Channel estimation in a CDMA receiver using multiple access interference cancellation

Method for channel estimation and receiver

The invention relates to a method for estimating a channel in a receiver of a cellular radio system, in which the received signal is subjected to multi-use interference cancellation.

The method according to the invention can be generally applied in any cellular radio system in which multi-use interference cancellation is applied, but it is particularly suitable for use in a cellular radio system utilizing the CDMA multi-use method.

CDMA is a multi-use method based on spread spectrum technology that has recently been introduced in cellular radio systems alongside the previous FDMAm and TDMAx.

CDMA has several advantages over previous methods, such as the simplicity of frequency planning and spectrum efficiency.

In the CDMA method, a user's narrowband data signal is multiplied by a spreading code that is significantly wider than the data signal over a relatively wide band. The bandwidths used in known test systems are, for example, 1.25 MHz, 10 MHz and 25 MHz. When multiplied, the data signal is spread over the entire band used. All users transmit using the same frequency band at the same time. Each connection between a base station and a mobile station uses its own spreading code, and user signals can be distinguished at the receivers based on each user's spreading code. The aim is to select the spreading codes so that they are orthogonal to each other, i.e. they do not correlate with each other.

In conventional conventionally implemented CDMA receivers, the correlators synchronize to the desired signal, which is identified by the spreading code. The Da35 level signal is returned to the original band at the receiver to be retransmitted with the same spreading code as in the transmission phase. Those signals multiplied by another spreading code do not ideally correlate and return to the narrow band. Thus, they appear as noise for the desired signal. The aim is thus to detect the signal of the desired user among a number of interfering signals. In practice, spreading codes are not uncorrelated and signals from other users make it difficult to express the desired signal by distorting the received signal. This interference between users causes what is called a multi-use disorder.

The more concurrent users there are in the system, the greater the multi-use failure. Thus, the capacity of a CDMA cellular radio system is limited by the mutual interference caused by users to each other as described above. Interference can be reduced by trying to keep the signal power levels of the terminals at the base station receiver as equal as possible through precise power control. In this case, the base station monitors the powers of the received signals, and sends power control commands to the terminals. Another known way to reduce interference is to use a variety of active multi-use interference cancellation methods as well as multi-user simultaneous detection methods.

In a multipath channel, it is essential that the complex amplitude as well as the delay can be estimated for each received signal component with a significant power level so that the corresponding interference component can be removed from the desired signal. This measures amplitude and delay30 which is called channel estimation.

Prior art solutions for channel estimation have not taken into account interference cancellation, but have been considered as separate solutions. As a result, the estimation results are thus made of a signal comprising interfering signals of many users, and the estimation results thus obtained are not the best possible.

It is therefore an object of the present invention to provide a method for estimating a channel which takes into account the elimination of multifunction interference to be performed.

This is achieved by a method of the type described in the introduction, which is characterized in that the channel estimation is performed on a signal which has been subjected to multi-use interference cancellation.

The invention further relates to a receiver in a cellular radio system, wherein the receiver comprises means for performing multi-use interference cancellation on the received signal. The receiver according to the invention is characterized in that the receiver comprises means for calculating channel estimates from a signal which has been subjected to multi-use interference cancellation.

In the method according to the invention, the channel estimator thus takes into account the elimination of the multi-use interference, and since the estimation is performed, the channel estimates obtained from the interference-purified signal 20 are considerably of better quality than has previously been possible.

The method of the invention is suitable for use with any channel estimation algorithm. Correspondingly, the solution according to the invention does not take a position on the multi-use interference cancellation method.

In a preferred embodiment of the invention, the channel parameters are estimated both before and after interference cancellation. The channel estimates calculated from the interference-cleaned signal can be fed back to the first estrator, where they can be utilized.

The invention can also be applied to multistage receivers in which, in the second and later stages, the channel parameters can be estimated after interference cancellation in the earlier stages.

The invention will now be described in more detail with reference to the examples accordin g to the accompanying drawings, in which Figure 1 shows a part of a cellular radio system in which the method according to the invention can be applied, Figure 2 illustrates the structure of a receiver according to the block diagram, Figure 3 illustrates an alternative structure Fig. 4 illustrates the structure of a multistage receiver according to the invention by means of a block diagram, Fig. 5 illustrates the structure of a receiver stage and Fig. 6 illustrates the overlap of frame structures in Asynchronous traffic .

In the following, the invention will be described using, but not limited to, the CDMA cellular radio system as an example. It is not per se relevant to the invention which multi-purpose system is used. The only condition is to use one of the multi-use Troubleshooting methods. In addition to CDMA systems, multi-use interference cancellation methods can also be applied, for example, in TDMA systems to eliminate co-channel interference.

Figure 1 shows a part of a cellular radio system in which the method according to the invention can be applied. The figure shows a base station 10 communicating with Subscriber terminals 11 14 in its coverage area. In a CDMA system, subscriber terminals thus transmit their own signal to the base station 15 to 18 in the same frequency band and thus interfere with each other to some extent depending on the cross-correlation characteristics of the spreading codes used. Correspondingly, the base station transmits to all terminals using the same frequency band. In addition, the recei vers may receive signals from a neighboring cell.

Figure 2 illustrates the basic structure of a receiver 977180 implementing a preferred embodiment of the method according to the invention, essentially for the invention by means of a block diagram. The receiver comprises first estimating means 20, the input of which is a received and Digitized signal, in which means a preliminary channel Estimation is performed by some known estimation method. The complex amplitude and delay are usually estimated from the channel. The receiver further comprises detector means 21 for calculating preliminary symbol estimates for the received transmission.

The receiver according to the invention further comprises means 22 for performing interference cancellation on the received signal by some known interference cancellation method. The receiver according to the invention further comprises second estimating means 23 in which the channel parameters are estimated from the interference-free signal, whereby better estimates are obtained than in the first estimating means performing the preliminary estimation. Also in other estimation means, the estimation can be performed by some known estimation method.

The receiver can also utilize feedback 24 from the second estimating means 23 to the first estimating means 20. In this case, the estimation results obtained from the second estimating means can be utilized when calculating preliminary estimation decisions, for example by_of channel parameters are exported to the first estimation means as default values ​​when calculating the initial channel parameters for the next sample b_{n + 1}.

Figure 3 illustrates the basic structure of a receiver implementing a second embodiment of the method according to the invention by means of a block diagram essentially for the invention. The receiver comprises a plurality of matched filters or RAKE receivers 31, 37, 44, each adapted to receive and demodulate a single user signal that can be distinguished from one another based on a spreading code. The received transmission is applied to a first estimating means 30, where the estimation of the preliminary channel parameters is performed. From the estimating means, the signal is applied to a first matched filter 31 where the desired signal is demodulated and further to a first detector 32 where a bit decision is made. The signal 33 from the detector 32, thus comprising an estimate of the transmission of the first user, is passed on to other parts of the receiver, as well as to the first regeneration means 35, where the detected signal is regenerated, i.e. multiplied by a spreading code. The obtained regenerated signal is passed on to the first adder means 36, where it is subtracted from the received transmission introduced to the adder means 36 via the first delay means 34.

The signal from the first adder 36 thus comprises a received transmission minus the effect of the signal demodulated by the first matched filter 31, i.e. typically the strongest signal. Said signal is applied to a second estimating means 38 and a second matched filter 37. Thus, in the second estimating means 38, the channel parameters are re-estimated from the signal which has been subjected to interference cancellation, i.e. the effect of the strongest signal has been removed therefrom. The estimates thus obtained are, of course, of better quality than the estimates obtained from the interfering signal. The re-estimated signal is demodulated in a second matched filter 37, which is typically matched to the second strongest signal. As above, the received signal is detected in the second detector 39 and the detected signal 40 is passed on to other stages of the receiver. The signal 40 is also applied to a second regeneration means 42, where the detected signal is regenerated, i.e. multiplied by a spreading code. The regeneration uses the estimation results obtained in the second estimation medium 38. The obtained regenerated signal is passed on to the second adder 43, where it is subtracted from the received transmission input to the adder 43 via the second delay means 41.

Accordingly, at the receiver, all signals are estimated, demodulated, detected, and regenerated until all signals are processed. Typically, the operations are performed on the signals in order of intensity, with the last matched filter 44 and detector 45 processing the signal with the weakest strength, but the processing order may be different. However, in the receiver according to the invention, it is essential that the channel parameters of the signals are estimated from a signal which has been disturbed, i.e. in which, in the example described above, the effect of signals from users stronger than the user in question has been removed.

The method according to the invention can also be applied in a multistage receiver, in which all the users to be received are treated in parallel and the symbol estimates are refined in successive receiver stages by repeating the reception procedure after the interference cancellation. Similarly, channel estimates can be refined iteratively at each stage. Figure 4 illustrates the basic structure of a multistage receiver. The receiver of the figure comprises three consecutive receiver stages 47 to 49, in which the symbol estimates of the incoming signal are estimated in each stage. The latter degrees always give a more accurate result compared to the previous degrees. In the receiver according to the invention, multi-use interference can be performed at each stage and the channel parameters of the incoming signal can be estimated. In the latter stages, the parameter estimation is thus performed on a signal of better quality than in the previous stages, and the estimates obtained are more accurate.

Figure 5 illustrates in more detail the possible structure of one receiver stage at the block diagram level. Each degree can in principle be similar in structure. The receiver stage comprises a plurality of matched filters 51a to 51c and estimating means 52a to 52c, the input 50 of which is either a received signal or a signal from a previous stage. In each estimating means and matched filter, the signal of one received user is processed. In the case of a second or later stage, the channel estimation is thus performed after the interference cancellation performed in the previous stage. The signal from each matched filter 51a to 51c is applied to a corresponding detector 53a to 53c, where a symbol estimate for the received signal is calculated. The detected signal is passed to regeneration means 54a-54c, where the detected signal is regenerated, and in which means the information obtained from the estimating means on the complex amplitudes and phases of the signals is utilized. The regenerated signals are further applied to interference cancellation means 55a to 55c, which can be implemented, for example, by adders to which the transmission received via the delay element 56 is applied, and in which interference from other desired signals is reduced from said transmission. The received signal is passed on to other stages of the receiver.

The solution according to the invention is suitable for use in connection with any known channel estimation algorithm. Channel estimation can be implemented, for example, by means of a correlator following the peaks of the simple channel impulse response or a decision feedback estimator utilizing the already obtained bit estimates.

The channel estimator may also be associated with filtering, in which case, for example, successive channel estimates are averaged, whereby the stochastic variation of the channel estimates can be smoothed. Filtering may comprise processing the signal by, for example, linear, non-linear, adaptive or time-varying processing methods.

The channel estimator can also utilize a prediction procedure, in which case the estimator seeks to monitor and anticipate changes related to the channel. The prediction procedure can be implemented, for example, with a signal processor, which calculates default values ​​for future parameters on the basis of the obtained estimates.

For the sake of simplicity, the method according to the invention and the receiver 10 have been described above mainly using synchronous traffic as an example. However, the solution according to the invention is correspondingly suitable for use in asynchronous traffic. In this case, the requirements caused by asynchrony must be taken into account. Asynchronous traffic is illustrated in Figure 6, which shows portions of two simultaneously received signals comprising time slots 60 to 63 and 64 to 67, respectively. Thus, when calculating and eliminating interference, for example, for time slot 61, those portions of time slots 65 and 66 that overlap with time slot 61 must be taken into account.

Although the invention has been described above with reference to the example according to the accompanying drawings, it is clear that the invention is not limited thereto, but can be modified in many ways within the scope of the inventive idea set forth in the appended claims.