GB2401004A

GB2401004A - Weighing signals according to estimated interference before combination in a rake receiver.

Info

Publication number: GB2401004A
Application number: GB0309094A
Authority: GB
Inventors: Yuk Ching Chow; Yan Qing Bian; Michael Philip Fitton
Original assignee: Toshiba Research Europe Ltd
Current assignee: Toshiba Europe Ltd
Priority date: 2003-04-22
Filing date: 2003-04-22
Publication date: 2004-10-27
Anticipated expiration: 2023-04-22
Also published as: GB2401004B; GB0309094D0

Abstract

A wireless signal receiver, in particular but not exclusively a combiner for Rake finger signals. There is provided apparatus for use with a Rake receiver having a plurality of rake fingers coupled to a combiner which combines the outputs of said fingers based on a determined weight corresponding to each finger; the apparatus comprising: means for determining an interference parameter of a signal at the output of each finger; means for determining the weights according to said parameters. The means for determining an interference parameter of a signal at the output of each finger comprises means for comparing each said signal with an estimated symbol corresponding to said signal.

Description

1 240 1004

RAKE RECEIVER

FIELD OF THE INVENTION

The present invention relates to wireless signal receivers, and in particular but not exclusively to combiners for Rake finger signals.

BACKGROUND OF THE INVENTION

Rake receivers are utilised especially in CDMA based receivers such as those implemented at wireless bases stations and mobile terminals. Rake receivers comprise a number of Rake fingers each having different delays arranged to correspond to expected multipath signal components of a wanted signal arriving at the receiver with corresponding delays. The signal components received by each Rake finger are then combined to provide an improved received wanted signal for further processing within the receiver. An example Rake receiver design is illustrated in figure 1.

Rake receivers typically use RSSI (received signal strength indicator) measurements to determine combination weightings or the ratio by which to combine the various signal components on the Rake fingers. This is intended to minimise the effects on the combined received signal of high interference on one or some of the received signal components.

However the error performance of a rake receiver using RSSI-based combining degrades under the situation that the interference signal level of each rake finger is different to the other. This is because the combining weights used in RSSI-based combining are not a function of SINR (signal-to-interference plus noise ratio).

US 6,192, 066 B1 (Asanuma) discloses a system for determining combination weightings based partly on received signal strength on each finger and also partly on a interference level determination on each finger. The interference determination is made by a comparison of the received power of a wanted signal on the wanted path (the rake finger in question) with that received on the other paths (the other rake fingers).

In D. Noneaker, "Optimal combining for rake reception in mobile cellular CDMA forward links", IEEE MILCOM, pp.842-846, Oct., 1998, combining weights based on RSSI and SINR are compared. However no implementation is derived.

Similarly S. Tantikovit, A. Sheikh and M. Wang, "Combining schemes in rake receiver for low spreading factor long-code W-CDMA systems", IKE Electronics Lett., pp.1872- 1874, Oct., 2000, also compares the performance of RSSI and SINR based rake finger combining techniques.

W. Kuo "Iinprove rake receiver using finger variance weight", IEEE ICC, pp.1163- 1167, Jun., 2001, proposes using combination weights based on measured pilot signal strengths at the different fingers. The weights used are based on a ratio of the measured pilot strength to the measured variance at different fingers. In particular it is proposed to use the channel gains of the multipath components for SINR measurement, i.e., the SINR for fib path is SINRf (n) = Kf (n) |Kk (n)l k=] Of Eq. 1 However this does not take into account interference components which are caused by the imperfect auto- and cross-correlations of spreading codes.

US6285861 discloses a rake receiver combining arrangement utilising two antennas with a rake receiver each, and arranged such that the interference correlated between a finger associated with the first antenna and the corresponding finger of the second antenna is minimised with respect to the desired signal.

SUMMARY OF THE INVENTION

In general terms the invention provides a method of estimating the SiNR value for the signal component received by each of a number of Rake fingers, especially in order to determine weight values for combining these signal components. The SINR values are determined by estimating the symbol received by the Rake fingers and comparing this with the received signal component on each rake finger. Preferably the symbol is an unexpected symbol such as a data symbol. The symbol may be estimated by using the output of a standard RSSI weighted combining Rake receiver and a symbol detector or estimator. By subtracting the signal on each rake finger from the expected signal based on this symbol, the SINR value for each finger can be calculated. An SINR based rake combining technique can then be used to provide an improved rake output for further processing, including re-estimation of the symbol. Alternatively or additionally the S1NR for each finger could be used to disconnect a rake finger from the output combiner if its estimated SINR is too low, in order to prevent degradation of the combined signal quality.

This provides for improved accuracy of SINR estimation and in one of the embodiments improved SINR based rake finger combination. In turn this provides for improved reception quality, especially with respect to wireless communications between mobile terminals and a base station. In addition, this arrangement allows for the reduction of the base-station and terminal transmission powers given the improved accuracy of reception, hence an overall reduction in interference in the network.

This can advantageously be combined with SINR estimation based on a comparison of each Rake finger received signal with an expected signal based on a known symbol such as a pilot symbol. This further improves reception quality, power reduction and overall interference levels.

In one embodiment a modified rake receiver utilises tentative user data symbols, for example from a symbol detector or estimator, to estimate SINR in each rake finger.

This is preferably achieved by decoding the received symbol using the combined rake receiver signal determined without using SiNR to set the combination weights, for example by using RSSI or some of the other techniques described above. Then, combiner weights that are based on the SINR of each rake finger can be calculated.

In particular in one aspect the present invention provides an apparatus for use with a Rake receiver having a plurality of rake fingers coupled to a combiner which combines the outputs of said fingers based on a determined weight corresponding to each finger; the apparatus comprising: means for determining an interference parameter of a signal at the output of each finger; means for determining the weights according to said parameters.

The weights may be applied to the received signals from which they were determined, or later received signals depending on the configuration of the apparatus. By applying, and updating, the weights to subsequent signals, storage requirements are reduced.

Preferably the means for determining an interference parameter of a signal at the output of each finger comprises means for comparing each said signal with an estimated symbol corresponding to said signal. The symbol may be an unknown data symbol for

example.

Preferably the apparatus further comprises means for estimating said symbol from the plurality of signals. This might include for example a hard or semi-hard symbol detector or decision block or algorithm.

Preferably there is a second combiner which combines the plurality of signals prior to said symbol detection or estimation. The second combiner may be configured to use weights for the various fingers based on a measure of the RSSI value for each respective signal. Preferably the rake finger signals are combined according to the ratio of each said parameter value to the combined parameter values. s

While a Rake receiver is preferred it is possible other receiver types receiving multiple time displaced signal components which are to be combined could alternatively be used.

In another aspect the present invention provides apparatus for determining a wanted signal from a received signal, the apparatus comprising: means for determining a number of time displaced signal; means for combining the signal components to determine a combined signal; means for estimating a symbol corresponding to the combined signal; means for determining an expected signal corresponding to said symbol; means for determining the difference between the expected signal and each time displaced signal component in order to determine an interference parameter associated with each time displaced signal component; means for re-combine the time displaced signal components depending on their respective interference parameters.

Preferably the time displaced signals are determined by respective fingers in a rake receiver.

Similarly a corresponding method is also provided.

In general terms in another aspect the present invention provides methods and apparatus for estimating interference parameters in received signals, and consequently it is also provided for determining a wanted component of the received signal. The received signal may be one or a combination of signal components from one or more rake fingers for example, or from some other type of receiver. A symbol corresponding to the received signal is detected or estimated, for example using hard or semi-hard symbol detectors. An expected signal corresponding to this estimated symbol is also estimated or otherwise determined. The difference between the expected and received signals is then determined in order to arrive at the interference parameter. Preferably this is the signal-tointerference-plus-noise ratio. Knowing this, the wanted signal, that is without this interference component can also be determined.

In particular in this other aspect the present invention provides a method of estimating an interference parameter in a received signal, the method comprising: receiving said signal; estimating a symbol corresponding to the signal; determining an expected signal corresponding to said symbol; determining the difference between the expected signal and the received signal.

There is also provided a method of estimating a wanted signal component in a received signal, the method comprising: receding said signal; estimating a symbol corresponding to the signal; determining said wanted signal component by determining an expected signal corresponding to said symbol.

Determining the expected signal comprises determining the channel characteristics associated with the expected signal. Typically this is achieved by comparing say a known pilot symbol to that symbol as received in a signal and from this determining the effect the channel is having on transmitted symbols.

Preferably receiving said signal comprises dispreading samples of a received CDMA signal. Preferably this further comprises receiving and dispreading a number of time displaced samples of said CDMA signal in a rake receiver.

The interference parameter may be the power of the interference-plusnoise component of the received signal, in which case a moving average of the interference-plus-noise signal strength over a predetermined period is determined.

Corresponding apparatus are also provided.

In addition, software corresponding to the above methods is also provided, the software including program code embodied on a code readable medium such as an optical disk, magnetic storage, or a transient medium such as a signal. Similarly there is also provided configurable and re- configurable hardware arranged to implement the above apparatus.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the present invention will now be described in detail with reference to the attached drawings, by way of example only and without intending to be limiting, in which: Figure 1 is a schematic showing a conventional rake receiver architecture; Figure 2 is a block diagram showing a modified architecture of the embodiment; Figure 3 is a flow chart showing operation of the embodiment of figure 2; Figure 4a is a schematic showing a circuit arrangement of the rake receiver of figure 2; Figure 4b is a schematic showing an alternative circuit arrangement of the rake receiver of figure 2; Figure 5a and 5b illustrate symbol detection; Figure 6 is a schematic showing a circuit arrangement of a rake receiver according to another embodiment; Figure 7 shows a signalling format for a pilot symbol based transmission; Figure 8 shows the average bit error probability of various types of 2-user transmission under equal paths Rayleigh fading channel; and Figure 9 shows the average bit error probability of various types under 3 equal paths Rayleigh fading channel.

DETAILED DESCRIPTION

Figure I shows a known rake receiver structure with M rake fingers utilising RSSI based combining. In RSSI-based combining, the weight for each rake finger, w(n), is proportional to the received signal strength of the rake finger or the channel-gain of the multipath that the rake finger is locked to.

The output of each rake finger, Zf (n), at nit symbol interval can be written in term of desired signal and interference signal components, Zf (n) = Zf DS(n)+Zf,+N (n), for f = 1.. Eq. 2

where M denotes the number of rake fingers in the rake receiver. Zf,Ds(n) and Zf /+N(n) represent the desired signal and interference plus noise signal (I+N) components of Zf (n). In a single-cell operating environment, Zf,+N(n) can expressed as Zf [+N (n) = Zf MPI (n) + Zf,IC! (n) + Zf,N (n) , for J = I..M Eq. 3 The subscripts DS, MPI, ICI and N represent the desired signal, multipath interference, multichannel interference and noise components of Zf (n). From Eq. 2, the S1NR of Zf (n) is a function of the desired signal to the interference plus noise signal ratio, i.e., SINRr(n)=funtZr ( )), forf=l...M Eq. 4 It can be seen from Eq. 2, that once the desired signal component, Zf Ds(n), is known, the (I+N) component, Zf,+N(n), can be calculated, hence the value of SINR,(n) can be estimated. However, Zf Ds(n) is a function of the transmitted user data symbol which is unknown to the receiver, and hence makes SINRJ(n) difficult to determine.

A modified rake receiver according to an embodiment is shown in Figure 2 and its operation is summarised below with reference to figure 3.

Block (1): A signal is received and a symbol corresponding to the signal is estimated.

This is achieved using a conventional rake receiver output and deriving a tentative user data symbol, for example by using a hard or semi-hard symbol detector.

Block (2): An expected signal corresponding to the estimated symbol is then determined. For each rake finger, an estimate for the received desired user signal is obtained by using the tentative user data symbol. This is achieved using the estimated channel characteristics in known manner.

Block (3): The difference between the received signal and the expected signal is then determined to obtain an interference parameter. For each rake finger, the interference plus noise component is calculated by using the result obtained in block (2) and the received signal of that rake finger.

Block (4): For each rake finger, average power of the interference plus noise signal that obtained in block (3) is calculated. The average power may be determined by a moving average or other methods.

Block (5): For each rake finger, using results given in blocks (2) and (4) , the weight for that rake finger is calculated. The weight is a function of desired user signal to interference plus noise signal power ratio.

Block (6): Weights each rake finger signal by the new weight, which is obtained as shown in block (5), and combines all weighted signals to form a modified rake receiver output.

Based on the SINR estimation method outlined above, a rake receiver structure that performs SINR-based combining is shown in Figure 4a. An Mfinger rake receiver is assumed in the structure. In this structure, the output of the conventional rake receiver is passed into an additional signal processor module V1 (SPM-V1), which consists of blocks (1) to (6), to performs Eq. 4 to 10 for interference plus noise (I+N) estimation and SINR-based combining. The operations of these blocks are described in more detail below.

Block (1): The tentative user symbol at the nth time interval, s(n), can be estimated from the conventional rake receiver output, Z(n) in Figure 4a. The conventional rake receiver uses RSSI based weights for combining the various rake signal components, however other methods of assessing the weights for the combiner could also be used.

In a first implementation s(n) is calculated by the following formula.

s(n) = a DecH[Z(n)], for f = 1..M Eq. 5 where a is the overall gain introduced by the transmitter and receiver processing.

D ecH[.] represents the hard symbol decision block (detector) that quantises its input into one of the nearest constellation points.

From Eq.2, in order to calculate the interference plus noise component, Z'f,+N(n), from the received signal, Z'f(n), the value of the desired signal component at the Rake finger, Z'f +(n), is required.

Since Z'f,Ds(n) = s(n) lK'ftn)l, therefore, the value of Z'f,Ds(n) can be obtained if s(n), the transmitted user symbol, is known. The hard symbol decision block is used to estimate s(n) from Z(n) [see Eq.5], where Z(n) is the output of the conventional Rake receiver.

The function of the hard symbol decision block is explained with reference to figures 5a and Sb. The diagram shows four possible transmit signal points (black circles) and the received signal (cross) is closest to the transmit signal points which has amplitude of A and phase angle of O. The hard decision block will assume that signal as a transmit signal. Such symbol detectors are well known to those skilled in the art.

Block (2): s(n) represents an estimate for the symbol transmitted by the transmitter.

However this has travelled through a multipath fading channel to reach the receiver.

Therefore the desired signal at the receiver (the product of the transmitted symbol and the channel) needs to be evaluated. The desired signal component at the nth symbol interval, Zf',Ds(n), is estimated by: Zf,Ds (n) = |K; (n)| s(n), for f = 1...M Eq. 6 where |Kf (n)| = |Kf (n)| represents the magnitude of the complex channel gain at th path, i.e. K(n) , and the asterisk represents the conjugate operation. The channel Awn) for each path is estimated using pilot symbols or channels and channel estimators in known manner.

Block (3): The estimated value of Zf (I) can be calculated by subtracting Zf Ds(n) from Zf (n), i.e., Zf it (n) = Zf (n) - Zf DS (n), for f = 1...M Eq. 7 Block (4): The average power of the (I+N) component, Pf +(n), is calculated as PJ,l+N (n) = |[Zf,l+N (n)]avg | , for f = 1...M Eq. 8 or PI, +N (n) = [|Zf,+N (n)| ] for f = 1 M Eq. 9 where [x(n)]aVg represents the averaging function of the signal x(n) within a certain averaging time interval. The optimal averaging time interval will be a function of channel conditions, interference signal strength and its rate of change in the time domain. In one arrangement an averaging length of 80 symbols is implemented. A moving averaging (MA) technique can be applied on x(n) such as a single FIR (finite impulse response) filter, however, other averaging techniques could also be applied, for example an infinite impulse response filter.

Block (5): Using Eq. 5 and 7 (or 8), a new set of weights for each finger at the nth symbol interval is obtained, wfS'NR(n), which is proportional to the SINRJ(n), Wf,S/NR (n) = p for f = I..M Eq. 10 Block (6): SlNRbased combining can then be performed by weighting Zf(n), with Wf,5iNR(n) and then combining with others, i.e.,

M

Z. ,,,d (n) = wf,SINR (n) Zf (n) f= Eq. 11 where the output of modified rake combiner, Zmod(n), can be use for further processing along the receiver chain. This includes for example determining an improved tentative user symbol from the output of modified rake combiner, ZmOd(n). This may be implemented by using the same decoder as used in block 1, in which case the output from the modified rake receiver is fed back to the beginning of the detector algorithm, and the algorithm rerun.

Figure 4b shows an alternative structure of the modified rake receiver that can be used when both phase compensation and amplitude weighting operations are performed inside the rake finger of a conventional rake receiver. Because of this combined operation, only Z'(n) is available to the additional signal processing function SPM-V1.

Blocks (1) to (5) are the same using Zen) instead of ZJ(n) but the following equation can be used in block (6) to calculate Zmod(n) Zrr'd (n) = Wf,sR (n) IKf; )1 Eq. 12 An Alternative method for tentative symbol estimation for Block (1) is described with respect to Figure 6.

Instead of using the hard symbol decision method, DecH[.], the estimated received user signal at the nth time interval, s(n), can be calculated by the following formula.

s(n) = a DecsH [Z(n)] for f = 1..M Eq. 13 where a is the overall gain introduced by the transmitter and receiver processing.

Dec,vH[.] represents the semi-hard symbol decision block that decides which symbol is transmitted at each symbol time interval.

In the DecH{.] implementation, its input is quantised to the nearest constellation point, i.e., both amplitude and phase of the input signal are quantised. In the case of DecsH[.], the phase of a input signal is quantised to the nearest constellation point, however, the amplitude is soft quantised within the nearest constellation point's amplitude. The advantage of using the semi-hard decision method is that the soft amplitude level provides an additional confidence level of the decision block. Referring to figure Sb, the semi-hard symbol decision of received signal #1 is A'exp(jOc), whereas the decision output is Acexp(lOc) for received signal #2. The semi-hard quantisation alternative can be used in either of the arrangements shown in figures 4a and 4b. As a further alternative, the amplitude but not the phase is quantised.

Figure 6 shows an alternative modified rake receiver structure with reduced processing complexity compared with the two structures of Figures 4a and 4b. In the structure of Figure 6, the rake combiner output, Zmod(n) ' is fed back to a different second digital signal processing function SPM-V2 for rake combiner weights calculation. The weights applied at each symbol interval are based on the weights derived from blocks (1) to (6) from the previous symbol period. The previously described equations are modified as follows: Wf SiNR (n) = J id, for f = 1..M Eq. 14 where Zf DS (n -1) can be determined as, Zf As (n -1) = |Kf (n -1)| s(n -1), for f = 1.. M Eq. 15 s(n -1) in Eq. 14 can be calculated as s(n -1) = Dec[Zrnr,d (n -1)], for f = 1...M Eq. 16 Either hard or semi-hard symbol decisions can be used in the above calculation.

The parameter P;+N(n-l) given in Eq.13 can be calculated by using Eq.7 or 8 with the use of Zf IN (n -1) instead of Zf,'+N (n) Whilst these embodiments can be used to determine SINR based combiner weights using unknown transmitted symbol estimates, the SINR estimation method and implementation arrangements described above are also suitable for transmission systems using pilot symbol-based transmission formats. This is because in the pilot symbol-based transmission, the transmitted pilot symbols, which are known by the receiver, are multiplexed periodically with the user data sequence in each slot before transmission. The signalling format is shown in Figure 7. Therefore, SINR estimation can be aided by the known pilot symbols, for example by reducing errors which might occur in consecutive data symbol estimates.

During the pilot symbol transmission intervals, the desired signal component Zf Ds(n) and (I+N) component Zf,+N(n) can be estimated more accurately at the receiver and hence provide accurate SINRJ(n) estimates. In this case there is no need to perform the process of block 1 because the pilot symbol is already known. Thus only blocks 2-6 are performed using the expected or known pilot symbol. This provides an accurate reference for estimating the SINR when applying the data symbol estimation techniques described above for the SINRin) calculation. Also SINRJ(n) estimation within a pilot symbol transmission interval can remove estimation error that occurs during the data symbol transmission interval from the pervious slot. Thus it can correct for estimation errors in SINR values that may occur consecutively if a sequence of tentative user symbols is incorrect.

In pilot-based transmission systems, the SINR estimation procedure discussed above is separated into two operations, one is used in pilot transmission intervals and the other is used in data transmission intervals.

During the pilot transmission interval, the algorithm performs the following steps at each symbol interval: 1. Based on the known pilot symbol, sp(n), calculates the estimates of desired signal component, Zf DS (n), for all rake fingers (see Eq.5, with s(n) replaced by spin)).

2. Calculates the estimates of (I+N) component, Zf '+N (n), for all rake fingers (see Eq.6).

3. Uses MA method to calculate I+N (n) for all rake fingers.

During the data transmission interval, the algorithm performs the following steps at each symbol interval (as described in more detail previously): 1. Calculates the estimates of user data symbol, s(n) . 2. Based on s^(n), calculates the estimates of desired signal component, Zf DS (n), for all rake fingers (see Eq.5).

3. Calculates the estimates of (I+N) component, Zf '+N (n), for all rake fingers (see Eq.6).

4. Uses MA (moving average) method to calculate I+N (n) for all rake fingers.

5. Calculates the new set of weights, we), for all rake fingers (see Eq.9) .

6. Performs weighting on Zf(n) and combine all weighted signals to produce modified rake receiver output (see Eq.10). If only Zen) is already available, Eq.11 can be used instead.

The embodiment can equally be configured just to operate on estimated data or other estimated symbols. For example this could be applied to pilot-less wireless systems. In this application the channel Kf(n) can be estimated through the use of differential coherent detection for example. For this a differential decoder is needed at the transmitter and Kf(n) is just a previous despread signal, for example Uf(n-l) in figure 4.

In a further alternative, other known or expected symbols, other than pilot symbols, could be used.

The embodiments described allow rake receivers to perform SINR-based combining instead of RSSI-based combining as used in conventional rake receiver. As discussed above, the drawback of the RSSI-based rake receiver is that bit error performance is degraded when the interference level in each rake finger is different to the others. This becomes particularly important with respect to inter-cellular interference in cellular networks, as well as in indoor environments with high levels of multipath interference.

The average bit error probability (BEP) of a conventional rake receiver, a modified rake receiver and a rake receiver with perfect SINR-based combining is given in Figures 8 and 9. Also the analytical single user BEP with maximum ratio combining is also given for reference.

Simulation Assumptions: À Sampling rate of 4 samples per chip is used in the simulation.

À 2 users transmission: Desired user has SF= 4 and interference user has SF= 8.

The transmit power of the interference user is 3dB lower than the desired user.

After the data spreading, a real value scrambling code is used for chip scrambling. The length of the scrambling code is 38400 chips.

À Channel coding is not applied in the simulation.

À B-PSK transmission is used for both users and transmit pulse shaping filter is root-raised cosine filter with roll-off factor of 0.22.

Channel models: o 2-path Rayleigh fading channel with velocity of 50m/s. Average power gains of two paths are equal. The delay between the two paths is 5 samples (1.25 chips interval) and is fixed during the simulation.

o 3-path Rayleigh fading channel with velocity of 50m/s. Average power gains of three paths are equal. The delay between the adjacent paths is 6 samples (1.5 chips interval) and is fixed during the simulation.

Number of fingers in the rake receiver is equal to number of multipath. Perfect synchronization at all levels and perfect channel estimation.

À Averaging length of 80 symbols is used for moving average. À Results are based on estimated data only (ie not including use of known

pilot symbols) It can be seen that the error performance of the modified rake receiver of the embodiments provides improvements over a conventional rake receivers at low BEP regions. For example, at average BEP of le-2 and 4e-3, an improvement of 0.2dB and 1.5dB is observed for the 2-path channel simulation, and improvement of 0. 2dB and 0.5dB for the 3-path channel simulation. These results also show that the performance gain of the perfect SINR-base rake receiver over the modified rake receiver is around 2dB in the 2-path channel simulation and 3dB in the 3-path channel simulation. If pilot symbols are available in the transmission, further improvement on the modified rake receiver can be expected.

The skilled person will recognise that the above-described apparatus and methods may be embodied as processor control code, for example on a carrier medium such as a disk, CD- or DVD-ROM, programmed memory such as read only memory (Firmware), or on a data carrier such as an optical or electrical signal carrier. For many applications embodiments of the invention will be implemented on a DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). Thus the code may comprise conventional programme code or microcode or, for example code for setting up or controlling an ASIC or FPGA. The code may also comprise code for dynamically configuring re-configurable apparatus such as re- programmable logic gate arrays. Similarly the code may comprise code for a hardware description language such as Verilog _ or VHDL (Very high speed integrated circuit Hardware Description Language). As the skilled person will appreciate, the code may be distributed between a plurality of coupled components in communication with one another.

The skilled person will also appreciate that the various embodiments and specific features described with respect to them could be freely combined with the other embodiments or their specifically described features in general accordance with the above teaching. The skilled person will also recognise that various alterations and modifications can be made to specific examples described without departing from the scope of the appended claims.

Claims

CLAIMS: 1. A method of estimating an interference parameter in a received

signal, the method comprising: receiving said signal; estimating a symbol corresponding to the signal; determining an expected signal corresponding to said symbol; determining the difference between the expected signal and the received signal.
2. A method according to claim 1 wherein determining the expected signal comprises determining the channel characteristics associated with the expected signal.
3. A method according to claim 1 or 2 wherein estimating the symbol comprises a hard symbol decision or a semi-hard symbol decision.
4. A method according to claim 1, 2 or 3 wherein receiving said signal comprises dispreading samples of a received CDMA signal.
5. A method according to claim 4 further comprising receiving and dispreading a number of time displaced samples of said CDMA signal in a rake receiver.
6. A method according to any one preceding claim wherein the interference parameter is the signal-to-interference-plus-noise ratio.
7. A method according to any one of claims I to 5 wherein the interference parameter is the power of the interference-plus-noise component of the received signal, and wherein the method further comprises determining a moving average of the interference-plus-noise signal strength over a predetermined period.
8. A method of estimating a wanted signal component in a received signal, the method comprising: receiving said signal; estimating a symbol corresponding to the signal; determining said wanted signal component by determining an expected signal corresponding to said symbol.
9. A method according to claim 6 wherein determining the expected signal comprises determining the channel characteristics associated with the expected signal.
10. A method according to claim 6 or 7 wherein estimating the symbol comprises a hard symbol decision or a semi-hard symbol decision.
11. A method according to claim 8, 9 or 10 wherein receiving said signal comprises dispreading samples of a received CDMA signal.
12. A method according to claim 11 further comprising receiving and dispreading a number of time displaced samples of said CDMA signal in a rake receiver.
13. A method of determining a wanted signal from a received signal, the method comprising: determining a number of time displaced signal; combining the signal components to determine a combined signal; estimating a symbol corresponding to the combined signal; determining an expected signal corresponding to said symbol; determining the difference between the expected signal and each time displaced signal component in order to determine an interference parameter associated with each time displaced signal component; re-combine the time displaced signal components depending on their respective interference parameters.
14. A method according to claim 13 wherein the time displaced signals are determined by respective fingers in a rake receiver.
15. Apparatus for estimating an interference parameter in a received signal, the apparatus comprising: means for receiving said signal; means for estimating a symbol corresponding to the signal; means for determining an expected signal corresponding to said symbol; means for determining the difference between the expected signal and the received signal.
16. An apparatus according to claim 15 wherein the means for determining the expected signal comprises means for determining the channel characteristics associated with the expected signal.
17. An apparatus according to claim 1S or 16 wherein the means for estimating the symbol comprises a hard symbol decision block or a semihard symbol decision block.
18. An apparatus according to claim 15, 16 or 17 wherein the means for receiving said signal comprises means for dispreading samples of a received CDMA signal.
19. An apparatus according to claim 18 further comprising a rake receiver for receiving and dispreading a number of time displaced samples of said CDMA signal.
20. An apparatus according to any one of claims 15 to 19 wherein the interference parameter is the signal-to-interference-plus-noise ratio.
21. An apparatus according to any one of claims 15 to 20 wherein the interference parameter is the power of the interference-plus-noise component of the received signal, and wherein the apparatus further comprises means for determining a moving average of the interference-plusnoise signal strength over a predetermined period.
22. An apparatus for estimating a wanted signal component in a received signal, the apparatus comprising: means for receiving said signal; means for estimating a symbol corresponding to the signal; means for determining said wanted signal component by determining an expected signal corresponding to said symbol.
23. An apparatus according to claim 22 wherein the means for determining the expected signal comprises means for determining the channel characteristics associated with the expected signal.
24. An apparatus according to claim 22 or 23 wherein the means for estimating the symbol comprises a hard symbol decision block or a semihard symbol decision block.
25. An apparatus according to claim 22, 23 or 24 wherein the means for receiving said signal comprises dispreading samples of a received CDMA signal.
26. An apparatus according to claim 25 further comprising a rake receiver for dispreading a number of time displaced samples of said CDMA signal.
27. An apparatus for determining a wanted signal from a received signal, the apparatus comprising: means for determining a number of time displaced signal; means for combining the signal components to determine a combined signal; means for estimating a symbol corresponding to the combined signal; means for determining an expected signal corresponding to said symbol; means for determining the difference between the expected signal and each time displaced signal component in order to determine an interference parameter associated with each time displaced signal component; means for re-combine the time displaced signal components depending on their respective interference parameters.
28. An apparatus according to claim 27 wherein the time displaced signals are determined by respective fingers in a rake receiver.
29. Apparatus for use with a Rake receiver having a plurality of rake fingers coupled to a combiner which combines the outputs of said fingers based on a determined weight corresponding to each finger; the apparatus comprising: means for determining an interference parameter of a signal at the output of each finger; means for determining the weights according to said parameters.
30. An apparatus according to claim 29 wherein the means for determining an interference parameter of a signal at the output of each finger comprises means for comparing each said signal with an estimated symbol corresponding to said signal.
31. An apparatus according to claim 29 or 30 wherein the symbol is an unknown data symbol.
32. An apparatus according to claim 29, 30 or 31 further comprising means for estimating said symbol from the plurality of signals.
33. An apparatus according to claim 32 further comprising a second combiner to combine said plurality of signals prior to said symbol estimation.
34. An apparatus according to claim 33 wherein the second combiner determines the weights of each said signal according to its measured RSSI value.
35. An apparatus according to any one of claims 29 to 34 wherein the means for determining the weights according to said parameters comprises means for determining the ratio of each said parameter value to the combined parameter values.
36. A Rake receiver having a plurality of rake fingers coupled to a combiner which combines the outputs of said fingers based on a determined weight corresponding to each finger and an apparatus according to any one of claims 29 to 35.