Interference Cancellation Technique

Abstract

A technique for cancelling interference in a cellular network (100) is provided. The cellular network (100) comprises intra-frequency cells (102, 104, 106) sending a first identifying signal indicative of a first identifier (ID1) and a second identifying signal indicative of a second identifier (ID2). Each of the cells is identified by a combination of the first identifier and the second identifier. As to a method aspect of the technique, first samples of the first identifying signal are extracted from a signal received by a User Equipment (120). The first samples include contributions from different cells (104, 106) of the cellular network (100). A first set of sample points is selected based on the first samples and a first candidate (104-1) for the first identifier (ID1). Second samples of the second identifying signal are extracted from the received signal. The second sample includes contributions from the different cells (104, 106). A second set of sample points is selected based on the second samples and a second candidate (104-2) for the second identifier (ID2). The interference of the first identifying signal and/or the second identifying signal is cancelled from the received signal based on both the first set and the second set.

Claims

1-20. (canceled)

21. A method of cancelling interference in a cellular network, the cellular network comprising intra-frequency cells sending a first identifying signal indicative of a first identifier and a second identifying signal indicative of a second identifier, wherein each cell is identified by a combination of the first identifier and the second identifier, the method comprising: extracting first samples of the first identifying signal from a signal received by a User Equipment, the first samples including contributions from different cells of the cellular network; selecting a first set of sample points based on the first samples and a first candidate for the first identifier; extracting second samples of the second identifying signal from the received signal, the second samples including contributions from the different cells; selecting a second set of sample points based on the second samples and a second candidate for the second identifier; and cancelling interference of at least one of the first identifying signal indicative of the first candidate and the second identifying signal indicative of the second candidate from the received signal based on both the first set and the second set.

22. The method of claim 21, wherein: the samples are extracted in the time domain; and the sets of sample points are selected in the time domain.

23. The method of claim 22, wherein at least one of: the first set of sample points is selected out of a predefined selection time window on a time axis of the first samples; and the second set of sample points is selected out of a predefined selection time window on a time axis of the second samples.

24. The method of claim 23, wherein at least one of: the first set of sample points is selected out of the first 10% of the sample points on a time axis of the first samples; and the second set of sample points is selected out of the first 10% of the sample points on a time axis of the second samples.

25. The method of claim 21, wherein the interference cancellation includes regenerating the corresponding identifying signal according to the first set and the second set.

26. The method of claim 25, wherein the interference is cancelled from the received signal by subtracting the regenerated identifying signal.

27. The method of claim 21, wherein at least one of the first set of sample points and the second set of sample points are represented by a bitmask, wherein each bit in the bitmask determines whether or not a corresponding sample belongs to the set.

28. The method of claim 27, wherein the interference cancellation is based on both the first set and the second set according to a bitwise conjunction of a bitmask representing the first set and a bitmask representing the second set.

29. The method of claim 21, wherein a number of candidates for the first identifier is greater than a number of candidates for the second identifier.

30. The method of claim 29, wherein: the interference is cancelled for the first identifying signal based on the first set and independently of the second set; and the interference is cancelled for the second identifying signal based on both the first set and the second set.

31. The method of claim 21, wherein the first and second sets of sample points are selected based on power of channel states in the time domain estimated for the first and second samples, respectively.

32. The method of claim 31, further comprising estimating the channel states in the frequency domain and transforming the estimated channel states to the time domain.

33. The method of claim 32, wherein the channel states are estimated for the first and second samples by: transforming the first and second samples to the frequency domain; and multiplying at least a portion of the first and second samples with a complex conjugate of first and second identifying signals reproduced according to the first and second candidates for the first and second identifiers, respectively.

34. The method of claim 21, wherein: the extraction of the first identifying signal is based on a timing of the first identifying signal; and the extraction of the second identifying signal is based on a timing of the second identifying signal.

35. The method of claim 21, wherein the different cells are synchronized.

36. The method of claim 21, wherein each cell sends the first identifying signal and the second identifying signal in Orthogonal Frequency-Division Multiplexing (OFDM) symbols.

37. The method of claim 21, wherein the first samples and the second samples are separated by at least a cyclic prefix.

38. A computer program product stored in a non-transitory computer readable medium for cancelling interference in a cellular network, the cellular network comprising intra-frequency cells sending a first identifying signal indicative of a first identifier and a second identifying signal indicative of a second identifier, wherein each cell is identified by a combination of the first identifier and the second identifier, the computer program product comprising software instructions which, when run on one or more processing circuits of a computing device, causes the computing device to: extract first samples of the first identifying signal from a signal received by a User Equipment, the first samples including contributions from different cells of the cellular network; select a first set of sample points based on the first samples and a first candidate for the first identifier; extract second samples of the second identifying signal from the received signal, the second samples including contributions from the different cells; select a second set of sample points based on the second samples and a second candidate for the second identifier; and cancel interference of at least one of the first identifying signal indicative of the first candidate and the second identifying signal indicative of the second candidate from the received signal based on both the first set and the second set.

39. A device for cancelling interference in a cellular network, the cellular network comprising intra-frequency cells sending a first identifying signal indicative of a first identifier and a second identifying signal indicative of a second identifier, wherein each cell is identified by a combination of the first identifier and the second identifier, the device comprising: processing circuitry; memory containing instructions executable by the processing circuitry whereby the processing circuitry is operative to: extract first samples of the first identifying signal from a signal received by a User Equipment, the first samples including contributions from different cells of the cellular network; select a first set of sample points based on the first samples and a first candidate for the first identifier; extract second samples of the second identifying signal from the received signal, the second samples including contributions from the different cells; select a second set of sample points based on the second samples and a second candidate for the second identifier; and cancel the interference of at least one of the first identifying signal indicative of the first candidate and the second identifying signal indicative of the second candidate from the received signal based on both the first set and the second set.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In the following, the present disclosure is described in more detail with reference to exemplary embodiments illustrated in the drawings, wherein

[0032] FIG. 1 schematically illustrates a cellular network comprising a plurality of intra-frequency cells;

[0033] FIG. 2 schematically illustrates a first embodiment of a device for cancelling interference in the cellular network of FIG. 1;

[0034] FIG. 3 shows a flowchart of a method of cancelling interference in the cellular network of FIG. 1;

[0035] FIG. 4 schematically illustrates a combination of a first identifier and a second identifier for identifying the intra-frequency cells of FIG. 1;

[0036] FIG. 5 schematically illustrates resource elements sent by cells of the cellular network of FIG. 1 in the range of reception of a User Equipment;

[0037] FIG. 6 schematically illustrates exemplary power profiles of channel states measured by the User Equipment;

[0038] FIG. 7 schematically illustrates a block diagram for a second embodiment of a device for cancelling interference in the cellular network of FIG. 1;

[0039] FIG. 8 schematically illustrates a block diagram of an interference cancellation entity instantiated twice in the device of FIG. 7;

[0040] FIG. 9 schematically illustrates a valid path selection performed by the entity of FIG. 8 based on both a first set and a second set of sample points; and

[0041] FIG. 10 schematically illustrates a signal regeneration performed by the entity of FIG. 8 based on the valid path selection of FIG. 9.

DETAILED DESCRIPTION

[0042] In the following description, for purposes of explanation and not limitation, specific details are set forth, such as specific device and network configurations and specific methods, steps and functions, in order to provide a thorough understanding of the technique presented herein. It will be appreciated that the technique may be practiced in other embodiments that depart from these specific details. While cellular networks and cell identifiers described herein are consistent with 3GPP Long Term Evolution (LTE), the technique is also applicable in cellular networks according to the Global System for Mobile Communications (GSM) and the Universal Mobile Telecommunications System (UMTS). Furthermore, the technique is applicable in networks using access technology other than 3GPP Radio Access Technologies. For example, a Wireless Local Area Network according to the family of standards IEEE-802.11 can be structured in hotspots that correspond to the cells described herein.

[0043] Those skilled in the art will further appreciate that the methods, steps and functions described herein may be implemented using individual hardware circuitry, using software functioning in conjunction with a programmed microprocessor or a general purpose computer, using one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs) and/or one or more Field Programmable Gate Arrays (FPGAs). It will also be appreciated that the technique disclosed herein may be embodied in a processor and a memory coupled to the processor, wherein the memory stores one or more programs that perform the methods, steps and functions described herein when executed by the processor.

[0044] FIG. 1 schematically illustrates a cellular network 100 including a plurality of intra-frequency cells 102, 104 and 106. Each of the cells 102 to 106 includes a base station 108, 110 and 112. Herein, intra-frequency means that each of the base stations 108 to 112 operates on the same carrier frequency.

[0045] Each of the cells 102 to 106 periodically sends a first identifying signal indicative of a first identifier ID1 and a second identifying signal indicative of a second identifier ID2. The first identifier ID1 and the second identifier ID2 of the cell 104 are indicated at reference signs 104-1 and 104-2, respectively. Analogously, reference signs 106-1 and 106-2 indicate the first and second identifiers of the cell 106.

[0046] When a User Equipment (UE) 120 is within the range of both the cell 104 and the cell 106, it may succeed receiving the identifying signals of the cell 104 and read the first and second identifier 104-1 and 104-2 (e.g., because signal strength of the cell 104 is greater than signal strength of the cell 106). The identifying signals of the cell 104 cause intra-frequency interference, e.g., when the UE 120 searches for the further cells 102 and 106. In case of strong intra-frequency interference caused by the cell 104, it is difficult for the UE 120 to detect the identifiers 106-1 and 106-2 of the weaker cell 106. Interference cancellation for the strong interference cell 104 is necessary before searching for weaker cells 102 and 106. Conventionally, the interference cancellation is performed separately for the first identifying signal indicative of the first identifier 104-1 and for the second identifying signal indicative of the second identifier 104-2.

[0047] A radio signal received at the UE 120 includes contributions of the second identifying signal from both the interfering cell 104 and the weaker cell 106. Depending on the number of possible second identifiers ID2, the neighboring cells 104 and 106 may have identical second identifiers 104-2 and 106-2. As a consequence, the conventional interference cancellation performed for the second identifying signal based on a second candidate, ID2=A, for the second identifier 104-2 of the interfering cell 104 also cancels the second identifying signal indicative of the second identifier 106-2 of the weaker cell 106 from the received signal. Consequently, the UE 120 cannot detect additional weaker cells, such as the cell 106, having the same second identifier 106-2 as the interfering cell 104.

[0048] In the case of an LTE implementation of the cellular network 100, the first identifying signal may be the Secondary Synchronization Signal (SSS). The second identifying signal may be the Primary Synchronization Signal (PSS). The combination of the first identifier and the second identifier is also referred to as a Physical Cell Identity (PCI). There are only three different candidates (also referred to as hypotheses) for the second identifier ID2. As a consequence, the conventional interference cancellation for the PSS is likely to eliminate in the received signal contributions that represent the PSS from different cells.

[0049] FIG. 2 schematically illustrates a block diagram of a first embodiment of a device 200 for cancelling interference in a cellular network, such as the cellular network 100. The device 200 includes a first extracting unit 202 for extracting first samples of a first identifying signal from a signal received by a UE, such as the terminal 120 in the cellular network 100. The first samples include contributions from different cells of the cellular network, i.e., from the cell 104 and the cell 106 of FIG. 1.

[0050] The device 200 further includes a first selecting unit 204 for selecting a first set of sample points out of the first samples based on the first samples and a first candidate, e.g., ID1=a, for the first identifier ID1.

[0051] The device 200 further includes a second extracting unit 206 and a second selecting unit 208. The first and second extracting units 202 and 206 can be implemented as first and second instances of an extracting object. The first and second selecting units 204 and 208 can be implemented as first and second instances of a selecting object.

[0052] The second extracting unit 206 is applied to the received signal and extracts second samples different from the first samples. The second samples represent the second identifying signal and include contributions from different cells, i.e., the cells 104 and 106. The second selecting unit 208 selects a second set of sample points out of the second samples based on the second samples and a second candidate, e.g., ID2=A, for the second identifier ID2.

[0053] The device 200 further includes a cancelling unit 210 for cancelling the interference due to the first identifying signal indicative of the first candidate and/or the second identifying signal indicative of the second candidate. The interference is cancelled from the received signal based on both the first set and the second set.

[0054] FIG. 3 shows a flowchart of a method 300 of cancelling interference in a cellular network comprising intra-frequency cells sending a first identifying signal indicative of a first identifier and a second identifying signal indicative of a second identifier. The first identifying signal and the second identifying signal are included in a signal received by a UE performing the method 300. Each of the cells within the range of the UE is identifiable based on a combination of the first identifier and the second identifier.

[0055] First samples of the first identifying signal are extracted from the signal received by the UE in a step 302. The first samples include contributions as to the first identifying signal from different intra-frequency cells of the cellular network. A first set of sample points is selected based on the first samples and on a given first candidate of the first identifier in a step 304.

[0056] Second samples of the second identifying signal are extracted from the received signal in a step 306. The second samples include contributions as to the second identifying signal from those cells that contributed to the first samples. In a step 308 of the method 300, a second set of sample points is selected based on the second samples and on a given second candidate for the second identifier.

[0057] The interference of at least one of the first identifying signal indicative of the first candidate and the second identifying signal indicative of the second candidate is cancelled from the received signal using both the first set and the second set in a step 310 of the method 300.

[0058] The units 202 to 210 of the device 200 may perform the steps 302 to 310, respectively. The device 200 and the method 300 are implemented in the UE 120. The cellular network may include the intra-frequency cell 102, 104 and 106, e.g., in addition to further inter-frequency and intra-frequency cells.

[0059] FIG. 4 illustrates the structure of a cell identifier N.sub.ID.sup.(cell), which is also referred to as the Physical Cell Identity (PCI) in LTE networks 100. The N.sub.ID.sup.(cell) results from a combination of the first identifier N.sub.ID.sup.(1) included in the first identifying signal SSS and the second identifier IN.sub.ID.sup.(2) included in the second identifying signal PSS. The combination of PSS and SSS identifies one out of 502 different PCIs. The SSS indicates one out of 168 cell-identity groups, N.sub.ID.sup.(1)[0,167]. The PSS indicates one out of 3 different cell identities, N.sub.ID.sup.(2)[0,2], within each of the cell-identity groups according to:

N.sub.ID.sup.(cell)=3N.sub.ID.sup.(1)+N.sub.ID.sup.(2).

[0060] The mobile device 120 uses PSS and SSS jointly for searching available cells and reading cell parameters corresponding to the cell identified by the PCI.

[0061] In a typical cell search procedure, the PSS is detected. The PSS is included in the last symbol 504 of the first slot in each radio frame sent by the cell 104, as is schematically illustrated in the frequency-time grid 500 in FIG. 5 for an Frequency Division Duplex (FDD) implementation in LTE. Frequency is shown along the vertical direction and time along the horizontal direction of the grid 500. Frequency is structured in Resource Blocks (RB), each of which includes 12 subcarriers. Time is structured in subframes, each of which encompasses 1 millisecond subdivided into 2 slots. Each slot includes (in the case of normal cyclic prefix) 7 Orthogonal Frequency-Division Multiplexing (OFDM) symbols.

[0062] The signal received at each of the one or more receive antennas of the UE 120 is sampled (after down-conversion) at a sampling rate of, e.g., 1.92 MHz. For a useful OFDM symbol duration of 66.7 microseconds, each of the first samples and the second samples include 128 samples. The total duration of one OFDM symbol further includes a normal cyclic prefix (CP) length of 4.7 microseconds corresponding to 9 further samples. In a variant operating at a sampling rate equal to 30.72 MHz, the first samples include 66.7 microseconds30.72 MHz=2048 samples. In further variants, the number of the first samples and the number of the second samples is in the range between 128 and 2048, e.g., 256, 512 or 1024.

[0063] Since the PSS is repeated in the last symbol of the first slot in the sixth subframe, the UE 120 is synchronized with the cell 104 on a 5-millisecond basis. Furthermore, the UE 120 reads the second identifier N.sub.ID.sup.(2).

[0064] The UE 120 then detects the first identifying signal SSS indicated at reference sign 502 in the frequency-time grid 500. The PSS 504 and the SSS 502 are sent by the cell 104 in subsequent OFDM symbols in the case of an FDD implementation. Based on the SSS 502, the UE 120 receives the frame timing, the Cyclic Prefix (CP) length and the first identifier NI.sub.ID.sup.(1) specifying the cell-identity group. The technique also applies to TDD implementations.

[0065] FIG. 6 schematically illustrates power 602 of channel states as a function of time (which is also referred to as a channel profile) for the first identifying signal 502. The origin of the time axis is aligned with the beginning of the OFDM symbol allocated to the first identifying signal 502. The position of the first arrow (on the left-hand side) of the first channel profile 602 corresponds to the first of the first samples. The channel states represent a response function of the channel. The length of the first arrow schematically represents the power 602 of the channel state, e.g., the square of the absolute value of the channel response function, at the first sample of the first identifying signal 502, and accordingly for the further first samples.

[0066] The time axis of the first channel state power 602 shown in FIG. 6 approximately encompasses the first 10 samples of the first samples. Black circles 604 on the time axis in FIG. 6 indicate the first set of sample points. The first set of sample points 604 is selected in the step 304 by selecting those first samples, the channel state power 602 of which exceeds a pre-defined selection threshold.

[0067] Similarly, a profile of power 606 of the channel states for the second identifying signal 504 is indicated by vertical arrows. The length of each of the arrows indicates the square of the amplitude of the channel response function at the corresponding points 608 on the time access representing the second set. The origin of the time axis for the profile of channel state power 606 for the second identifying signal 504 is aligned with the beginning scheduled for the OFDM symbol 504 allocated to the second identifying signal.

[0068] Since the first identifiers 104-1 and 106-1 differ for the cells 104 and 106, the first set 604 of sample points selected based on the first candidate 104-1 indicates contributions (also referred to as paths) of the cell 104 exclusively. The second identifier is less specific than the first identifier. Since the second set 608 is selected based on the second candidate 104-2, which coincides with the second identifier 106-2 of the cell 106, the second set 608 includes contributions from both the cell 104 (indicated by the solid-line envelope) and the cell 106 (indicated by the dashed-line envelope).

[0069] Since the time axis for the first set 604 selected out of the first samples and the time axis for the second set 608 selected out of the second samples is aligned with the OFDM symbols 502 and 504 including the first and second identifying signals, respectively, the first set 604 and the second set 608 indicate a time shift or a delay due to signal propagation. Therefore, the first set 604 and the second set 608 are also referred to as paths. The time shifts or paths represented by the first set 604 relate to the first identifying signal filtered according to the first candidate for the first identifier. The second set 608 represents time shifts or paths of the second identifying signal filtered according to the second candidate for the second identifier.

[0070] FIG. 7 schematically illustrates a block diagram of a second embodiment of the device 300 for cancelling interference in a cellular network. The second embodiment shown in FIG. 7 differs from the first embodiment shown in FIG. 3 in that the interference cancellation is performed for both the first identifying signal SSS and the second identifying signal PSS by a first interference cancellation entity 702 and a second interference cancellation entity 706, respectively.

[0071] A receive signal buffer realizes the unit 202 by extracting the first samples that are provided to the first interference cancellation entity 702. The receive signal buffer also realizes the unit 06 by extracting the second samples that are provided to the second interference cancellation entity 706.

[0072] The first interference cancellation entity 702 includes the selecting unit 204 for selecting the first set 604 based on the extracted first samples and the first candidate included in configuration information 703. The first set 604 is represented by channel information that is provided to the second interference cancellation entity 706. The second interference cancellation entity 706 includes the selecting unit 208 for selecting the second set 608 based on the extracted second samples and the second candidate included in configuration information 705.

[0073] The functionality of the cancelling unit 210 is implemented in the first interference cancellation entity 702 for cancelling the interference of the first identifying signal SSS indicative of the first candidate, e.g., the first candidate 104-1. The functionality of the cancelling unit 210 for cancelling the interference of the second identifying signal PSS indicative of the second candidate is implemented in the second interference cancellation entity 706.

[0074] The symmetric functional arrangement of the first interference cancellation entity 702 and the second interference cancellation entity 706 enables implementing the entities 702 and 706 as instances of the same object. The interference cancellation entities 706 (for the PSS) and 702 (for the SSS) share the same functional structure shown in FIG. 8 with different configurations (i.e., different input values and parameters).

[0075] FIG. 8 schematically illustrates a block diagram including the receive signal buffer realizing the functionality of the second extracting unit 206 and the instance for the second interference cancellation entity 706. The receive signal buffer 206 extracts the second samples representing the OFDM symbol 502. The second samples are provided to the second selecting unit 208.

[0076] The second selecting unit 208 includes a channel estimation sub-unit 802 and a valid path selection sub-unit 804. The configuration information 705 includes a frequency domain representation 705A of an ideal second identifying signal indicative of the second candidate. Using the frequency domain representation 705A, the channel states h.sub.0, h.sub.1, h.sub.2, . . . (collectively referred to by reference sign 803) are estimated based on the extracted second samples. In the time domain, the channel state h.sub.0 represents the estimate for the first sample of the second samples, the channel state h.sub.1 represents the estimate for the second sample of the second samples, etc. The channel states 803 are provided in the time domain to the valid path selection sub-unit 804.

[0077] The squares 606 of the absolute values of the individual channel states 803 are computed for those points in time falling within a predefined selection window 900, if the corresponding sample is included in the first set 604. In other words, only those samples that belong to the first set 604 are considered for the cancellation of the second identifying signal. FIG. 9 schematically illustrates the time domain of the second identifying signal 504 and the selection window 900 at the beginning of the symbol duration. The power 606 of the channel states 803 is schematically illustrated by vertical arrows.

[0078] The first set 604 is included in the channel information provided to the second interference cancellation entity 706. The first set 604 indicates samples by listing the corresponding points in time, by listing the corresponding sample numbers or by means of a bitmap (as indicated in FIG. 9). The first set 604 is extrinsic information from the perspective of the second interference cancellation entity 706. The first set 604 indicates valid paths selected based on the first identifying signal 502.

[0079] Furthermore, it is required that the square 606 of the absolute value of each of the channel states 803 exceeds the selection threshold 902. Determining whether or not the value of the individual channel states 803 exceeds the selection threshold 902 is an implementation of the second selecting step 308. Otherwise, the corresponding point in time is not considered for the second interference cancellation (i.e., the corresponding samples is not included in the second set 608). The sample-wise conjunction 812 of the first set and the second set are provided to the cancelling unit 210. The second set 608 does not have to be computed explicitly and/or does not have to be stored in memory. For example, determining the conjunction 812 may be part of the second selecting step 308. For instance, only the samples indicated by the first set 604 are checked for the selection threshold 902.

[0080] In the second embodiment of the device 200 shown in FIG. 7, the first instance for the first interference cancellation entity 702 ignores extrinsic information or sets all bits of the bitmap representing the extrinsic information to one.

[0081] A first variant of the second embodiment for the device 200 uses mutual extrinsic information. The extrinsic information that is input to the first interference cancellation entity 702 includes the second set 608 provided by the second interference cancellation entity 706. The first set 604 provided by the first interference cancellation entity 702 is input to the second interference cancellation entity 706 (as shown in FIG. 7).

[0082] Optionally, the selection threshold 902 depends on a noise level. For example, the selection threshold 902 depends linearly on the noise level. The time domain of the OFDM symbol 504 including the second samples is sub-divided into the selection time window 900 for computing the power 606 of the channel states 803 and a window 904 for computing the noise level, e.g., power of noise and interference.

[0083] The second interference cancellation entity 706 comprises the cancelling unit 210. The cancelling unit 210 includes a regeneration sub-unit 806 and a subtraction sub-unit 808. The combination 812 of the first set and the second set are output as combined channel information to the regeneration sub-unit 806.

[0084] Optionally, the valid path selection sub-unit 804 provides a clean channel profile 814 to the regeneration sub-unit 806. The clean channel profile 814 includes coefficients (h.sub.0, h.sub.1, h.sub.2, . . . ) of the channel states 803 only for those sample points, i.e., points in time, for which the valid path selection sub-unit 804 determined a point in time (i.e., if the point in time is indicated by the combination 812 of the first set 604 and the second set 608).

[0085] The operation of the regeneration sub-unit 806 is schematically illustrated in FIG. 10. Time domain representations 705B of the ideal second identifying signals indicative of the second candidate are combined in the time domain. There is one copy of the time domain representation 705B of the ideal second identifying signal for each point in time (i.e., for each valid path) indicated in the combination 812. Each copy of the time domain representation 705B is weighted according to the coefficient (e.g., included in the clean channel profile 814) representing the channel state 803 at the corresponding point in time (e.g., the sample number indicated by the combination 812). The weighted ideal second identifying signals 705B are time-shifted to the corresponding point in time and added up. The addends are time-shifted according to the temporal position indicated by the combination 812 relative to the beginning of the OFDM symbol 504.

[0086] In the subtraction sub-unit 808, the regenerated second identifying signal 1000 is subtracted from the second samples representing the received OFDM symbol 504. The interference caused by the cell 104 is thus removed in the resulting second samples without removing the PSS signal contributions from cell 106. The interference-free second samples are written back to the receive signal buffer for further analysis of the received signal as to the cell 106. During the interference regeneration 806 and the subtraction 808, a signal included in the CP is also considered.

[0087] An exemplary implementation of the computation performed by the interference cancellation entities 702 and 706 is described. Firstly, the first interference cancellation entity 702 executes interference cancellation for the SSS and outputs the first set 604 to indicate which of the sample points correspond to a valid path. Secondly, the second interference cancellation entity 706 executes interference cancellation for the PSS. The channel estimation 802 based on the PSS is restricted to those sample points indicated by the first set 604 derived from the SSS. In other words, a sample point (i.e., a path) is kept in the channel estimation 802 for the PSS, only if the first set 604 derived from the SSS also indicates this sample point. The channel estimation sub-unit 802 computes the channel estimation 803 in the time domain. The channel estimation 802 for the PSS is used for the PSS regeneration 806.

[0088] For conciseness, the common structure underlying the first and second interference cancelation entities 702 and 706 is described, wherein XSS denotes the SSS for the first interference cancelation entity 702 and the PSS for the second interference cancellation entity 706. Similarly, the word samples refers to the first samples in the instance for the entity 702 and to the second samples in the instance for the entity 706.

[0089] Firstly, the target XSS signal is extracted from the received signal for each antenna according to the steps 302 and 306:

.sub.T.sup.(q)={circumflex over (r)}.sup.(q)(t:t+L.sub.FFT), q=0N.sub.RX1,

[0090] Herein, {circumflex over (r)}.sup.(q) is the received signal on the q.sup.th receive antenna. The received signal is mixed including signal contributions from different cells. The length of the received signal is L.sub.sig. N.sub.RX is the number of receive antennas. The parameter t is the timing of the XSS. .sub.T.sup.(q) denotes the extracted samples of the XSS in the time domain of length L.sub.FFT, e.g., 128 samples. An offset is the sample number to be in advance for data extraction.

[0091] Secondly, the channel estimation sub-unit 802 transforms the samples .sub.T.sup.(q) to the frequency domain by Fast Fourier Transformation (FFT),

.sub.F.sup.(q)=FFT(.sub.T.sup.(q)),

[0092] wherein .sub.F.sup.(q) represents the XSS in the frequency domain.

[0093] Thirdly, a complex conjugate of the frequency domain representation 705A of the ideal identifying signal S.sub.F.sup.(XSS)(m) is multiplied with the received and extracted XSS samples, resulting in a channel estimation in the frequency domain:

.sub.F.sup.(q)(m)=.sub.F.sup.(q)(m)(S.sub.F.sup.(XSS)(m))*,

wherein

[00001]$m^{}=\{\begin{array}{c}m+L_{\mathrm{FFT}}-31,0m30\\ m-30,31m61\end{array},m=0,1,\mathrm{.Math.}.Math..Math.61.$

[0094] Herein, S.sub.F.sup.(XSS)(m), m=0, 1, . . . 61 is the ideal (i.e. originally transmitted) XSS sequence in the frequency domain and .sub.F.sup.(q)(m) is the channel estimation for the XSS in the frequency domain.

[0095] Fourthly, the sequence .sub.F.sup.(q)(m) is partly set to zero:

.sub.F.sup.(q)(m)=0, m=0, 32, 33, . . . L.sub.FFT32.

[0096] Fifthly, an Inverse FFT (IFFT) transforms the channel estimation from the frequency domain to the time domain:

.sub.F.sup.(q)(m)=IFFT(.sub.F.sup.(q)/{square root over (L.sub.FFT)},

wherein .sub.T.sup.(q) is the channel estimation 803 for the XSS in the time domain with length L.sub.FFT.

[0097] The valid path selection sub-unit 804 picks up the first set 604 indicative of the valid paths and calculates channel information 812 to indicate which path is valid.

[0098] Firstly, the channel profile of the one or more receive antennas is combined:

{circumflex over (P)}(n).sub.q=0.sup.N.sup.RX.sup.1g.sub.q|.sub.T.sup.(q)(n)|.sup.2.

[0099] Herein, g.sub.q is the weight of the different antennas and {circumflex over (P)}(n) is the combined channel profile 602/606 in terms of power.

[0100] Secondly, N.sub.max, paths that are greatest in power are selected in the selection window 900 for the first/second set. The selection window 900 encompasses a range [0, 1, . . . L.sub.win1] in terms of first/second samples, i.e., the first L.sub.win samples of the sequence of the first/second samples fall within the selection window 900 for searching valid paths. The first/second set is represented by N.sub.max indices, [i.sub.0, i.sub.1, . . . i.sub.Nmax1], each of which identifies one sample in the set.

[0101] Thirdly, the noise power is computed based on the channel state power {circumflex over (P)} of those samples that are out of the selection window 900:

[00002]${}^2=\frac{\underset{n=L_{\mathrm{win}}}{\overset{L_{\mathrm{FFT}}-1}{.Math.}}.Math.\hat{P}\left(n\right)}{L_{\mathrm{FFT}}-L_{\mathrm{win}}}.$

[0102] Fourthly, the sub-unit 804 verifies that each of the N.sub.max paths exceeds the selection threshold, .sup.2, (which corresponds to the first/second set), and that the N.sub.max paths are indicated by the first set 604, e.g., represented by the bitmask f.sub.in, from outside:

[00003]$\hat{P}\left(i_k\right)=\{\begin{array}{c}0,\mathrm{if}.Math..Math.\hat{P}\left(i_k\right)<{}^2\\ \hat{P}\left(i_k\right).Math.f_{\mathrm{in}}\left(i_k\right),\mathrm{otherwise}\end{array},k=0,1,\mathrm{.Math.}.Math..Math.N_{\max }-1.$

[0103] For the interference cancellation of the SSS, the first interference cancellation entity 702 initializes the bitmask f.sub.in by an all-ones-sequence with length L.sub.win. For the interference cancellation of the PSS, the second interference cancellation entity 706 is provided with the first set 604 by means of the bitmap f.sub.in from the first interference cancellation entity 702.

[0104] Fifthly, a clean channel profile 814 is determined for output to the regeneration sub-unit 806:

[00004]$h_T^{\left(q\right)}\left(i_k\right)=\{\begin{array}{c}0,\mathrm{if}.Math..Math.\hat{P}\left(i_k\right)=0\\ {\hat{h}}_T^{\left(q\right)}\left(i_k\right),\mathrm{otherwise}\end{array},k=0,1,\mathrm{.Math.}.Math..Math.N_{\max }-1$

[0105] Herein, h.sub.T.sup.(q) is the clean channel profile 814, which is initialized by an all-zeros-sequence with length L.sub.win.

[0106] Sixthly, channel information f.sub.out is determined for output 812 to the regeneration sub-unit 806:

[00005]$f_{\mathrm{out}}\left(i_k\right)=\{\begin{array}{c}0,\mathrm{if}.Math..Math.\hat{P}\left(i_k\right)=0\\ 1,\mathrm{otherwise}\end{array},k=0,1,\mathrm{.Math.}.Math..Math.N_{\max }-1$

[0107] In the case of the first interference cancellation entity 702, the channel information f.sub.out is output to the second interference cancellation entity 706.

[0108] The interference of the XSS is regenerated by the sub-unit 806. To remove the interference as much as possible, the minimum CP length .sub.cp.sub._.sub.min is taken into account. E.g., .sub.cp.sub._.sub.min=9 in case of L.sub.FFT=128. So the length of the ideal XSS (i.e., the originally sent XSS) sequence in the time domain has a length L.sub.XXS=L.sub.FFT+.sub.cp.sub._.sub.min.

[0109] Firstly, a regenerated interference, I.sup.(q), is initialized as an all-zeros-sequence with length L.sub.XSS+L.sub.win1.

[0110] Secondly, interference regeneration is performed path by path. An exemplary implementation of the regeneration sub-unit 806 is outlined by below pseudo-code:

TABLE-US-00001 for k = 0: N.sub.max 1 if f.sub.out(i.sub.k) == 0 I.sup.(q)(i.sub.k: i.sub.k + L.sub.XSS) = I.sup.(q)(i.sub.k: i.sub.k + L.sub.XSS 1) + h.sub.T.sup.(q)(i.sub.k)S.sub.T.sup.(XSS), q = 0~N.sub.RX 1 end end

[0111] Above implementation of the regeneration sub-unit 806 is an approximation that avoids a convolution (in the time domain). Advantageously, the complexity of the implementation only depends on the number of valid paths.

[0112] The interference is removed from the received signal by the subtraction sub-unit 808 to get clean first/second samples.

[0113] Firstly, the subtraction sub-unit 808 checks if the signal to be cancelled is out of the signal range before cancellation. A typical start point is

t.sub.start=t.sup.(XSS).sub.cp.sub._.sub.min.

[0114] The start point is changed, if above value is smaller than 0:

t.sub.start=max(t.sup.(XSS).sub.cp.sub._.sub.min, 0).

[0115] The offset caused by the change relative to the typical start point is .sub.start:

.sub.start=t.sub.start(t.sup.(XSS).sub.cp.sub._.sub.min)

[0116] A typical end point is

t.sub.endt.sup.(XSS)+L.sub.FFT+L.sub.wim1.

[0117] The end point is changed, if above value is greater than L.sub.sig1:

t.sub.end=min(t.sup.(XSS)+L.sub.FFT+L.sub.win1, L.sub.sig1)

[0118] The offset caused by the change relative to the typical end point is .sub.end:

.sub.end=t.sup.(XSS)=+L.sub.FFT+L.sub.win1t.sub.end.

[0119] Secondly, the interference is subtracted from the received signal within the range [t.sub.start, t.sub.end]:

{circumflex over (r)}.sup.(q)(t.sub.start:t.sub.end)={circumflex over (r)}.sup.(q)(t.sub.start:t.sub.end)I.sup.(q)(.sub.start:L.sub.XSS+L.sub.win1.sub.end),

q=0N.sub.RX1.

[0120] As has become apparent based on above exemplary embodiments, at least some embodiments of the technique allow cancelling signals of an interfering cell, even if the signal to be cancelled does not uniquely identify the interfering cell. Cells in a cellular network uniquely identified by two or more identifiers can be efficiently cancelled based on the combination of the identifiers for each interfering cell. The interference cancellation can be performed for one or all of the signals indicative of the corresponding identifier.

[0121] In the foregoing, particular principles, preferred embodiments and various modes of implementing the techniques disclosed herein have exemplarily been described. However, the present invention should not be construed as being limited to the particular principles, embodiments and modes discussed above. Rather, it will be appreciated that variations and modifications may be made by a person skilled in the art without departing from the scope of the present invention as defined in the following claims.