Data Transmitter and Data Receiver
Description
Embodiments of the present invention relate to a data transmitter for transmitting a data
packet to a data receiver, in particular to a data transmitter which, for the purpose of
synchronizing the data packet in the data receiver and of equalizing channel-encoded data
packets, generates data packets having two reference sequences and transmits same to the
data receiver via a communication channel.
For transmitting small amounts of data, for example of data or payload data of a sensor,
e.g. of a heating, electricity or water sensor, a radio transmission system may be employed.
To this end, the sensor typically has a measuring means including a radio transmitter (data
transmitter) attached to it which transmits the data to a data receiver by means of a burst
type data packet. In digital radio transmission systems, or radio communication systems,
training sequences, or pilots, are typically used for synchronizing the data packet in the
data receiver. Training sequences are deterministic or pseudorandom binary data
sequences, e.g. PRBS (pseudorandom bit stream) sequences transmitted to the data
receiver by the data transmitter along with the actual payload data within the data packet.
The training sequences are known to the data receiver. By correlating a receive data stream
with the known training sequences, the data receiver may determine the temporal position
of the known training sequences in the receive data stream. In this context, the correlation
function comprises, at the location of the training sequence in the receive data stream, a
correlation peak which is the higher, or the larger, the better the receive data stream
matches the known training sequences. The more the receive data stream, or a transmission
signal, is superimposed by noise, however, the lower or smaller the correlation peak of the
correlation function will be.
In the publication "A Concept for Data-Aided Carrier Frequency Estimation at Low
Signal-to-Noise Ratios" by Susanne Godtmann, Niels Hadaschik, Wolfgang Steinert and
Gerd Ascheid, the training sequence is subdivided into two parts spaced apart from each
other, which enables performing improved frequency estimation.
In a radio transmission system wherein a code gain or coding gain is accomplished by
means of code combining, i.e., by combining two or more data packets, by transmitting
redundant information in several different data packets at different times, it is necessary to
detect the individual data packets even when the signal-to-noise ratio (SNR) is very low
and is not sufficient for (fully) decoding an individual data packet. Depending on the code

gain by means of said combination of several data packets, the signal-to-noise ratio
necessary, or required, for decoding at which the data can still be detected decreases at the
data receiver. However, for realizing the code gain it is required that the individual data
packets can be found or determined or partly be decoded, even incorrectly, even in the case
of a low signal-to-noise ratio, in the receive data stream.
Thus, the present invention is based on the object of providing a concept enabling
transmission of a data packet from a data transmitter to a data receiver via a
communication channel even in the case of poor signal-to-noise ratios.
This object is achieved by a data transmitter as claimed in claim 1, a data receiver as
claimed in claim 10, a method of transmitting a data packet as claimed in claim 22, a
method of receiving a data packet as claimed in claim 23, or a computer program as
claimed in claim 26.
The invention provides a data transmitter for transmitting a data packet to a data receiver
via a communication channel, comprising a means for generating the data packet and a
means for transmitting the data packet. The means for generating the data packet is
configured to generate a data packet having a first data block and a second data block and a
predefined first reference sequence and second reference sequence for synchronizing the
data packet in the data receiver, wherein the first reference sequence is longer than the
second reference sequence, and wherein in the data packet, the second data block is located
between the first reference sequence and the second reference sequence, and the first
reference sequence is located between the first data block and the second data block. The
means for transmitting the data packet is configured to transmit the data packet to the data
receiver via the communication channel.
The invention further provides a data receiver for receiving a data packet from a data
transmitter via a communication channel, the data packet comprising a first data block and
a second data block and a predefined first reference sequence and second reference
sequence for synchronizing the data packet in the data receiver, wherein, in the data
packet, the second data block is located between the first reference sequence and the
second reference sequence, and the first reference sequence is located between the first
data block and the second data block. The data receiver comprises a means for receiving
the data packet which is configured to localize the first reference sequence and the second
reference sequence of the data packet in the receive data stream and to determine, or detect
(e.g. equalize), the data packet on the basis of a determinable transmission parameter

which may be derived from the first reference sequence and from the second reference
sequence.
In embodiments, the data transmitter generates a data packet having a first reference
sequence and a second reference sequence. The first and second reference sequences are
known to the data receiver, which enables the data receiver to localize the first reference
sequence and the second reference sequence and, thus, the data packet in a receive data
stream. Because of the configuration of the inventive data packet, which has a first long
reference sequence located, in the data packet, between the first data block and the second
data block, and which has a second reference sequence which is shorter than the first
reference sequence and is located in the data packet such that it is spaced apart, by means
of the second data block, from the first reference sequence, it is possible to localize, or
detect, the first reference sequence and the second reference sequence in the receive data
stream even in the case of a low signal-to-noise ratio.
Moreover, the means for generating the data packet of the data transmitter may be
configured to subdivide, in the data packet, the first reference sequence into a first
reference subsequence and a second reference subsequence, the first reference subsequence
and the second reference subsequence each having the length of the second reference
sequence.
In a preferred embodiment, the means for generating the data packet of the data transmitter
is configured to provide, in the data packet, the first reference subsequence, the second
reference subsequence, and the second reference sequence as ML sequences (MLS
maximum length sequences), respectively, each having an additional binary element.
The means for generating the data packet of the data transmitter may further be configured
to derive the first data block and the second data block of the data packet from a first base
data block.
In addition, the means for generating a data packet of the data transmitter may be
configured to derive a third data block and a fourth data block from a second base datgl
block and to provide the third data block and the fourth data block in the data packet, the
third data block in the data packet being located at a data packet end, and the fourth data
block in the data packet being located at a data packet start.
In a preferred embodiment, the means for receiving the data packet of the data receiver
may further be configured to correlate the receive data stream with the first reference

sequence and with the second reference sequence, which are known to the data receiver, so
as to localize the first reference sequence and the second reference sequence of the data
packet in the receive data stream.
In addition, the data receiver may comprise a means for equalizing the data blocks of the
data packet, said means being configured to perform equalization for the first data block on
the basis of the first reference sequence so as to obtain an equalized first data block, and to
perform equalization for the second data block on the basis of the first reference sequence
or of a reference subsequence adjacent to the second data block and of the second
reference sequence so as to obtain an equalized second data block.
The means for equalizing the data blocks of the data receiver may be configured to decode
the equalized first data block and second data block so as to obtain a decoded first data
block and second data block. Moreover, the means for equalizing the data blocks of the
data receiver may be configured to encode the decoded first data block or second data
block so as to obtain an encoded first data block or second data block. Equalization for a
third data block may be performed on the basis of the encoded first data block if the first
data block in the data packet has a smaller temporal distance from the third data block than
does the second data block. Alternatively, equalization for the third data block may be
performed on the basis of the encoded second data block if the second data block in the
data packet has a smaller temporal distance from the third data block than does the first
data block. Moreover, equalization for a fourth data block may be performed on the basis
of the encoded first data block if the first data block in the data packet has a smaller
temporal distance from the fourth data block than does the second data block
Alternatively, equalization for the fourth data block may be performed on the basis of the
encoded second data block if the second data block in the data packet has a smaller
temporal distance from the fourth data block than does the first data block. Furthermore
the third data block and the fourth data block may be derived from a second base data
block, the third data block in the data packet being located at a data packet end, and the
fourth data block in the data packet being located at a data packet start.
Additionally, the means for equalizing the data blocks of the data receiver may be
configured to perform equalization for the first data block, the second data block, the third
data block, and the fourth data block while using a frequency estimation, phase estimation
or channel estimation.

Further embodiments of the invention additionally relate to methods of transmitting a data
packet and of receiving a data packet as well as to computer programs for performing the
inventive methods.
Embodiments of the present invention will be explained in more detail below with
reference to the accompanying figures, wherein:
Fig. 1 shows a schematic view of an embodiment of a data transmitter as well as
an embodiment of a data packet transmitted by the data transmitter;
Figs. 2a, b, c show three embodiments of the data packets transmitted by the data
transmitter, in schematic views in each case;
Figs. 3a, b show embodiments of the correlation function for three successive ME
sequences and for a separate third ML sequence, in schematic views in each
case;
Figs. 4a, b show two embodiments of the data packets transmitted by the data
transmitter and having a third data block, in schematic views in each case;
Fig. 5 shows a schematic view of an embodiment of a method of generating a data
packet from a first base data block and a second base data block;
Fig. 6 shows a schematic view of an embodiment of a data packet transmitted by
the data transmitter, the first reference sequence comprising two ML
sequences each having a binary element, and the second reference sequence
comprising an ML sequence having a binary element;
Fig. 7 shows a schematic view of an embodiment of a data receiver and of a data
packet received by the data receiver;
Fig. 8 shows a schematic view of an embodiment of a method of equalizing the
data blocks of a data packet.
In the following description of the embodiments of the invention, elements that are equal
or equal in function are provided with the same reference numerals in the figures, so the
their descriptions in the different embodiments are interchangeable.

Fig. 1 shows a schematic view of an embodiment of a data transmitter 100 and an
embodiment of a data packet 102 transmitted by the data transmitter 100. The data
transmitter 100 is configured to transmit a data packet 102 to a data receiver via a
communication channel. To this end, the data transmitter 100 comprises a means 104 for
generating the data packet 102 and a means 106 for transmitting the data packet 102.
The means 104 for generating the data packet 102 is configured to generate a data packet
102 having a first data block 108, a second data block 110, a predefined first reference
sequence 112 and a predefined second reference sequence 114 for synchronizing the data
packet in the data receiver, wherein the first reference sequence 112 is longer than the
second reference sequence 114, and wherein in the data packet 102, the second data block
110 is located between the first reference sequence 112 and the second reference sequence
114, and the first reference sequence 112 is located between the first data block 108 and
the second data block 110. According to the invention, thus, what is generated is riot a data
packet having only one training sequence or reference sequence but a data packet 102
wherein the reference sequence is subdivided into a first reference sequence 112 and a
second reference sequence 114, the first reference sequence 112 and the second reference
sequence 114 having different lengths and being spaced apart from each other in the data
packet 102 by means of a data block, e.g., by means of the second data block 110. In a
hardware implementation, the means 104 for generating the data packet 102 may be a
microprocessor or microcontroller, whereas the means 106 for transmitting the data packet
102 may be a transmission unit.
What follows is a description, by means of a time axis 118, of a temporal occurrence of a
data block and/or a reference sequence in the data packet 102. The times TO to TN of the
time axis 118 may characterize a time sequence of transmitting the data blocks and/or
reference sequences.
The first data block 108, the first reference sequence 112 and the second data block 110
form a data packet core area 116, the second reference sequence 114 being adjacent, in the
data packet 102, to the data packet core area 116. In this context, the second reference
sequence 114 may be located, in the data packet, at a data packet start between times T0
and T1, i.e., temporally before the data packet core area 116 extending from time T1 to T5
Alternatively, the second reference sequence 114 may be located, in the data packet 102, at
a data packet end, i.e., temporally after the data packet core area 116. Optionally, further n
data blocks (n can be an element of natural numbers) may be provided, in the data packet.
102, temporally before and/or after the core area 116.

The means 104 for generating the data packet 102 may further be configured to provide, in
the data packet 102, the first reference sequence 112 and the second reference sequence
114 as pseudorandom binary sequence in each case, e.g., as a PRBS sequence or as an ML
sequence. ML sequences are pseudorandom binary sequences of the length (2m - 1), the
number of binary ones of the ML sequence being higher by one, by definition, than the
number of binary zeros. In the frequency range, the representation of an ML sequence is
similar to white noise. In embodiments, the first reference sequence 112 and the second
reference sequence 114 may be derived from a long pseudorandom sequence, e.g., from an
ML sequence, it being possible for the first reference sequence to correspond to a first part
and for the second reference sequence 114 to correspond to a second part, of the long
pseudorandom sequence.
For obtaining the payload data, the means 104 for generating the data packet 102 may
comprise an interface configured to obtain, e.g., payload data from a sensor. The sensor
may be a heating, electricity or water meter, for example. However, it shall be noted that
the inventive approach is applicable to any radio transmission systems for transmitting
(e.g., channel-coded) data packets. A data block, e.g., the first data block 108 or the second
data block 110, may be derived from the payload data. In addition, a data block may be
provided with redundancy, e.g., with CRC (cyclic redundancy check) bits, for a
redundancy check. In addition, a data block may comprise the payload data in an encoded
form; in one embodiment, the payload data in the data block are encoded such that the
encoded payload data have redundant components.
Moreover, the first data block 108 and the second data block 110 may have different
lengths, it being possible for the first data block 108 to be longer than the second data
block (or vice versa). In embodiments, a data block may thus comprise a limited number of
bits which is predefined in each case. Alternatively, the first data block 108 and the second
data block 110 may be equal in length, it being possible for the length of the first data
block 108 and of the second data block 110 to be specified and/or predefined in advance
e.g., during evaluation and processing of the payload data. In addition, the lengths of the
respective data blocks may be dynamically adapted, e.g., in dependence of a payload data
density or amount of payload data, to the payload data and/or information, as a result of
which, e.g., a data packet 102 having data blocks of different lengths may arise
Furthermore, the redundancy of the respective data blocks may be adapted to a priority
and/or relevance of the payload data or information; thus, e.g., information of higher
relevance may be provided with more redundancy than information of lower relevance.

The means 106 for transmitting the data packet 102 is configured to transmit the data
packet 102 to the data receiver via the communication channel, e.g., in the form of a radio
transmission path. In embodiments, the means 106 for transmitting the data packet 102
may transmit the data packet 102 to the data receiver via the communication channel e.g.
by means of MSK modulation (MSK = minimum shift keying), PSK modulation (PSK =
phase shift keying, digital phase modulation), QAM modulation (QAM = quadrature
amplitude modulation), FSK modulation (FSK = frequency shift keying, digital frequency
modulation) or by means of a different analog or digital modulation at a corresponding
carrier frequency.
Figs. 2a to 2c show three embodiments of the data packets 102 transmitted by the data
transmitter 100, in schematic views in each case. In the embodiment shown in Fig. 2a, the
first data block 108 is temporally located at the data packet end between the times T4 and
T5. The first reference sequence 112 is located, in the data packet 102, between the first
data block 108 and the second data block 110 and extends from time T2 to T4, the first
reference sequence 112, the first data block 108 and the second data block 110 forming a
data packet core area 116 extending from time T1 to T5. The second reference sequence
114, which is shorter than the first reference sequence 112, is located, in the date packe
102, temporally before the second data block 110 between times T0 and T1 and, thus, at the
data packet start.
In the embodiment shown in Fig. 2b, the first reference sequence 112 is also located
between the second data block 110 and the first data block 108. However, the second
reference sequence 114 is located at the data packet end between times T4 and T5, so that
the data packet core area 116 extends from time To to T4. In the data packet 102 shown in
Fig. 2b, the first data block 108, the first reference sequence 112, the second data block
110 and the second reference sequence 114 thus are located, with respect to the data packe
shown in Fig. 2a, in a temporally reversed sequence in the data packet 102.
The means 104 for generating the data packet 102 may further be configured to implement,
in the data packet 102, the first reference sequence 112 to be (precisely) double the length
of the second reference sequence 114. In addition, the means 104 for generating the data
packet may be configured to subdivide, in the data packet 102, the first reference sequence
112 into a first reference subsequence 112a and a second reference subsequence 112b, the
first reference subsequence 112a and the second reference subsequence 112b each having
the length of the second reference sequence 114.

In the embodiment shown in Fig. 2c, the first reference subsequence 112a, the second
reference subsequence 112b, and the second reference sequence 114 are equal in length, or
equal in size. In addition, in the embodiment shown in Fig. 2c, the first data block 108
forms the first reference subsequence 112a, the second reference subsequence 112b, and
the second data block 110 forms a data packet core area 116, the second reference
sequence 114 being located, in the data packet 102, temporally before the data packet core
area 116. Alternatively, the second reference sequence 114 may be located, e.g.,
temporally after the data packet core area 116 and, thus, at the data packet end. Moreover!,
the first reference subsequence 112a, the second reference subsequence 112b, and the
second reference sequence 114 may be random or pseudorandom binary sequences.
In embodiments, the means 104 for generating the data packet 102 may be configured to
generate a data packet 102 in that the first reference subsequence 112a, the second
reference subsequence 112b, and the second reference sequence 114 are identicaly: i.e., the
first reference subsequence 112a is identical with the second reference subsequence 112b
and is identical with the second reference sequence 114.
In addition, the first reference subsequence 112a, the second reference subsequence 112b,
and the second reference sequence 114 may each be configured as an ML sequence having
an additional binary element, so that the data packet 102 comprises three ML sequences of
the length (2m - 1), each of which has been extended by one binary element. The binary
element may be, e.g., a binary "one" or a binary "zero", it being possible for the binary
element to be located, within the reference sequence and/or reference subsequence,
temporally before or after the ML sequence. The ML sequence may thus be extended by
one bit by means of the binary element, e.g., with a zero bit. Moreover, the first reference
subsequence 112a, the second reference subsequence 112b, and the second reference
sequence 114 are not transmitted successively, but in two parts, as the first reference
sequence 112 and the second reference sequence 114, which are spaced apart from each
other in the data packet by means of a data block. To this end, two ML sequences of
identical length are combined and are separated, as the first reference sequence 112, from
the third ML sequence and/or the second reference sequence 114 by means of a data block
The first data block 108 and the second data block 110 may be derived from a base data
block, as will be described in more detail below with reference to the description of the
embodiment shown in Fig. 5. A first part of the base data block, e.g., the first data block
108, may be temporally located on a first side of the first reference sequence 112. A second
part of the base data block, e.g., the second data block 110, is arranged, in the data packet
to be temporally located on the other, or second, side of the first reference sequence 112

and/or of the pair of ML sequences, so that the base data block is temporally located before
(on the left on the time axis 118) and temporally after (on the right on the time axis 118)
the ML sequence tuple. Preferably, the base data block is subdivided into two parts and/or
data blocks identical in size. However, equivalent subdivision is not mandatory.
Extension of the ML sequences by one bit may be effected since the bits are modulated
with a (2^n)-stage modulation, for example with a (non-differential) MSK modulation. If a
sequence or reference sequence or reference subsequence consists of an even number of
bits, that bit which follows said reference sequence or reference subsequence will be
mapped, in the MSK modulation, to the same axis of a constellation diagram as the first bit
of the first reference sequence or reference subsequence. By means of the even number of
bits of the first reference subsequence 112a, of the second reference subsequence 112b and
of the second reference sequence 114 as well as of the data block between the first
reference sequence 112 and the second reference sequence 114, i.e., of the second data
block 110, one achieves that all of the reference sequences, that is the first reference
subsequence 112a, the second reference subsequence 112b, and the second reference
sequence 114, have the same constellation points following modulation. The complex
baseband representation therefore contains three identical reference sequence portions
and/or training sequence portions.
For simple implementation into, e.g., a microcontroller, the length of the data blocks which
belong together may be selected to be an integer multiple of 8 bits. The extension of the
ML sequences by one bit correspondingly results in that in one embodiment, a training
sequence, e.g., the first reference subsequence 112a, the second reference subsequence
112b, and the second reference sequence 114, each are exactly four bytes in length.
For localizing the first and second reference sequences 112 and 114 in a data stream, or
receive data stream, the receive data stream may be correlated with the known first and
second reference sequences 112 and 114. The amount of the correlation function
comprises, at the location or temporal position, a correlation peak which is the higher, or
the larger, the better the receive data stream matches the known first and second reference
sequences 112 and 114. The more the signal is superimposed by noise, the smaller the
amount of the correlation peak will be. In the correlation, antipodal ML sequences may be
used which have the property that the result of the correlation function approaches a delta
function (correlation peak) when an ML sequence is correlated with a periodically
continued version of the same ML sequence, i.e., when there is a so-called periodic
autocorrelation function. Antipodal sequences s may be generated from binary sequences
by the following mapping:

x_k=0 becomes s_k=-l and
x_k=l becomes s_k=+l.
In embodiments it is thus possible to detect the data packet 102 by means of the first and
second reference sequences 112 and 114 "in noise", i.e., at a low signal-to-noise ratio.
Fig. 3 shows a schematic view of a distribution of correlation peaks of a correlation of
three ML sequences with the receive data stream, the third of the three ML sequences
being separated, whereas Fig. 3b shows a schematic view of a distribution of correlation
peaks of a correlation of three successive ML sequences with the receive data stream, the
time being plotted on the abscissa, and the normalized amount of the correlation function
being plotted on the ordinate.
Due to the subdivision of the reference sequence into a first reference sequence 112 and a
second reference sequence 114, which are separated from each other by means of a data
block, secondary peaks Nl to N6, which are smaller in terms of amount, are yielded in the
synchronization in the data receiver for the correlation function as compared to only one
reference sequence. In addition, the secondary peaks Nl to N6 have the same (e.g.,
normalized) amount and/or the same amplitude. In the distribution of correlation peaks,
which is shown in Fig. 3a, all of the secondary peaks Nl to N6 have an amount of "1",
whereas the main peak H has an amount of "3". Without the separation of the second
reference sequence 114, the secondary peaks Nl to N6 would have an amount increasing
toward the main peak H, as is depicted in Fig. 3b. Starting from the secondary peaks N1
and N4 with the amount of "1", the amount increases to an amount of "2" for the
secondary peaks N2 and N3, whereas the main peak H also comprises an amount of "3".
This may complicate identification of the main peak H in particular in the event of noise,
i.e., in the event of a poor signal-to-noise ratio.
In addition, in the distribution of correlation peaks which is shown in Fig. 3a, the main
peak H has three times the height of the secondary peaks Nl to N6. Due to the inventive
arrangement of the first and second reference sequences 112 and 114, the main peak H
may thus be more easily distinguished and/or separated from the secondary peaks Nl and
N6 in the data receiver by setting a threshold value. Thus, the threshold value for the
distribution of correlation peaks which is shown in Fig. 3a may be set to, or specified to
have, the value of "two", whereas for the distribution of correlation peaks which is shown
in Fig. 3b, a threshold value of the value of "2" is already reached by the secondary peaks
N3 and N4. For the distribution of correlation peaks which is shown in Fig. 3, the threshold

value would have to be set to, or specified to have, the value of "2.5", which would result
in a threshold value increased by a factor of 1.25.
Due to the use of three ML sequences of the length m, resulting reference sequences
having lengths not divisible by two may additionally be generated. If, for example, a
training sequence and/or reference sequence is required which is longer than 127 bits
(corresponding to 27-l), the next ML sequence in length would already be 255 bits
(corresponding to 28-l) in length. Utilization of three ML sequences of the length 63
(corresponding to 26-l) results in a reference sequence of the length 189, i.e., a length
which is half-way between the ML sequences of lengths of 127 bits and 255 bits. Due to
the subdivision of the reference sequence, same behaves similarly to an imagined ML
sequence of the length of 3*m with a correlation peak, or main peak, H and small
secondary peaks Nl to N6.
In schematic views, Figs. 4a and 4b show two embodiments of the data packets 102
transmitted by the data transmitter 100, the data packets 102 each having a third data
block. The data packets 102 shown in Figs. 4a and 4b thus correspond to the data packet
102 of Fig. 2a, which comprises an additional third data block 120. The means 104 fc
generating a data packet 102 may be configured to provide a third data block in the data
packet 102.
In the data packet 102, the third data block 120 may be located, as is shown in Fig. 4a, at a
data packet end between times T5 and T6, whereas that part of the data packet 102 which is
known from Fig. 2a extends from time To to T5. Alternatively, the third data block 120
may be located, in the data packet 102, at a data packet start between times T0 and T1':
Accordingly, the second reference sequence 114 is located between times Ti and T2, and
the data packet core area 116 extends from time T2 to T6. Moreover, the means 104 fc
generating the data packet 102 may be configured to provide n further data blocks in the
data packet. Starting from the data packet core area 116, the n data blocks may be arranged
with increasing values of n from the data packet core area 116, on the time axis 118 to the
right, left or alternately to the right and to the left of the data packet core area 116 and the
second reference sequence 114.
Fig. 5 shows a schematic view of an embodiment of a method of generating a data packet
102 from a first base data block 124 and a second base data block 126. In a first step, the
first data block 108 and the second data block 110 may be derived from the first base data
block 124 and be provided, in the data packet 102, on the time axis 118 to the right and the
left of the first reference sequence 112. The first reference sequence may be located;

accordingly, between times T3 and T5 in the data packet 102 shown in Fig. 5, whereas the
first data block 108 may be located between times T5 and T6, and the second data block
110 may be located between times T2 and T3. The second reference sequence 114 may be
provided, in the data packet, between times T] and T2. In a second step, the third data block
120 and the fourth data block 122 may be derived from the second base data block 126. In
this context, the third data block 120 may be provided between times T6 and T7, and the
fourth data block may be provided between times TO and Tl in the data packet 102.
Moreover, the means 104 for generating the data packet 102 may be configured to perforr
the method shown in Fig. 5. The means for generating the data packet 102 may be
configured to derive the first data block 108 and the second data block 110 of the data
packet 102 from a first base data block 124. In addition, the means 104 for generating the
data packet may be configured to derive a third data block 120 and a fourth data block 122
from a second base data block 126 so as to provide the third data block 120 and the fourth
data block 122 in the data packet 102, the third data block 120 being located, in the data
packet 102, at a data packet end, and the fourth data block 122 being located, in the data
packet 102, at a data packet start. The means 104 for generating the data packet may obtain
the base data blocks, e.g., the first base data block 124 and the second base data block 126.
directly or in the form of payload data. In addition, the means for generating the data
packet 102 may be configured to process the payload data so as to obtain a data packet 102
comprising corresponding data blocks.
Fig. 6 shows a schematic view of an alternative embodiment of a data padket 102
transmitted by the data transmitter 100, the first reference sequence 112 comprising two
ML sequences 130a and 130b, each having one binary element 132a and 132b,
respectively, and the second reference sequence 114 comprising one ML sequence 130c
having one binary element 132c. Thus, the first reference sequence 112 is subdivided into.
first reference subsequence 112a and a second reference subsequence 112b, which are
identical with the second reference sequence 114. The three ML sequences 130a to 130c
have binary elements 132a to 132b added to them, respectively, the binary element 132a to
132c each being a binary zero in the data packet 102 shown in Fig. 6. Alternatively, the
binary element 132a to 132c may be, e.g., a binary one and/or may temporally precede the
respective ML sequence 130a to 130c.
Furthermore, the means 104 for generating the data packet 102 may be configured to
provide further (2*n) data blocks in the data packet 102, so that the data packet 102 show
in Fig. 6 comprises further (2*n) data blocks. In this context, a (2*n-l)th data block 134 and
a (2*n)th data block 136 may be derived from an nth base data block, the (2*n-l)th data

block 134 being located, in the data packet 102, at the data packet end, and the (2*n)th data
block 136 being located, in the data packet 102, at the data packet start (or vice versa). It
shall also be noted that the means for generating the data packet 102 may be configured, as
was already described with reference to Fig. 4a, to generate a data packet having n data
blocks.
Fig. 7 shows a schematic view of an embodiment of a data receiver 150 as well as of a data
packet 102 received by the data receiver 150. The data receiver 150 is configured to
receive a data packet from a data transmitter 100 via a communication channel. The data
packet 102 comprises a first data block 108 and a second data block 110 and a predefine4
first reference sequence 112 and second reference sequence 114 for synchronizing the data
receiver 150 and for equalizing the received data packet which is disturbed by the
communication channel; in the data packet 102, the second data block 110 is located
between the first reference sequence 112 and the second reference sequence 114, and the
first reference sequence 112 is located between the first data block 108 and the second data
block 110. The data receiver 150 comprises a means 152 for receiving the data packet 102,
said means being configured to localize the first reference sequence 112 and the second
reference sequence 114 of the data packet 102 in the receive data stream, and to determine
the data packet 102 on the basis of a determinable transmission parameter which may be
derived from the first reference sequence 112 and from the second reference sequence 114.
The transmission parameter may be a frequency, a frequency shift, a phase, a phase shift, a
group run time, or a frequency-dependent attenuation of the communication channel.
The means 154 for receiving the data packet 102 may receive the receive data stream via
an interface, e.g., an antenna, the means 154 for receiving the data packet 102 being
configured to localize the data packet 102, or, in particular, the first reference sequence
112 and the second reference frequency 114 of the data packet 102 in the receive data
stream.
In one embodiment, the means 154 for receiving the data packet 102 may be configured to
correlate the receive data stream with the first reference sequence 112 and with the the
second reference sequence 114, which are known to the data receiver 150, so as to localize
the first reference sequence 112 and the second reference sequence 114 of the data packet
102 in the receive data stream. In addition, following localization of the first reference
sequence 112 and of the second reference sequence 114, the data packet 102 may be
determined, or obtained, on the basis of a determinable transmission parameter.
Furthermore, the means 154 for receiving the data packet may be configured to determine
the transmission parameter on the basis of a frequency estimation of a phase estimation. In

embodiments, the transmission parameter may thus be a frequency, or carrier frequency, at
which the data packet 102 was transmitted from the data transmitter 100 to the data
receiver 150 via the communication channel, or may be a phase or phase shift between the
data transmitter 100 and the data receiver 150.
For synchronization with the data packet 102, the data receiver 150 may thus localize the
first reference sequence 112 and the second reference sequence 114 in the receive data
stream, and may determine, e.g. on the basis of a comparison between the received first
and second reference sequences and the known first and second reference sequences 112
and 114, the transmission parameter, e.g., a carrier frequency by means of which the data
packet 102 wa transmitted via the communication channel.
The inventive data transmitter 100 and/or data receiver 150 exhibit advantages, in
particular, in such radio transmission systems wherein data packets 102 must be detected at
low signal-to-noise ratios, but need not necessarily be fully decoded. This is the case, e,g.,
in radio transmission systems comprising code combining, wherein payload data is
differently coded and is sent out as data packets 102 at different times. In the data receive
150, a high code gain may result from combining the differently encoded data packets 102,
i.e., the useful information (payload) may still be decoded even at very low signal-to-noise
ratios. To make this possible, the burst-type data packets 102 must be found, or located, in
the receive data stream, i.e., the data receiver 150 must be able to synchronize itself to a
received data packet and to equalize channel effects.
The data receiver 150 may further comprise a means 152 for equalizing the data blocks.of
the data packet 102, said means being configured to perform equalization for the first data
block 108 on the basis of the first reference sequence 112 so as to obtain an equalized fir,u
data block, and to perform equalization for the second data block 110 on the basis of the
first reference frequency 112 or of a reference subsequence adjacent to the second data
block 110, and of the second reference sequence 114 so as to obtain an equalized second
data block. The mode of operation of the means 152 for equalizing the data blocks of the
data packet 102 will be explained in more detail below in the embodiment shown in Fig. 8:
Fig. 8 shows a schematic view of an embodiment of a method of equalizing the data blocks
of a data packet 102. Equalization may be a correction of a frequency shift, phase shift, o$
channel equalization which is caused, e.g., in the transmission of the data packet 102 froui
the data transmitter 100 to the data receiver 150 via the communication channel. The data
packet 102 shown in Fig. 8 corresponds to the data packet 102 of Fig. 6, wherein the first
reference sequence 112 is subdivided into a first reference subsequence 112a and a second

reference subsequence 112b, and wherein the first reference subsequence 112a and the
second reference subsequence 112b are identical with the second reference sequence 114,
and wherein the first reference subsequence 112a, the second reference subsequence 112b,
and the second reference sequence 114 comprise ML sequences 130a to 130c as well as a
binary element 132a to 132c, respectively.
In accordance with the invention, the first reference sequence 112 and the second reference
sequence 114 may be used not only for synchronizing the data receiver 150, i.e., for
localizing the first and second reference sequences 112 and 114 in the receive data stream
and for determining the transmission parameter, but in addition also for channel
equalization of the first base data block 124, or of the first data block 108, and the second
data block 110. Since the reference sequences 112 and 114 are spaced apart and/or are
separated from each other by means of a data block, frequency estimation may be
performed with a small estimation error. In addition, the frequency and phase estimation or
an estimation of a different transmission parameter may be performed separately for the
first data block 108 and for the second data block 110. Thus, synchronization or
equalization may be performed on the basis of the knowledge of the transmission
parameter.
The phase estimation for the first data block 108 may be averaged over the first reference
subsequence 112a and the second reference subsequence 112b and/or over the first ML
sequence 130a having the binary element 132a and the second ML sequence 130b having
the binary element 132b. Thus, a long known training sequence and/or reference sequence
is available for the phase estimation of the first data block 108, whereby a higher
estimation accuracy may be achieved. The estimation for the second data block 110 is
effected while using the second reference subsequence 112b and the second reference
sequence 114 and/or while using the second ML sequence 130b having the binary element
132b and the third ML sequence 130c having the binary element 132c. Thus, two reference
sequences are available also for the phase estimation of the second data block 110,
whereby the estimation accuracy can be increased.
For decoding n further base data blocks, which are divided and are joined, at the data
packet start and data packet end, respectively, to the transmit telegraph and/or to the data
packet core area 116, no further reference sequence is required for equalization as well a?
frequency and phase estimation. For equalizing the nth base data block, the information of a
correctly decoded preceding base data block, e.g., of the (n-l)th, (n-2)th or (n-3)th base data^
block, may be used in accordance with the invention, the information of the correctly
decoded (n-l)th (i.e., immediately preceding) base data block being used in a preferred

embodiment. Alternatively, one may also use the information of several preceding
correctly decoded base data blocks. The information of the data blocks is thus re-encoded
and compared to the receive sequence. The channel equalizations, e.g. frequency and phase
shift, may be determined from a difference or a comparison between the encoded
information and the receive sequence, that is, between the re-encoded or newly encoded
data block and the received data block. The individual data blocks are encoded by
themselves and may be used, following decoding and encoding or re-encoding, again as
references for equalizing further data blocks in the data packet 102. Thus, there is the
possibility of decoding the data blocks while starting at the first reference sequence 112,
and of subsequently encoding or re-encoding same. The encoded data blocks may be used
as reference data for channel estimation, frequency estimation, phase estimation or SNR
estimation. Neighboring data blocks may therefore be decoded with the time*variable
parameters. This design of the slots and/or of the arrangement of the data blocks in the data
packet 102 enables tracking of the parameter estimation over time and, thus, adaptation to
a time-variable channel or transmission channel. Due to the inventive design of the data
packet 102, known reference sequences within the data blocks that are required specifically
for parameter estimation are not necessary.
In embodiments, error propagation during decoding and re-encoding of a data packet 102
with subsequent parameter estimation has no repercussions since only error-free data
packets 102 or data blocks are processed further. The absence of errors in a data packet
102 or in a data block may be ascertained, e.g., by means of a CRC check sum. The CRC
check sum may be calculated over one data block, over several data blocks or over the
entire data packet 102. In addition, each data block may comprise one or more CRC bits!
so that decoding may be aborted as early as at the occurrence of a first corrupt data block
so as to save computing time, for example. If a data packet 102 cannot be decoded in a
completely error-free manner, this data packet may be combined, when using code
combining, with a different data packet 102 of the same transmitter that was sent out, e.g.,
at a later point in time. With combined data packets 102, equalization and decoding may
also be effected in an iterative manner. Alternatively, any correctly decodable data blo£M
of the data packet 102 may be used immediately, whereas the non-correctly decodable data
blocks are temporarily stored (latched) and are decoded at a later point in time while using
code combining, or by combining them with a further or different data packet 102 of the
same transmitter that was sent out, e.g., at a later point in time.
Additionally, it is also possible for data blocks to be coded differently. For example, data
blocks which are temporally closer to the first reference sequence 112 may be put under
higher protection and/or be provided with more redundance so as to be able to correctly

decode said data blocks without code combining with a higher probability than those data
blocks that are temporally located further apart from the first reference sequence 142. Thijs
may be exploited for correctly decoding important information as early as at reception of a
data packet 102 without having to wait for a further data packet 102 in order to be able to
perform code combining.
The means 152 for equalizing the data blocks of the data packet 102 may further be
configured to perform the above-described inventive method shown in Fig. 8. The means
152 for equalizing the data blocks of the data packet 102 may be configured to perform
equalization for the first data block 108 on the basis of the first reference subsequence
112a and of the second reference subsequence 112b so as to obtain an equalized first dat£
block 160. The second data block 110 may be equalized on the basis of the second
reference subsequence 112b and of the second reference sequence 114 so as to obtain an
equalized second data block 162. Equalization for the first data block 108 and the second
data block 110 may be effected while using a frequency estimation, phase estimation or
channel estimation.
In addition, the means 152 for equalizing the data blocks may be configured to decode the
equalized first data block 160 and the second data block 162 so as to obtain a decoded first
data block 164 and second data block 166. If it was possible to correctly decode the
equalized first data block 160, i.e., if the decoded first data block 164 comprises valid
information, this information may be re-encoded so as to be used as a reference sequence
for equalizing the third data block 120. By analogy, the information of a correctly decoded
second data block may be used for equalizing a fourth data block 122. Moreover, th»
means 152 for equalizing the data blocks may be configured to encode the decoded first
data block 164 or second data block 166 so as to obtain an encoded first data block 168 or
second data block 170.
In addition, the means 152 for equalizing the data blocks may be configured to^perfoni
equalization for a first data block 108 and a second data block 110, the first data block 108
and the second data block 110 being derived from a first base data block 124.
Additionally, the means for equalizing the data blocks of the data receiver may be
configured to decode the equalized first data block and second data block so as to obtain a
decoded first data block and second data block. Moreover, the means for equalizing the
data blocks of the data receiver may be configured to encode the decoded first data block
or second data block so as to obtain an encoded first data block or second data block.
Equalization for a third data block may be performed on the basis of the encoded first data

block if the first data block in the data packet has a smaller temporal distance from the
third data block than does the second data block. Alternatively, equalization for the third
data block may be performed on the basis of the encoded second data block if the second
data block in the data packet has a smaller temporal distance from the third data block than
does the first data block. Also, equalization for a fourth data block may be performed on
the basis of the encoded first data block if the first data block in the data packet has
smaller temporal distance from the fourth data block than does the second data block:
Alternatively, equalization for the fourth data block may be performed on the basis of the
encoded second data block if the second data block in the data packet has a smaller
temporal distance from the fourth data block than does the first data block. In addition, the
third data block and the fourth data block may be derived from a second base data blocks
the third data block in the data packet being located at a data packet end, and the fourth
data block in the data packet being located at a data packet start. Equalization for the third
data block 120 and the fourth data block 122 may also be effected while using a frequency
estimation, phase estimation or channel estimation.
Furthermore, an equalized third data block 172 and an equalized fourth data block 174 may
be decoded so as to obtain a decoded third data block 176 and a decoded fourth data block
178. The decoded third data block 176 and the decoded fourth data block 178 may
subsequently be encoded so as to obtain an encoded third data block 180 and an -encoded
fourth data block 182. The means 152 for equalizing the data blocks may be configuredrtd
perform equalization for a fifth data block (not shown) on the basis of the encoded third
data block 180 and to perform equalization for a sixth data block (not shown) on the basis
of the encoded fourth data block 182.
Equalization of a third data block 120 may be effected in accordance with the embodiment!
shown in Fig. 4a or 4b, of a data packet 102. The means 152 for equalizing the data blocks
may be configured to perform equalization for the third data block 120 on the basis of the
encoded first data block 168 if the first data block 108 in the data packet 102 has a-smaller
temporal distance from the third data block 120 than does the second data block 110, onto
perform equalization for the third data block 120 on the basis of the encoded second data
block 170 if the second data block 110 in the data packet 102 has a smaller temporal
distance from the third data block 120 than does the first data block 108, the third data
block 120 in the data packet 102 being located at a data packet start or data packet end.
Equalization for the third data block 120 may be effected while using a frequency;
estimation, phase estimation or channel estimation.
The inventive concept shall be described once again in summary below.

The present invention addresses synchronization and equalization of channel-coded data
packets 102 in a radio transmission system, e.g., with code combining. In a radio
transmission wherein the transmission reliability is to be increased with the aid of code
combining in that redundant information in several different data packets 102 is
transmitted at different times, it is necessary to detect the individual data packets 102 eve
when the signal-to-noise ratio (SNR) is very low and does not suffice for decoding the
individual packet or data packet 102. Depending on the code gain due to the combination
of several receive packets or data packets 102, the required signal-to-noise ratio at the
receiver or data receiver 150 at which the data can still be detected decreases. So that this
code gain can be realized, it must be possible to find and even partly (even if erroneously;
for example) decode the individual data packets 102 at this low signal-to-noise ratio. For
finding and partly decoding the individual data packets 102, synchronization and
equalization are therefore necessary. Embodiments of the invention therefore describe a
method of synchronizing individual burst-type transmission packets or data packets 102i:
a radio transmission system (e.g. with code combining) without it being mandatory to be
able to fully decode the individual packet or data packet 102.
For synchronization, the reference sequence or training sequence is split up into two parts
112 and 114 which are different in size and are separated by a data block 110. In addition^
a first base data block 124 is equalized with the aid of the training sequence and/or
reference sequence, wherein, additionally, a frequency and phase estimation may be
effected by the training sequence and/or reference sequence 112 and 114. Further base data
blocks are equalized with the aid of the previously received re-encoded base data block, ^
being possible for a frequency and phase estimation to be effected by the correctly received
data.
For synchronization in the receiver or data receiver 150, a training sequence and/or
reference sequence is typically used in the transmission of a burst-type transmit packet^
data packet. In the present invention, the training sequence and/or reference sequence is
structured as is shown in Fig. 6 and as will be described below. The training sequence or
reference sequence is subdivided, e.g., into three identical subsequences. Said
subsequences are, e.g., maximum length sequences (ML sequences) 130a to 130c havin6
the length (2m-l), which are extended by one bit (zero bit) 132a to 132c. They are not
transmitted successively, but in two fragments 112 and 114. For example, two ML
sequences 130a and 130b of identical length are combined, and are separated from the
third ML sequence 130c of identical length by a data block 110 and/or by part of the first
base data block 124. The second part of the first base data block 124 is arranged on ftie

other side of the pair of ML sequences, so that the base data block 124 will be located to
the left and right of the ML sequence tuple 130a and 130b. Subdivision of the base data
block is (ideally) effected, e.g., such that it yields two parts identical in size. Equivalent
subdivision is not mandatory.
Extension of the ML sequences 130a to 130c by one bit was effected since the bits are
modulated, e.g., with a (2An)-stage modulation, for example with a (non-differential) MSK
modulation. If a sequence consists of an even number of bits, that bit which follows said
sequence will be mapped to the same axis of the constellation diagram as the first bit of the
first sequence. By means of the even number of bits in the first training sequence and/on
reference sequence 112 and 114 as well as of the data block 110 located between the
training sequences and/or reference sequences 112 and 114, one achieves that all of the
training sequences and/or reference sequences 112 and 114 have the same constellation
points following modulation. The complex baseband representation therefore contain*
three identical training sequence portions and/or reference sequence portions.
For simple implementation into, e.g., a microcontroller, the length of the data blocks or
base data blocks which belong together is selected to be, e.g., an integer multiple of 8 bits.
The extension of the ML sequences by 1 bit correspondingly results in that e.g. straining
sequence is exactly 4 bytes in length.
Due to the subdivision of the training sequences and/or reference sequences into two
pieces separated from data or from a data block 110, secondary peaks, which are smaller i~
terms of amount as compared to three successive training sequences, are yielded in the
synchronization. Additionally, the secondary peaks have the same amount. Without the last
training sequence 114, secondary peaks of increasing sizes would arise, as may be seen in
Fig. 3b. Fig. 3a, by contrast, shows a schematic representation of the correlation function
peaks with a separated third training sequence 114. One can see that all of the secondary
peaks Nl to N6 have the same amplitudes. The main peak H has three times the height
relative to the secondary peaks Nl to N6. By means of this arrangement, the main peak H
in the receiver or data receiver 150 may easily be distinguished or separated from the
secondary peaks Nl to N6 by setting a threshold value.
Due to the use of three ML sequences of the length (2m-l), one can also generate resulting
training sequences and/or reference sequences 112 and 114 having lengths not divisible by
2. If, for example, a sequence which is longer than 127 bits (corresponding to 27-l) is used,
the next ML sequence in length would already be 255 bits (corresponding to>28-l) lift
length. Utilization of 3 ML sequences of the length 63 (corresponding to 26-l) results ir#a

training sequence and/or reference sequence of the length 189, i.e., a length which is halft
way between both lengths of ML sequences. Due to the subdivision of the sequence, same
behaves similarly to an imagined ML sequence of the length of 3*(2m-l) with a correlation
peak and small secondary values.
In the inventive approach, the training sequence is subdivided so as to exploit it
simultaneously for channel equalization of the first base data block 124. Since the training
sequences and/or reference sequences 112 and 114 are spaced apart, frequency estimation
may be performed with a smaller estimation error. The frequency and phase estimation
may be performed separately for both parts of the first base data block 124, i.e. forsthe fixst
data block 108 and the second data block 110.
The phase estimation for the first data block 108 may be averaged over the first ML
sequence 130a and the second ML sequence 130b. Thus, a long known sequence 112 h
available for the first phase estimation, whereby a higher estimation accuracy may be
achieved. The estimation for the second data block 110 is effected while using the second
ML sequence 130b and the third ML sequence 130c. Thus, 2 training sequences and/or
reference sequences are available also for this phase estimation, whereby the estimation
accuracy can be increased. For decoding further base data blocks n, which are divided and
are joined, on the left and right, respectively, to the transmit telegraph, no further training
sequence is used for equalization as well as frequency and phase estimation. For equalizing
the nth base data block, the information of the correctly coded base data block n-lis used.
Said information is re-encoded and compared to the receive sequence. The channel
equalizations and the frequency and phase shift may be determined from the difference
between the encoded information and the receive sequence. The individual data blocks are1
encoded by themselves and may be used, following decoding and re-encoding, again as
references for equalizing further data blocks in the data packet 102. Thus, there is the
possibility of decoding the data blocks while starting at the ML sequence!* and * ctf
subsequently re-encoding same. The n-coded blocks may be used as reference data Ha*
channel estimation, frequency estimation, phase estimation or SNR estimation.
Neighboring data blocks may be decoded with the time-variable parameters. This design of
the slots enables tracking of the parameter estimation over time and, thus, adaptation to ■■
time-variable channel. Due to this data packet design, known training sequences and/or
reference sequences within the data blocks that are required specifically for parameter
estimation are not necessary. Error propagation during decoding and re-encoding of a data
packet 102 with subsequent parameter estimation has no repercussions in this system since
only error-free data packets 102 are processed further. If a data packet 102 cannot fei
decoded in a completely error-free manner, this data packet 102 may be combined, wHfen

using code combining, with a different data packet 102 of the same transmitter or daft
transmitter 100 that was sent out, e.g., at a later point in time (code combining). With these
combined data packets 102, equalization and decoding may then also be effected in ah
iterative manner.
Additionally, it is possible for data blocks to be coded differently. For example, data
blocks which are located closer to the training sequence and/or reference sequence might
be put under higher protection so as to correctly decode said data without code combining
with a higher probability than the outer data. This may be exploited for correctly decoding
important information as early as at reception of a data packet 102 without having to wait
for a further data packet 102 in order to be able to perform code combining. Due to the
specific subdivision of the training sequence, the method described enables achieving both
good phase and frequency estimations for channel equalization while obtaining good
correlation properties therewith so as to detect data packets 102, in particular with radi
transmission systems with code combining, from the noise, even if they are only partly
decodable. Iterative decoding enables iterative tracking the phase and frequency
estimations while starting from the training sequence in the center of the data packet 102 in
and outward manner. To this end, the actual data is used as the training sequence.
Even though some aspects have been described within the context of a device, it is
understood that said aspects also represent a description of the corresponding method, sb
that a block or a structural component of a device is also to be understood as a
corresponding method step or as a feature of a method step. By analogy therewith, aspect
that have been described in connection with or as a method step also represent a
description of a corresponding block or detail or feature of a corresponding device. Some
or all of the method steps may be performed while using a hardware device, such as a
microprocessor, a programmable computer or an electronic circuit?. In some embodiments,
some or several of the most important method steps may be performed by such a device.
Depending on specific implementation requirements, embodiments of the invention may
be implemented in hardware or in software. Implementation may be effected while using a
digital storage medium, for example a floppy disc, a DVD, a Blu-ray disc, a CD, a ROM,
PROM, an EPROM, an EEPROM or a FLASH memory, a hard disc or any other magnetic
or optical memory which has electronically readable control signals stored thereon which
may cooperate, or cooperate, with a programmable computer system such that the
respective method is performed. This is why the digital storage medium may be computer-
readable.

Some embodiments in accordance with the invention thus comprise a data carrier which
comprises electronically readable control signals that are capable of cooperating with a
programmable computer system such that any of the methods described herein is
performed. Generally, embodiments of the present invention may be implemented as a
computer program product having a program code, the program code being effective to
perform any of the methods when the computer program product runs on a computer.
The program code may also be stored on a machine-readable carrier, for example. Other
embodiments include the computer program for performing any of the methods described
herein, said computer program being stored on a machine-readable carrier.
In other words, an embodiment of the inventive method thus is a computer program which
has a program code for performing any of the methods described herein, when the
computer program runs on a computer. A further embodiment of the inventive methods
thus is a data carrier (or a digital storage medium or a computer-readable medium) on
which the computer program for performing any of the methods described herein is
recorded.
A further embodiment of the inventive method thus is a data stream or a sequence of
signals representing the computer program for performing any of the methods described
herein. The data stream or the sequence of signals may be configured, for example, to be
transferred via a data communication link, for example via the internet. A further
embodiment includes a processing means, for example a computer or a programmable
logic device, configured or adapted to perform any of the methods described herein. A
further embodiment includes a computer on which the computer program for performing
any of the methods described herein is installed.
A further embodiment in accordance with the invention includes a device or a system
configured to transmit a computer program for performing at least one of the methods
described herein to a receiver. The transmission may be electronic or optical, for example.
The receiver may be a computer, a mobile device, a memory device or a similar device, for
example. The device or the system may include a file server for transmitting the computer
program to the receiver, for example.
In some embodiments, a programmable logic device (for example a field-programmable
gate array, an FPGA) may be used for performing some or all of the functionalities of the
methods described herein. In some embodiments, a field-programmable gate array may
cooperate with a microprocessor to perform any of the methods described herein;

Generally, the methods are performed, in some embodiments, by any hardware, device.
Said hardware device may be any universally applicable hardware such as a computer
processor (CPU), or may be a hardware specific to the method, such as an ASIC.
The above-described embodiments merely represent an illustration of the principles of the
present invention. It is understood that other persons skilled in the art will appreciate any
modifications and variations of the arrangements and details described herein. This is why
the invention is intended to be limited only by the scope of the following claims rather than
by the specific details that have been presented herein by means of the description and the
discussion of the embodiments.

We Claim:
1. Data transmitter (100) for transmitting a data packet (102) to a data receiver via a
communication channel, comprising:
a means (104) for generating the data packet (102) having a first data block (108)
and a second data block (110) and a predefined first reference sequence (112) and
second reference sequence (114) for synchronizing the data packet in the data
receiver, wherein the first reference sequence (112) is longer than the second
reference sequence (114), and wherein in the data packet, the second data block
(110) is located between the first reference sequence (112) and the second reference
sequence (114), and the first reference sequence (112) is located between the first
data block (108) and the second data block (110); and
a means (106) for transmitting the data packet (102) to the data receiver via the
communication channel;
wherein the means (104) for generating the data packet (102) is configured to
provide, in the data packet (102), the first reference sequence (112) to be double the
length of the second reference sequence (114); and
wherein the means (104) for generating the data packet (102) is configured to
subdivide, in the data packet (102), the first reference sequence (112) into a first
reference subsequence (112a) and a second reference subsequence (112b), the first
reference subsequence (112a) and the second reference subsequence (112b) each
having the length of the second reference sequence (114).
2. Data transmitter (100) as claimed in claim 1, wherein the means (104) for
generating the data packet (102) is configured to generate a data packet (102)
wherein the first reference subsequence (112a), the second reference subsequence
(112b) and the second reference sequence (114) are identical.
3. Data transmitter (100) as claimed in any of the previous claims, wherein the means
(104) for generating the data packet (102) is configured to provide, in the data
packet (102), the first reference sequence (112) and the second reference sequence
(114) as pseudorandom sequences or ML sequences.

4. Data transmitter (100) as claimed in claim 2, wherein the means (104) for
generating the data packet (102) is configured to provide, in the data packet (102),
the first reference subsequence (112a), the second reference subsequence (112b),
and the second reference sequence (114) as ML sequences (130), respectively, each
having an additional binary element (132).
5. Data transmitter (100) as claimed in any of the previous claims, wherein the means
(104) for generating the data packet (102) is configured to provide a third data
block (120) in the data packet (102), the third data block (120) in the data packe
(102) being located at a data packet start or a data packet end.
6. Data transmitter (100) as claimed in any of the previous claims, wherein the means
(104) for generating the data packet (102) is configured to derive the first data
block (108) and the second data block (110) of the data packet (102) from a first
base data block (124).
7. Data transmitter (100) as claimed in claim 6, wherein the means (104) for
generating the data packet (102) is configured to derive a third data block (120) and
a fourth data block (122) from a second base data block (126) and to provide the
third data block (120) and the fourth data block (122) in the data packet (102), the
third data block (120) in the data packet (102) being located at a data packet end,
and the fourth data block (122) in the data packet (102) being located at a data
packet start.
8. Data receiver (150) for receiving a data packet (102) from a data transmitter via a
communication channel, the data packet (102) comprising a first data block (108)
and a second data block (110) and a predefined first reference sequence (112) and
second reference sequence (114) for synchronizing the data receiver (150), wherein,
in the data packet (102), the first data block (110) is located between the first
reference sequence (112) and the second reference sequence (114), and the first
reference sequence (112) is located between the first data block (108) and the
second data block (110), the first reference sequence (112) being double the length
of the second reference sequence (114), and the first reference sequence (112)
being subdivided into a first reference subsequence (112a) and a second reference
subsequence (112b), the first reference subsequence (112a) and the second
reference subsequence (112b) each having the length of the second referenc
sequence (114), comprising

a means (154) for receiving the data packet (102), said means being configured to
localize the first reference sequence (112) and the second reference sequence (114)
of the data packet (102) in the receive data stream and to determine the data packet
(102) on the basis of a determinable transmission parameter which may be derived
from the first reference sequence (112) and from the second reference sequence
(114).
9. Data receiver (150) as claimed in claim 8, wherein the means (154) for receiving
the data packet (102) is configured to correlate the receive data stream with the first
reference sequence (112) and with the second reference sequence (114), which are
known to the data receiver (150), so as to localize the first reference sequence (112)
and the second reference sequence (114) of the data packet (102) in the receive data
stream.
10. Data receiver (150) as claimed in claim 8 or 9, wherein the means (154) for
receiving the data packet (102) is configured to determine the transmission
parameter on the basis of a frequency estimation or a phase estimation.
11. Data receiver (150) as claimed in any of claims 8 to 10, further comprising a means
(152) for equalizing the data blocks of the data packet (102), said means being
configured to perform equalization for the first data block (108) on the basis of the
first reference sequence (112) so as to obtain an equalized first data block (160),
and to perform equalization for the second data block (110) on the basis of the first
reference sequence (112) or of a reference subsequence adjacent to the second data
block (110) and of the second reference sequence (114) so as to obtain an equalized
second data block (162).
12. Data receiver (150) as claimed in claim 11, wherein the means (152) for equalizing
the data blocks is configured to perform the equalization for the first data block
(108) and the second data block (110) while using a frequency estimation, phase
estimation or channel estimation.
13. Data receiver (150) as claimed in claim 11 or 12, the means (152) for equalizing the
data blocks being configured to decode the equalized first data block (160) and
second data block (162) so as to obtain a decoded first data block (164) and second
data block (166).

14. Data receiver (150) as claimed in any of claims 11 to 13, the means (152) for
equalizing the data blocks being configured to encode the decoded first data block
(164) or second data block (166) so as to obtain an encoded first data block (168) or
second data block (170).
15. Data receiver (150) as claimed in any of claims 11 to 14, the means (152) for
equalizing the data blocks being configured to perform equalization for a third data
block (120) on the basis of the encoded first data block (168) if the first data block
(108) in the data packet (102) has a smaller temporal distance from the third data
block (120) than does the second data block (110), or to perform equalization for
the third data block (120) on the basis of the encoded second data block (170) if the
second data block (110) in the data packet (102) has a smaller temporal distance
from the third data block (120) than does the first data block (108), the third data
block (120) being located, in the data packet (102), at a data packet start or data
packet end.
16. Data receiver (150) as claimed in claim 15, wherein the means (152) for equalizing
the data blocks is configured to perform the equalization for the third data block
(120) while using a frequency estimation, phase estimation or channel estimation.
17. Data receiver (150) as claimed in any of claims 11 to 14, wherein the means (152)
for equalizing the data blocks is configured to perform equalization for a first data
block (108) and a second data block (110), the first data block (108) and the second
data block (110) being derived from a first base data block (124).
18. Data receiver (150) as claimed in claim 17, the means (152) for equalizing the data
blocks being configured to perform equalization for a third data block (120) on the
basis of the encoded first data block (168) if the first data block (108) in the data
packet (102) has a smaller temporal distance from the third data block (120) than
does the second data block (110), or to perform equalization for the third data block
(120) on the basis of the encoded second data block (170) if the second data block
(110) in the data packet (102) has a smaller temporal distance from the third data
block (120) than does the first data block (108), and to perform equalization for a
fourth data block (122) on the basis of the encoded first data block (168) if the first
data block (108) in the data packet (102) has a smaller temporal distance from the
fourth data block (122) than does the second data block (110), or to perform
equalization for the fourth data block (122) on the basis of the encoded second data
block (170) if the second data block (110) in the data packet (102) has a smaller

temporal distance from the fourth data block (122) than does the first data block
(108), the third data block (120) and the fourth data block (122) being derived from
a second base data block (126), the third data block (120) in the data packet (102)
being located at a data packet end, and the fourth data block (122) in the data
packet (102) being located at a data packet start.
19. Data receiver (150) as claimed in claim 18, wherein the means (152) for equalizing
the data blocks is configured to perform equalization for the third data block (120)
and the fourth data block (122) while using a frequency estimation, phase
estimation or channel estimation.
20. Method of transmitting a data packet to a data receiver via a communication
channel, comprising:
generating the data packet having a first data block and a second data block and a
predefined first reference sequence and second reference sequence for
synchronizing the data receiver, wherein the first reference sequence is longer than
the second reference sequence, and wherein in the data packet, the first data block
is located between the first reference sequence and the second reference sequence,
and the first reference sequence is located between the first data block and the
second data block, the first reference sequence being double the length of the
second reference sequence, and the first reference sequence being subdivided into a
first reference subsequence and a second reference subsequence, the first reference
subsequence and the second reference subsequence each having the length of the
second reference sequence; and
transmitting the data packet to the data receiver via the communication channel.
21. Method of receiving a data packet from a data transmitter via a communication
channel, the data packet comprising a first data block and a second data block and a
predefined first reference sequence and second reference sequence for
synchronizing the data receiver, wherein, in the data packet, the first data block is
located between the first reference sequence and the second reference sequence,
and the first reference sequence is located between the first data block and the
second data block, the first reference sequence being double the length of the
second reference sequence, and the first reference sequence being subdivided into a
first reference subsequence and a second reference subsequence, the first reference

subsequence and the second reference subsequence each having the length of the
. second reference sequence, comprising:
localizing the first reference sequence and the second reference sequence in the
receive data stream; and
determining the data packet in the receive data stream on the basis of a
determinable transmission parameter which may be derived from the first reference
sequence and from the second reference sequence.
22. Method of receiving a data packet as claimed in claim 21, wherein during
localization of the first reference sequence and of the second reference sequence,
the receive data stream is correlated with the first reference sequence and the
second reference sequence, which are known to the data receiver, so as to localize
the first reference sequence and the second reference sequence of the data packet in
the receive data stream.
23. Method of receiving a data packet as claimed in claim 21 or 22, wherein during
determination of the data packet, the transmission parameter is determined on the
basis of a frequency estimation or a phase estimation.
24. Computer program for performing any of the methods as claimed in any of claims
20 to 23, when the computer program runs on a computer or microprocessor.