EP2248268A2

EP2248268A2 - Method for the detection and generation of a useful signal and associated devices and communications system

Info

Publication number: EP2248268A2
Application number: EP08878341A
Authority: EP
Inventors: Jaroslaw Kussyk
Original assignee: Siemens AG
Current assignee: Siemens AG
Priority date: 2008-02-26
Filing date: 2008-02-26
Publication date: 2010-11-10
Also published as: CN102027680B; CN102027680A; WO2010078876A9; US8374229B2; US20110007791A1; HK1152418A1; WO2010078876A2

Abstract

The invention relates to a method for the generation of a useful signal and to a method for the detection of the useful signal, an element of the useful signal being functionally dependent on at least one further element of the useful signal. The invention further relates to associated devices and to a communications system comprising such a device.

Description

description

Method for detecting and generating a useful signal and associated devices and communication system

The invention relates to a method for detection and to a method for generating a useful signal and associated devices and to a communication system.

In broadband systems for transmitting information or for obtaining information, synchronization with a useful signal is usually required.

In this case, a receiver (for example a measuring device) often does not have a priori information as to whether and to what extent a useful signal is contained in a received signal or in an input signal. To make matters worse, that the received signal may be distorted by a disturbance or generally altered,

Noise often occurs as white noise (or Gaussian noise) in which a spectral noise power density in a certain more or less large frequency band is practically constant. Particularly problematic are disturbances which are different from such white noise and have strong time- and / or frequency-selective properties.

Thus, in a receiver or in a measuring device, a plurality of frequency channels may for a short time be disturbed in such a way that detection of a useful signal is no longer possible in a frequency channel.

The object of the invention is to avoid the abovementioned disadvantages and in particular to present an approach by means of which, even in the case of strong temporal and / or frequency-selective interference, a robust detection of a synchronization signal or a useful signal with high reliability is made possible,

This object is achieved according to the features of the independent claims. Further developments of the invention will become apparent from the dependent claims.

To achieve the object, a method for detecting a useful signal is specified,

- In which the useful signal is a multi-dimensional useful signal;

in which the useful signal comprises several elements,

- In which a functional dependence of an element of the Nutϊsignals is taken into account by at least one further element of the desired signal.

Thus, in the detection of the useful signal, elements of the useful signal in several dimensions, e.g. in a time-frequency domain, considered.

Preferably, the plurality of elements of the desired signal are functionally dependent on each other.

Furthermore, the multiple elements of the useful signal are advantageously designed to be redundant, so that a part of the plurality of elements can be used for a reconstruction or detection of the useful signal.

Preferably, the plurality of elements of the payload signal are split among the multiple dimensions of the payload signal so that disturbance of the payload signal upon transmission from a transmitter to a receiver is tolerable. A development is that different elements of the useful signal are processed with different carrier signals and / or on different subchannels,

The useful signal can be divided into different subchannels, which are modulated in particular with different carrier signals.

Another development is that the functional dependency is determined using a predetermined n-dimensional matrix or can be determined.

In particular, this n-dimensional matrix can be given to specify the functional dependencies. Thus, it is possible that the n-dimensional array predetermines the functional dependencies such that, depending on the nature of the disturbance, it is possible to detect the useful signal.

In particular, it is a further development that the functional dependency is determined on the basis of a frequency modulation, a differential modulation and / or a coding.

The dependencies of the elements of the NutEsignals each other can be determined by a differential modulation. In this case, information about state changes is preferably transmitted, not by absolute reference values. Examples of differential modulations are: D-PSK (differential phase shift keying), delta modulation.

It is also a development that the functional dependence is determined in a time range and / or in a frequency range.

In particular, the elements of the useful signal can be distributed in the time domain and / or in the frequency domain may be arranged, wherein the functional dependencies may be in one or both of these areas.

Furthermore, it is a development that the useful signal is an n-dimensional useful signal. In particular, that can

Useful signal to be a two-dimensional useful signal, in particular in a time-frequency range.

As an additional development, the functional dependence of an element of the useful signal of at least one further element of the useful signal in a receiver, in particular in the context of a decoding and / or a demodulation, is taken into account.

A further development is that the functional dependence of an element of the useful signal of at least one further element of the useful signal is taken into account by determining individual elements of a synchronization matrix based on a hard and / or soft decision.

Here, a variety of different methods for determining a hard decision or a soft decision can be used. In particular, threshold comparisons or stochastic decisions are possible, for example, a given context, e.g. consider a time duration of given length.

An embodiment is that the functional

Dependence of an element of the useful signal is taken into account by at least one further element of the same subchannel and / or from at least one further subchannel of the multidimensional useful signal.

An alternative embodiment is that for at least one element a corresponding Sound signal estimate or an estimate of the signal strength of the respective element is taken into account.

A next embodiment is that after taking into account the functional dependence of a

Correlation, in particular an autocorrelation of the signal with itself, with at least partially disjoint parts of the signal or a cross-correlation of the signal with another signal, is performed.

In particular, the coefficients of the correlator can be adapted to the synchronization matrix.

It is also an embodiment that, after taking into account the functional dependence, a filtering is carried out by means of a filter. The coefficients of the filter can be adapted to the synchronization matrix.

A refinement is that a correlation coefficient is determined on the basis of the filter between the filter's coefficients adapted to the synchronization matrix and a time-limited section of the received signal.

An additional embodiment is that the useful signal is detected if the correlation coefficient at the output of the correlator or filter reaches and / or exceeds a predetermined threshold value,

Another embodiment is that it is checked in a subsequent iteration whether a next correlation coefficient is better than a previous correlation coefficient and in this case the useful signal is detected on the basis of the better correlation coefficient. It is also a possibility that, before taking into account the functional dependence, a reduction of a noise signal component is carried out comprising the steps:

(a) transform an input signal into a multi-dimensional signal;

(b) estimating a noise component of the multi-dimensional signal, wherein the estimation of the noise component is based on at least a subset of the multi-dimensional signal; (c) comparison of the noise component with the multi-dimensional signal; (d) at least partially suppressing the multi-dimensional signal by comparison,

It should be noted that it is in the

Interference signal component can be any interference signal of any degree or intensity, in particular can be reduced more or less as part of the reduction of the Störsignalanteils this. By way of example, it is also possible to (almost) completely reduce the interference signal component.

Advantageously, the approach presented here makes it possible to reduce the interference signal component in an output signal in that one recognized as severely disturbed

Input signal strongly suppressed completely and thus not further processed.

This has the advantage, for example, that false alarms due to incorrectly recognized data can be prevented.

In particular, this method is suitable for detecting useful signals, wherein the useful signals are preferably transmitted redundantly at different locations of the multi-dimensional signal. Signals at strongly disturbed places can be faded out while eg signals at hardly (or not) disturbed places are allowed to pass through for further processing.

Another advantage of the presented approach is that it does not need to know where which useful signals are present in the multi-dimensional signal. Rather, the reduction of the Störsignalanteils advantageously without closer knowledge of the nature or the location of the useful signals, in other words, the reduction of the Störsignalanteils takes place regardless of whether straight payload signals are transmitted or not.

It is also an advantage that a dynamic adaptation of the reduction of the interference signal component is dependent on actual channel interference. This is particularly noticeable in the case of temporary (strong) disturbances, such as those that occur in power grids when they are used as communication networks.

A development is that the transformation comprises a serial-parallel conversion of the input signal.

In this case, advantageously, a serial signal can be transformed into a multi-dimensional signal. For example, a time-frequency transformation or a wavelet transformation can be used. Furthermore, these and other transformations can be used to obtain the multidimensional signal, in particular an n-dimensional signal.

It is also a development that the input signal comprises a multi-dimensional signal and / or a complex signal.

Another development is that the transformation comprises a transformation into a time-frequency domain. In particular, it is a development that the multi-dimensional signal is stored in a memory.

Preferably, the memory may be embodied as a plurality of parallel shift registers coupled line by line. Also, the memory may be implemented as a conventional memory.

For example, the memory allows further processing units access to a multi-dimensional temporally limited section of the multidimensional signal,

It is also a development that the memory comprises a field of coupled shift registers, which is synchronized with the transformation of the input signal.

Furthermore, it is a development that the estimation of the interference component of the multidimensional signal is carried out by providing at least one value of a predetermined function and / or a statistic and / or a statistical variable and / or a quantity or function derived therefrom for the at least one subset of the multidimensional Input signal is determined.

Here, the statistic may include an n-dimensional cumulative distribution function or an n-dimensional probability (density) function. The statistical quantity includes, for example, an average, a variance, an amount, or a statistical moment. The derived quantity includes, for example, an excess, a skewness, a median, and / or a characteristic function. Also, combinations among the above are possible.

In particular, the estimate for at least one feature and / or for at least one parameter of the Störsignalanteils and / or the Nutzsignalanteils be determined.

Examples of such features or parameters are: signal amplitude, signal power, signal magnitude, signal level, phase, frequency, and combinations thereof.

The subset may include one or more patterns or sections of the multi-dimensional signal.

In particular, these sections may be provided at different locations of the multi-dimensional signal. The subset itself can have different form (s). It is also possible that the subset has a shape that covers a multi-dimensional area and thereby has a multi-dimensional recess within this area. In the example of a two-dimensional time-frequency surface, this corresponds to a shape with a hole or an opening in the mold.

In the context of an additional development, the at least one subset has a periodicity.

Thus, multiple subsets or patterns may be considered at particular times, with the subsets having a particular time interval from one another.

A next development is that the comparison of the noise component with the multi-dimensional signal is performed by using the noise component at least one signal barrier is set.

Thus, the comparison allows the setting of at least one signal barrier based on the estimated noise component. Thus, a threshold for a useful signal can be dynamically adjusted, ie a signal above the threshold can be further processed, a signal below the threshold is suppressed and not further processed.

An embodiment is that the multi-dimensional signal is at least partially suppressed, provided that this has no predetermined distance to the interference component.

The type of suppression of the multi-dimensional signal may advantageously depend on the signal strength and does not have to be designed as a hard decision.

For example, the more clearly it is possible to pass the multidimensional signal (stronger), the clearer it is from the disturbance (the greater, for example, a distance from the disturbance).

An alternative embodiment is that an estimate of a groove signal strength of the multi-dimensional signal is performed, in particular by at least one value of a predetermined function and / or a statistic and / or a statistical variable and / or a derived quantity or function for the at least one subset determined

An embodiment is that the estimate for at least one feature and / or for at least one

Parameter of the useful signal component is determined or carried out.

In particular, it is optionally possible to additionally estimate the useful signal strength and to take this into account in the reduction of the interference signal component.

A next embodiment is that the comparison of the interference component with the multi-dimensional signal is carried out taking into account the useful signal level. It is also an embodiment that a useful signal is detected in the multi-dimensional signal, provided that this has a predetermined distance to the interference component.

Both the estimation of the interference component and the estimation of the useful signal component can be multidimensional and / or complex like the receiver signal itself.

Accordingly, the method for reducing the noise signal component can be used to detect a useful signal.

A development consists in that at least one further channel is taken into account in the reduction of the interference signal component.

Accordingly, the approach may be implemented as a multiple-input-multiple-output (MIMO) system, where each channel may have an impact on the other channels of the system and account for all or part of these effects.

An additional embodiment is that following step (d) in a step (e) an inverse transformation to step (a) is performed.

Another embodiment is that a parallel serial conversion of the output signal is performed in an output signal sequence.

Accordingly, the output signal in a serial form in the original time domain and the estimated noise signal in serial form in a timely manner can be provided.

It is also a further embodiment that the useful signal is a synchronization signal. The above object is also achieved by a method for generating a useful signal,

in which a useful signal is composed of a plurality of carrier signals, each carrier signal in a subchannel being modulated by a separate element, the elements being generated according to a predetermined synchronization matrix,

- In which an element of the slot signal of at least one further element of the useful signal has a functional dependence.

In particular, the generated groove signal is a multidimensional useful signal as described above. Also, the synchronization matrix is preferably a multi-dimensional synchronization matrix.

An embodiment consists in that the useful signal is transmitted substantially at the beginning and / or several times during an information transmission.

It is also a further embodiment that the generated useful signal is a synchronization signal.

Another embodiment is that the useful signal is determined based on a plurality of carrier signals, wherein each carrier signal is modulated in a subchannel by a separate element.

Furthermore, a solution for detecting a useful signal comprising a processor unit, which is set up in such a way that the method can be carried out on the processor unit as described herein, is provided for achieving the above-mentioned object.

In addition, to solve the above object, a device is specified for generating a Useful signal comprising a processor unit and / or an at least partially hardwired circuit arrangement, which is set up such that the method can be carried out as described herein,

Said processor unit may be or include any type of processor or computer or computer with correspondingly necessary peripherals (memory, input / output interfaces, input devices, etc.). Such a processor unit can in particular in one

Communication device may be provided, which in particular has a transmitter, receiver (receiver) or a transceiver.

Furthermore, a hardwired circuit unit, e.g. an FPGA or ASIC or other integrated circuit may be provided. In particular, electronic, electromagnetic, acoustic or other elements may be provided to detect and / or process different signals.

In particular, the device may thus comprise a unit for parallel processing of signals and / or a unit for serial processing of signals.

The device may comprise or be embodied as: a measuring device, a diagnostic device, a counter, an information acquisition device, a control device, a meter and / or a corresponding system.

The device can be used in power engineering.

It is possible that the signal comprises different physical quantities:

an electrical variable,

an electromechanical size, an electromagnetic variable,

an acoustic size,

a thermal size,

a mechanical (in particular a hydraulic or pneumatic) size,

a chemical quantity,

- an optical size.

Combinations of the aforementioned sizes are possible as a signal (e).

It is a further development that the device is a communication device, wherein the communication device exchanges signals with another communication device via a communication link which at least partially comprises a power network.

Furthermore, in order to achieve the object, a system is specified comprising a device as described herein,

Embodiments of the invention are illustrated and explained below with reference to the drawings.

Show it:

Fig.l a block diagram comprising units for signal processing, which allow generation of a synchronization signal at a transmitter as well as an element-wise reception or a detection of the element-by-element received synchronization signal at a receiver;

2 shows a block diagram comprising units for

Signal processing enabling a reduction of a noise component; 3 is a sketch illustrating that several

Subareas of a two-dimensional signal are used to determine a noise component,

The following notation is used in particular:

a payload signal;

a Nutesignal or synchronization signal

(on the transmitter side) or an element of the useful signal or synchronization signal;

Carrier signal per channel j;

a reference signal (receiving side) per channel j;

a detected element of the synchronization signal;

an n-dimensional synchronization matrix;

a coefficient matrix of the n-dimensional filter or the correlator;

an element of the synchronization matrix an element of the coefficient matrix a correlation coefficient or a correlation matrix;

a disorder;

an estimate of the disturbance n (t, f); an input signal;

an input signal after one

Suppression;

a degree that indicates to what extent that

Useful signal s (t, f) is included in the input signal r (t, f);

an estimate or a measurement of the

wanted signal;

a multi-dimensional (input) signal of a first channel;

a multi-dimensional (input) signal of a first channel after interference suppression or passing through a signal barrier;

a subset of the multidimensional

Input signal used for the disturbance;

a subset of the multi-dimensional input signal used for the payload signal;

a function for determining the

Noise component based on a function for determining the Nutssignalanteils based on

The signals mentioned here include or are real or complex, in particular multi-dimensional signals. 1 shows a block diagram comprising units for signal processing which enable a modulation or coding of a synchronization signal at a transmitter 120 as well as a demodulation or a detection of the coded synchronization signal at a receiver 130.

The following is an example of a

Synchronization signal assumed, with any other type of useful signal is coded or decoded.

The transmitter 120 has a unit for bit and / or symbol modulation 101 or coding, which is connected to a unit 102 on the basis of which an inverse time-frequency transformation or a signal synthesis by means of different carrier signals he follows. The output signal of the unit 102 is subjected to parallel-to-serial conversion in a unit 103 and sent to a receiver 130 via a transmission channel 104. The transmission channel has a fault which may in particular be a non-Gaussian disturbance.

At the transmitter 120 becomes a synchronization signal from several different carrier signals composed, each by the

Subsynchronization signals or, elements of the synchronization signals be modulated in subchannels j.

The synchronization signals are based on a given two-dimensional synchronization matrix generated. This can be done for example by a frequency modulation and / or by a differential modulation.

Here, preferably, an element the synchronization signal is a functional dependency comprising at least one other element from the same subchannel j and / or with at least one other element from a subchannel A synchronization signal generated in this way can be sent at the beginning of an information transmission and optionally several times during the information transmission.

The receiver 130 comprises the following components: The input signal r (t) is supplied to an optionally provided interference suppression 105 and then subjected to a serial-to-parallel conversion 106. In a subsequent unit 107, a time-frequency transformation and a frequency channel separation by means of several different reference signals The input signal generated at the output of the unit 107 is fed either to a unit 108 for reducing a noise signal in the time-frequency domain or directly to a unit 109 sur demodulation bsw, bit / symbol decision.

As such, the unit 108 is optional. Should it be present, it supplies an input signal reduced by a noise signal component and optionally additionally an estimated noise signal component or, optionally, additionally an estimated useful signal strength to the unit 109. The operation of the unit 108 and possibly the unit 105 will be explained in more detail below in particular.

The unit 109 provides at its output individual elements of a synchronization matrix prepared based on a hard and / or soft decision (hard decision / soft decision).

From these elements, in a matched filter or, in a correlator 110, a correlation coefficient determined and supplied to a unit 111 for signal detection with a threshold value. Subsequently, a signal synchronization takes place in a unit 112, at the output of which the best synchronization signal which could be determined during a predetermined period of time is provided. This synchronization signal is advantageously used to detect further user data,

At the receiver 130, a time-frequency transformation of the input signal ≤ in individual time-frequency signal components performed based on the reference signals Such a transformation can be, for example, a short-time Fourier transformation, a wavelet transformation or carried out by means of a filter bank.

The signal obtained from the unit 107 is supplied either directly or via the unit 108 of the unit 109.

At unit 108, some of the time-frequency input signal components may be partially or completely canceled depending on the actual disturbance and the remaining input signal components with corresponding noise signal estimates and / or provided with corresponding estimates of a strength of a useful signal. Interference suppression may alternatively also be provided before the time-frequeriz transformation 107.

At the output of the unit 109 are preferably at each discrete processing time from the received time-frequency elements of the suspected synchronization signal individual element the synchronization matrix based on a hard and / or a soft one

Decision made. In particular, the individual elements the synchronization matrix with each other made comparable on the basis of said hard and / or soft decisions.

A hard decision includes, for example, an assignment of a fixed quantities to a predetermined value. This can be done for example by a comparison with the predetermined value. On the other hand, a soft decision is made, in particular, if the final decision is deferred. As such, an input, if it has a value that does not allow a clear decision, may be stored in a context, e.g. their timing, in order to obtain greater certainty in the final decision. In the soft decision, in particular temporal courses of an input variable or

Probability distributions considered.

In the processing of the signals in the unit 109, in particular in the context of pemodulation, in particular a functional dependence of a single element (in particular, each such element) of the synchronization signal from at least one other element from the same channel j and / or with at least one other element from a subchannel considered. Optionally, in addition, the aforementioned estimate of the noise component and / or the estimation of the strength of the useful signal be taken into account.

The received individual elements of the

synchronization matrix are supplied to the unit 110, wherein the individual coefficients of the 2-dimensional filter or the correlator OrS to the received elements of the synchronization matrix are adjusted. At the output of the filter or correlator 109, a correlation coefficient is determined in particular for each time point U. This correlation coefficient is determined by the two-dimensional coefficient matrix and the Section of the data stream of the elements Reaches and / or exceeds the value of

correlation coefficient a predetermined threshold, the synchronization signal can be regarded as detected (detection in the unit 111).

Additionally, in unit 112 in the next

synchronization phase be checked if the value of the correlation coefficient has still improved. In this case, the better value of the correlation coefficient used to the timing of the synchronization signal even more accurate to capture. Preferably, a predetermined period of time is waited in the unit 112 and the best time position of the synchronization signal during this Duration determined.

The unit 108 will be explained in more detail below.

It should be noted here that the interference signal component and the strength of the useful signal component can be at least partially dependent on one another.

Below, in particular, the approach presented here for interference suppression will be explained.

in Fig.2 is an input signal a serial-to-parallel conversion 201 and then subjected to a time-frequency transformation 202. Alternatively, other (multi-dimensional) transformations can take place. The result of the time-frequency transformation 202 is a multi-dimensional signal which is stored in a memory 203. Preferably, the memory 203 is implemented as a field (or array) of line-by-line parallel shift registers. A column of the coupled parallel shift registers comprises elements of a current spectrum after a time-frequency transformation, a line comprises a time course in a spectral channel.

A memory size allows storage of a time period T _c . The current calculation may be beneficial at a time take place, so that for the calculation, a context before the current time t, ie from a time to to the time t and a context from the time t to the time T _c are available.

The following units can access all entries of the shift register at a time access.

Thus, the memory 203 performs the function of a multi-channel delay or a sliding window for a plurality of parallel data series, which can be analyzed and / or further processed by the subsequent units accordingly.

Alternatively to the embodiment of the

Parallel shift register can also be provided a conventional memory. The function of the sliding window can be implemented by special address pointers, or a corresponding incrementation of the same.

Thus, the input signal is first subjected to a decomposition by means of a short-time Fourier transformation or a wavelet transformation or another transformation. Then it becomes the time-frequency distribution of the multidimensional signal preferably continuously

(eg at given discrete times) an estimate a storsignalanteils determined and in particular for each of those components of the input signal which are used for further processing. This particular non-linear estimation of the interference signal component takes place in a unit 204, which is downstream of the memory 203.

For determining the estimation of the interference signal component can for an i-th or j-th component of the input signal by means of the function depending on the type or multi-dimensional pattern (for example, according to a given time-frequency pattern) of the suspected or assumed disorder of the subset of individual components of the multi-dimensional signal be used.

It should be noted that this subset is an arbitrary pattern of or a choice from the multi-dimensional signal can correspond. Is it, for example, the multi-dimensional signal by a two-dimensional time-frequency consideration, this subset may comprise at least one section in this plane, in particular a plurality of sections, possibly also periodically recurring sections.

The function as well as the function may each include a determination of an energy density, power, or average signal amplitude. In particular, the subset represent an environment of the signal to be recognized, without the signal to be recognized itself should be included.

Preferably, the useful signal component provides in the input signal even no or only a negligible contribution to the estimated noise component

For example, if a disturbance in the time-frequency domain has a recurring pattern with a period T _N, or if this disturbance is stationary with the same period T _N , then the estimate of the noise component based on subsets of individual components of the time-frequency distribution of the multi-dimensional signal be formed, these subsets from each other around the period are removed.

3 shows examples of such subsets 301, 302, 303 (also called "ranges" or "patterns") used to determine or estimate the noise component serve. The respective subset can have the most varied forms. By way of example, an elliptical shape is shown in FIG. 3 for each subset. The shown input signal is exemplary within subset 302, but not within subsets

301 and 303. By considering a plurality of subsets 301 to 303 for the estimation of the interference signal component, it is possible to determine the interference signal component in the subset 302 and thus to determine whether the signal to be detected is strongly or weakly disturbed. If the signal is severely disturbed (eg with a received signal below a predetermined threshold), the signal will not be further processed, but suppressed. This has the advantage that only largely undisturbed signals are further processed and those signals that are transmitted for further processing can be correctly decoded, recognized, demodulated or measured on a severely disturbed channel.

Preferably, those signal components dme appear interesting for further processing or detection, subjected to a comparison in a unit 205. One in such a comparison may be the signal components Signal barriers (especially non-linear signal barriers) are supplied, wherein signal passing properties of the signal barriers based on the estimated star signal components or optionally based on the estimated useful signal strength controlled and / or adjusted.

Are for example the signal components greater than or equal to a threshold resulting from the estimated noise components results, so are the signal components for further processing as an external signal pass through.

The lower the signal components iαi ratio to the threshold, the sooner or the stronger these are suppressed. In other words, the signal components should be as trouble-free as an output signal be further processed.

Optionally, based on a unit for (optional non-linear) useful signal strength estimation 206 of components of the input signal a portion of a useful signal estimated. Accordingly, in the unit 205, the output signal additionally based on the estimated useful signal strength be determined. This can be achieved, for example, in that a difference between the estimated amplitude or the strength of the useful signal and the estimated noise is determined and fed to a threshold comparison: the greater the difference, the less the reception signal suppressed. The smaller the difference, the stronger the received signal becomes suppressed. Optionally, with a correspondingly small or negative difference, the received signal be completely suppressed.

The respective output signal remains the corresponding estimated interference signal and / or the estimated magnitude of the wanted signal assigned as an indicator of the quality of the useful signal.

The output signal and the estimated noise can either be further processed directly or respectively transformed back into the time domain (eg by means of an inverse short-time Fourier transformation or an inverse wavelet transformation), see units 207 and 208 in FIG.

Correspondingly, the units 207 and 2OS can each be followed by a unit for parallel-to-serial conversion 209 or 210, at whose outputs in each case an output signal sequence or an estimated interference signal sequence provided.

Optionally, the estimated useful signal or its strength or level Ai (U, fj) via a unit 212 for inverse time-frequency transformation into a signal Ä \ (u) and on by means of a unit 213 for parallel-to-serial conversion into a serial signal be implemented.

Further shown in FIG. 2 is a block 211 which illustrates that optionally for at least one further channel K with an input signal the arrangement described above can be provided. The operation of the optional block 211 is analogous to the above statements concerning the first channel with the input signal ,

In addition, the effects of each channel on all other channels can be taken into account. This is exemplified in Fig. 2 by showing the effects of channel K on channel 1 (see dotted line).

So can a stored multi-dimensional signal in unit 205. Furthermore, certain components (or pattern) of the further channel K in the estimated noise signal detection unit 204 (j) taken into account. Optionally, components can also be used the further channel K in the Nutzsignalstärkeschätzung 206 are taken into account.

Accordingly, it is possible for the channel 1 to be an output as discussed above _x and an estimated noise component and an estimated useful signal strength of the channel K influenced by the signals and ^be taken into account ^{in the} corresponding units of the channel K (this case is not shown in Figure 2 for the sake of clarity).

Furthermore, it is possible that a plurality of channels 1..K are provided, each channel comprising the units for Nutzsignalstärkeschätzung, for determining the Output signal and for the estimation of the noise component of each other channel influenced.

In connection with the unit 108, the following advantages result in particular:

a. It is possible to suppress a single interfering signal or a group of interfering signals.

b. In particular, with this approach, disturbances in the input signal can generally be suppressed. Only the useful signal components, which are distinguished in the time-frequency distribution of the input signal from the currently estimated interference signal components, are not suppressed or further processed and / or evaluated.

c. Thus, the interference suppression remains effective even if the useful signal is not contained in the input signal or the interference situation or the type of interference has changed.

d. This approach can be used in particular independently of any further processing or independently of the type of further processing of the signals, in particular because the time-frequency distribution of the interference-suppressed signal can be transformed back into the time domain as required.

e. Possible fields of application or applications for the solution presented here are all types of signal processing in which a noise suppression can be beneficial. Other advantages:

From the sender are the individual elements transmit the synchronization signals distributed over multiple frequency channels (or frequency sub-channels), with the individual elements the Synehronisationssignale have a functional dependence on each other. Such a functional dependency can in particular be given for a time U either in a time range and / or in a frequency range.

Advantageously, in the approach presented here, the individual elements the synchronization signals from at least one other element from the same subchannel j and / or from at least one other element from a subchannel. Such Redundancy allows error-free reception of individual elements of the synchronization signal or useful signal even if some of the elements of the synchronization signal or useful signal are disturbed.

To further improve the robustness of the proposed approach, a noise suppression can be performed in the receiver prior to the actual signal reception (see unit 105 in Fig.1). This also works if it is not known whether the synchronization signal or the useful signal is contained in the input signal.

The interference suppression 105 can in particular as a

Barrier filter or an amplitude limiter be executed.

A further advantage of the approach proposed here is that when digitally receiving each individual element of the synchronization matrix Gkj, a decision is made about the assignment of this element (hard and / or soft decision). This can be advantageous a recognition of the synchronization signal or useful signal also take place if the transmission channel has disturbances other than white noise,

Thus, in several subchannels already individually demodulated elements the synehronisationsmatrix G _kj a two-dimensional signal matched filter, in particular a FIR filter, or a correlator are supplied, where these are preferably processed together or together and in particular with each other. Accordingly, a large number of individual elements can with high probability in different time-frequency domains of the synchronization _matrix G _kj .

Also, the approach presented here can be successfully applied to channels with frequency-selective and / or time-selective fading.

Claims

claims

1. A method for detecting a useful signal,

- wherein the slot signal is a multi-dimensional slot signal;

in which the useful signal comprises several elements,

- In which a functional dependence of an element of the Nut2signals is considered by at least one further element of the useful signal.

2. Method according to claim 1, in which different elements of the useful signal are processed with different carrier signals and / or on different subchannels.

3. The method according to any one of the preceding claims, wherein the functional dependence is determined based on a predetermined n-dimensional matrix.

4. The method according to any one of the preceding claims, wherein the functional dependence on the basis of a frequency modulation, a different llen modulation and / or a coding is determined.

5. The method according to any one of the preceding claims, wherein the functional dependence in a time range and / or in a frequency range is determined.

6. The method according to any one of the preceding claims, wherein the useful signal is an n-dimensional useful signal,

7. The method according to any one of the preceding claims, wherein the useful signal is a two-dimensional useful signal, in particular in a time-frequency range.

8. The method according to any one of the preceding claims, wherein the functional dependence of an element of the useful signal is taken into account by at least one further element of the useful signal at a receiver, in particular in the context of a decoding and / or a demodulation,

9. The method according to any one of the preceding claims, wherein the functional dependence of an element of the useful signal is taken into account by at least one further element of the Nutasignals by individual elements of a synchronization matrix are determined based on a hard and / or a soft decision.

10. The method according to any one of the preceding claims, wherein the functional dependence of an element of the groove signal of at least one further element of the same subchannel and / or at least one further subchannel of the multi-dimensional useful signal is taken into account.

11. The method according to any one of the preceding claims, wherein for at least one element, a corresponding Störsignalschat2ung and in particular a corresponding estimate of the useful signal component of the respective element is taken into account.

12. The method according to any one of the preceding claims, wherein after taking into account the functional

Dependence a correlation is performed.

13. Verfa.hren according to any one of the preceding claims, wherein after taking into account the functional dependence filtering is carried out by means of a filter.

14. The method of claim 13, wherein the coefficients of the filter are adapted to a synchronization matrix.

15. The method according to any one of claims 13 or 14, wherein based on the filter, a correlation coefficient is determined between the at the

Synchronization matrix matched coefficients of the filter and a time-limited portion of the received signal.

16. The method of claim 15, wherein the useful signal is detected if the correlation coefficient reaches and / or exceeds a predetermined threshold.

17. Method according to claim 16, wherein it is checked in a subsequent iteration whether a next correlation coefficient is better than a previous correlation coefficient and in this case the useful signal is recognized on the basis of the better correlation coefficient.

18. The method according to any one of the preceding claims, wherein before taking into account the functional

Dependence a reduction of a noise signal component is carried out comprising the steps:

(a) transform an input signal into a multi-dimensional signal; (b) estimating a noise component of the multi-dimensional signal, wherein the estimation of the noise component is based on at least a subset of the multi-dimensional signal;

(c) comparison of the noise component with the multi-dimensional signal;

(d) At least partial suppression of the multidimensional signal by comparison.

19. The method of claim 18, wherein the transform comprises a serial-to-parallel conversion of the input signal.

20. The method according to any one of claims 18 or 19, wherein the input signal comprises a multi-dimensional signal and / or a complex signal.

21. The method according to any one of claims 18 to 20, wherein the transformation comprises a transformation into a time-frequency domain,

22. The method according to any one of claims 18 to 21, wherein the multi-dimensional signal is stored in a memory.

23. The method of claim 22, wherein the memory comprises a field of coupled shift registers that is synchronized with the transformation of the input signal,

24. Method according to one of claims 18 to 23, in which the estimation of the interference component of the multidimensional signal is carried out by providing at least one value of a predetermined function and / or a statistic and / or a statistical variable and / or a quantity or function derived therefrom the at least one subset is determined

25. The method of claim 24, wherein the estimate is determined for at least one feature and / or for at least one parameter of the noise component,

26. The method according to any one of claims 18 to 25, wherein the at least one subset has a periodicity.

27. The method according to any one of claims 18 to 26, wherein the comparison of the interference component with the multi-dimensional signal is performed by using the interference component at least one signal barrier is set.

28. The method according to claim 27, wherein the multi-dimensional signal is at least partially suppressed, provided that it has no predetermined distance from the interference component.

29. Method according to one of claims 18 to 28, in which an estimate of a groove signal component of the multidimensional signal is carried out, in particular by at least one value of a predetermined function and / or a statistic and / or a statistical variable and / or a variable derived therefrom or Function for the at least one subset is determined.

30. The method of claim 29, wherein the estimate for at least one feature and / or for at least one parameter of the useful signal component is determined.

31. The method according to any one of claims 29 or 30, wherein the comparison of the noise component is performed with the multi-dimensional signal taking into account the Nutssignalanteils.

32. The method according to any one of claims 18 to 31, wherein a useful signal is detected in the multi-dimensional signal, provided that this has a predetermined distance to the interference component.

33. The method according to any one of claims 18 to 32, wherein at least one further channel is taken into account in the reduction of the interference signal component.

34. The method according to any one of claims 18 to 33, wherein following step (d) in a step (e) to the step (a) inverse transformation is performed.

35. The method of claim 34, wherein a parallel-serial conversion of the output signal is performed in an output signal sequence.

36. The method according to any one of the preceding claims, wherein the useful signal is a synchronization signal.

37. The method according to any one of the preceding claims, wherein the useful signal is determined based on a plurality of carrier signals, each carrier signal is modulated in a subchannel by a separate element,

38. Method for generating a useful signal,

- In which an element of the useful signal of at least one further element of the useful signal has a functional dependence.

39. The method according to claim 38, wherein the useful signal is transmitted substantially at the beginning and / or several times during an information transmission,

40. Method according to claim 38, wherein the useful signal is a synchronization signal.

41. An apparatus for detecting a useful signal comprising a processor unit and / or an at least partially hard-wired Sσhaltungsanordnung which is arranged such that the method according to one of claims 1 to 37 can be carried out.

42. The apparatus of claim 41, wherein the device is a communication device, wherein the communication device with another Kommunikationsgerat signals via a communication link, which at least partially comprises a power network exchanged.

43. Device for generating a useful signal comprising a processor unit and / or an at least partially hard-wired

Circuit arrangement arranged such that the method according to one of Claims 38 to 40 can be carried out.

44. The apparatus of claim 43, wherein the device is a communication device, wherein the communication device with another communication device signals via a

Communication link, which at least partially includes a power grid, exchanges.

45. System comprising a device according to one of claims 41 to 44.