DE102008064381A1 - Method for coding data signal in orthogonal frequency-division multiplexing based system, involves splitting two specified number of data, where one data has even index and another data has odd index - Google Patents

Abstract

The method involves splitting two specified number of data, where one data has an even index and another data has an odd index. The former data is transformed by inverse discrete Fourier transform at a certain signal value, where the latter data is transformed by inverse discrete Fourier transform at another signal value. Independent claims are also included for: (1) a method for reconstructing a data signal; (2) a device for coding a data signal; and (3) a device for reconstructing a data signal.

Description

The The invention relates to methods and apparatus for coding and Reconstruct a data signal in an OFDM-based system.

One OFDM (OFDM - Orthogonal Frequency Division Multiplexing) Data transfer system will be available today used in a variety of applications. Exemplary applications are WLAN based on IEEE 812.11 (WLAN - Wireless Local Area Network, IEEE institutes of Electrical and Electronics Engineers), DVB-T (DVB-T Digital Video Broadcasting Terrestrial) and DAB (DAB - Digital Audio Broadcasting).

Out A document [1] is known that the transmission technology OFDM very sensitive to frequency errors is due to frequency differences between each local oscillators in a transmitter and receiver. From a document [2] goes according to chapter II.A shows that in OFDM systems three synchronization problems may occur, which the symbol synchronization, the carrier frequency synchronization and affect the sample clock synchronization.

Consequently it is the object of the present invention methods and devices for coding or reconstructing a data signal in an OFDM-based System indicating a reduction of a signal disturbance due to a carrier frequency shift achieve.

These Task is by the independent claims solved. Further developments of the present application are the dependent claims remove.

• – Aufteilen der N Daten in erste Daten und zweite Daten, wobei die ersten Daten die Daten mit einem geradzahligen Index und die zweiten Daten die Daten mit einem ungeradzahligen Index umfassen;
• – Transformieren der ersten Daten mit einer N/2-dimensionalen inversen diskreten Fourier Transformation in erste Signalwerte und Transformieren der zweiten Daten mit einer N/2-dimensionalen inversen diskreten Fourier Transformation in zweite Signalwerte;
• – Zuordnen der ersten Signalwerte zu ersten empfangenen Signalwerten und Zuordnen der zweiten Signalwerte zu zweiten empfangenen Signalwerten;
• – Generieren von N/2 ersten Hilfssignalwerten durch jeweilige Subtraktion des zweiten empfangenen Signalwerts von dem ersten empfangenen Signalwert, wobei jeweils die empfangenen Signalwerte mit einem identischen Index subtrahiert werden;
• – Generieren von N/2 zweiten Hilfssignalwerten durch jeweilige Addition des zweiten empfangenen Signalwerts und des ersten empfangenen Signalwerts, wobei jeweils die empfangenen Signalwerte mit einem identischen Index addiert werden;
• – Erzeugen von Hilfssignalwerten durch Zuordnen der ersten Hilfssignalwerte den ersten N/2 Hilfssignalwerten und durch Zuordnen der zweiten Hilfssignalwerte den zweiten N/2 Hilfssignalwerten;
• – Generieren von empfangenen Daten durch Transformation der N Hilfssignalwerte mittels einer N-dimensionalen diskreten Fourier Transformation.

The invention relates to a method for encoding a data signal with N data,
where the following steps are performed:

• Dividing the N data into first data and second data, wherein the first data comprises the data with an even-numbered index and the second data comprises the data with an odd-numbered index;
• Transforming the first data with an N / 2-dimensional inverse discrete Fourier transform into first signal values and transforming the second data with an N / 2-dimensional inverse discrete Fourier transform into second signal values;
• Assigning the first signal values to first received signal values and assigning the second signal values to second received signal values;
• Generating N / 2 first auxiliary signal values by respectively subtracting the second received signal value from the first received signal value, wherein each of the received signal values is subtracted with an identical index;
• Generating N / 2 second auxiliary signal values by respective addition of the second received signal value and the first received signal value, wherein the received signal values are respectively added with an identical index;
• Generating auxiliary signal values by associating the first auxiliary signal values with the first N / 2 auxiliary signal values and by assigning the second auxiliary signal values to the second N / 2 auxiliary signal values;
• Generation of received data by transformation of the N auxiliary signal values by means of an N-dimensional discrete Fourier transformation.

By The present method allows data obtained from transmit a transmitter to a receiver should be less susceptible to interference a carrier frequency shift are, as the methods known today. Another advantage of Method is to see that this without increased computational effort across from can be implemented and executed by known methods. As data, each Understood kind of information that is presented in digital form can be which data may be encoded and uncoded. For example, the Data Voice data encoding according to a coding standard, d. H. compressed, are. In particular, encoded / compressed data especially vulnerable against carrier frequency shift, because of the reduction of redundancy information from the data underlying information is a reconstruction on occurrence is made difficult by mistakes. The invention also increases in coded / compressed Data the quality the reconstructed data. Other types of data that the procedure can process are, for example, image and video data, web content or IP based packets (IP-IETF Internet Protocol).

• – Aufteilen der N Daten in erste Daten und zweite Daten, wobei die ersten Daten die Daten mit einem geradzahligen Index und die zweiten Daten die Daten mit einem ungeradzahligen Index umfassen;
• – Transformieren der ersten Daten mit einer N/2-dimensionalen inversen diskreten Fourier Transformation in erste Signalwerte und Transformieren der zweiten Daten mit einer N/2-dimensionalen inversen diskreten Fourier Transformation in zweite Signalwerte;
• – Zuordnen der ersten Signalwerte zu ersten empfangenen Signalwerten und Zuordnen der zweiten Signalwerte zu zweiten empfangenen Signalwerten;

bei dem folgende Schritte durchgeführt werden:

• – Generieren von N/2 ersten Hilfssignalwerten durch jeweilige Subtraktion des zweiten empfangenen Signalwerts von dem ersten empfangenen Signalwert, wobei jeweils die empfangenen Signalwerte mit einem identischen Index subtrahiert werden;
• – Generieren von N/2 zweiten Hilfssignalwerten durch jeweilige Addition des zweiten empfangenen Signalwerts und des ersten empfangenen Signalwerts, wobei jeweils die empfangenen Signalwerte mit einem identischen Index addiert werden;
• – Erzeugen von Hilfssignalwerten durch Zuordnen der ersten Hilfssignalwerte den ersten N/2 Hilfssignalwerten und durch Zuordnen der zweiten Hilfssignalwerte den zweiten N/2 Hilfssignalwerten;
• – Generieren von empfangenen Daten durch Transformation der N Hilfssignalwerte mittels einer N-dimensionalen diskreten Fourier Transformation.

The invention also comprises a method for reconstructing a data signal with N data in the form of received data from first received signal values and from second received signal values, wherein the first and second received signal values were generated by the following steps:

• Dividing the N data into first data and second data, wherein the first data includes the data with a ge Radar index and the second data include the data with an odd index;
• Transforming the first data with an N / 2-dimensional inverse discrete Fourier transform into first signal values and transforming the second data with an N / 2-dimensional inverse discrete Fourier transform into second signal values;
• Assigning the first signal values to first received signal values and assigning the second signal values to second received signal values;

where the following steps are performed:

• Generating N / 2 first auxiliary signal values by respectively subtracting the second received signal value from the first received signal value, wherein each of the received signal values is subtracted with an identical index;
• Generating N / 2 second auxiliary signal values by respective addition of the second received signal value and the first received signal value, wherein the received signal values are respectively added with an identical index;
• Generating auxiliary signal values by associating the first auxiliary signal values with the first N / 2 auxiliary signal values and by assigning the second auxiliary signal values to the second N / 2 auxiliary signal values;
• Generation of received data by transformation of the N auxiliary signal values by means of an N-dimensional discrete Fourier transformation.

The Reconstruction method is characterized in that this can be used in a receiver unit is. The advantage and the fields of application are analogous to the method for coding.

Further Part of the invention is an apparatus for coding a data signal with N data that includes the following units:

• - First Unit for dividing the N data into first data and second data where the first data is the data with an even-numbered index and the second data comprises the data with an odd index;
• - Second Unit for transforming the first data with an N / 2-dimensional inverse discrete Fourier transform into first signal values and to transform the second data with an N / 2-dimensional one inverse discrete Fourier transform into second signal values;
• - Third Unit for allocating the first signal values to first received ones Signal values and for assigning the second signal values to second received signal values;
• - Fourth A unit for generating N / 2 first auxiliary signal values by respective ones Subtracting the second received signal value from the first received Signal value, wherein in each case the received signal values with a identical index to be subtracted;
• - Fifth unit for generating N / 2 second auxiliary signal values by respective ones Addition of the second received signal value and the first received Signal value, wherein each of the received signal values with a identical index are added;
• - Sixth A unit for generating auxiliary signal values by assigning the first one Auxiliary signal values to the first N / 2 auxiliary signal values and by assigning the second auxiliary signal values the second N / 2 auxiliary signal values;
• - Seventh Unit for generating received data by transformation of the N auxiliary signal values by means of an N-dimensional discrete Fourier transformation.

The Encoder is an implementation of the method for coding, the advantages and the fields of application of this Device analogous to the method of coding.

• – Aufteilen der N Daten in erste Daten und zweite Daten, wobei die ersten Daten die Daten mit einem geradzahligen Index und die zweiten Daten die Daten mit einem ungeradzahligen Index umfassen;
• – Transformieren der ersten Daten mit einer N/2-dimensionalen inversen diskreten Fourier Transformation in erste Signalwerte und Transformieren der zweiten Daten mit einer N/2-dimensionalen inversen diskreten Fourier Transformation in zweite Signalwerte;
• – Zuordnen der ersten Signalwerte zu ersten empfangenen Signalwerten und Zuordnen der zweiten Signalwerte zu zweiten empfangenen Signalwerten;

die folgende Einheiten umfasst:

• – Vierte Einheit zum Generieren von N/2 ersten Hilfssignalwerten durch jeweilige Subtraktion des zweiten empfangenen Signalwerts von dem ersten empfangenen Signalwert, wobei jeweils die empfangenen Signalwerte mit einem identischen Index subtrahiert werden;
• – Fünfte Einheit zum Generieren von zweiten Hilfssignalwerten durch jeweilige Addition des zweiten empfangenen Signalwerts und des ersten empfangenen Signalwerts, wobei jeweils die empfangenen Signalwerte mit einem identischen Index addiert werden;
• – Sechste Einheit zum Erzeugen von Hilfssignalwerten durch Zuordnen der ersten Hilfssignalwerte den ersten N/2 Hilfs signalwerten und durch Zuordnen der zweiten Hilfssignalwerte den zweiten N/2 Hilfssignalwerten;
• – Siebte Einheit zum Generieren von empfangenen Daten durch Transformation der N Hilfssignalwerte mittels einer N-dimensionalen diskreten Fourier Transformation (N-DFT).

In addition, the invention comprises an apparatus for reconstructing a data signal with N data in the form of received data from first received signal values and from second received signal values, the first and second received signal values being generated as follows:

• Dividing the N data into first data and second data, wherein the first data comprises the data with an even-numbered index and the second data comprises the data with an odd-numbered index;
• Transforming the first data with an N / 2-dimensional inverse discrete Fourier transform into first signal values and transforming the second data with an N / 2-dimensional inverse discrete Fourier transform into second signal values;
• Assigning the first signal values to first received signal values and assigning the second signal values to second received signal values;

the following units include:

• - fourth unit for generating N / 2 first auxiliary signal values by respectively subtracting the second received signal value from the first received signal value, wherein in each case the received signal values are subtracted with an identical index;
• - fifth unit for generating second auxiliary signal values by respective addition of the second received signal value and the first received signal value, wherein the received signal values are respectively added with an identical index;
• - Sixth unit for generating auxiliary signal values by assigning the first auxiliary signal values to the first N / 2 auxiliary signal values and by assigning the second auxiliary signal values to the second N / 2 auxiliary signal values;
• Seventh unit for generating received data by transforming the N auxiliary signal values by means of an N-dimensional discrete Fourier transformation (N-DFT).

The Apparatus for reconstructing the data signal provides an implementation of the method for reconstructing. Advantages and applications are analogous to the method of reconstruction.

Finally is a use according to one of the preceding methods or according to one of the preceding devices part of the invention, wherein These methods and apparatus for transmitting data in one Telecommunication system, in particular for voice transmission, or in a data transmission system, in particular for transmission of data in the Inter- and / or Intranet are usable. Especially at the use of the methods and devices in the field of data transmission result in particular advantage, in particular by the method or Devices a signal quality of reconstructed data at carrier frequency shift elevated.

The The invention and its developments are explained in more detail with reference to figures. It show in detail:

1 an OFDM-based transmission method according to the prior art;

2 a comparison of the signal-to-interference ratio of the prior art method and the method according to one of the embodiments of the invention;

3 an embodiment of the invention;

4A / B Comparison of signal spectra of the prior art method and the method according to one of the embodiments of the invention.

elements with the same function and mode of action are in the figures with provided the same reference numerals.

mit k = 0, ..., N – 1, j als imaginäre Einheit und π als Kreiszahl. Anschließend werden die diskreten transformatierten Daten dd(.) einem Digital-Analog-Wandler DAC (DAC – Digital-to-Analog Converter) mit einem Interpolationsfilter h_DAC(t) und einer Sampling Frequenz f_s (= Abtastfrequenz) in ein zeitkontinuierliches Signal ss(t) für den Datenblock gewandelt. Der Parameter t beschreibt dabei die Zeitkonstante. Das zeitkontinuierliche Signal ss(t) wird gemäß Gleichung (2) ermittelt:

Based on 1 For example, the prior art OFDM (OFDM - Orthogonal Frequency Division Multiplexing) transmission system will transmit N data c (0), c (N-1) of a data block, ie OFDM blocks, of a data stream (DS) from a transmitter SEN via a channel CHA to a receiver EMP described in more detail. Received data d (0),..., D (N-1), ie the data reconstructed after the transmission, which are available on the basis of intercarrier interference (ICI) and / or intersymbol interference (ISI) from the Data c (.) May differ. In the specification and claims, variables and reference numerals occurring as a vector, such as the data c (0), ..., c (N-1), are also written in abbreviated notation as c (.). A block with the N data c (.) Is encoded as discrete transformed data dd (0),..., Dd (N-1) using discrete N-complex subcarriers in the OFDM baseband, the coding according to equation (1) can be represented in the form of an inverse discrete Fourier transformation (IDFT):

with k = 0, ..., N - 1, j as imaginary unit and π as circle number. Subsequently, the discrete transformed data dd (.) A digital-to-analog converter DAC (DAC - Digital-to-Analog Converter) with an interpolation filter h _DAC (t) and a sampling frequency f _s (= sampling frequency) in a continuous-time signal ss (t) converted for the data block. The parameter t describes the time constant. The time-continuous signal ss (t) is determined according to equation (2):

The time-continuous signal ss (t), ie the complex signal, is then at a carrier frequency f ₀ , z. F ₀ = 2.4 GHz, modulated to give a signal xs (t) to be transmitted:

den Realanteil und

den Imaginäranteil des jeweiligen Arguments.This designates

the real part and

the imaginary part of the respective argument.

In the transmission of the signal to be transmitted xs (t) this is transformed into a received signal xe (t) due to an impulse response h _Ch (t) of the channel CHA. xe (t) = xs (t) ⊗ h ch (t) (4)

The Symbol ⊗ represents a convolution operation.

In the receiver EMP, a demodulation of the high-frequency received signal xe (t) with the carrier frequency f _{0 is then to be} carried out. In practice, however, the respective carrier frequencies f ₀ in the transmitter and receiver are slightly different, resulting in a frequency shift. Thus, the carrier frequency is f ₀ f assumed as the estimated carrier frequency fs ₀ = ₀ + ₀ .DELTA.f on the side of the receiver, wherein .DELTA.f ₀ represents the frequency shift. It is assumed that the frequency shift over the entire OFDM block is constant with the N data. A continuous-time received signal se (t) thus results in: se (t) = 1 2 [ss (t) ⊗ h ch (T) exp (j2πf 0 t)] exp (j2πΔf 0 t) (5)

wobei

als diskrete empfangene Daten mit l = 0, ..., N – 1 bezeichnet wird. Ferner beschreibt h(t) eine Impulsantwort des gesamten Übertragungssystems gemäß h(t) = 12 hDAC(t) ⊗ hCH(t)exp(j2πf0t) ⊗ hADC(t) (7)wobei ein Dezimationsfilter durch die Impulsantwort des Analog-Digital-Wandlers h_ADC(t) berücksichtigt wird. In einem nachfolgenden Schritt werden die diskreten empfangenen Daten de(.) einer N-dimensionalen diskreten Fourier Transformation DFT (DFT – Discrete Fourier Transformation) unterzogen. Am Ausgang der diskreten Fourier Transformation werden die empfangenen Daten d(0), ..., d(N – 1) ausgegeben. Die Gleichungen (8a) und (8b) zeigen diesen Verarbeitungsschritt.

wobei β = NΔf₀/f_s repräsentiert und m = 0, ..., N – 1 ist. Es wird noch hingewiesen, dass in Gleichung (8b) die empfangenen Daten d(.) als Funktion der Daten c(.) dargestellt worden sind. Nach der Entzerrung jedes Zwischenträgers kann die Gleichung (8b) formalisiert geschrieben werden als

mit den Equalizerkoeffizienten E_m,m und mit

und mitAn analog-to-digital converter (ADC) is used to sample the continuous-time received signal at an estimated sampling frequency fs _s = (1 + α) f _s . Here, it is assumed that the parameter α describing a real number is constant over the duration of the OFDM block. The sampled signal form of the continuous-time received signal se (t) results according to equation (6):

in which

is referred to as discrete received data with l = 0, ..., N - 1. Further, h (t) describes an impulse response of the whole transmission system according to FIG h (t) = 1 2 H DAC (t) ⊗ h CH (T) exp (j2πf 0 t) h ADC (t) (7) wherein a decimation filter is taken into account by the impulse response of the analog-to-digital converter h _ADC (t). In a subsequent step, the discrete received data de (.) Are subjected to an N-dimensional discrete Fourier transformation DFT (DFT). At the output of the discrete Fourier transformation, the received data d (0), ..., d (N-1) are output. Equations (8a) and (8b) show this processing step.

where β = NΔf ₀ / f _s and m = 0, ..., N - 1. It should be noted that in equation (8b) the received data d (.) Has been represented as a function of the data c (.). After equalization of each subcarrier, equation (8b) can be written formalized as

with the equalizer coefficients E _{m, m} and with

and with

Equation (8b) can also be expressed in matrix notation as d ~ (.) = c (.) KHE = c (.) A (12) where the diagonal matrixes H and E with the elements H _{m, m} and E _{m, m represent} the influence of the system frequency response H and the equalizer E. The matrix K with the elements K _{n, m} represents the interference of the intermediate carriers (ICI = Inter Carrier Interference). It is noteworthy that ideal synchronization, ie, α = β = 0, reduces the matrix K to the unit matrix. In addition, if the changes in α and β are slow so that they can be tracked by the training algorithm of the equalizer, the perturbations of the ICI can also be represented by the matrix A as follows

For measuring the signal disturbances due to frequency offsets, the signal-to-interference ratio SIR (SIR) per subcarrier is used, which is defined as

In 2 For example, the signal-to-noise ratio SIR is plotted on the ordinate as 10 log ₁₀ (SIR) above the parameter β as the curve labeled MA.

The use of an equalizer is exemplary for a person skilled in the art on page 193 i. V. m. 7.9 from [3], so that will not be discussed further.

With the aid of an embodiment according to 2 the invention will be explained in more detail below. First, the data c (0), ..., c (N-1) are converted into first data d1 (0), ..., d1 (N / 2 - 1) and second data d2 (0), ... , d2 (N / 2 - 1), where the data c (0), c (2), ..., c (N-2) with even-numbered indices are the first data d1 (.) and the data c (1 ), c (3), ..., c (N - 1) with odd-numbered indices are assigned to the second data d2 (.). So d1 (0) = c (0), d1 (1) = c (2), ..., d1 (N / 2 - 1) = c (N - 2), d2 (0) = c (1 ), d2 (1) = c (3), ..., d2 (N / 2 - 1) = c (N - 1). This step is performed by a first unit M1.

In one next Step, which is realized by a second unit M2 will be the first data d1 (.) by an N / 2-dimensional inverse Fourier transform IDFT transformed into first signal values SW1 (0), ..., SW1 (N / 2 - 1). Analogous the second data d2 (.) are converted into second signal values SW2 (0),. SW (N / 2 - 1) transformed.

Analogous to the procedure according to 1 the first and second signal values SW1 (.), SW2 (.) are supplied to the digital-to-analogue converter DAC, which forms from them the associated continuous-time signal ss (t).

According to Equation 15, the time-continuous signal ss (t) of this embodiment results

After the modulation of the continuous-time signal ss (t) with the carrier frequency f ₀ , the transmission of the resulting signal xs (t) to be transmitted via the channel CHA, which has an impulse response h _CH , and after demodulation of the after transmission via the channel received signal xe (t) by means of the estimated carrier frequency fs ₀ results in the continuous-time received signal se (t) analogous to equation 5.

In a subsequent step, the continuous-time received signal se (t) is transformed into discrete received data de (.) By means of the analog-to-digital converter ADC. Since in this embodiment the data c (.) Has been transformed by means of two N / 2-dimensional inverse discrete Fourier transforms N / 2-IDFT units, the discrete received data de (.) Results

The discrete received data de (0), ..., de (N / 2 - 1) represent the first signal values SW1 (0), ..., SW1 (N / 2 - 1) and represent the discrete received data de (N / 2), ..., de (N-1) the second signal values SW2 (0), ..., SW2 (N / 2 - 1). The discrete received Data de (0), ..., de (N / 2 - 1) are used as first received signal values SE1 (0), ..., SE1 (N / 2 - 1) and the discrete received data de (N / 2), ..., de (N-1) as second received signal values SE2 (0), ...., SE2 (N / 2 - 1) in Referred to below.

In a next step, first auxiliary signal values SH1 (0),..., SH1 (N / 2-1) are generated by a fourth unit M4 from the first and second received signal values SE1 (.), SE2 (.). In each case, by subtracting the second received signal value SE2 (IX) from the first received signal value SE1 (IX) in each case, the first auxiliary signal value SH1 (IX) is generated, wherein the first and second received signal values are subtracted with an identical index IX. Mathematically, this is describable as SH1 (IX) = SE1 (IX) - SE2 (IX) with 0 ≤ IX <N / 2 - 1 (17a)

Furthermore, in a further step, a fifth unit M5 generates second auxiliary signal values SH2 (0),..., SH2 (N / 2-1) from the first and second received signal values SE1 (.), SE2 (.). In each case, the respective second auxiliary signal value SH2 (IX) is generated by pairwise adding the second received signal value SE2 (IX) and the first received signal value SE1 (IX), the first and second received signal values being summed with an identical index IX. Mathematically, this is describable as SH2 (IX) = SE1 (IX) + SE2 (IX) with 0 ≤ IX <N / 2 - 1 (17b)

The first auxiliary signals SH1 (.) are orthogonal to the second auxiliary signals SH2 (.).

Subsequently, the first and second auxiliary signal values SH1 (.) And SH2 (.) Are assigned to the auxiliary signal values SH (.). This is done by respectively copying the first auxiliary signal values SH1 (.) To the auxiliary values SH (0),..., SH (N / 2 - 1). The second auxiliary signal values SH2 (.) Are also copied to the auxiliary signal values SH (N / 2),..., SH (N-1). Equation (18) illustrates this

The auxiliary signal values SH (.) Are transferred in the subsequent step to the N-dimensional discrete Fourier transformation unit N-DFT, which reconstructs the received data d (.) From this. Equations (19a) and (19b) represent the received data in algebraic form:

wobei H_m,m gemäß Gleichung (10) ist. Nach der Gewichtung wird die Gleichung (22) zu d ~(.) = c(.)KHE = c(.)A (23)mit

und mit

und

After inserting the equation (16) into the equation (19b), the equation (19b) can be expressed according to the equations (20) and (21) as follows:

where H _{m, m is} according to equation (10). After weighting, equation (22) becomes d ~ (.) = c (.) KHE = c (.) A (23) With

and with

and

If the parameters α and β change slowly, they can be tracked by the Equalizer training algorithm, so that results

in this connection results in the SIR value too

In 2 the SIR value is mapped according to equation (29) in the curve MA. It turns out that with the same parameter β, the presented method according to 3 clearly gives a lower ie better signal-to-interference ratio SIR per carrier than the method known in the art. The use of an equalizer is exemplary for a person skilled in the art on page 193 i. V. m. 7.9 from [3], so that will not be discussed further.

4 shows a comparison of the received data d (.) encoded as QPSK-Quadrature Phase Shift Keying (QPSK) signal for a parameter configuration α = 0 and β = 0.2. In this case, the received data are represented by their real part R and complex part I. In the 4A are the received data of the method according to 1 and in 4B the received data according to the method in 3 to see. This shows that the procedure according to 3 compared to the method according to 1 results in much more concentrated signals, allowing a better separation of the transmitted signals in the receiver. This ultimately means that the received data according to the procedure 3 with significantly fewer errors than the received data using the procedure according to 1 ,

The embodiment according to 3 is on the basis of a communication system for transmitting the data as speech signals, for example. According to a standard AMR (AMR - Adaptive Multirate) are coded be sent. A unit M3 makes the association between the first and second signal values SW1 (.) And SW2 (.) And the first and second received signal values (SE1 (.) And SE2 ( 3 In one possible embodiment, the unit M3 comprises a plurality of units. However, the present embodiment is not to be considered as limiting. Rather, the assignment can also be accomplished by other units.

In the 1 and 3 the essential procedural steps have been explained. In this case, the units N-IDFT, DAC, ADC, N-NDFT, N / 2-IDFT, M1-M7 used there can be designed as software units and / or hardware or in combination as software and hardware. For example, these units are executed by a signal processor. In addition, a realization of the presented method may require a buffering of individual data or signals. This is in the 1 and 3 not shown for reasons of simplification, but can still be inserted depending on the requirement. For example, a memory for temporarily storing the discrete data, ie the first and / or second data d1 (.), D2 (.) And / or the discrete received data de (.) May be necessary.

• – N-IDFT/N/2-IDFT: N bzw. N/2 dimensionale inverse diskrete Fourier-Transformationseinheit; /N/2-IDFT wird auch als zweite Einheit M2 bezeichnet;
• – DAC: Digital-zu-Analog-Convertereinheit für komplexe Signale
• – P1: Erste Modulationseinheit (= IQ-Modulator) zum Modulieren des zeitkontinuierlichen Signals ss(t) mit der Trägerfrequenz f₀.
• – P2: Zweite Modulationseinheit (= IQ-De-Modulator) zum Demodulieren des empfangenen Signals xe(t) mit der geschätzten Trägerfrequenz fs₀;
• – ADC: Analog-zu-Digital-Convertereinheit für komplexe Signale
• – N-DFT: N-Dimensionale diskrete Fourier-Transformationseinheit; auch als siebte Einheit M7 bezeichnet;
• – M1: erste Einheit zum Aufteilen der Daten c(.) in erste und zweite Daten d1(.), d2(.);
• – M3: dritte Einheit zum Zuordnen der ersten und zweiten Signalwerte SW1(.), SW2(.) zu den ersten und zweiten empfangenen Signalwerten SE1(.), SE2(.);
• – M4: vierte Einheit zum Generieren der ersten Hilfssignalwerte SH1(.);
• – M5: fünfte Einheit zum Generieren der zweiten Hilfssignalwerte SH1(.);
• – M6: sechste Einheit zum Zuordnen der ersten und zweiten Hilfssignalwerte SH1(.), SH2(.) zu den Hilfssignalwerten SH(.);

In the realization of the method steps according to 1 in the form of a first device V1 and / or the realization of the method steps according to FIG 3 in the form of a second device V2, the following modules or units are in use:

• N-IDFT / N / 2-IDFT: N or N / 2 dimensional inverse discrete Fourier transformation unit; / N / 2 IDFT is also referred to as second unit M2;
• - DAC: digital-to-analog converter unit for complex signals
• P1: first modulation unit (= IQ modulator) for modulating the time-continuous signal ss (t) with the carrier frequency f ₀ .
• P2: second modulation unit (= IQ-de-modulator) for demodulating the received signal xe (t) with the estimated carrier frequency fs ₀ ;
• - ADC: Analog-to-digital converter unit for complex signals
• N-DFT: N-dimensional discrete Fourier transformation unit; also referred to as the seventh unit M7;
• M1: first unit for dividing the data c (.) Into first and second data d1 (.), D2 (.);
• M3: third unit for assigning the first and second signal values SW1 (.), SW2 (.) To the first and second received signal values SE1 (.), SE2 (.);
• - M4: fourth unit for generating the first auxiliary signal values SH1 (.);
• M5: fifth unit for generating the second auxiliary signal values SH1 (.);
• - M6: sixth unit for assigning the first and second auxiliary signal values SH1 (.), SH2 (.) To the auxiliary signal values SH (.);

The Device V2 can for transmission data in a telecommunications system, in particular for voice or image data transmission, or in a data transmission system, in particular for transmission data on the Internet and / or intranet.

In the embodiment the invention has been explained with reference to a complete transmission scheme, with Transmitter, channel and receiver. It goes without saying that the invention only the transmitter-side Units or receiver side Claim units, as these designed in a transmitter or receiver Device implemented or can be operated.

Bibliography:

• [1] J. Armstrong, Analysis of New and Existing Methods of Reducing Intercarrier Interference Due to Carrier Frequency Offset in OFDM, IEEE Transactions on Communications, Vol. 3, March 1999
• [2] Ai Bo et al., Patent on Synchronization Techniques in Wireless OFDM Systems, Recent Patents on Electrical Engineering 2008, 1, 14-21, Bentham Science Publishers Ltd.
• [3] P. Golden, H. Dedieu, K. Jacobsen, Fundamentals of DSL technology, Auerbach Publications, published in 2006, ISBN 0-8493-1913-7

Claims (5)

Method for coding a data signal (DS) with N data (c (0), ..., c (N-1)), in which the following steps are performed: - splitting the N data (c (0), ... , (c (N-1)) into first data (d1 (0), ..., d1 (N / 2 - 1)) and second data (d2 (0), ..., (N / 2 - 1 )), wherein the first data (d1 (.)) the data (c (0), c (2), ..., c (N / 2 - 2)) with an even-numbered index and the second data (d2 (d2) .)) comprise the data (c (1), c (3), ..., c (N / 2 - 1)) with an odd-numbered index; Transforming the first data (d1 (.)) With an N / 2-dimensional inverse discrete Fourier transformation (N / 2-IDFT) into first signal values (SW1 (0),..., SW1 (N / 2-1) ) and transforming the second data (d2 (.)) with an N / 2-dimensional inverse discrete Fourier transform (N / 2-IDFT) into second signal values (SW2 (0), ..., SW2 (N / 2 - 1 )); Assigning the first signal values (SW1 (.)) To first received signal values (SE1 (0),..., SE1 (N / 2-1)) and assigning the second signal values (SW2 (.)) To second received signal values ( SE2 (0), ..., SE2 (N / 2 - 1)); Generating N / 2 first auxiliary signal values (SH1 (0), ..., SH1 (N / 2-1)) by respectively subtracting the second received signal value (SE2 (IX)) from the first received signal value (SE1 (IX) ) SH1 (IX) = SE1 (IX) - SE2 (IX), wherein in each case the received signal values (SE1 (IX), SE2 (IX)) are subtracted with an identical index (IX); Generating N / 2 second auxiliary signal values (SH2 (0), ..., SH2 (N / 2-1)) by respective addition of the second received signal value (SE2 (IX)) and the first received signal value (SE1 (IX) ) SH1 (IX) = SE1 (IX) + SE2 (IX), wherein in each case the received signal values (SE1 (IX), SE2 (IX)) are added with an identical index (IX); Generating auxiliary signal values (SH (0),..., (N / 2-1)) by assigning the first auxiliary signal values (SH1 (.)) To the first N / 2 auxiliary signal values (SH (0),..., SH (N / 2 - 1)) and by assigning the second auxiliary signal values (SE2 (.)) To the second N / 2 auxiliary signal values (SH (N / 2), ..., SH (N-1)); - Generating received data (d (0), ..., d (N-1)) by transforming the N auxiliary signal values (SH (0), ..., SH (N-1)) by means of an N-dimensional discrete Fourier transformation (N-DFT). Method for reconstructing a data signal (DS) with N data (c (0), ..., c (N - 1)) in the form of received data (d (0), ..., d (N-1)) from first received Signal values (SE1 (0), ..., SE (N / 2 - 1)) and second received Signal values (SE2 (0), ..., SE2 (N / 2 - 1)), where the first and second received signal values (SE1 (.) and SE2 (.)) by the following Steps were created: - split up of N data (c (0), ..., (c (N - 1)) into first data (d1 (0), ..., d1 (N / 2 - 1)) and second data (d2 (0), ..., (N / 2 - 1)), wherein the first data (d1 (.)) the data (c (0), c (2), c (N / 2 - 2)) with an even-numbered index and the second data (d2 (.)) the data (c (1), c (3), ..., c (N / 2 - 1)) comprising an odd-numbered index; - Transform the first Data (d1 (.)) With an N / 2-dimensional Inverse Discrete Fourier Transformation (N / 2 IDFT) into first signal values (SW1 (0), ..., SW1 (N / 2 - 1)) and transforming the second data (d2 (.)) with an N / 2-dimensional inverse Discrete Fourier Transformation (N / 2-IDFT) into second signal values (SW2 (0), ..., SW2 (N / 2 - 1)); - Assign the first signal values (SW1 (.)) to first received signal values (SE1 (0), ..., SE1 (N / 2 - 1)) and Assigning the second signal values (SW2 (.)) To second received Signal values (SE2 (0), ..., SE2 (N / 2 - 1)); at the following Steps performed become: - To generate of N / 2 first auxiliary signal values (SH1 (0), ..., SH1 (N / 2 - 1)) respective subtraction of the second received signal value (SE2 (IX)) from the first received signal value (SE1 (IX)) SH1 (IX) = SE1 (IX) - SE2 (IX), wherein in each case the received signal values (SE1 (IX), SE2 (IX)) with be subtracted from an identical index (IX); - To generate of N / 2 second auxiliary signal values (SH2 (0), ..., SH2 (N / 2 - 1)) respective addition of the second received signal value (SE2 (IX)) and the first received signal value (SE1 (IX)) SH1 (IX) = SE1 (IX) + SE2 (IX), whereby in each case the received signal values (SE1 (IX), SE2 (IX)) to be added with an identical index (IX); - Produce of auxiliary signal values (SH (0), ..., (N / 2 - 1)) by assigning the first one Auxiliary signal values (SH1 (.)) To the first N / 2 auxiliary signal values (SH (0), ..., SH (N / 2 - 1)) and by assigning the second auxiliary signal values (SE2 (.)) the second one N / 2 auxiliary signal values (SH (N / 2), ..., SH (N-1)); - To generate of received data (d (0), ..., d (N-1)) by transformation of the N auxiliary signal values (SH (0), ..., SH (N-1)) by means of an N-dimensional one Discrete Fourier Transformation (N-DFT). Device (V2) for coding a data signal (DS) with N data (c (0), ..., c (N-1)), comprising the following units: - First unit (M1) for splitting the N data (c (0), ..., (c (N-1)) into first data (d1 (0), ..., d1 (N / 2 - 1)) and second data (d2 (0), ... , (N / 2 - 1)), where the first data (d1 (.)) Is the data (c (0), c (2), ..., c (N / 2 - 2)) with an even-numbered index and the second data (d2 (.)) comprises the data (c (1), c (3), c (N / 2 - 1)) with an odd-numbered index; second unit (M2) for transforming the first data (M2) d1 (.)) with an N / 2-dimensional inverse discrete Fourier transform (N / 2-IDFT) into first signal values (SW1 (0), ..., SW1 (N / 2-1)) and transforming the second one Data (d2 (.)) Having an N / 2-dimensional inverse discrete Fourier transform (N / 2 IDFT) into second signal values (SW2 (0), ..., SW2 (N / 2-1)); - Third unit (M3) for assigning the first signal values (SW1 (.)) To first received signal values (SE1 (0), ..., SE1 (N / 2 - 1)) and for assigning the second signal values (SW2 (. )) to second received signal values (SE2 (0), ..., SE2 (N / 2 - 1)); - fourth unit (M4) for generating N / 2 first auxiliary signal values (SH1 (0), ..., SH1 (N / 2 - 1)) by respectively subtracting the second received signal value (SE2 (IX)) from the first received one Signal value (SE1 (IX)) SH1 (IX) = SE1 (IX) - SE2 (IX), wherein in each case the received signal values (SE1 (IX), SE2 (IX)) are subtracted with an identical index (IX); Fifth unit (M5) for generating N / 2 second auxiliary signal values (SH2 (0), ..., SH2 (N / 2-1)) by respectively adding the second received signal value (SE2 (IX)) and the first received one Signal value (SE1 (IX)) SH1 (IX) = SE1 (IX) + SE2 (IX), wherein in each case the received signal values (SE1 (IX), SE2 (IX)) are added with an identical index (IX); - Sixth unit (M6) for generating auxiliary signal values (SH (0), ..., (N / 2 - 1)) by assigning the first auxiliary signal values (SH1 (.)) To the first N / 2 auxiliary signal values (SH (0) , ..., SH (N / 2 - 1)) and by assigning the second auxiliary signal values (SE2 (.)) The second N / 2 auxiliary signal values (SH (N / 2), ..., SH (N - 1 )); Seventh unit (M7) for generating received data (d (0), ..., d (N-1)) by transforming the N auxiliary signal values (SH (0), ..., SH (N-1)) by means of an N-dimensional discrete Fourier transformation (N-DFT). Device (V3) for reconstructing a data signal (DS) with N data (c (0), ..., c (N - 1)) in the form of received data Data (d (0), ..., d (N-1)) from first received signal value (SE1 (0), ..., SE (N / 2 - 1)) and from second received signal values (SE2 (0), ..., SE2 (N / 2 - 1)), where the first and second received signal values (SE1 (.) and SE2 (.)) as follows were generated: - split up of N data (c (0), ..., (c (N - 1)) into first data (d1 (0), ..., d1 (N / 2 - 1)) and second data (d2 (0), ..., (N / 2 - 1)), wherein the first data (d1 (.)) the data (c (0), c (2), ..., c (N / 2 - 2)) with an even-numbered index and the second data (d2 (.)) the data (c (1), c (3), ..., c (N / 2 - 1)) comprising an odd-numbered index; - Transform the first Data (d1 (.)) With an N / 2-dimensional Inverse Discrete Fourier Transformation (N / 2 IDFT) into first signal values (SW1 (0), ..., SW1 (N / 2 - 1)) and transforming the second data (d2 (.)) with an N / 2-dimensional inverse Discrete Fourier Transformation (N / 2-IDFT) into second signal values (SW2 (0), ..., SW2 (N / 2 - 1)); - Assign the first signal values (SW1 (.)) to first received signal values (SE1 (0), ..., SE1 (N / 2 - 1)) and Assigning the second signal values (SW2 (.)) To second received Signal values (SE2 (0), ..., SE2 (N / 2 - 1)); the following units includes: - Fourth Unit (M4) for generating N / 2 first auxiliary signal values (SH1 (0), ..., SH1 (N / 2 - 1)) by respective subtraction of the second received signal value (SE2 (IX)) from the first received signal value (SE1 (IX)) SH1 (IX) = SE1 (IX) - SE2 (IX), wherein in each case the received signal values (SE1 (IX), SE2 (IX)) with be subtracted from an identical index (IX); - Fifth unit (M4) for generating N / 2 second auxiliary signal values (SH2 (0), ..., SH2 (N / 2 - 1)) by respective addition of the second received signal value (SE2 (IX)) and the first received signal value (SE1 (IX)) SH1 (IX) = SE1 (IX) + SE2 (IX), whereby in each case the received signal values (SE1 (IX), SE2 (IX)) to be added with an identical index (IX); - Sixth Unit (M4) for generating auxiliary signal values (SH (0), ..., (N / 2 - 1)) Associating the first auxiliary signal values (SH1 (.)) With the first N / 2 auxiliary signal values (SH (0), ..., SH (N / 2 - 1)) and by assigning the second auxiliary signal values (SE2 (.)) the second one N / 2 auxiliary signal values (SH (N / 2), ..., SH (N-1)); - Seventh Unit (M7) for generating received data (d (0), ..., d (N-1)) Transformation of the N auxiliary signal values (SH (0),..., SH (N-1)) by means of an N-dimensional discrete Fourier transform (N-DFT). Use of any of the methods of claim 1 or 2 or one of the devices according to claim 3 or 4 for transmission of data in a telecommunication system, in particular for voice transmission, or in a data transmission system, in particular for transmission of data in the internet and / or intranet.