the frequency estimate from an extreme value associated with the polynomial.

[0215] According to a forty-ninth aspect with reference to the forty-eighth aspect, the signal evaluation device 12 performs the interpolation with a second-degree polynomial.

[0216] According to a fiftieth aspect with reference to the forty-ninth aspect, the signal evaluation device 12 performs the interpolation with a polynomial of the following form: y(x) = y 0 - c(xx 0 ) 2< , and wherein the free parameters y 0 , c and x 0 are determined from the maximum and the neighboring points.

[0217] According to a fifty-first aspect with reference to one of the thirty-seventh to fiftieth aspects, the signal evaluation device 12 determines the time estimate and uses it for a renewed determination of the frequency difference by using, in a renewed determination of the values ​​of the decision variables di [k], samples of the respective partial pilot sequences shifted by the time estimate or by shifting reference symbols of a reference sequence in time by the time estimate.

[0218] According to a fifty-second aspect with reference to the fifty-first aspect, the signal evaluation device 12 interpolates the sample values ​​of the respective partial pilot sequences for the renewed determination of the frequency difference.

[0219] According to a fifty-third aspect with reference to at least one of the fifty-first to fifty-second aspects, the signal evaluation device 12, after the renewed determination of the frequency difference, carries out a frequency estimation for the frequency difference in that the signal evaluation device 12 determines a frequency estimate based on a maximum value obtained by determining the maximum value and a positive comparison with the decision threshold and at least one value adjacent to a frequency variable.

[0220] A fifty-fourth aspect relates to a method for synchronizing a receiver 10 with a transmitter 1, wherein at least one signal transmitted by the transmitter 1 is received by the receiver 10, wherein, starting from a pilot sequence or a partial pilot sequence of the received signal, a synchronization of the receiver 10 with the transmitter 1 is carried out, wherein the receiver 10 receives a plurality of telegram fragments transmitted by the transmitter 1, wherein the telegram fragments each have a partial pilot sequence, wherein the telegram fragments complement each other to form a telegram containing data transmitted by the transmitter 1, wherein the telegram fragments are shorter than the telegram, wherein the partial pilot sequences complement each other to form a pilot sequence, wherein, starting from the partial pilot sequences, the receiver 10 is synchronized with the transmitter 1,Starting from the partial pilot sequences and using the DFT method, a value for a frequency difference between a transmission frequency of the transmitter 1 and a reception frequency of the receiver 10 is determined by determining values ​​of decision variables di [k] for each partial pilot sequence, by combining the determined values ​​of the decision variables di [k] across all partial pilot sequences, and by determining a maximum value using the combined values.

[0221] A fifty-fifth aspect relates to a computer program having a program code for carrying out the method according to one of the twenty-fourth or fifty-fourth aspects.

[0222] Although some aspects have been described in the context of a device, it should be understood that these aspects also represent a description of the corresponding method, so that a block or component of a device can also be understood as a corresponding method step or as a feature of a method step. Analogously, aspects described in the context of or as a method step also represent a description of a corresponding block, detail, or feature of a corresponding device. Some or all of the method steps may be performed by (or using) a hardware apparatus, such as a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, some or more of the key method steps may be performed by such an apparatus.

[0223] Depending on specific implementation requirements, embodiments of the invention may be implemented in hardware or in software, or at least partially in hardware or at least partially in software. The implementation may be carried out using a digital storage medium, for example a floppy disk, a DVD, a Blu-ray disc, a CD, a ROM, a PROM, an EPROM, an EEPROM, or a FLASH memory, a hard disk, or other magnetic or optical storage device on which electronically readable control signals are stored that can interact or interact with a programmable computer system such that the respective method is carried out. Therefore, the digital storage medium may be computer-readable.

[0224] Some embodiments according to the invention thus comprise a data carrier having electronically readable control signals capable of interacting with a programmable computer system such that one of the methods described herein is carried out.

[0225] In general, embodiments of the present invention may be implemented as a computer program product having a program code, wherein the program code is effective to perform one of the methods when the computer program product is run on a computer.

[0226] The program code can, for example, also be stored on a machine-readable medium.

[0227] Other embodiments include the computer program for performing one of the methods described herein, wherein the computer program is stored on a machine-readable medium. In other words, one embodiment of the method according to the invention is thus a computer program that has program code for performing one of the methods described herein when the computer program is executed on a computer.

[0228] A further embodiment of the method according to the invention is thus a data carrier (or a digital storage medium or a computer-readable medium) on which the computer program for performing one of the methods described herein is recorded. The data carrier or the digital storage medium or the computer-readable medium is typically tangible and / or non-transitory.

[0229] A further embodiment of the method according to the invention is thus a data stream or a sequence of signals that represents the computer program for carrying out one of the methods described herein. The data stream or the sequence of signals can be configured, for example, to be transferred via a data communication connection, for example, via the Internet.

[0230] A further embodiment comprises a processing device, for example a computer or a programmable logic device, which is configured or adapted to carry out one of the methods described herein.

[0231] A further embodiment comprises a computer on which the computer program for performing one of the methods described herein is installed.

[0232] A further embodiment according to the invention comprises a device or system designed to transmit a computer program for performing at least one of the methods described herein to a recipient. The transmission can be electronic or optical, for example. The recipient can be, for example, a computer, a mobile device, a storage device, or a similar device. The device or system can, for example, comprise a file server for transmitting the computer program to the recipient.

[0233] In some embodiments, a programmable logic device (e.g., a field-programmable gate array, an FPGA) may be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field-programmable gate array may interact with a microprocessor to perform any of the methods described herein. In general, in some embodiments, the methods are performed by any hardware device. This may be general-purpose hardware such as a computer processor (CPU) or method-specific hardware such as an ASIC or, for example, a microprocessor, e.g., in the form of an ARM architecture.

[0234] The above-described embodiments are merely illustrative of the principles of the present invention. It is understood that modifications and variations of the arrangements and details described herein will be apparent to others skilled in the art. Therefore, it is intended that the invention be limited only by the scope of the following claims and not by the specific details presented in the description and explanation of the embodiments herein. References

[0235] [1] Z. Y. Choi and Y. H. Lee, "Frame synchronization in the presence of frequency offset," Communications, IEEE Transactions on, vol. 50, no. 7, pp. 1062-1065, 2002. [2] Sust, M.K. ; Kaufmann, R. F. ; Molitor, F. ; Bjornstrom, G.A.: Rapid acquisition concept for voice activated CDMA communication. In: IEEE Global Telecommunications Conference, 1990 Bd. 3, 1990, S. 1820#1826 [3] Internationale Patentanmeldung "Optimized Preamble and Methods for Interference Robust Packet Detection for Telemetry Applications" (PCT / EP2016 / 057014) [4] G. Kilian, H. Petkov, R. Psiuk, H. Lieske, F. Beer, J. Robert, and A. Heuberger, "Improved coverage for low-power telemetry systems using telegram splitting," in Proceedings of 2013 European Conference on Smart Objects, Systems and Technologies (SmartSysTech), 2013 [5] G. Kilian, M. Breiling, H. H. Petkov, H. Lieske, F. Beer, J. Robert, and A.Heuberger, “Increasing Transmission Reliability for Telemetry Systems Using Telegram Splitting,” IEEE Transactions on Communications, vol. 63, no. 3, pp. 949-961, Mar. 2015. [6] Wolfgang Koch, script for the seminar receiver synchronization at Fraunhofer IIS, June 10, 2015 - June 15, 2015 [7] Uwe Lambrette, Ralf Mehlan and Heinrich Meyr, Comparison of Demodulation Techniques for MSK, RWTH Aachen, https: / / www.ice.rwth-aachen.de / fileadmin / publications / Lambrette95TIRR.pdf, last retrieved: September 19, 2016 [8] Kay, Steven M.: Fundamentals of Statistical Signal Processing: Detection theory. Upper Saddle River, NJ: Prentice Hall PTR, 1998. ISBN 9780135041352 [9] Umberto Mengali, Aldo N. D'Andrea: "Synchronization Techniques for Digital Receivers," Plenum Press, 1997, ISBN 0-306-45725-3

[10] Walter Kellermann: "Digital Signal Processing," lecture notes from winter semester 2016 / 17, Chair of Multimedia Communication and Signal Processing (LMS), Friedrich-Alexander University Erlangen-Nuremberg.

[11] Steven M.Kay: "Fundamentals of Statistical Signal Processing - Vol. 2: Detection Theory", Prentice Hall, 1998, ISBN: 0-13-345711-7

[12] Z. Y. Choi and Y. H. Lee, "Frame synchronization in the presence of frequency offset", IEEE Transactions on Communications, vol. 50, no. 7, pp. 1062-1065, 2002.

Claims

1. Receiver (10), wherein the receiver (10) is configured to receive at least one signal transmitted by a transmitter (1), wherein the receiver (10) comprises a synchronization device (11), wherein the synchronization device (11) is configured to synchronize the receiver (10) with the transmitter (1) on the basis of the received signal, wherein the receiver (10) comprises a signal evaluation device (12), wherein the signal evaluation device (12) determines data for the synchronization from the received signal and transmits them to the synchronization device (11) for the synchronization, wherein the receiver (10) is configured to receive a plurality of subpackets, transmitted by the transmitter (1), onto which a telegram, containing data transmitted by the transmitter (1), is subdivided, wherein the subpackets are transmitted with different frequencies at different times, wherein the subpackets each comprise a subpilot sequence, wherein subpackets are shorter than a single telegram comprising the transmitted data, wherein the subpilot sequences complement each other to form a pilot sequence, and wherein the signal evaluation device (12) retrieves, at each time k, L sample values in the symbol distance and multiplies them with L conjugated complex symbols of the subpilot sequences, and, on the basis of the result of the multiplication and using the DFT method, a value for a frequency difference between a transmission frequency of the transmitter (1) and a receive frequency of the receiver (10), by the signal evaluation device (12) determining values of decision variables (di[k]) for each subpilot sequence, combining the determined values of the decision variables (di[k]) with one another via the subpilot sequences and carrying out a determination of a maximum value with the combined values.

2. Receiver (10) as claimed in claim 1, wherein the signal evaluation device (12) determines the values of the decision variables (di[k]) for each subpilot sequence, by the signal evaluation apparatus (12) forming multiplication values of a multiplication of samples of the respective subpilot sequence with conjugated complex values of an equal number of reference symbols of a known reference sequence and with samples of the complex exponential oscillation for a plurality of frequencies of a complex exponential oscillation referred to as frequency hypotheses, and summing up the multiplication values.

3. Receiver (10) as claimed in claim 1 or 2, wherein the signal evaluation device (12) combines the values of the decision variable (di[k]) determined for the subpilot sequences together for the subpilot sequences by the signal evaluation device (12) adding the amounts of the values of the decision variable (di[k]).

4. Receiver (10) as claimed in claim 1 or 2, wherein the signal evaluation device (12) combines the values of the decision variable (di[k]) determined for the subpilot sequences together for the subpilot sequences by the signal evaluation device (12) forming a sum of the magnitudes of a real part and an imaginary part of the values of the decision variable (di[k]).

5. Receiver (10) as claimed in claim 1 or 2, wherein the signal evaluation device (12) combines the values of the decision variable (di[k]) determined for the subpilot sequences together for the subpilot sequences by the signal evaluation device (12) summing up the magnitude squares of the values of the decision variable (di[k]).

6. Receiver (10) as claimed in claim 1 or 2, wherein the signal evaluation device (12) coherently combines the values of the decision variable (di[k]) determined for the subpilot sequences together for the subpilot sequences taking into account the amount and phase of the values of the decision variable (di[k]).

7. Receiver (10) as claimed in one of claims 1 to 6, wherein the signal evaluation device (12) combines the determined values of the decision variables (di[k]) with one another taking into account weighting factors.

8. Receiver (10) as claimed in one of the claims 1 to 6, wherein the signal evaluation device (12) combines the determined values of the decision variables (di[k]) with inclusion of weighting factors relating respectively to the subpilot sequences.

9. Receiver (10) as claimed in one of claims 1 to 8, wherein the signal evaluation device (12) carries out a time estimate for a time offset between a reference point in time of a transmission of a respective subpacket and a value assumed and / or estimated for the evaluation of the respective subpacket for the reference point in time, by the signal evaluation device (12) determining a time estimate value for a time offset on the basis of a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a time variable.

10. Receiver (10) as claimed in one of claims 1 to 9, wherein the signal evaluation device (12) carries out a frequency estimate for the frequency difference by the signal evaluation device (12) determining a frequency estimate based on a maximum value obtained by determining the maximum value and at least one value adjacent to the maximum value with respect to a frequency variable.

11. Method for synchronizing a receiver (10) with a transmitter (1), wherein at least one signal transmitted by the transmitter (1) is received by the receiver (10), wherein the receiver (10) is synchronized with the transmitter (1) on the basis of a pilot sequence or a subpilot sequence of the received signal, wherein the receiver (10) receives a plurality of subpackets, transmitted by the transmitter (1), onto which a telegram, containing data transmitted by the transmitter (1), is subdivided, wherein the subpackets are transmitted with different frequencies at different times, wherein the subpackets each comprise a subpilot sequence, wherein subpackets are shorter than a single telegram comprising the transmitted data, wherein the subpilot sequences complement each other to form a pilot sequence, wherein the receiver (10) is synchronized with the transmitter (1) starting from the subpilot sequences, wherein, at each time k, L sample values in the symbol distance are retrieved and multiplied with L conjugated complex symbols of the subpilot sequences, and, on the basis of the result of the multiplication and using the DFT method, a value for a frequency difference between a transmission frequency of the transmitter (1) and a receive frequency of the receiver (10), determining values of decision variables (di[k]) for each subpilot sequence, combining the determined values of the decision variables (di[k]) with one another via the subpilot sequences and carrying out a determination of a maximum value with the combined values.

12. Computer program with a program code causing, when the program is carried out by a computer, the same to carry out the method as claimed in claim 11.