Transmitter, receiver and corresponding method
阅读说明：本技术 发送器、接收器以及对应方法 (Transmitter, receiver and corresponding method ) 是由 雅可比·科内瑟尔 格尔德·基利安 约瑟夫·伯恩哈德 约尔格·罗伯特 约翰尼·韦切斯勒 多米 于 2018-04-06 设计创作，主要内容包括：一种发送器(1),被配置为发送信号,每个信号具有导频序列,所述导频序列包括多个导频序列符号,其中,发送器(1)包括信号产生器(2),其中,信号产生器(2)被配置为基于基本序列来提供导频序列,基本序列包括多个基本序列符号,其中,信号产生器(2)基于连续重复(R-1)次的基本序列符号来提供导频序列符号,并且其中,R是大于或等于2的自然数,其中,基本序列被配置为使得导频序列与由导频序列形成的发送信号的相关性具有尽可能窄的主最大值和/或尽可能小的次最大值。(Transmitter (1) configured to transmit signals, each signal having a pilot sequence comprising a plurality of pilot sequence symbols, wherein the transmitter (1) comprises a signal generator (2), wherein the signal generator (2) is configured to provide the pilot sequence based on a base sequence comprising a plurality of base sequence symbols, wherein the signal generator (2) provides the pilot sequence symbols based on base sequence symbols repeated (R-1) times in succession, and wherein R is a natural number greater than or equal to 2, wherein the base sequence is configured such that a correlation of the pilot sequence with a transmitted signal formed by the pilot sequence has a primary maximum value as narrow as possible and/or a secondary maximum value as small as possible.)
1. A transmitter (1) for transmitting a signal,
wherein the transmitter (1) is configured to transmit signals, each of the signals having a pilot sequence comprising a plurality of pilot sequence symbols,
wherein the transmitter (1) comprises a signal generator (2),
wherein the signal generator (2) is configured to provide the pilot sequence based on a base sequence comprising a plurality of base sequence symbols,
wherein the signal generator (2) provides the pilot sequence symbol based on the base sequence symbol repeated (R-1) times in succession, and wherein R is a natural number greater than or equal to 2,
wherein the base sequence is configured such that the correlation of the pilot sequence with a transmitted signal formed by the pilot sequence comprises a primary maximum which is as narrow as possible and/or a secondary maximum which is as small as possible.
2. Transmitter (1) according to claim 1,
wherein the signal generator (2) generates symbol blocks in the pilot sequence, each of the symbol blocks comprising a base sequence symbol and (R-1) repetitions of the base sequence symbol, an
Wherein the signal generator (2) generates the pilot sequence such that symbol blocks of the base sequence symbols immediately follow each other in the order of the base sequence symbols within the base sequence.
3. Transmitter (1) according to claim 1 or 2,
wherein, in case the pilot sequence comprises eight pilot sequence symbols, the base sequence has one of the following forms:
0010. 1101, 0100 or 1011, and
where 0 and 1 are binary base sequence bits, respectively.
4. Transmitter (1) according to claim 1 or 2,
wherein, in case the pilot sequence comprises twelve pilot sequence symbols, the base sequence has one of the following forms or forms obtained from the following forms by reversing the bit order:
000101, 001011, 001101, 010001, 111010, 110100, 110010, or 101110, and
where 0 and 1 are binary base sequence bits, respectively.
5. Transmitter (1) according to one of the claims 1 to 4,
wherein the signal generator (2) provides symbol blocks in the pilot sequence, each of the symbol blocks comprising a base sequence symbol and (R-1) repetitions of the base sequence symbol, an
Wherein the signal generator (2) provides the base sequence symbols with phase factors such that the phase factor of the respective i-th occurrence of a base sequence symbol in a symbol block is the same for all symbol blocks, and
wherein i is a natural number between 1 and R.
6. Transmitter (1) according to claim 5, wherein the phase factor is a component of a modulation alphabet.
7. Transmitter (1) according to claim 5 or 6,
wherein, in case the pilot sequence comprises eight pilot sequence symbols, the pilot sequence has one of the following forms or forms obtained from the following forms by reversing the bit order:
01011001, 10100110, 10011010, 01100101, 00001100, 11110011, 00110000, 11001111, and
where 0 and 1 are pilot binary sequence bits, respectively.
8. Transmitter (1) according to claim 5 or 6,
wherein, in case the pilot sequence comprises twelve pilot sequence symbols, the pilot sequence has one of the following forms or forms obtained from the following forms by reversing the bit order:
000000110011, 111111001100, 1100110000, 001100111111, 010101100110, 101010011001, 011001101010, 100110010101, 000011001111, 111100110000, 010110011010, 101001100101, 000011110011, 111100001100, 110011110000, 001100001111, 010110100110, 101001011001, 011001011010, 100110100101, 001100000011, 110011111100, 110000001100, 001111110011, 011001010110, 100110101001, 011010100110, or 100101011001, and
where 0 and 1 are pilot binary sequence bits, respectively.
9. Transmitter (1) according to one of the claims 1 to 8,
wherein the signal generator (2) provides the pilot sequence such that the pilot sequence has at least one supplementary symbol,
wherein the signal generator (2) provides symbol blocks in the pilot sequence, each of the symbol blocks comprising a base sequence symbol and (R-1) repetitions of the base sequence symbol, an
Wherein the signal generator (2) provides the pilot sequence in such a way that the at least one supplementary symbol precedes or follows the symbol block.
10. Transmitter (1) according to claim 9,
wherein the signal generator (2) generates the pilot sequence such that the pilot sequence comprises at least two supplementary symbols, an
Wherein the signal generator (2) provides the pilot sequence such that at least one of the at least two supplementary symbols precedes the symbol block and at least one other of the at least two supplementary symbols follows the symbol block.
11. Transmitter (1) according to claim 9 or 10,
wherein the at least one supplementary symbol or the at least two supplementary symbols are configured such that the pilot sequence has a primary maximum value which is as narrow as possible and/or a secondary maximum value which is as small as possible with respect to a transmitted signal formed by the pilot sequence.
12. Transmitter (1) according to one of the claims 9 to 11,
wherein, in the case that the pilot sequence comprises eight pilot sequence symbols, the base sequence is of the form 001 and there are two supplemental bits which together have one of the following forms:
01. 10, 00 or 11, and
where 0 and 1 are the binary base sequence bit and the binary supplemental symbol, respectively.
13. Transmitter (1) according to claim 12,
wherein the pilot sequence comprises one of the following forms or forms obtained from the following forms by reversing the bit order:
00001101, 11110010, 10110000, 01001111, 01011000, 10100111, 00011010, 11100101, 01000011, 1010011100, 11000010, 00111101, 00010110, 11101001, 01101000, 10010111, 10000110, 01111001, 01100001, 10011110, 10011110, 00101100, 11010011, 00110100, or 11001011, and
where 0 and 1 are pilot binary sequence bits, respectively.
14. Transmitter (1) according to one of the claims 9 to 11,
wherein, in case the pilot sequence comprises twelve pilot sequence symbols, the base sequence is of the form 00010 and there are two complementary bits which together have one of the following forms:
01. 10, 00 or 11, and
where 0 and 1 are the base sequence binary bit and the supplemental binary bit, respectively.
15. Transmitter (1) according to claim 14,
wherein the pilot sequence has one of the following forms or forms obtained from the following forms by reversing a bit order:
100000011001, 011111100110, 100110000001, 011001111110, 001010110011, 110101001100, 110011010100, 001100101011, 000000110010, 111111001101, 010011000000, 1011001111, 010101100111, 101010011000, 111001101010, 000110010101, 010000001100, 101111110011, 001100000010, 1100111101, 000101011001, 111010100110, 100110101000, or 011001010111, and
where 0 and 1 are pilot binary sequence bits, respectively.
16. A method for transmitting a signal, the method comprising,
wherein the signals are transmitted each with a pilot sequence comprising a plurality of pilot sequence symbols,
wherein the pilot sequence is provided based on a base sequence comprising a plurality of base sequence symbols,
wherein the pilot sequence symbols are provided in accordance with (R-1) consecutive repetitions of a base sequence symbol, where R is a natural number greater than or equal to 2; and
wherein the base sequence is configured such that the pilot sequence has a correlation with a transmission signal formed by the pilot sequence with a primary maximum value as narrow as possible and/or a secondary maximum value as small as possible.
17. A receiver (10, 20, 30),
wherein the receiver (10, 20, 30) is configured to receive at least one signal from the transmitter according to any of claims 1 to 16 and to evaluate the signal for a pilot sequence,
wherein the receiver (10) comprises a signal evaluation device (11),
wherein the signal evaluation device (11) is configured to perform a first evaluation of the received signal,
wherein the signal evaluation device (11) at least partially samples the received signal at a first sampling rate and/or processes samples of the received signal using only every ith sample during the first evaluation, wherein i is a natural number greater than or equal to 2,
wherein the signal evaluation device (11) produces an evaluation result for the pilot sequence during the first evaluation,
wherein the signal evaluation device (11) performs a second evaluation of the received signal depending on the evaluation result,
wherein the signal evaluation device (11) at least partially samples the received signal at a second sampling rate and/or further processes samples of the received signal using only every kth sample during the second evaluation, and
wherein the second sampling rate is greater than the first sampling rate and/or wherein k is a natural number less than i.
18. Receiver (10) according to claim 17,
wherein in the first evaluation the signal evaluation device (11) calculates a reduced pilot sequence consisting of a plurality of symbols from the received signal, wherein the number of symbols of the reduced pilot sequence is equal to the number of symbols of the base sequence, and
wherein the signal evaluation device (11) for generating the evaluation result compares the calculated reduced pilot sequence with the stored base sequence.
19. Receiver (10) according to claim 17 or 18, wherein the signal evaluation device (11) determines a correlation with the stored base sequence during sample processing.
20. A method for receiving at least one signal transmitted by the method of claim 16,
wherein the received signal is evaluated for a pilot sequence,
wherein the received signal is subjected to a first evaluation in which,
(i) at least partially sampling the received signal at a first sampling rate and/or further processing samples of the received signal using only every ith sample, where i is a natural number greater than or equal to 2, and
(ii) generating an evaluation result related to the pilot sequence, an
Wherein the received signal is subjected to a second evaluation depending on the evaluation result, in which second evaluation,
at least partially sampling the received signal at a second sampling rate and/or further processing samples of the received signal using only every kth sample value, wherein the second sampling rate is greater than the first sampling rate and/or k is a natural number less than i.
21. The method of claim 20, wherein the first and second portions are selected from the group consisting of,
wherein the pilot sequence of the signal is divided into at least two sub-regions that partially overlap each other,
wherein at least two of the sub-regions are correlated with sub-regions of a reference sequence and a partial result is produced in each case, an
Wherein an overall result for the pilot sequence is generated from the partial result.
22. The method according to claim 20 or 21,
wherein a Fourier transform is determined separately for each case of at least two sub-groups or at least two sub-pilot sequences of the pilot sequence,
wherein the determined Fourier transforms are incoherently added and an addition result is produced, an
Wherein an evaluation result for the pilot sequence is generated based on the addition result and based on a reference sequence.
23. A system (50) for signal transmission, wherein the system (50) comprises at least one transmitter (1) according to any one of claims 1 to 15 and at least one receiver (10, 20, 30) according to claims 17 to 19.
24. A computer program comprising program code for performing the method of any of claims 16 or 20 to 22.
25. A receiver (10, 20, 30),
wherein the receiver (10, 20, 30) is configured to receive at least one signal and to evaluate the at least one signal against a pilot sequence,
wherein the receiver (10) comprises a signal evaluation device (11),
wherein the signal evaluation device (11) is configured to perform a first evaluation of the received signal,
wherein the signal evaluation device (11) at least partially samples the received signal at a first sampling rate and/or processes samples of the received signal using only every ith sample during the first evaluation, wherein i is a natural number greater than or equal to 2,
wherein the signal evaluation device (11) produces an evaluation result for the pilot sequence during the first evaluation,
wherein the signal evaluation device (11) performs a second evaluation of the received signal depending on the evaluation result,
wherein the signal evaluation device (11) at least partially samples the received signal at a second sampling rate and/or further processes samples of the received signal using only every kth sample during the second evaluation, and
wherein the second sampling rate is greater than the first sampling rate and/or wherein k is a natural number less than i.
26. Receiver (10) according to claim 25,
wherein in the first evaluation the signal evaluation device (11) calculates a reduced pilot sequence consisting of a plurality of symbols from the received signal, wherein the number of symbols of the reduced pilot sequence is equal to the number of symbols of the base sequence, and
wherein the signal evaluation device (11) for generating the evaluation result compares the calculated reduced pilot sequence with the stored base sequence.
27. Receiver (10) according to claim 25 or 26, wherein the signal evaluation device (11) determines a correlation with the stored base sequence during sample processing.
28. A method for receiving at least one signal,
wherein the received signal is evaluated for a pilot sequence,
wherein the received signal is subjected to a first evaluation in which,
(i) at least partially sampling the received signal at a first sampling rate and/or further processing samples of the received signal using only every ith sample, where i is a natural number greater than or equal to 2, and
(ii) generating an evaluation result related to the pilot sequence, an
Wherein the received signal is subjected to a second evaluation depending on the evaluation result, in which second evaluation,
at least partially sampling the received signal at a second sampling rate and/or further processing samples of the received signal using only every kth sample value, wherein the second sampling rate is greater than the first sampling rate and/or k is a natural number less than i.
29. The method of claim 28, wherein the first and second portions are selected from the group consisting of,
wherein the pilot sequence of the signal is divided into at least two sub-regions that partially overlap each other,
wherein at least two of the sub-regions are correlated with sub-regions of a reference sequence and a partial result is produced in each case, an
Wherein an overall result for the pilot sequence is generated from the partial result.
30. The method according to claim 28 or 29,
wherein a Fourier transform is determined separately for each case of at least two sub-groups or at least two sub-pilot sequences of the pilot sequence,
wherein the determined Fourier transforms are incoherently added and an addition result is produced, an
Wherein an evaluation result for the pilot sequence is generated based on the addition result and based on a reference sequence.
31. A computer program comprising program code for performing a process according to any one of claims 28 to 30.
Technical Field
The present invention relates to a transmitter, a receiver and corresponding methods for transmitting and receiving signals.
Background
In many data transmission systems, pilot sequences (also referred to as training sequences or synchronization sequences) are inserted into the data stream to be transmitted for signal detection or parameter estimation. This can be either an uninterrupted transmission of a data stream, in which the pilot sequences are scattered at certain intervals, or a packet-oriented transmission, in which each packet (also referred to as telegram) usually contains exactly one pilot sequence. The pilot sequence is also referred to as a preamble or midamble if located at the beginning or in the middle of the packet. However, the pilot sequence may also be distributed within the packet in the form of two or more subsequences.
In telemetry systems, sensor networks, and applications that are keyed to the "internet of things" (IoT), asynchronous packet transmissions with long transmission stalls between packets typically occur.
In asynchronous packet transmission, the transmitter and the receiver are not synchronized, i.e. the receiver does not know the transmission time slot of the respective data packet. In order not to lose a packet, it must continuously check whether a packet is present in its received signal and estimate its time position with a certain accuracy throughout the reception latency.
To exacerbate the situation, the actual carrier frequency of the transmitted signal may deviate significantly from the nominal frequency and vary over time. The center frequency of the receive filter may also deviate from the nominal frequency. From the receiver point of view, the frequency difference between the carrier frequency of the transmitted signal and the center frequency of the receiving filter (hereinafter referred to as frequency offset) is decisive. For data detection, an estimation of the instantaneous frequency offset is required, and in the case of coherent detection methods, a phase estimation is also required.
The receiver must process two subjects in sequence in total:
1. and (3) detection: the packets are identified and their temporal position is at least roughly estimated, if necessary also taking into account the spectral position.
2. Synchronizing, comprising:
time synchronization: the exact time position of the packet is estimated,
frequency synchronization: estimating and correcting frequency offset, an
Phase synchronization: after frequency correction, the phase is estimated.
Due to the use of asynchronous systems, it is necessary to perform the detection of the telegrams by means of pilot sequences. The receiver must continuously search in its received signal whether the sensor node has sent a telegram. The decision of the receiver as to whether the received signal is caused by noise or by the transmitted signal is called telegram detection, or simply detection. For this reason, a pilot sequence with fixedly defined pilot symbols (commonly referred to as a "marker" in the english literature) is usually transmitted before the data to be transmitted.
Disclosure of Invention
It is an object of the invention to propose a transmitter and a receiver which use pilot sequences for data communication and which simplify the detection and/or processing of the pilot sequences.
This object is achieved by a transmitter.
The transmitter is configured to transmit signals, each signal having a pilot sequence comprising a plurality of pilot sequence symbols. The transmitter includes a signal generator. The signal generator provides a pilot sequence based on a base sequence comprising at least one base sequence symbol. In particular, the signal generator provides the pilot sequence symbols based on the base sequence symbols repeated (R-1) times in succession, and wherein R is a natural number greater than or equal to 2,
in one configuration, a signal generator provides a pilot sequence based on a base sequence having a plurality of base sequence symbols.
Accordingly, the transmitter transmits signals each including a pilot sequence. The pilot sequence has a plurality of pilot sequence symbols. The signal generator thus provides a corresponding pilot sequence based on a base sequence having at least one base sequence symbol (and in one arrangement a plurality of base sequence symbols). This includes, for example, utilizing stored pilot sequences or generating pilot sequences based on the base sequence. For example, the providing further includes generating symbols of the signal to be transmitted according to a mapping (e.g., MSK).
The base sequence symbols are repeated (R-1) times in succession in the pilot sequence so that the base sequence symbols occur R times. Thus, the base sequence symbols in the pilot sequence are repeated (R-1) times in succession such that each base sequence symbol occurs R times. This means that for detection it is sufficient to record and evaluate only every R-th symbol at the receiver side. This results in a reduction of the total expected length of the pilot sequence due to the reduced number of pilot sequence symbols and at the same time reduces the effort for initial evaluation of the received signal. This means that by sub-sampling at the receiver side, an optimized pilot sequence can be used for computationally optimized detection.
In one configuration, a sequence of base sequence symbols repeated R times is mapped with the modulation alphabet in each case (and in some configurations possibly also other symbols). In one configuration, a digital modulation method is used. In one variant, this is, for example, binary phase shift keying (binary PSK, BPSK). For example,
In one configuration, a signal generator generates at least one symbol block in a pilot sequence that includes at least one base sequence symbol and (R-1) repetitions thereof.
In one configuration, the signal generator provides a pilot sequence based on a base sequence comprising a plurality of base sequence symbols and provides symbol blocks in the pilot sequence, each symbol block comprising a base sequence symbol and a repetition thereof. Furthermore, the signal generator provides the pilot sequence in such a way that the symbol blocks of the base sequence symbols follow each other in the order of the base sequence symbols within the base sequence. In this configuration, the base sequence symbols repeated R times in the pilot sequence are referred to as symbol blocks, and the symbol blocks appear in the pilot sequence in the order of the base sequence symbols in the base sequence. Furthermore, the symbol blocks follow each other immediately.
In one configuration, the pilot sequence is specified to have L pilot sequence symbols, where L is a natural number. The number of base sequence symbols is then equal to the quotient of L divided by R. If the pilot sequence has a length of eight symbols and each basic sequence symbol is repeated once, i.e. if R ═ 2, four basic sequence symbols are required.
In one configuration, each base sequence symbol repeats once, such that R ═ 2.
One configuration is that the base sequence is configured such that the correlation of the pilot sequence with the transmitted signal formed by the pilot sequence has a primary maximum value that is as narrow as possible and/or a secondary maximum value that is as small as possible. This refers to the selection of the base sequence. More precisely, this configuration describes the correlation of the actual pilot sequence with the modulated (time-continuous) signal formed by it and transmitted by the transmitter.
In one configuration, it is provided that, in the case of a pilot sequence having eight pilot sequence symbols, the base sequence has one of the following forms: 0010. 1101, 0100 or 1011. Thus, 0 and 1 are binary base sequence bits, respectively. The sequence given here and in the following in one configuration is particularly relevant for Minimum Shift Keying (MSK), which is a mapping of binary symbols to signal segments to be actually generated therefrom. Alternatively or additionally, the sequences presented here and below are applicable to other linear or quasi-linear modulation methods, such as MSK or GMSK.
The described bits are converted into actual symbols by corresponding modulation and/or by mapping.
One configuration is that, in the case where the pilot sequence has twelve pilot sequence symbols, the base sequence has one of the following forms or forms obtained from the following forms by reversing the bit order: 000101, 001011, 001101, 010001, 111010, 110100, 110010, or 101110. Thus, 0 and 1 are binary base sequence bits, respectively. When selecting a base sequence, it is generally known that bit inversion and reversal of the bit order does not change the correlation properties of the sequence.
In one configuration, it is provided that the signal generator provides at least one symbol block in the pilot sequence, the symbol block comprising a base sequence symbol and (R-1) repetitions thereof, wherein the base sequence symbol has a phase factor. This configuration relates both to the case where the base sequence includes only one base sequence symbol and to the case where the base sequence includes a plurality of base sequence symbols. Thus, in the above configuration, the base sequence symbols occur in the pilot sequence R times consecutively, but with a phase factor.
In one configuration, it is provided that the signal generator provides a pilot sequence based on a base sequence comprising a plurality of base sequence symbols. Thus, in the pilot sequence, the signal generator provides symbol blocks, each symbol block comprising the base sequence symbols and (R-1) repetitions thereof, so that each symbol block is R symbols in length. The signal generator provides the phase factor for the base sequence symbol such that the phase factor for the ith occurrence of the base sequence symbol in the symbol block is the same for all symbol blocks, i being a natural number between 1 and R and representing the position of the base sequence symbol in the respective symbol block. The ith occurrence of i-1 is the zeroth repetition. Further explanation can be found in the following description. In particular, one configuration provides for the repeated symbols to have different phase factors.
In one configuration, the phase factor is a component of a modulation alphabet. Thus, the modulation alphabet is typically used to convert digital symbols for analog processing and/or current signaling.
Thus, a first base sequence symbol has the same phase factor in all symbol blocks, while a corresponding second (i.e. located at a second position) base sequence symbol has the same phase factor in all symbol blocks, respectively, which in one arrangement is different from the phase factor of the corresponding first base sequence symbol. Thus, when all base sequence symbols are considered, the step size of the same phase factor is given by repeating the base sequence symbols.
In one arrangement, the phase factor is derived fromrThe product of the identified phase and an imaginary number j (an exponent which is a function of the natural exponent), where the phase phirBetween 0 and 2 pi or 0 deg. and 360 deg..
One configuration is that, in the case of a pilot sequence having eight pilot sequence symbols, the pilot sequence has one of the following forms or forms obtained from the following forms by reversing the bit order: 01011001, 10100110, 10011010, 01100101, 00001100, 11110011, 00110000, 11001111. Thus, 0 and 1 are binary pilot sequence bits, respectively. In the pilot sequence, the symbols are transmitted in a specified order.
In one configuration, it is provided that, in the case of a pilot sequence having twelve pilot sequence symbols, the pilot sequence has one of the following forms or forms obtained from the following forms by reversing the bit order:
000000110011, 111111001100, 1100110000, 001100111111, 010101100110, 101010011001, 011001101010, 100110010101, 000011001111, 111100110000, 010110011010, 101001100101, 000011110011, 111100001100, 110011110000, 001100001111, 010110100110, 101001011001, 011001011010, 100110100101, 001100000011, 110011111100, 110000001100, 001111110011, 011001010110, 100110101001, 011010100110, or 100101011001.
Thus, 0 and 1 are binary pilot sequence bits. In one configuration, the above pilot sequences are related to MSK mapping.
The signal generator must provide a pilot sequence such that the pilot sequence has at least one supplemental symbol. The signal generator provides at least one symbol block in the pilot sequence, the symbol block comprising the base sequence symbol and its (R-1) repetitions. The signal generator provides a pilot sequence such that at least one supplemental symbol precedes or follows the symbol block.
In one configuration, it is provided that the signal generator provides a pilot sequence such that the pilot sequence has at least one supplementary symbol. The signal generator thus provides blocks of symbols in the pilot sequence, each block of symbols comprising a base sequence symbol and its repetition. In addition, the signal generator provides a pilot sequence such that at least one supplemental symbol precedes or follows the symbol block. In one configuration, the symbol blocks immediately follow each other such that there are no supplemental symbols between the symbol blocks, and the supplemental symbols are located only before or only after the symbol blocks.
In one configuration, it is provided that the signal generator provides the pilot sequence such that the pilot sequence has at least two supplementary symbols. Thus, the signal generator provides the pilot sequence such that at least one of the at least two supplemental symbols precedes the symbol block and at least one other of the at least two supplemental symbols follows the symbol block. Thus, the pilot sequence includes: a core formed by R repeated base sequence symbols; and a periphery formed by the supplemental symbols.
One configuration is that at least one supplementary symbol or at least two supplementary symbols are configured such that the correlation of the pilot sequence with the transmitted signal formed by the pilot sequence has a primary maximum value that is as narrow as possible and/or a secondary maximum value that is as small as possible. The selection of one or more supplementary symbols is therefore based on the fact that the resulting pilot sequence can be identified and/or synchronized as much as possible.
In one configuration it is provided that in the case of a pilot sequence having eight pilot sequence symbols, the base sequence is of the form 001 and there are two supplemental bits which together have one of the following forms: 01. 10, 00 or 11. Thus, 0 and 1 are the binary base sequence bit and the binary supplemental bit, respectively. Like the other bits already mentioned, the complementary bits are converted into complementary symbols by a corresponding mapping. As are the base sequence bits.
A complementary or alternative arrangement to the above configuration is that the pilot sequence has one of the following forms or forms obtained from the following forms by reversing the bit order:
00001101, 11110010, 10110000, 01001111, 01011000, 10100111, 00011010, 11100101, 01000011, 1010011100, 11000010, 00111101, 00010110, 11101001, 01101000, 10010111, 10000110, 01111001, 01100001, 10011110, 10011110, 00101100, 11010011, 00110100, or 11001011.
Thus, 0 and 1 are binary pilot sequence bits. In one configuration, the above pilot sequences are related to MSK mapping.
In one configuration, it is provided that in the case of a pilot sequence having twelve pilot sequence symbols, the base sequence has the form 00010 and there are two supplemental bits, which together have one of the following forms:
01. 10, 00 or 11. Thus, 0 and 1 are the binary base sequence bit and the binary supplemental bit, respectively.
A complementary or alternative arrangement to the above configuration is that the pilot sequence has one of the following forms or forms obtained from the following forms by reversing the bit order:
100000011001, 011111100110, 100110000001, 011001111110, 001010110011, 110101001100, 110011010100, 001100101011, 000000110010, 111111001101, 010011000000, 1011001111, 010101100111, 101010011000, 111001101010, 000110010101, 010000001100, 101111110011, 001100000010, 1100111101, 000101011001, 111010100110, 100110101000, or 011001010111.
Thus, 0 and 1 are binary pilot sequence bits. In one configuration, the above pilot sequences are related to MSK mapping.
The invention also relates to a method for transmitting a signal.
Thus, the method comprises at least the steps of:
each of said signals having a pilot sequence when transmitted, said pilot sequence comprising a plurality of pilot sequence symbols, an
Providing a pilot sequence based on a base sequence having at least one base sequence symbol such that the base sequence symbols are repeated at least consecutively to form the pilot sequence symbols.
The above configuration of the transmitter can be realized by correspondingly configured steps of the method, so that duplicate embodiments are omitted here.
The invention also achieves the object by means of a receiver.
The receiver is configured to receive at least one signal and evaluate it against a pilot sequence. In one configuration, the receiver evaluates using a stored or generally known reference sequence.
In one configuration, a reference sequence known to the receiver corresponds to a pilot sequence used by the transmitter for transmission, which the receiver uses to estimate the pilot sequence comprising the received signal.
The following configurations relate to the evaluation or a specific variant of the signal processing for the evaluation, respectively. Therefore, an assisting or auxiliary component of a receiver for evaluating against pilot sequences is specifically described.
In one configuration, the receiver includes a signal evaluation device.
The signal evaluation device is configured to cause an initial evaluation of the received signal. Thereby, the signal evaluation device samples the received signal at least partly at the first sampling rate during the first evaluation. Alternatively or additionally, the signal evaluator processes samples of the received signal using only every ith sample. Therefore, i is a natural number greater than or equal to 2. In addition, the signal evaluation device produces an evaluation result with respect to the pilot sequence during the first evaluation.
Depending on the result of the evaluation, the signal evaluation device performs a second evaluation of the received signal. To this end, the signal evaluation device samples the received signal at least partially at a second sampling rate during a second evaluation. Alternatively or additionally, the signal evaluation device further processes the samples of the received signal in the second evaluation using only every kth sample. Thus, the second sampling rate is greater than the first sampling rate, and/or k is a natural number less than i.
Thus, in the first evaluation, the signal evaluation device performs sub-sampling by setting the sampling rate accordingly during sampling or by processing fewer samples. In particular, if the signal from the above-mentioned transmitter is used together with a pilot sequence based on a base sequence, sub-sampling is sufficient due to repetition of symbols.
Thus, already during the first evaluation an evaluation result may be generated, which provides information on whether a pilot sequence has been detected. If this is the case, a second evaluation is made in one configuration, in which a higher sampling rate is used. In one configuration, this allows verification of a positive evaluation result of the first evaluation. For example, in the second evaluation, it may be determined whether each symbol of the base sequence does occur R times.
In one configuration, provision is made for the signal evaluation device to recognize, in a first evaluation, a reduced pilot sequence consisting of a plurality of symbols from the received signal, the number of symbols of the reduced pilot sequence being equal to the number of symbols of the basic sequence. Thus, to generate an evaluation result, the signal evaluation device compares the identified reduced pilot sequence with the stored base sequence (instead of specifying a reference base sequence). If the first sampling rate corresponds to the repetition rate of the base sequence symbols, the result is that the reduced pilot sequence is ideally the same as the base sequence used to provide the pilot sequence. For this purpose, in one configuration, the possible basic sequences are made available to the signal evaluation device by being stored in a corresponding data memory.
One arrangement is that during sample processing, the signal evaluation device determines a correlation or an approximation of the correlation with the stored base sequence. Thus, the stored base sequences are also those base sequences that are typically used to generate pilot sequences and may also be referred to as reference base sequences, for example.
In another configuration of the receiver, the receiver comprises a processing device. The processing device may be present instead of or in addition to the signal evaluation device.
The processing device is configured to divide the pilot sequence of the received signal into at least two partially overlapping sub-regions. The processing device correlates at least two sub-regions of the pilot sequence with sub-regions of the reference sequence and generates a partial result in each case. Finally, the processing device generates an overall result of the received signal for the pilot sequence based on the partial results.
In one configuration, the reference sequence is stored in a data store.
In one configuration, it is provided that the processing device adds the partial results incoherently to obtain an overall result.
According to one configuration, the processing device weights the symbols of at least two sub-regions before performing the correlation, depending on how many sub-regions the symbols belong to. If the symbols thus belong to overlapping regions, their weights are different from those when they belong to non-overlapping regions.
In another configuration of the receiver, the receiver comprises a transform device. The transformation device may be present instead of or in addition to the signal evaluation device and/or the processing device.
The transformation device is configured to determine a fourier transform for at least two sub-packets or at least two sub-pilot sequences of the pilot sequence, respectively. The transformation device incoherently adds the determined Fourier transforms and generates an addition result. Further, the transformation device generates an evaluation result based on the addition result. In one configuration, the evaluation result is based on an application to the reference sequence. In one configuration, the evaluation result is applied to the pilot sequence, and alternatively to at least two sub-pilot sequences belonging to a common pilot sequence.
Thus, according to this configuration, the pilot sequence is divided into sub-packets or received in the form of sub-pilot sequences. For example, a complete and coherent pilot sequence is received and divided into subsequences during receiver evaluation.
In one configuration, two subpackets belong to two received signals. The receiver thus receives at least two signals, at least one sub-packet belonging to each signal.
In another configuration, two sub-packets belong to one received signal.
According to one configuration, the transforming device pads the sub-packet or sub-sequence to be transformed with zeros at the beginning or end of the sub-packet or sub-sequence before determining the fourier transform.
One configuration provides that the transformation device, after determining the fourier transformation, performs an interpolation between the maximum value of the sub-packet or sub-sequence to be transformed and the adjacent position of this maximum value.
In another configuration, it is provided that the transformation device performs interpolation between a maximum value of the sub-packet to be transformed or the sub-sequence to be transformed and an adjacent position of the maximum value after generating the addition result.
According to one configuration, the transformation device performs interpolation using a quadratic polynomial.
One configuration provides that the transformation device performs interpolation using a polynomial of the form:
y(x)＝y0-c(x-x0)2
wherein the free parameter y0C and x0Is determined from the maximum of the adjacent positions. The parameters are chosen such that they each pass through a maximum and an adjacent position.
The transformation device determines the maximum value of the interpolation curve by the following function:
wherein x0Is the abscissa value of the maximum value of the polynomial, y (0) is the maximum value, and y (-1) and y (1) are adjacent positions.
In one configuration, the form is y (x) y0-c(x-x0)2Is used as the interpolation function.
Abscissa value x of maximum polynomial0For example to represent an improved time estimate (normalized to the sampling interval T/N).
Abscissa value x according to maximum value of polynomial0The improved frequency estimate may be determined by:
according to one configuration, the incoherent summing comprises: the determined magnitude of the fourier transform, the square of the magnitude, or the addition of an approximation of the magnitude. The approximation of the quantity is for example the sum of the real and imaginary quantities.
One configuration provides that the transformation device is configured to perform a fast fourier transform or a discrete fourier transform to evaluate the pilot sequence with respect to frequency and/or phase.
The object is also achieved by a method for receiving at least one signal, wherein the received signal is evaluated for a pilot sequence.
The above configuration of the receiver can be realized by correspondingly configured steps of the method, so that repeated embodiments are omitted here.
In one arrangement, the method comprises the steps of:
an initial evaluation of the received signal is performed,
wherein the received signal is at least partially sampled at a first sampling rate,
and/or
Wherein only every ith sample value is used for further processing the samples of the received signal, an
Wherein an evaluation result related to the pilot sequence is generated,
wherein i is a natural number greater than or equal to 2.
-performing a second evaluation of the received signal on the basis of the evaluation result,
wherein the received signal is at least partially sampled at a second sampling rate,
and/or
Wherein only every kth sample is used for further processing of samples of the received signal, an
Wherein the second sampling rate is greater than the first sampling rate and/or k is a natural number less than i.
The following steps are provided in alternative or complementary configurations:
the pilot sequence of the signal is divided into at least two partially overlapping parts.
Correlating at least two subregions with subregions of the reference sequence and producing a partial result in each case.
From the partial results, an overall result for the pilot sequence is generated.
In an equivalent alternative or complementary configuration, the method comprises at least the following steps:
at least two subgroups or at least two sub-pilots for a pilot sequence
Sequences, respectively, determining the fourier transform.
Incoherently adding the determined Fourier transforms and producing an addition result.
Based on the addition result and on the reference sequence, an evaluation result for the pilot sequence is generated.
Furthermore, the invention relates to a signal transmission system having at least one transmitter in one of the above-mentioned configurations and at least one receiver in one of the above-mentioned configurations.
Finally, the invention relates to a computer program with a program code to perform the above-mentioned method according to an arrangement.
Drawings
In particular, there are many possibilities to configure and develop transmitters, receivers, systems and corresponding methods. For this purpose reference is made, on the one hand, to the claims and, on the other hand, to the following description of embodiments taken in conjunction with the accompanying drawings, in which:
figure 1 shows a schematic diagram of signal processing according to a correlation method for detecting pilot sequences,
figure 2 shows the aperiodic autocorrelation function of the binary sequence 10010111,
figure 3 shows the time-continuous cross-correlation function of sequence 10010111 with MSK modulation and a matched filter for MSK modulation,
figure 4 shows a schematic diagram of signal processing with unknown frequency offset,
figure 5 shows a schematic diagram of a system with multiple transmitters and receivers,
figure 6 shows the generation of a visual representation of a pilot sequence from a base sequence,
fig. 7 shows an amplitude diagram of the correlation function with MSK modulation of a matched filter receiver for a pilot sequence length of eight symbols,
fig. 8 shows an amplitude diagram of the correlation function with MSK modulation of a matched filter receiver for a pilot sequence length of 12 symbols,
fig. 9 shows a graph of the magnitude of the correlation function for MSK modulation with matched filter receiver for a pilot sequence length of eight symbols with supplemental symbols,
fig. 10 shows a graph of the magnitude of the correlation function for MSK modulation with matched filter receiver for a pilot sequence length of 12 symbols with supplemental symbols,
figure 11 shows a schematic diagram of the division of the pilot sequence into two non-overlapping sub-regions,
figure 12 shows a schematic diagram of the division of a pilot sequence into two sub-regions with overlap,
fig. 13 shows a schematic diagram of the division of the pilot sequence into three sub-regions with overlap, and
fig. 14 shows a schematic diagram of an example of the modulation used.
Detailed Description
Hereinafter, the problem of the asynchronous data communication system will be discussed again. Thereby, the transmitter transmits a signal provided with a pilot sequence. The receiver receives the signal and evaluates it against the pilot sequence, i.e. the receiver checks whether the received signal has a pilot sequence. For this purpose, the receiver is partially referenced to a known reference sequence.
The sampled received signal is usually present in the receiver with a certain oversampling. In the receiver, for each time step k of the oversampled signal, the probability of the presence of a synchronization sequence in the immediately preceding time window of the received signal is evaluated. For this purpose, the function f is usedNP(k) Is applied to the received signal sample for each time step and its output value is compared to a threshold value. If the function value exceeds the threshold, it is assumed that the pilot sequence has been transmitted at this time. The theoretical basis of this method is in the so-called "detection theory" (Neyman-Pearson detector, [8 ]]) As discussed herein.
Up to now, correlation methods have been used in practical implementations of detection, where the received signal is permanently correlated with the pilot sequence. For detection, the amount of correlation results is estimated. Subsequently, threshold detection or Maximum Likelihood (ML) detection or a combination of both as just described is performed.
Up to now, when selecting a pilot sequence to be used, a decision is usually made with reference to an autocorrelation function (ACF). The following sequences were selected: the secondary correlation peak of the ACF is smallest and the ACF falls as steeply as possible on both sides of the main peak. This allows the exact synchronization time to be determined very accurately. Furthermore, due to the lower secondary correlation peaks, the number of false detections at the position of these secondary peaks is reduced.
In DE 102011082098 a1, a method for a battery-powered transmitter is described, in which a data packet is divided into several sub-packets, each of which transmits only a part of the total information (so-called "telegram splitting"). Such sub-packets are called "hops". Several information symbols are sent in one hop. These hops are transmitted on one frequency or distributed over several frequencies (so-called "hopping"). There are some pauses between these hops where no transmission occurs.
In one configuration, such hops may be used by a transmitter or receiver as described below.
The receiver uses the pilot sequences present in each telegram to perform the subject matter mentioned in the summary: detection and synchronization.
The pilot sequence consists of L modulation symbols (also referred to as pilot symbols or herein as pilot sequence symbols) and is typically transmitted compactly at the beginning (preamble) or in the middle (midamble) of the message. Alternatively, the pilot sequence may be arbitrarily scattered among the data symbols. It is common practice to extract 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 a priori by the receiver or stored appropriately.
When using telegram splitting, i.e. in splitting a telegram into several sub-packets (also called segments), each segment contains its own pilot sequence. Thus, each segment typically contains the same pilot sequence.
In modern receivers of radio-based systems, it is common practice to mix the received signal down into baseband after bandpass filtering, and to use analog-to-digital converters (ADCs) for sampling and quantization at equal distances in time. Each sample value is a complex value consisting of a real part and an imaginary part. Thus, the sampling is at least at the symbol clock or generally at integer multiples thereof (oversampling). One purpose of the detection is therefore to search for the signal part with the pilot sequence in the sample sequence. Various methods are known for this purpose, which are briefly described below.
Low frequency offset correlation method
A sequence of samples of a received signal is correlated with a sequence of symbols of a known reference sequence.
To determine a correlation value at time k, the sample value at time k and the previous L-1 samples are taken from the received signal at symbol intervals. When oversampling by N times is performed, only every nth sample is acquired.
These L values are multiplied by the conjugate complex symbol of the pilot sequence. These products are then added together. Based on the amount of correlation values thus obtained, it is determined whether the L samples contain the complete pilot sequence. The amount of correlation values is therefore also referred to as decision variables.
An example of a complete signal processing at a sampling time k is shown in fig. 1.
Thus, the received signal r (t) is first passed through a receive filter (e.g., an optimal filter, also referred to as a matched filter) to obtain a time correlation function x (t). By k0To identify the marked estimated end of the pilot sequence. Further, T is the symbol interval, or 1/T is the symbol rate. N is an oversampling factor.
The following notation is generally used:
the time variable in parentheses is always time-continuous. For example, r (t) represents a time-continuous received signal.
The time variable in brackets is always time discrete and usually represents a continuous number of samples. For example, x [ k ] denotes the kth value of the (time-continuous) signal x (t) after the receive filter.
In order to determine whether a pilot sequence is present, basically two methods are known, which are often performed in succession:
1. first, threshold detection is typically performed. Will decide on the variable | d [ k ]]L and threshold dthrA comparison is made. If the decision variable is above the threshold, then the pilot sequence is deemed detected and the time index k marks the last sample of the sequence. Thus, a rough estimate of the time position of the pilot sequence is already available. In FIG. 1, when k is set0This is the case when k. In the negative case, the counterk increases: k becomes k + 1.
2. Based on the detected pilot sequence, maximum detection can be selected to be performed within a specified time window after the time from the first detection. For this purpose, the sample value with the largest amount of decision variables is used as the detection time. The time window is typically less than the telegram duration. This step improves the accuracy of the time estimation, which is particularly advantageous for pilot sequences with unfavorable correlation properties.
Selection of pilot sequences:
for symbol alphabets with M symbols, there is MLA possible pilot sequence. For a binary symbol alphabet (M2) and a pilot sequence length L8, there are a total of 28256 possible sequences.
For detection with correlation methods, the properties of the aperiodic autocorrelation function (AKF) of the pilot sequence are of crucial importance. Mathematically, this is defined as follows:
having a [ I ]]0 for I < 1 or I ≧ L
The maximum value is when i ═ 0 and L for all sequences. All sequences are equally applicable if only detection of pilot sequences is considered.
However, in order to estimate the time as accurately as possible, it is desirable that the size of the amplitude values of all ACFs when i ≠ 0 is as small as possible with respect to the maximum value. These values are also referred to as correlation secondary peaks.
If the relevant secondary peak of CKF is zero, the ACF is described as ideal. Unfortunately, no sequence has an ideal ACF.
Currently, it is common practice to use a pilot sequence that includes the smallest possible secondary peak. An example of a binary sequence of length eight is 10010111. If
Furthermore, instead of using time-discrete ACFs, the time-continuous cross-correlation function (CCF) between the modulated and filtered signal and the ACF of the pilot sequence is typically used to select the pilot sequence. However, their form is mainly determined by the ACF of the pilot sequence, but also depends on the modulation pulse and the impulse response of the reception filter.
For linearly modulated signals, and signals that can be represented by linearly modulated signals, such as Minimum Shift Keying (MSK) or Gaussian Minimum Shift Keying (GMSK), it can be shown that the cross-correlation function is given in normalized representation by:
here, h (t) is formed by the modulation pulse g (t) and the impulse response g of the receiving filterrAnd (t) is obtained by convolution.
For accurate time estimation, sequences with CCFs before and after the main maximum as close as possible to the ideal CCF are preferred. The ideal CCF is derived from the above formula and the theoretically ideal ACF for the pilot sequence. It therefore has the form of a pulse h (t). The above sequence 10010111 satisfies this property (see fig. 3).
In fig. 3, the MSK and the ideal CCF of the matched filter are also drawn as dashed lines. The normalized CCF is plotted on the y-axis. The time offset k is plotted on the x-axis. Thus, a continuous CCF of sequence 10010111 with MSK modulation and matched filter is shown. The dotted line is the function h (t).
The detection method of unknown frequency offset comprises the following steps:
a disadvantage of the above-described correlation method is that the detection is reliable only for very small frequency differences (and thus between the carrier frequency of the transmitted signal and the center frequency of the receive filter). Accordingly, other methods are described below.
The FFT method comprises the following steps:
a method that is also applicable to large frequency offsets is described in [9 ]. This can be seen as a generalization of the above-described method. The basic functions are summarized as follows.
Before summing, the values x [ k-l ] in FIG. 1]a*[L-l+1]Multiplied by a sample of the complex exponential oscillation. This operation is performed a number of times for different vibration frequencies, referred to as frequency hypotheses.
For each frequency hypothesis, a separate decision variable | d [ k, i ] is obtained]The decision variable depends not only on k but also on the index i, where i represents the ith frequency. All decision variables | d [ k, i determined for a time step k are selected]The maximum value of | is. The associated frequency index is called i0[k]。
The maximum value is then compared to a threshold value. If the maximum value is above the threshold, then a pilot sequence is detected and compared with i0[k]The index associated frequency may be used as a rough estimate of the frequency offset.
For equidistant frequency hypothesis, this corresponds to L values x [ k ]]a*[L-1]、x[k-N]a*[L-2]To x [ k- (L-1) N]a*[0]Discrete Fourier Transform (DFT).
Given that L is a power of 2, the DFT can be performed particularly efficiently in the form of the well-known Fast Fourier Transform (FFT). If L is not a power of 2, then the DFT length is rounded to the next larger power of 2 and a corresponding number of zeros are added to the L values so that the FFT can be applied.
To improve the reliability of the detection, the L FFT input values can still be supplemented with any number of additional zeros.
This method is illustrated in fig. 4.
This method is suitable for frequency offsets up to almost half the symbol rate.
However, if a matched filter (so-called optimum filter) is used, an energy loss of about 3dB must be accepted at a frequency offset of 0.5. This loss can be greatly reduced by expanding the bandwidth of the receive filter (e.g., by a factor of 1.2). However, this results in some loss at low frequency offsets (about 0.8dB for a bandwidth extension of 1.2 times).
The FFT method has a disadvantage of relatively high calculation amount. For the FFT only, in the best case (if L is a power of 2 and no zeros are inserted), approximately 5L (1+ ldL) floating point operations (FLOP) need to be performed per time step k [10 ]. This is well in excess of the 2L times of FLOP required for summation in the correlation method at low frequency offsets.
The FFT method is theoretically considered optimal according to the Neyman-Pearson criterion [11 ].
Correlation of phase difference:
- 上一篇：一种医用注射器针头装配设备
- 下一篇：终端装置、基站装置以及通信方法