The present invention relates to a data transmission method, in which an encoded data signal is transmitted in the form of a data stream of data bursts between a transmitter and a receiver, and to a corresponding data transmission apparatus.
Although in principle the present invention can be employed for arbitrary data transmissions, it will be explained along with the problems it seeks to solve with regard to a cellular CDMA (Code Division Multiple Access) data transmission system.
In particular, by means of CDMA, a plurality of data streams can be transmitted simultaneously over a joint frequency band, as known for instance from K. D. Kammeyer, Nachrichtenübertragung [Communications Transmission], 2nd Edition, in the series entitled Informationstechnik, published by Teubner, Stuttgart, 1996.
In CDMA, a simultaneous transmission of multiple data is effected by encoding the data with codes that are as much as possible uncorrelated, and by superposition of the encoded signals.
In data transmission, among other things, data transmitted in succession or simultaneously interfere with one another; that is, troublesome interference typically occurs in the transmission, in particular intersymbol interference (ISI) from multi-path transmission, and multiple access interference (MAI) by correlated codes.
The following methods are known for interference treatment:
In this type of data transmission, preequalization in the transmitter is an important interference treatment. Joint preequalization in the transmitter of data signals to be transmitted makes simple data detectors possible. In transmission channels that change quickly, however, preequalization in the transmitter leads to higher error rates than with interference treatment techniques in the receiver. For preequalization, the transmitter must know the pulse response of the transmission channel to be used. The TDD (Time Division Duplex) method makes it possible to achieve this knowledge. Accordingly, the channel is estimated before the data transmission.
The following have proved to be disadvantages of the prior art:
A concept fundamental to the present invention is that a combination of joint preequalization in the transmitter and interference treatment in the receiver is performed. The signals to be transmitted are accordingly jointly preequalized in such a way that for some of the data to be transmitted, interference can be eliminated in the transmitter, and the interference of the other data can be treated in the receiver.
This has the particular advantage of making joint preequalization and interference treatment in the receiver possible in a single system. As a result, a simple receiver with a channel that changes slowly is possible, and at the same time, transmission with a channel that changes quickly is possible.
Another concept fundamental to the present invention is that in an interference treatment in the transmitter, the channel estimate required for the preequalization is predicted from a plurality of prior channel estimates.
This has the particular advantage that the channel estimate better corresponds to the channel at the instant of transmission. This improves the transmission when the transmission channel is changing quickly.
Another concept fundamental to the present invention is that an estimate of the channel pulse response used by the transmitter is performed by the receiver, and the data are detected using this channel estimate.
This also has the particular advantage that the transmission when the transmission channel is changing quickly is improved.
Advantageous refinements of and improvements to the applicable subject of the invention are found in the dependent claims.
In a preferred refinement, the interference treatment is performed in terms of at least one code in the transmitter, and the interference treatment of the data to be transmitted is performed in terms of at least one further code in the receiver.
In a further preferred refinement, the data signal is CMDA-encoded.
In a further preferred refinement, the codes of the data signal to be transmitted are pre-distortion-suppressed jointly, and for some codes, the pulse function is used as a channel estimate.
In a further preferred refinement, the data with a channel estimate performed as a pulse function are detected in the receiver by a rake receiving device.
In a further preferred refinement, in an interference treatment in the transmitter, the channel estimate required for the preequalization is predicted from a plurality of prior channel estimates.
In a further preferred refinement, the channel estimate to be predicted is calculated from the prior channel estimates by linear extrapolation.
In a further preferred refinement, an estimate of the channel pulse response used by the transmitter is performed by the receiver, and the data are detected using this channel estimate.
In a further preferred refinement, the received signal is filtered in accordance with the estimated pulse response, and then the data are detected by means of a rake receiving device.
In a further preferred refinement, the estimated pulse response is used for pseudoinverse detection.
One exemplary embodiment of the invention is shown in the drawing and will be explain in further detail in the ensuing description.
Shown are:
In the drawings, the same reference numerals identify components that are the same or functionally the same.
In
In this first embodiment, a combination of joint preequalization and rake reception is employed.
By means of a decision making device in the transmitter, as a function of at least one data transmission parameter, a decision is made as to whether an interference treatment of the data signal to be transmitted is to be performed in the transmitter 1 or in the receiver. The data transmission parameter can for instance be the rate of change of the data channel, or a measure of it.
If for certain data codes the decision is that the interference treatment will be performed in the transmitter 1, then a joint preequalization takes place. In it, channel estimates for connections with interference treatment in the receiver are selected as a pulse function (delta function—pulse response of an ideal channel). The detection of the data with pulse function channel estimate is done by means of a rake receiving device in the receiver.
In
For preequalization of the data signals to be transmitted, the pulse response of the current transmission channel in the transmitter 1 must be known. However, the channel can be estimated only a certain length of time before the transmission. The faster the channel changes, the more the estimate deviates from the current channel. This makes the data transmission highly erroneous, for instance when there are high relative speeds between the receiver and the transmitter.
In this second embodiment, the pulse response of the current channel is predicted from a plurality of prior channel estimates, or extrapolated linearly by the extrapolator 45, for instance. This leads to a marked improvement in the transmission when the channel is changing quickly.
The estimation of the channel pulse responses by the preequalizing transmitter 1 is done during time segments when this station is receiving data. The linear extrapolation of the channel pulse responses estimated during the data reception is applied to the transmission time period. The preequalization of the transmitted signal is done in accordance with the extrapolated channel estimate.
In
The channel pulse response used for the preequalization is estimated in the receiver 2, in this embodiment. Using this estimate, the preequalization is taken into account in the receiver, and the data are detected by interference-treating methods.
In a first example of this embodiment, there is one point for multi-point preequalization with rake reception, as illustrated in
This is a TDD mode, with regularly alternating reception and transmission time segments EZ/SZ, with code division multiple access (CDMA).
In it, an estimate of the channel pulse response used by the transmitter (the preequalizing station) is made by the receiver (rake station). This estimate of the channel is performed during time segments that surround the time segment that is used in the preequalizing transmitting station for the channel estimate (transmitting time segment of the rake station=receiving time segment of the preequalizing station). Next, averaging of the two channel estimates is done.
The filter 60 is used on the received signal, using this estimated pulse response.
Finally, data detection takes place by means of the rake receiving device 70.
In a second example of this embodiment, there is one point for multi-point preequalization with pseudoinverse detection, using the averaged channel estimate instead of the rake detection.
A discrete-time CDMA transmission system with block transmission is assumed. Let d(k)=(d(k)1, . . . , d(k)M), where k=1, . . . , K is the vector of the M data symbols to be transmitted of the kth user. CDMA encoding and preequalization are linear projections of the data vectors d(k) onto the signal vectors s(k) to be transmitted. These vectors are added together to make the total signal s and are broadcast by the transmitter:
in which d(k)T stands for the transposed vector d(k). B(k) is the (M·Q+W−1)×M projection matrix, containing encoding and preequalization, with the spread factor Q of the CDMA codes and having the length W for the channel estimates used in the preequalization.
The signal s is transmitted to the kth user via the kth transmission channel. Let h(k)=(hl(k), . . . hw(k)) be the pulse response of this channel, and let
be the corresponding convolution matrix. If there is additive noise n(k) of the channel, the kth user then receives the signal
With the reception matrix R(k)=(B(k)H·H(k)H·H(k)·B(k))−1·B(k)H·H(k)H, the receiver obtains from the estimates d^ of the transmitted data in accordance with the equation
{circumflex over (d)}(k)
in which H(k)H stands for the conjugated complex, transposed matrix H(k).
Although the present invention has been described above in terms of preferred exemplary embodiments, it is not limited to them but instead can be modified in manifold ways.
The invention can be employed wherever signals to be transmitted have to be preequalized and at the same time connections over quickly changing channels are required, especially in TDD radio transmission systems with preequalization.
Number | Date | Country | Kind |
---|---|---|---|
199 50 021 | Oct 1999 | DE | national |
The present application is a continuation application of U.S. patent application Ser. No. 10/380,583, filed Mar. 14, 2003. The invention described and claimed herein is also described in PCT/DE00/03604, filed on Oct. 13, 2000 and DE 19950021.5, filed Oct. 16, 1999. This German Patent Application, whose subject matter is incorporated here by reference, provides a basis for a claim of priority of invention under 35 U.S.C. 119(a)-(d).
Number | Name | Date | Kind |
---|---|---|---|
3593041 | Labarre | Jul 1971 | A |
4995057 | Chung | Feb 1991 | A |
5432816 | Gozzo | Jul 1995 | A |
5499272 | Bottomley | Mar 1996 | A |
5761237 | Petersen et al. | Jun 1998 | A |
5905946 | Lilleberg et al. | May 1999 | A |
6243425 | Langberg et al. | Jun 2001 | B1 |
20020114297 | Karna et al. | Aug 2002 | A1 |
Number | Date | Country |
---|---|---|
4140742 | Jun 1993 | DE |
19623665 | Apr 1997 | DE |
19747457 | Apr 1999 | DE |
19747369 | May 1999 | DE |
19818215 | Nov 1999 | DE |
69325201 | Nov 1999 | DE |
19909921 | Sep 2000 | DE |
0564937 | Oct 1993 | EP |
4-2226 | Jan 1992 | JP |
9502297 | Jan 1995 | WO |
9905807 | Feb 1999 | WO |
Entry |
---|
Bosch, “Summary of Joint Predistortion,” TSG-RAN Working Group1 meeting #3, TSGR1#3(99)156, 4 pages (19999). |
Bosch, “Tx Diversity with Joint Predistortion,” TSG-RAN Working Group 1 Meeting No. 6, TSGR1#6(99)918, 9 pages (1999). |
Esmailzadeh, Riaz et al., “Pre-RAKE Diversity Combination for Direct Sequence Spread Spectrum Communications Systems,” IEEE International Conference on Communications, pp. 463-467 (1993). |
Esmailzadeh, Riaz et al., “Pre-RAKE Diversity Combination for Direct Sequence Spread Spectrum Mobile Communications Systems,” IEICE Trans. Commun. vol. E76-B(8):1008-1015 (1993). |
Jeong, Dong Geun et al., “Effects of Channel Estimation Error in MC-CDMA/TDD Systems,” 2000 IEEE 51st Vehicular Technology Conference Proceedings, VCT 2000, vol. 3:1773-1777 (2000). |
Klein, Anja et al., “Zero Forcing and Minimum Mean-Square-Error Equalization for Multiuser Detection in Code-Division Multiple-Access Channels,” IEEE Transactions on Vehicular Technology, vol. 45(3):276-287 (1996). |
Siemens AG, “Comments on Joint Predistortion,” 3GPP TSG RAN WG1, TSGR1#9(99)106, 3 pages (1999). |
Number | Date | Country | |
---|---|---|---|
20140349590 A1 | Nov 2014 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 10380583 | US | |
Child | 14451007 | US |