RECEIVER APPARATUS

Abstract

A receiver apparatus (1) for receiving a signal over a fading channel comprises a frequency domain interpolation unit (16) with a filter unit (10), a power comparison unit (20), and a processing unit (21). The filter unit (10) comprises a first filter element (11), which is a reference filter, and a second filter element (12). The first filter element (11) is arranged as large-band filter. The power comparison unit (20) compares the power of the signal filtered with said second filter element (12) with the power of the signal filtered with said first filter element (11). The processing unit (21) determines an appropriate filter length for a third filter element (13) of the filter unit (10) on the basis of this comparison. Thereby, a trade-off is made between an additional power received due to a long echo and an additional Gaussian noise power. The signal filtered with the third filter element (13) is sent to a signal processing unit (22) for further signal processing and then output at said output (4) of the receiver apparatus (1).

Description

FIELD OF THE INVENTION

The present invention relates to a receiver apparatus for receiving a signal over a fading channel, to a wireless network comprising such a receiver apparatus and to a method for transmission of data. More particularly, the present invention relates to a wireless network on the basis of a modulation technique such as orthogonal frequency division multiplexing (OFDM) as used, for example, in DVB-T or DVB-H.

BACKGROUND OF THE INVENTION

EP 0 838 928 A2 describes an apparatus for receiving digital signals transmitted over a transmissive channel as a sequence of OFDM multi-carrier symbols. Thereby, the transmitted signal is of the type including a reference symbol transmitted at each predetermined group of OFDM symbols. The apparatus known comprises means for generating an estimate of a channel transfer function by dividing the demodulated elementary components of a previously received OFDM symbol by the respective estimates of the same previously transmitted symbol, as obtained after equalization, forward error correction, re-encoding and re-mapping. Further, the initial estimate of the channel transfer function is obtained by dividing the demodulated components of a predetermined OFDM reference symbol by the respective components of the same transmitted symbol as stored in the receiver. The known apparatus comprises a frequency-domain filter for filtering the estimates of the channel transfer function with an equivalent “bandwidth” that is greater than the time duration of the guard interval but smaller than the time distance between the periodical repetitions of the channel impulse response due to sub-sampling in the frequency domain.

The apparatus known from EP 0 838 928 A2 has the disadvantage of a complex implementation having a very long feedback loop through the Viterby decoder into the channel estimation. Further, the quality of the signal received may degrade for channels with fast changing impulse responses.

OBJECT AND SUMMARY OF THE INVENTION

It is an object of the invention to provide a receiver apparatus, a wireless network and method for transmission of data with an improved performance, especially with an improved adaption to a channel with a changing impulse response.

This object is solved by a receiver apparatus as defined in claim 1, by a wireless network as defined in claim 8, and by a method as defined in claim 9. Advantageous developments of the invention are mentioned in the dependent claims.

It is noted that the first filter element, the second filter element, and the third filter element are not necessarily implemented as distinct elements. For example, a filter unit may provide a general filter element. This general filter element can then perform the operation of the first filter element, the second filter element, and the third filter element, respectively.

It is advantageous that the power comparison unit calculates a power ratio as a fraction value having a numerator which is the power of the signal filtered with the second filter and a denominator which is the power of the signal filtered with the first filter, and that the processing unit determines the appropriate filter length for the third filter element on the basis of this power ratio. This has the advantage that the determination of the appropriate filter length can be made with reference to a relative value, which measures the strength of the signal filtered with the second filter with reference to the strength of the signal filtered with the first filter.

Further, it is advantageous that the processing unit varies the filter length of the second filter element so that the power ratio is at least nearly 100% so as to provide a good reception in terms of received signal power versus received signal noise. In case that the filter length of the second filter is at most equal to the filter length of the first filter, the power ratio is at most 100%. In this case, the processing unit can vary the filter length of the second filter element so as to achieve the appropriate power ratio.

Thereby, it is further advantageous that a large amount of signal power is received, while the received power of noise is relatively small. Hence, it is advantageous that the shortest filter length of the second filter element is used for which the power ratio is at least nearly 100%.

It is advantageous that the processing unit varies the filter length of the second filter element so as to detect at least a long echo of the signal as long as the power of this long echo is superior to a certain threshold. This enables reception of a large amount of the signal power that can be utilized.

Thereby, a trade-off must be made between reception of an additional long echo of the signal and an additional power of a noise added due to the increase in filter length. Therefore, it is advantageous that the processing unit determines the appropriate filter length on the basis of the power of the long echo and an additional power of a noise, especially a Gaussian noise, added due to the increase in filter length so as to receive the long echo.

It is advantageous that the first filter element is a large-band filter. This has the advantage that reference can be made to the total amount of signal power.

It is also advantageous that the first filter element is a narrow-band filter, which is used as a reference. This has the advantage that reference can be made to the power of the main part of the signal. Hence, it is possible to reduce on-the fly the filter length of the third filter, which is the bandwidth of the equalization filter, by detecting the absence of long echoes.

These and other aspects of the invention will be apparent from and elucidated with reference to the embodiments described hereinafter.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will become readily understood from the following description of preferred embodiments thereof made with reference to the accompanying drawing, in which:

FIG. 1 shows a schematic view of a receiver apparatus of a wireless network according to an embodiment of the present invention; and

FIG. 2 shows a block diagram of a frequency domain interpolation unit of the receiver apparatus shown in FIG. 1.

DESCRIPTION OF EMBODIMENTS

FIG. 1 shows a receiver apparatus 1 of a wireless network 2. The receiver apparatus 1 comprises an input 3 and an output 4. The input 3 is connected with an antenna 5. The wireless network 2 also comprises a sender 6 with an antenna 7. The wireless network 2 can be based on a standard with a modulation scheme such as orthogonal frequency division multiplexing (OFDM). In such a standard the signal comprises pilots placed at certain frequencies that allow getting a right estimation of the channel in order to deal easily with the problem of echoes in a mobile environment. In DVB-T like standards, some pilots are only active during some symbols turning to data during the others. These pilots are arranged as scattered pilots, which are used for channel estimation and equalization. But, the receiver apparatus 1 and the wireless network 2 of the invention can also be used in other situations, especially with other modulation schemes.

The receiver apparatus 1 comprises a discrete Fourier transform (DFT) unit 8, a decoder unit 9, and a signal processing unit 22. The decoder unit 9 comprises a frequency domain interpolation unit 16 so as to measure the power with different interpolation filters during a frequency domain interpolation.

To get an estimation of the channel for every subcarrier of every OFDM symbol, the scattered pilots are interpolated both in time and in frequency domains. In frequency domain, a strategy to reject a maximum possible noise is to use the narrowest possible interpolation filter regarding the channel impulse response. But, in a mobile environment the bandwidth of the filter may not be chosen large enough to equalize a long echo that may appear suddenly.

The frequency domain interpolation unit 16 is described in further detail with reference to FIG. 2.

FIG. 2 shows a schematic block diagram of the frequency domain interpolation unit 16. The frequency domain interpolation unit 16 comprises an input 17 so as to input the estimates for the pilot positions after the DFT. The frequency domain interpolation unit 16 outputs the interpolated estimation for the channel for every symbol, especially the data symbols, with respect to the frequency domain.

A Gaussian noise model may be used to describe the noise which is added to the signal during the transmission from sender 6 to receiver apparatus 1. Such a Gaussian noise adds a specific noise energy to the signal which depends on a filter length used by a filter unit 10 of the receiver apparatus 1. The filter unit 10 comprises a first filter element 11, a second filter element 12, and a third filter element 13. In an embodiment of the present invention, the filter length of the first filter element 11 is fixed or at least set to a fixed value for a relatively long time. Further, the filter length of the first filter element 11 is selected so as to make the first filter element 11 a broad-band filter. The filter length of the second filter element 12 is variable. Further, it is advantageous that the filter length of the second filter element 12 is smaller or at most equal to the filter length of the first filter element 11, because the filter length of the first filter element 11 is relatively long.

It is noted that the filter length of the second filter element 12 may also be variable so that the fixed filter length of the first filter elements 11 is within the range of filter length of second filter element 12. In such a case it is advantageous that the filter length of the first filter element 11 is not too long. As a further alternative, the filter length of the first filter element 11 may be chosen relatively short. The first filter element 11 is then arranged as a narrow-band filter. In this case, it is advantageous that the filter length of the second filter element 12 is equal or greater than the length of the first filter element 11.

In the following it is assumed that the first filter element 11 is arranged as a broad-band filter and that the filter length of the second filter element 12 is equal or smaller than the length of first filter element 11. The other cases can be derived accordingly.

Despite the great disturbances possibly encountered in terrestrial channels, the Gaussian noise remains usually one of the most limiting ones, impacting every functioning limit. As the effect of Gaussian noise is particularly critical in the equalization process, a mean to decrease the amount of noise take into account in the bandwidth of the interpolation filter is simply to reduce its bandwidth. This can increase the performance, especially when no long echo is detected. Then, a narrower bandwidth can be used and gives much better performance in terms of noise rejection. But, when a long echo occurs, a relatively fast detection of the power of the new long echo is necessary. Otherwise, when using a filter with a reduced bandwidth to reject more noise, the equalizer may not be able to equalize and a lot of the equalized data is degraded.

The receiver apparatus 1 according to an embodiment of the invention enables a fast detection of at least a new long echo so that the appropriate filter length for signal reception, especially equalization, can be determined and used as a filter length of the third filter element 13, which is used for filtering the signal sent to output 4.

The filter unit 10 processes the signal from input 3 with the filter elements 11, 12, 13. This processing may be performed in parallel or in serial according to the processing speed of the filter unit 10. In the embodiment shown in FIG. 1 a parallel processing is shown. The signal filtered with the first filter element 11 is sent through a power determination unit 14, which measures the power of the filtered signal. The result of this power measurement is then sent to a power comparison unit 20. The signal is also filtered with the second filter element 12, and the power of this filtered signal is determined in a power determination unit 15. The power determined by the power determination unit 15 is also sent to the power comparison unit 20. The power comparison unit 20 compares the power of the signal filtered with the first filter element 11, which is received from the power determination unit 14, with a signal power of the signal filtered with the second filter element 12, which is received from the power determination unit 15. For example, the power comparison unit 20 calculates a power ratio as a fraction value having a numerator which is the power of the signal filtered with the second filter element and a denominator which is the power of the signal filtered with the first filter element. The power comparison unit 20 is connected with a processing unit 21. The processing unit 21 determines an appropriate filter length for the third filter element 13 on the basis of the comparison made by the power comparison unit 20, especially on the basis of the power ratio determined. The processing unit 21 may also vary the filter length of the second filter element 12 so as to determine the appropriate filter length for the third filter element 13. For example, the processing unit 21 can vary the filter length of the second filter element 12 so that the power ratio determined by the power comparison unit 20 is at least nearly 100%.

The processing unit 21 can also vary the filter length of the second filter element 12 and determine on the basis of the power ratio determined by the power comparison unit 20 the position of a long echo of the signal received. The processing unit 21 can include such a long echo, when the power of the long echo is superior to a certain threshold.

It is noted that the selection of the appropriate filter length for the third filter element 13 is a trade-off between the additional signal power and the additional noise power received, when the filter length is increased. Hence, it is advantageous that the power comparison unit 20 switches back to a larger filter length to avoid degrading the equalized data, when the determination of the power determination unit 14 indicates the need to switch back to a large-band interpolation filter. This is a possible way to secure the reduction of the interpolation filters bandwidth to reject Gaussian noise.

If the first filter element 11 is a large-band filter and the filter length of the second filter element 12 is smaller or at most equal to the filter length of the first filter element 11, the power determined by the power determination unit 14, which is the power of the signal filtered with the first filter element 11, is always greater as the power output by the power determination unit 15, because it includes more Gaussian noise. However, if the ratio between the output of the power determination unit 15 and the output of the power determination unit 14 is superior to a certain threshold, then it shows that this excessive amount of power attenuated by the used narrow-band filter of the second filter element 12 is due to a miss-equalized echo.

It is noted that the receiver apparatus can also be applied using a narrow-band filter for the first filter element 11 as a reference, to detect a possibility to reduce on the fly the bandwidth of the third filter element 13 by detecting the absence of long echoes.

When the appropriate filter length for the third filter element 13 is determined and set in the third filter element 13, the signal received is filtered by the third filter element 13 and sent to the output 18.

The decoder unit 9 decodes the signal and outputs the decoded signal to the signal processing unit 22 for further signal processing. After further signal processing a signal is output at the output 4 of the receiver apparatus 1.

Although an exemplary embodiment of the invention has been disclosed, it will be apparent to those skilled in the art that various changes and modifications can be made which will achieve some of the advantages of the invention without departing from the spirit and scope of the invention. Such modifications to the inventive concept are intended to be covered by the appended claims in which the reference signs shall not be construed as limiting the scope of the invention. Further, in the description and the appended claims the meaning of “comprising” is not to be understood as excluding other elements or steps. Further, “a” or “an” does not exclude a plurality, and a single processor or other unit may fulfill the functions of several means recited in the claims.

Claims

1. Receiver apparatus for receiving a signal over a fading channel, wherein symbols are modulated onto subcarrier frequency channels, and wherein pilots are placed at some of the subcarrier frequency channels, which receiver apparatus comprises: a frequency domain interpolation unit to interpolate the pilots at the subcarrier frequency channels, wherein said frequency domain interpolation unit comprises a filter unit for filtering the signal received in a frequency domain, and a power comparison unit, wherein said filter unit comprises a first filter element, a second filter element, and at least a third filter element,wherein a length of said second filter element is variable with respect to a length of said first filter element, wherein said power comparison unit compares a signal power of the signal filtered with said second filter element with a signal power of the signal filtered with said first filter element, wherein a processing unit determines an appropriate filter length for said third filter element based on an output of said power comparison unit, and wherein said third filter element filters the signal received with a filter length of said third filter element set to said appropriate filter length determined.

2. Receiver apparatus according to claim 1, characterized in that said power comparison unit calculates a power ratio as a fraction value having a numerator which is said power of the signal filtered with said second filter element and a denominator which is the power of the signal filtered with said first filter element, and that said processing unit determines said appropriate filter length for said third filter element based on the power ratio.

3. Receiver apparatus according to claim 2, characterized in that said processing unit varies said filter length of said second filter element so that the power ratio is at least approximately 100%.

4. Receiver apparatus according to claim 3, characterized in that said processing unit determines said appropriate filter length of said third filter element as the shortest filter length of said second filter element for which said power ratio is nearly 100%.

5. Receiver apparatus according to claim 2, characterized in that said processing unit varies said filter length of said second filter element so as to detect at least a long echo of the signal, and that said processing unit determines said appropriate filter length so that the long echo of the signal is within said appropriate filter length, when a power of the long echo is superior to a threshold.

6. Receiver apparatus according to claim 5, characterized in that said processing unit determines said appropriate filter length based on the power of the long echo and an additional power of a Gaussian noise added due to the increase in filter length so as to receive the long echo.

7. Wireless network comprising a sender and at least a receiver apparatus according to claim 1, wherein said sender sends a signal comprising transmitted symbols modulated onto subcarrier frequency channels over a fading channel to said receiver apparatus.

8. Method for transmission of data from a sender to a receiver apparatus, wherein symbols are modulated onto subcarrier frequency channels, and wherein pilots are placed at some of the subcarrier frequency channels, which method comprises the steps of: filtering a received signal comprising pilots of subcarrier frequency channels with a first filter length in a frequency domain;filtering the received signal with at least a second filter length in said frequency domain;comparing a signal power of the received signal filtered with said first filter length and a signal power of the received signal filtered with said second filter length;determining an appropriate filter length for filtering the received signal based on the comparison; andfiltering the received signal with an appropriate filter length in said frequency domain.

9. Method according to claim 8, characterized in that at least a filter length is a large-band filter length.

10. Method according to claim 8, characterized in that at least a filter length is a narrow-band filter length.