DE10122643B4 - Determination of the detection time of pre-equalized data signals - Google Patents

Abstract

Device for sending a pre-equalized transmission signal over one or more transmission channels with a transmission device for sending the transmission signal to a receiver, a channel information device (KS) for providing channel impulse responses and a data processing device for evaluating the channel impulse responses, characterized in that the data processing device is based on the Channel impulse responses and optimization criteria a detection time (ODZP) can be determined at which the receiver advantageously detects the transmission signal.

Description

The present invention relates to an apparatus for transmitting an equalized transmit signal over a plurality of transmission channels. Moreover, the present invention relates to a corresponding method for transmitting predistorted transmission signals.

Code Division Multiple Access (CDMA) allows multiple data streams to be simultaneously transmitted over a common frequency band. According to KD Kammeyer: "Nachrichtertrag", 2nd ed., Series Information Technology, Teubner, Stuttgart, 1996, 622-624, modulates the symbols of the data streams to be transmitted with so-called spreading codes. The spreading of several data streams is in 1 symbolically represented. The data streams d ⁽¹⁾ ... d ^{(K) of} the users (1) ... (K) are multiplied by respective spreading codes c ⁽¹⁾ ... c ^(K) and sent together by the sender to the respective receivers.

Cellular radio channels generally consist of several paths. When transmitting CDMA signals over such multipath or multipath channels, the simultaneously transmitted data streams interfere with each other. However, the disturbances can be eliminated by different techniques. On the one hand this can be done in the receiver by joint detection (JD) according to A. Klein, GK Kaleh and PW Baier: "Zero Forcing and Minimum Mean-Square-Error Equalization for Multiuser Detection in Code-Division Multiple-Access Channels", IEEE Trans. Vehic. Tech., Vol. 45 (1996), 276-287. Since this common detection in the receiver is very expensive, a different method for troubleshooting is often used. This method is called Joint Predistortion (JP) in the transmitter and is in the WO 99/56441 A1 and in the WO 00/27046 A1 described in more detail. The principle of common prior equalization is in 2 shown. The data streams d ⁽¹⁾ ... D ^(K) are jointly pre-equalized in a common pre-equalizer GV on the basis of the codes c ⁽¹⁾ ... C ^(K) and on the basis of impulse responses h ⁽¹⁾ ... H ^(K) , The impulse responses h ⁽¹⁾ ... H ^(K) are responses that the transmitter receives to channel-estimated pulses or reference signals sent by the receiver. When in 2 On the right-hand side of the receiver, the common data stream is despread by the conjugate complex despreading code c ^{(k) *} assigned to the receiver in order to obtain the detected data DD (1 ≦ k ε K). The transmission signals are pre-equalized in the transmitter so that they can be optimally detected at a given time. That is why very high transmission powers are usually required.

From the document DE 100 01 491 A1 For example, a method and a device for the synchronization of data transmission devices are known which, while taking into account the principles described above, can still be optimized further.

The object of the present invention is thus to be able to securely transmit pre-equalized signals with the lowest possible transmission power.

According to the invention, this object is achieved by a device for transmitting an equalized transmission signal over a plurality of transmission channels with a transmission device for transmitting the transmission signal to a receiver, a receiver device for receiving a reference signal and a data processing device for evaluating the reference signal, wherein with the data processing device on the basis of the reference signal a detection time can be determined at which the receiver conveniently detects the transmission signal.

Furthermore, the above-mentioned object is achieved by a method for transmitting a predistorted transmission signal over a plurality of transmission channels by receiving a reference signal from a receiver, evaluating the reference signal for channel estimation and transmitting the transmission signal to the receiver taking into account the evaluation of the reference signal, wherein when evaluating the reference signal a detection timing is set at which the receiver conveniently detects the transmission signal.

Advantageous developments of the device according to the invention and of the method according to the invention can be found in the subclaims.

The invention is based on the idea that the predistorted transmission signals differ depending on the time difference between the transmission time and the assumed detection time. The transmission signals are generally not only temporally shifted from each other, but also different in their course.

If the time difference between the transmission and detection time is chosen unfavorably, the pre-equalized signal from the transmitter may contain very high energies. This is the case, for example, when, at a detection time, little of the transmission signal is forwarded through the transmission channel to the receiver, but precisely this detection time point has been selected.

In 3 are exemplary the amplitudes of the impulse response of a transmission channel shown. Upon an impulse of the channel estimation receiver, the transmission channel ideally provides, for three transmission paths, three differently delayed pulses of different amplitude as impulse response. 4 shows the resulting signal energies with common pre-equalization. In this case, the transmission signal SS, the reception signal ES and the detection area DB are shown over the time ZT. It can be clearly seen that the cumulative energy over all three transmission paths at the receiver corresponding to the delays 3 at the beginning of the transmission and decreases gradually at the end of the transmission.

Prior art predistortion systems are typically designed to avoid small receive signals by using high transmit energies. In general, this results in both higher overall transmit power and higher peak-to-average ratio (PAR) ratios. According to the invention, the time difference between the transmission of pre-equalized signals and the detection time of these signals is selected such that an optimization criterion is met. In particular, the time difference can be selected so that the transmission signals have lower powers. As a result, the interference of the transmission system can be kept lower or with correspondingly increased transmission power better transmission qualities (given by transmission error rates) can be achieved.

The time difference between the transmission of pre-equalized signals and the detection instant can advantageously also be selected such that the transmission signals have smaller maximum-to-average power ratios. This allows more efficient transmit power amplifiers to be used.

Alternatively, the detection time may also be set as the time at which the cumulative energy of the received signal formed from a plurality of offset partial signals has reached half of its maximum value.

The time difference between the transmission of pre-equalized signals and the detection time of these signals is therefore not predetermined according to the invention, but is chosen as a function of the transmission channel so that an optimization criterion is met.

The present invention will now be explained in more detail with reference to the accompanying drawings, in which:

1 a spreading of several data streams;

2 a common prior equalization;

3 Amplitudes of a channel impulse response;

4 Signal energies with common pre-equalization;

5 a burst structure;

6 a device for pre-equalization;

7 a timing in TDD operation with pre-equalization;

8th an example of transmit signal power values and maximum-to-average power ratios at various detection times; and

9 a cumulative impulse response over time.

The following embodiment represents a preferred embodiment of the present invention and is reproduced for the UMTS TDD mode extended by the possibility of common pre-equalization.

A cellular radio system is required. The radio system operates in TDD mode, ie forward link (from base station BS to mobile station MS) and reverse link (from MS to BS) use the same frequency band. The data is transmitted per burst by CDMA. Each burst BU exists according to 5 from a reference signal REF for synchronization or channel estimation and two blocks DB1, DB2 of data symbols to be transmitted. Each user receiving data in the forward link transmits a reference signal in the reverse link. On the basis of the reference signal, the impulse responses of the individual mobile radio channels are estimated in the reverse link. The channel estimates are used to pre-equalize the forward link transmit signals.

6 shows the transmitting and receiving device for channel estimation in the reverse link and for transmitting the pre-equalized signals. Data DA are fed into a modulator MOD of the predistortion device and modulated there with spreading codes. Due to a channel estimation from a channel estimator KS, a pre-equalizer VE equalizes the modulated signal. About a switch or switch SW the pre-equalized signal is transmitted by TDD radio link TF to a receiver, not shown. The channel estimator KS receives from the switch SW the impulse response from the reverse link for channel estimation.

7 shows the timing of TDD operation with pre-equalization. From the mobile station MS, a reference signal is transmitted by transmission S to the base station BS for channel estimation KS. After the channel estimation KS, the common predistortion GV takes place in the base station. Parallel to the common pre-equalization GV in the base station BS, the reception E in the mobile station MS proceeds.

In the following, the procedure for optimizing the predistorted transmission signals is described in the patent application PCT / DE99 / 03329 described algorithm described. Code-matched filters are used to detect the pre-equalized signals. The algorithm calculates the transmit signals of several transmit antennas and is described in baseband.

The data is transmitted block by block. Let d ^(k) = (d ^(k) ₁ , ..., d ^(k) _M ), k = 1, ..., K be the vector of M within a block of data symbols of the k-th user to be transmitted. d = ( d ⁽¹⁾ , ..., d ^(K) ) denote the summary of all data symbols to be transmitted. From these data symbols, the transmission signals s ^{(n) of} the transmission antennas (n = 1, ..., N, N = number of antennas) are formed by linear transformation: s ^T = P · d ^T s = ( s ⁽¹⁾ , ..., s ^(N) )

empfängt der k-te Empfänger des Systems das Signal

Let h ^{(k, n)} = (h ₁ ^{(k, n)} , ..., h _W ^{(k, n)} ) (W = impulse response length) be the impulse responses and n ^{(k, n)} = (n ^{(k, n )} ₁ , ..., n ^{(k, n)} _{L + W-1} ) (L = length of the transmission signals s ⁽ⁿ⁾ ) the additive noise signals of the mobile radio channels from the nth antenna to the kth user. With the matrices

the kth receiver of the system receives the signal

demoduliert das Empfangssignal zu

konjugiert transponierte Matrix R^(k)(t_k)The code matched filter receiver to the k-th user code c ^(k) = (c ^(k) ₁ , ..., c ^(k) _Q ) (Q = length of the spreading codes with M * Q = L) and detection time t _k :

demodulates the received signal

conjugate transposed matrix R ^(k) (t _k )

erhält man als Gesamtvektor ḏ ^ = (d^ (1) / 1 ,..., d ^ (1) / M , ...,d ^ (K) / 1 , ..., d ^ (K) / M ) aller demodulierter Signale: ḏ ^ = R^H(t)·H·s ^T+R^H(t)·n ^T = R^H(t)·H·P·d ^T + R^H(t)·n ^T With the summaries

one obtains as total vector ḏ ^ = (d ^ (1) / 1 , ..., d ^ (1) / M , ..., d ^ (K) / 1 , ..., d ^ (K) / M ) of all demodulated signals: ḏ ^ = R ^H ( t ) · H · s ^T + R ^H ( t ) · n ^T = R ^H ( t ) · H · P · d ^T + R ^H ( t ) · n ^T

We choose now P = A ^t A = R ^H ( t ) · H

Where A ^{t is} the pseudoinverse of the matrix A.

Under realistic conditions, A has maximum rank. Then A · A ^{H is} invertible when K · M ≦ N · Q · M is satisfied, and it holds A ^t = A ^H · [A • A ^H ] ^-1

It follows ḏ ^ ^T = d ^T + R ^H ( t ) · n ^T R ^H ( t ) thus provides the transmitted data symbols d ^T and additive noise.

According to the invention, the vector of the detection times t is now optimally selected. t can z. B. be chosen so that the energy of the pre-equalized transmission signals is minimal.

The minimization of the transmission energy can be carried out by determining the associated transmission signals for all possible t ε {0,..., W-1} ^K , and selecting the transmission signals to the t with the lowest transmission signal energy.

8th shows an example in which the transmission signal energies SSL are shown as a function of the detection time DZP. In the figure, only the energies for the same detection times DZP of all users are shown ( t = (t, t, t)). Lowest transmission energy is achieved for t = (1, 1, 1).

An easier way is to derive t from the channel impulse responses used for the pre-equalization. For this purpose, the cumulative impulse response energies e ^{(k) of} the user channels are determined ( 9 shows the channel impulse response of 3 belonging cumulative impulse response energies):

For the detection time t _k , the value of i is then selected, for the e (k) / i closest to max ( e ^(k) ) / 2 less 1 ( 9 ):

9 shows the gradual increase of the cumulative energy KE in 3 stages over the time ZT to the maximum cumulative energy MKE. The optimal detection time ODZP results in this embodiment in the half maximum cumulative energy HMKE.

The detection times can also be determined such that the maximum to average power ratio MML (also referred to as PAR below) of the transmission signals becomes minimal. 8th shows MML or PAR values for the example given above of the K = 3 users. The lowest PAR value is achieved for t = (3, 3, 3).

The optimization criteria can also be combined so that the individual criteria (eg transmission energy and PAR) are not optimally fulfilled, but the overall criterion is.

Instead of the algorithm shown here, other algorithms can be used for the pre-equalization.

LIST OF REFERENCE NUMBERS

• BS

  base station

  BU

  burst

  THERE

  dates

  DB, DB1, DB2

  detection range

  DD

  detected data

  DZP

  Detection time

  e

  reception

  IT

  receive signal

  GV

  Common pre-equalizer

  HMKE

  half cumulative energy

  KE

  cumulative energy

  KS

  channel estimator

  MKE

  maximum cumulative energy

  MML

  Maximum-to-average power ratio

  MOD

  modulator

  MS

  mobile station

  ODZP

  Optimal detection time

  REF

  reference signal

  S

  broadcast

  SS

  send signal

  SSL

  Transmission signal power

  SW

  Switch

  TF

  TDD radio link

  VE

  Vorentzerrungseinrichtung

  ZT

  Time

Claims (16)

Device for transmitting a pre-equalized transmission signal via one or more transmission channels to a transmission device for transmitting the transmission signal to a receiver, a channel information device (KS) for providing channel impulse responses and a data processing device for evaluating the channel impulse responses, characterized in that the data processing device based on the channel impulse responses and optimization criteria, a detection time (ODZP) can be determined, to which the receiver conveniently detects the transmission signal. Apparatus according to claim 1, wherein with the data processing device on the basis of the channel impulse responses a transmission signal power of the transmitting device can be estimated and the detection time (ODZP) based on the transmission signal power can be determined. Apparatus according to claim 1 or 2, wherein with the data processing device on the basis of the channel impulse responses, a maximum-to-average power ratio of the transmitter estimable and the detection time based on the maximum-to-mean power ratio can be determined. Apparatus according to claim 2 or 3, wherein the detection time is determined by a minimum of the transmission signal power or the maximum-to-mean power ratio. Apparatus according to any one of claims 1 to 4, wherein the energies of the channel impulse responses in the data processing device are time-cumulative energy and the time at which the cumulative energy reaches a predetermined fraction of the maximum cumulative energy is determinable as the detection instant. The device of claim 5, wherein the fraction of the maximum cumulative energy is one-half. Apparatus according to any one of claims 1 to 6, further comprising pre-equalizer means (VE) for commonly predistorting the transmit signal based on the channel impulse responses. Device according to one of claims 1 to 7, wherein the channel information means is a memory means from which the channel impulse responses are readable, a channel impulse receiving means for receiving measured in the forward channel impulse channel responses or a channel estimator, with the impulse responses in the return channel can be measured. A method of transmitting a predistorted transmit signal over one or more transmission channels by providing channel impulse responses Evaluating the channel impulse responses and Sending the transmission signal to the receiver taking into account the evaluation of the channel impulse responses, characterized in that when evaluating the channel impulse responses taking into account optimization criteria, a detection time (ODZP) is set at which the receiver conveniently detects the transmission signal. The method of claim 9, wherein based on the channel impulse responses, the transmission signal power of the transmitter is estimated and the detection time (ODZP) is determined based on the transmission signal power. The method of claim 9 or 10, wherein based on the channel impulse responses, the maximum-to-average power ratio of the transmission signal of the transmitter is estimated and the detection time is determined based on the maximum-to-mean power ratio. The method of claim 10 or 11, wherein the detection time is determined by a minimum of the transmit signal power or the maximum-to-mean power ratio. The method of any one of claims 9 to 12, wherein the energies of the channel impulse responses are integrated over time into a cumulative energy and the time at which the cumulative energy reaches a predetermined fraction of the maximum cumulative energy is set as the detection time. The method of claim 13, wherein the fraction of the maximum cumulative energy is one-half. Method according to one of claims 9 to 14, wherein the transmission signal is pre-equalized before transmission. The method of claim 9, wherein providing the channel impulse responses comprises reading from a memory, receiving channel impulse responses measured in forward channel, and / or measuring the channel impulse responses in the back channel.

Images