Method and test apparatus for determining a pre-equalization matrix

Abstract

A method and test apparatus for determining a pre-equalization matrix are provided. A radio communication tester having a base station emulator and a channel emulator is provided. A device under test having at least two branches is provided. An NxM multiple-input multiple-output (MIMO) connection is established between the radio communication tester and the device under test. The NxM MIMO connection comprises at least two channels. A per-branch reference signal received power measurement is continuously forwarded from the device under test to the radio communication tester. A pre-equalization matrix is determined by the radio communication tester, wherein the per-branch reference signal received power measurement and the presence of additive white Gaussian noise (AWGN) are taken into account when calculating the pre-equalization matrix. The pre-equalization matrix compensates for crosstalk between the at least two channels and equalizes the branches such that the device under test receives equal power at its receive antennas.

Description

Technical Field

[0001] This invention relates to a method for determining a pre-equalization matrix to be used for testing. Furthermore, this invention relates to a testing apparatus for testing a device under test. Background Technology

[0002] In existing technologies, it is known to determine the over-the-air (OTA) measurement of the device under test (DUT) using a pre-equalization matrix. The pre-equalization matrix is ​​then applied to pre-equalize the radio channel established between the radio communication tester and the DUT. Therefore, the signal output from the radio communication tester is adapted through the pre-equalization matrix so that the signal received by the DUT has been equalized. In effect, the pre-equalization matrix and the channel matrix (also called the transfer matrix) compensate for each other, where the channel matrix describes the effects occurring in the radio channel established between the DUT and the radio communication tester.

[0003] Therefore, the channel matrix of the corresponding radio channel must be determined and mathematically inverted to obtain a pre-equalization matrix that can be used to equalize the radio channels. Then, by applying the pre-equalization matrix to equalize the effects of the radio channels, quasi-conducted test conditions for the device under test (DUT) are obtained. Since the quasi-conducted test conditions are met, OTA compliance testing of the DUT can be performed. A system and method for inverting the channel matrix are described in EP 3 565 135A1.

[0004] However, the known concept in the prior art is very time-consuming because the channel matrix must be determined and inverted, which may require a mathematically complex process.

[0005] Therefore, a simple and fast method is needed to establish quasi-conductive test conditions for the device under test (DUT) so that OTA conformance testing of the DUT can be performed. Summary of the Invention

[0006] This invention provides a method for determining a pre-equalization matrix to be used for testing, the method comprising the following steps:

[0007] - Provides a radio communication tester with a base station simulator and a channel simulator.

[0008] - Provide the device under test with at least two branches.

[0009] - Establish an NxM multiple-input multiple-output (MIMO) connection between the radio communication tester and the device under test. The NxM MIMO connection includes at least two channels.

[0010] - The measured reference signal received power (RSRP-B) of each branch is continuously relayed from the device under test to the radio communication tester, and

[0011] - A pre-equalization matrix is ​​determined using a radio communication tester, wherein the received power measurement of the reference signal for each branch and the presence of additive white Gaussian noise (AWGN) are taken into account when calculating the pre-equalization matrix, and wherein the pre-equalization matrix compensates for crosstalk between at least two channels and balances the branches such that the device under test receives equal power at its receiving antenna.

[0012] Therefore, quasi-conducted test conditions for the device under test (DUT) were realized, enabling over-the-air (OTA) compliance testing, which takes into account additive white Gaussian noise (AWGN), particularly the corresponding effects associated with non-inherent noise. Additive white Gaussian noise (AWGN) simulates the effects of many random processes occurring in nature. For example, AWGN contains many different natural noise sources, such as thermal vibrations of atoms in a conductor (also known as thermal noise), shot noise, or any other natural noise source. In fact, AWGN involves the fundamental noise model used in information theory, which assumes that the sum of many random processes has a Gaussian distribution. Therefore, AWGN is Gaussian because the noise has a normal distribution in the time domain. Furthermore, AWGN is white because the power of the noise is uniformly distributed across the entire frequency range. Moreover, AWGN is additive because the corresponding noise is added to any inherent noise of the test apparatus. For example, AWGN can be calculated and generated accordingly, where AWGN is (artificially) introduced into the test chamber, such as an anechoic chamber. Alternatively, AWGN is calculated and (only) mathematically considered when determining the pre-equalization matrix.

[0013] Since the device under test (DUT) forwards the RSRP-B measurement to the radio communication tester, particularly the channel simulator, a corresponding feedback from the DUT is provided, which is taken into account by the radio communication tester when calculating the pre-equalization matrix. Therefore, the effect of additive white Gaussian noise (AWGN) is appropriately considered, as the effect of AWGN influences the RSRP-B measurement fed back to the radio communication tester.

[0014] Generally, the Reference Signal Received Power per Branch (RSRP-B) is a measurement of the received power level of each branch (i.e., each receiving branch) of the device under test. The Reference Signal Received Power (RSRP) is a measurement of the Received Signal Strength Indicator (RSSI) type.

[0015] In fact, the received power per branch reference signal (RSRP-B) corresponds to the linear average of the power contribution of the resource element. The received power per branch reference signal (RSRP-B) is also called the power received per branch reference signal (RSRP-B).

[0016] The device under test (DUT) may have at least two (receive) antennas associated with at least two branches of the DUT. The at least two (receive) antennas establish a MIMO connection with the radio communication tester (specifically, at least two transmit antennas of the radio communication tester).

[0017] Generally, the power dynamic range can be maximized because the (residual) branches of the device under test (DUT) are balanced using a precalibration matrix. Since crosstalk between channels can be compensated for using the precalibration matrix, the isolation between the receiving antennas of the DUT is maximized, resulting in improved test conditions.

[0018] Therefore, when a pre-equalization matrix is ​​applied during testing of the device under test (DUT), crosstalk in the channels can be minimized, and isolation between branches of the DUT, or more precisely, between antennas, can be improved.

[0019] Because the radio communication tester includes a base station simulator and a channel simulator, each channel is a simulated channel. These simulated channels can simulate the real radio communication channel established between the device under test and the real base station.

[0020] The base station simulator can be configured to simulate different types of base stations. Furthermore, the channel simulator can be configured to simulate different types of channels. Therefore, the same radio communication device can be simulated under different conditions (e.g., ideal channels, real channels, and / or jammed channels).

[0021] Radio communication testers, particularly channel simulators, are configured to selectively activate a certain number of channels during testing / calibration to determine a pre-equalization matrix. Therefore, only one channel can be activated at a time, or two channels can be activated simultaneously. In fact, the corresponding number of channels activated simultaneously can vary, as it depends on the steps performed to determine the pre-equalization matrix.

[0022] Specifically, at least two distinct steps are performed when determining the pre-equalization matrix, wherein the at least two distinct steps are associated with different numbers of activated channels, such as a single channel in the first step and two or more channels in the second step.

[0023] The device under test (DUT) can be a 5G user equipment, such as a tablet or mobile phone (e.g., a smartphone). Furthermore, the base station emulator can be a 5G base station emulator configured to establish a 5G connection with the DUT.

[0024] Generally, the channel matrix is ​​estimated based on feedback provided by the device under test (i.e., the quantized power of each branch). Based on the feedback from the device under test, the pre-equalization matrix is ​​determined, and in particular approximated, to minimize crosstalk between channels and maximize the power dynamic range on each OTA channel.

[0025] Furthermore, an NxM MIMO connection means that N and M are equal to or greater than 2 to establish a MIMO connection. Specifically, an NxM MIMO connection is a 2x2 connection, involving two transmit antennas of the radio communication tester and two receive antennas of the device under test. In fact, the 2x2 connection ensures that the corresponding pre-calibration matrix can be easily determined.

[0026] On the one hand, it provides a way to verify whether the calculated pre-equalization matrix solves the optimization problem min||A·GI||2, where A relates to the transfer matrix, G to the pre-equalization matrix, and I to the identity matrix. Verification can be performed continuously, especially once the pre-equalization matrix is ​​available, even if the available pre-equalization matrix does not correspond to the optimized matrix. As mentioned above, when considering AWGN, the pre-equalization matrix is ​​determined, or more precisely estimated, based on the per-branch reference signal received power (RSRP-B) measurement, such that the determined / estimated pre-equalization matrix typically does not correspond to the inverted channel matrix. Therefore, verifying whether the estimated / determined pre-equalization matrix approximates the inverted channel matrix reduces the results of solving the aforementioned optimization problem under the condition ||A·GI||2.

[0027] Because AWGN is considered when calculating the pre-equalization matrix, the pre-equalization matrix does not correspond to the inverse of the transfer matrix.

[0028] Specifically, the transfer matrix describes the NxM MIMO connection between the radio communication tester and the device under test. The transfer matrix can describe the (ideal) channel established between the radio communication tester and the device under test.

[0029] On the other hand, the pre-equalization matrix is ​​obtained by multiplying the inverse of the compensation matrix (with compensation parameters) with the balance matrix (with balance parameters). Therefore, the pre-equalization matrix, based on two matrices with different purposes (i.e., compensation and balance), ensures that the pre-equalization matrix, when applied during testing of the device under test (DUT), compensates for crosstalk between at least two channels and simultaneously balances the branches of the DUT. Consequently, the DUT receives equal power at its receiving antenna.

[0030] For example, an initialization phase occurs, during which the corresponding parameters of the compensation matrix and the balance matrix are initialized. During the initialization phase, the compensation parameters of the compensation matrix and the balance parameters of the balance matrix are initialized with initial values. For example, the compensation parameters and balance parameters are equal to 1 during the initialization phase.

[0031] Furthermore, an initial calibration of the compensation matrix can be performed, thereby initially calibrating the amplitude values ​​of the compensation parameters. This initial calibration can be part of an initialization phase, where the corresponding amplitude values ​​of the compensation parameters are initially determined starting from an initial value (e.g., 1). Therefore, the initial calibrated amplitude values ​​are obtained after the initial calibration and may differ from the initial values.

[0032] Specifically, during initial calibration, the amplitude value of the compensation parameter is determined by considering the received power measurement of the reference signal for each branch. The corresponding measurements fed back from the device under test are considered by the radio communication tester (especially a channel simulator). The received measurements are processed accordingly to determine the amplitude value of the compensation parameter in order to obtain an initial calibration amplitude value, i.e., a starting value, that may differ from the initial value.

[0033] On the other hand, it provides that during the initial calibration, only one channel is activated at a time, and the received power measurement of the reference signal for each branch is collected by each branch. Therefore, the corresponding amplitude value of the compensation parameter can be obtained from... Determined, among which RSRP ij Is it only activating the corresponding channel T? i Simultaneously from branch R j RSRP-B. For each channel T i This is done so that all compensation parameters can be determined continuously.

[0034] According to another aspect, the compensation matrix is ​​determined during the optimization phase, thereby optimizing the compensation parameters of the compensation matrix. During the optimization phase, the initial values ​​are optimized to determine the optimal values ​​of the compensation matrix, so that the compensation matrix is ​​determined in an improved manner.

[0035] Specifically, during the optimization phase, the optimal phase and / or optimal amplitude values ​​of the compensation parameters are determined. An appropriate search interval is determined for the phase values ​​of the compensation parameters, and this search interval is used to find the optimal phase values. Once the optimal phase values ​​are determined, the amplitude values ​​can then be optimized.

[0036] Therefore, the optimization phase can include a two-step approach, as the phase and amplitude values ​​of the compensation parameters are subsequently optimized, thus fully optimizing the compensation parameters. The optimization of the amplitude values ​​ensures that they are optimized relative to the values ​​obtained during the initialization phase, particularly during the initial calibration.

[0037] For example, the optimal phase value is determined by activating two channels in pairs while deactivating all other channels. The optimal phase value provides the highest mutual (paired) isolation between the receiving antennas of the device under test. This isolation corresponds to destructive interference.

[0038] For each possible transmitter, pairwise activations are performed on consecutive repeating channels, i.e., the activation of transmitter pairs. By doing so, the entire compensation matrix is ​​optimized accordingly.

[0039] When the amplitude value was subsequently optimized, the mutual (paired) isolation between the branches of the device under test was further improved.

[0040] As mentioned above, radio communication testers, particularly channel simulators, are configured to selectively activate different numbers of channels during testing / calibration to determine the pre-equalization matrix. During the initialization phase (step one), only a single channel is activated at a time, while during the optimization phase (step two), two channels are activated at a time. Therefore, two different numbers of channels are activated during calibration to determine the pre-equalization matrix.

[0041] Generally speaking, the modified standard search algorithm can be applied to determine the optimal phase value and / or the optimal amplitude value.

[0042] Specifically, the optimization phase occurs after the initialization phase and / or initial calibration. During the initialization phase, the individual compensation parameters of the compensation matrix are set to initial values ​​corresponding to the starting point. During the initial calibration, the amplitude values ​​for the initial calibration are determined. In the subsequent optimization phase, the compensation parameters are optimized.

[0043] Once the optimization phase is complete, the obtained compensation matrix can be inverted to obtain its inverse. The inversion of the compensation matrix can be performed using numerical methods (depending on the rank of the compensation matrix).

[0044] Generally, the inverse compensation matrix and the balance matrix are multiplied together to obtain the pre-balance matrix.

[0045] On the other hand, a balancing matrix is ​​determined during the conditioning phase, thereby adjusting the balancing parameters. During the conditioning phase, the balancing matrix is ​​optimized for its objective: to provide equal power distribution at the branches of the device under test (DUT). In other words, the DUT receives the same power at its receiving antenna.

[0046] For example, the adjustment phase occurs after the optimization phase. The balancing parameters are then adjusted. Therefore, crosstalk is minimized by optimizing the compensation matrix. Subsequently, the balancing matrix is ​​changed by adjusting the balancing parameters to ensure that each branch of the device under test receives equal power.

[0047] Generally, the optimized pre-equalization matrix corresponds to the matrix obtained after performing the steps mentioned above (i.e., the optimization phase and the adjustment phase). As mentioned above, the optimized pre-equalization matrix does not necessarily correspond to the inverse of the channel matrix, or more precisely, the transition matrix, since it is derived from the RSRP-B measurements.

[0048] Additionally, an intermediate pre-equilibrium matrix can be determined after the initialization and / or optimization phases, which can be used for the optimization problems mentioned above.

[0049] Therefore, if the intermediate pre-equalization matrix has solved the optimization problem given above, the solution can be obtained very quickly, resulting in a significant speed advantage compared to known techniques in the prior art. Even if the intermediate pre-equalization matrix does not solve the optimization problem given above, the method is still significantly faster because it does not require determining the inverted channel matrix, or more precisely, the transition matrix.

[0050] Generally, the received power measurement of each branch reference signal forwarded by the device under test is processed by a radio communication tester, especially a channel simulator, in order to determine / estimate the pre-equalization matrix, for example, to calculate the compensation matrix based on the pre-equalization matrix.

[0051] Furthermore, the present invention provides a test apparatus for testing a device under test (DUT). The test apparatus includes a radio communication tester and the DUT. The test apparatus is configured to perform the methods described above. The features and advantages mentioned above also apply to the test apparatus in a similar manner. Therefore, please refer to the explanation given above.

[0052] The test setup may also include an anechoic chamber surrounding the device under test (DUT) during testing and determination of the pre-equalization matrix. In other words, the DUT is located within the anechoic chamber when it receives a signal transmitted by the radio communication tester. The DUT measures RSRP-B, thereby obtaining an RSRP-B measurement value, which is fed back to the radio communication tester, particularly the channel simulator. The radio communication tester, particularly the channel simulator, considers the various measurements in order to determine the pre-equalization matrix as described above.

[0053] In fact, when calculating the pre-equalization matrix, the received power measurement of the reference signal for each branch and the presence of additive white Gaussian noise are taken into account.

[0054] The device under test receives signals from the radio communication tester. s Related signals r Two signals pass through r =A·G· s + n They are interconnected, where A involves the (ideal) transfer matrix, G involves the pre-equilibrium matrix, and... n Additive white Gaussian noise is involved. The (optimal / optimized) pre-equalization matrix, determined as described above, equalizes the (ideal) transfer matrix A and the additive white Gaussian noise (AWGN). n .

[0055] Generally, the determined pre-equalization matrix can be applied using the test setup during the testing of the device under test (DUT). Therefore, OTA compliance testing of the DUT can be performed because quasi-transduction test conditions are achieved. Attached Figure Description

[0056] The foregoing aspects and numerous accompanying advantages of the claimed subject matter will become more readily understood with reference to the following detailed description taken in conjunction with the accompanying drawings, in which:

[0057] Figure 1 An overview of the testing apparatus according to the present invention is shown schematically;

[0058] Figure 2 A flowchart of a method for determining a pre-equilibrium matrix to be used for testing, according to the present invention, is shown;

[0059] Figure 3 It shows Figure 2 The detailed steps of the method shown in the figure; and

[0060] Figure 4 A preferred embodiment associated with the established 2×2 MIMO connection is shown. Figure 3 Detailed steps. Detailed Implementation

[0061] The detailed description set forth below in conjunction with the accompanying drawings is intended as a description of various embodiments of the disclosed subject matter and is not intended to represent only one embodiment, in which the same reference numerals refer to the same elements. Each embodiment described in this disclosure is provided by way of example or illustration only and should not be construed as preferred or superior to other embodiments. The illustrative examples provided herein are not intended to be exhaustive or to limit the claimed subject matter to the precise forms disclosed.

[0062] For the purposes of this disclosure, the phrase "at least one of A, B, and C" means, for example, (A), (B), (C), (A and B), (A and C), (B and C), or (A, B, and C), and includes all other possible permutations when more than three elements are listed. In other words, the term "at least one of A and B" generally refers to "A and / or B," i.e., "A only," "B only," or "A and B."

[0063] exist Figure 1 The image shows a test setup 10, which includes a radio communication tester 12 and a device under test (DUT) 14 that communicate with each other via an established radio communication connection 15.

[0064] The radio communication tester 12 includes a base station emulator 16 and a channel emulator 18. The base station emulator 16 and the channel emulator 18 are typically housed within the enclosure of the radio communication tester 12.

[0065] Additionally, the testing apparatus 10 includes an anechoic chamber 20. The device under test 14 is located inside the anechoic chamber 20.

[0066] In addition to the radio communication connection 15, a feedback connection 22 is also established between the device under test 14 and the radio communication tester 12, especially the channel simulator 18.

[0067] The device under test 14 is configured to perform a measurement to obtain a per-branch reference signal received power (RSRP-B) measurement.

[0068] The device under test 14 is able to forward the collected RSRP-B measurements to the radio communication tester 12, particularly the channel emulator 18, via feedback connection 22.

[0069] Therefore, the radio communication tester 12, especially the channel emulator 18, is able to take into account the RSRP-B measurement received from the device under test 14 via the feedback connection 22.

[0070] In the illustrated embodiment, the radio communication tester 12 has four transmitting antennas 22, while the device under test 14 has two receiving antennas 24. The receiving antennas 24 are associated with corresponding branches 26 of the device under test 14.

[0071] Generally, the radio communication tester 12 and the device under test 14 establish an NxM multiple-input multiple-output (MIMO) connection 15 between them. For example, the MIMO connection 15 corresponds to a 2x2 connection.

[0072] exist Figure 1 The image shows two active (expected) channels 28.

[0073] Figure 1 The test apparatus 10 shown is typically configured to perform Figure 2 and Figure 3 The method shown in the figure, wherein Figure 3 It shows Figure 2 The detailed steps of the method are shown in the figure.

[0074] In the first step S1, provide as follows Figure 1 The test setup 10 shown consists of a radio communication tester 12 and a device under test 14. Specifically, the device under test 14 is located inside an anechoic chamber 20.

[0075] In the second step S2, a MIMO connection 15, such as a 2×2 connection, is established between the radio communication tester 12 and the device under test 14.

[0076] In the third step S3, the radio communication tester 12 provides signals representing a certain base station and channel conditions, wherein these characteristics are defined by the base station simulator 16 and the channel simulator 18.

[0077] In the fourth step S4, the signal is transmitted to the anechoic chamber 20 and received by the device under test 14, which processes the signal accordingly.

[0078] Additive white Gaussian noise (AWGN) can be calculated and introduced into the anechoic chamber 20, so that the device under test 14 additionally receives signals related to the AWGN, such as... Figure 1 As illustrated. Alternatively, AWGN will be considered mathematically at a later stage, as will be described later.

[0079] In the fifth step S5, the device under test 14 performs a measurement on the received signal in order to obtain the measured value of the reference signal received power per branch (RSRP-B).

[0080] In the sixth step S6, the RSRP-B measurement value is forwarded to the radio communication tester 12 via the feedback line 22, which processes the information obtained from the device under test 14.

[0081] In step S7, the pre-equalization matrix G is determined. In other words, the pre-equalization matrix G is calculated by the radio communication tester 12, specifically the channel simulator 18.

[0082] When determining / compiling the pre-equalization matrix G, if the AWGN has not been previously introduced into the anechoic chamber 20, the AWGN can be considered mathematically.

[0083] The steps to be performed to determine / calculate the pre-equilibrium matrix G are as follows: Figure 3 The figure below illustrates this in more detail.

[0084] Generally, the pre-equilibrium matrix G can be calculated based on two different matrices: the compensation matrix B and the equilibrium matrix C. The inverse of the compensation matrix B is such that G = B. -1 ·C.

[0085] The compensation matrix B includes the compensation parameter ω ij This allows the compensation matrix B to be derived from... limited.

[0086] The balance matrix C includes the balance parameter c. i This allows the balance matrix to be derived from... limited.

[0087] Generally, the compensation matrix B compensates for crosstalk between at least two channels 28, while the balancing matrix C balances branch 26 so that the device under test 14 receives equal power at its receiving antenna 24.

[0088] Therefore, the pre-equalization matrix G obtained by multiplying the matrices B and C mentioned above compensates for the crosstalk between at least two channels 28 and balances the branches 26, so that the device under test 14 receives equal power at its receiving antenna 24.

[0089] In the first sub-step of determining / estimating the pre-equilibrium matrix G, an initialization phase may occur, such as... Figure 3 As shown. In the first step, parameter initialization is completed, for example by adjusting the compensation parameter ω. ij and balance parameter c i Set to 1. Generally, other initialization values ​​can be used, such as 0, 2, or 3. For example, predefined values ​​stored in the memory of the radio communication tester 12 can be used.

[0090] In the first step, calibration is performed to determine the amplitude value of the compensation parameter (|ω). ij In other words, the amplitude value of the compensation parameter is initially calibrated.

[0091] This is accomplished by activating only one channel at a time, where the received power measurement of the reference signal for each branch 26 is collected by each branch 26. Therefore, the initial calibration amplitude value of the compensation parameter can be obtained from... Confirmed, among which RSRP ij Is it only activating the corresponding channel T? i Simultaneously from branch R j RSRP-B. This is for each channel T. i This process allows the initial calibration amplitude values ​​for all compensation parameters to be determined in a continuous manner.

[0092] Therefore, during the initial calibration, the amplitude value of the compensation parameter is determined by taking into account the RSRP-B measurement value fed back from the device under test 14 to the radio communication tester 12.

[0093] Next, the first step is completed, which includes the initialization phase of initial calibration.

[0094] In the second sub-step of determining / estimating the pre-equalization matrix G, an optimization phase occurs, during which the compensation parameters of the compensation matrix are optimized because the optimal compensation parameters are determined, in particular by determining the optimal phase values ​​and / or optimal amplitude values ​​of the compensation parameters.

[0095] like Figure 3 As shown, the optimal phase value The optimal phase value is determined within a suitable search interval. It is determined by activating two channels 28 in pairs while deactivating all other channels. This is typically accomplished by activating two transmitter antennas 22 while simultaneously deactivating other transmitter antennas 22, thus activating a transmitter pair.

[0096] Then, a search algorithm is applied to determine the optimal phase value. These optimal phase values ​​provide the highest mutual isolation for the corresponding transmitter pairs of transmitter antenna 22.

[0097] Signals transmitted by the respective transmitter antennas 22 of the active transmitter pair can be received at once by a single receiving antenna 24 of the device under test 14. Then, different receiving antennas 24 are subsequently exposed to the respective transmitter pair. Alternatively, the receiving antennas 24 simultaneously receive signals transmitted by the respective transmitter pair, as they are associated with the distinguishable branch 26.

[0098] In any case, repeat the above-instructed procedure for all possible transmitter pairs so that the optimal phase values ​​for all compensation parameters can be determined.

[0099] Subsequently, amplitude adjustment can be performed to increase the isolation of branch 26, thereby optimizing the amplitude value of the compensation parameter (|ω). ij |), which optimizes the overall compensation parameters.

[0100] Therefore, the optimized compensation matrix B is obtained at the end of the second step, because compensation matrix B includes the optimized compensation parameter ω. ij .

[0101] Therefore, the radio communication tester 12, and in particular the channel simulator 18, is configured to selectively activate different numbers of channels 28 when the pre-equalization matrix G, i.e. the compensation matrix B which is part of it, is determined.

[0102] In the third sub-step, the (optimized) compensation matrix B is inverted to obtain the matrix to be multiplied by the balancing matrix. The inverted compensation matrix B is labeled D.

[0103] In the fourth sub-step, the balance parameter c is adjusted. i This ensures that the device under test 14 receives equal power at its receiving antenna 24. Therefore, the corresponding balance parameter c... i They are adapted to their initial values ​​set during the initialization phase described above.

[0104] In the fifth sub-step, the (optimal / optimized) pre-equilibrium matrix G is constrained by multiplying the inverse matrix of the compensation matrix B (i.e., matrix D) by the equilibrium matrix C.

[0105] As described above, the (optimized) pre-equalization matrix G compensates for crosstalk between at least two channels 28 and balances branches 26 so that the device under test 14 receives equal power at its receiving antenna 24.

[0106] Therefore, it is not necessary to determine the channel matrix / transfer matrix and invert it in order to pre-equalize (multiple) channels 28, because the pre-equalization matrix can be determined / calculated / estimated based on the steps described above, where AWGN and feedback from the device under test 14 are measured via the per-branch reference signal received power (RSRP-B).

[0107] In the steps indicated above, the pre-equilibrium matrix G can be determined / estimated / computed based on the available matrices, resulting in an intermediate pre-equilibrium matrix G that differs from the optimal pre-equilibrium matrix G. However, the intermediate pre-equilibrium matrix G is obtained earlier.

[0108] Furthermore, it can be verified whether the calculated (intermediate) pre-equalization matrix G has solved the optimization problem min||A·GI||2, where I relates to the identity matrix and A relates to the transfer matrix describing the NxM MIMO connection 15 between the radio communication tester 12 and the device under test 14.

[0109] exist Figure 4 In the example, a preferred embodiment associated with the established 2×2 MIMO connection is shown. Figure 3 The detailed steps are as follows. Therefore, the compensation parameters can be labeled ω1 and ω2, since only two compensation parameters are provided for crosstalk compensation. Therefore, they are labeled ω1 and ω2 instead of ω 12 and ω 21 shorten.

[0110] Therefore, in the first sub-step, i.e. the initialization phase, the balance parameters c1 and c2 are set to 1, while the phase value of the compensation parameter is set to zero.

[0111] Then, an initial calibration is performed to initially calibrate the amplitude values ​​(|ω1|,|ω2|) of the compensation parameters.

[0112] In the second sub-step, the optimized values ​​for the phase and amplitude of each compensation parameter are determined, particularly in a subsequent manner.

[0113] Therefore, the first step is to determine the optimal method for identification. The search interval is determined. Once the optimal phase value is determined for the corresponding compensation parameters, the amplitude value |ω1| is adjusted accordingly.

[0114] Then, the same steps are performed on the second compensation parameter to determine the optimal phase value of the second compensation parameter. And the optimal amplitude value |ω2|.

[0115] Therefore, the compensation matrix was obtained.

[0116] Then, in the third sub-step, the balance parameters c1 and c2 are adjusted as discussed above.

[0117] Then, the optimal / optimized pre-equilibrium matrix can be determined based on the optimized compensation matrix B and the balanced / adjusted equilibrium matrix C.

[0118] In any case, a fast method, i.e. a high-speed method, is provided for estimating the pre-equalization matrix (i.e., the transfer matrix) of the NxM OTA channel matrix, wherein this is done based on the quantized feedback from the device under test 14 and the presence of AWGN, since the RSRP-B measurement fed back to the radio communication tester 12 is taken into account.

[0119] Generally, the corresponding simulators 16 and 18 can be associated with the simulation module. Therefore, the radio communication tester 12 can be built using the module.

[0120] In this context and below, the term "module" is understood to describe suitable hardware, suitable software, or a combination of hardware and software configured to perform a specific function.

[0121] Hardware may include, in particular, CPUs, GPUs, FPGAs, ASICs, or other types of electronic circuit systems.

[0122] Certain embodiments disclosed herein (particularly corresponding modules) utilize circuit systems (e.g., one or more circuits) to implement the standards, protocols, methods, or techniques disclosed herein, to operatively couple two or more components, to generate information, process information, analyze information, generate signals, encode / decode signals, convert signals, transmit and / or receive signals, control other devices, etc. Any type of circuit system can be used.

[0123] In embodiments, the circuit system specifically includes one or more computing devices such as processors (e.g., microprocessors), central processing units (CPUs), digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), systems on a chip (SoCs), and any combination thereof, and may include discrete digital or analog circuit elements or electronic devices or combinations thereof. In embodiments, the circuit system includes hardware circuit implementations (e.g., implementations in analog circuit systems, implementations in digital circuit systems, and combinations thereof).

[0124] In embodiments, the circuit system includes a combination of circuitry and computer program products having software or firmware instructions stored on one or more computer-readable storage media that work together to cause a device to perform one or more protocols, methods, or techniques described herein. In embodiments, the circuit system includes circuitry that requires software, firmware, etc., for operation, such as a microprocessor or a portion thereof. In embodiments, the circuit system includes one or more processors or portions thereof, along with accompanying software, firmware, hardware, etc.

[0125] This application may refer to quantities and numbers. Unless otherwise stated, these quantities and numbers should not be considered limiting, but rather examples of possible quantities or numbers relevant to this application. Also in this respect, this application may use the term "multiple" to refer to a quantity or number. In this respect, the term "multiple" means any number more than one, such as two, three, four, five, etc. The terms "approximately," "about," "close to," etc., mean plus or minus 5% of the stated value.

Claims

1. A method for determining a pre-equalization matrix to be used for testing, the method comprising the steps of: - Provide a radio communication tester (12) with a base station simulator (16) and a channel simulator (18). - Provide the device under test (14) with at least two branches (26). - Establish a connection between the radio communication tester (12) and the device under test (14). Multiple-input multiple-output (MIMO) connection (15), the A MIMO connection (15) includes at least two channels (28). - The reference signal received power measurement value of each branch (26) is continuously forwarded from the device under test (14) to the radio communication tester (12), and - A pre-equalization matrix is ​​determined using the radio communication tester (12), wherein the measured power of the reference signal for each branch (26) and the presence of additive white Gaussian noise (AWGN) are taken into account when calculating the pre-equalization matrix, and wherein the pre-equalization matrix compensates for crosstalk between the at least two channels (28) and balances the branches (26) such that the device under test (14) receives equal power at its receiving antenna (24). The pre-equilibrium matrix is ​​obtained by multiplying the inverse of the compensation matrix with compensation parameters by the balance matrix with balance parameters. The compensation matrix is ​​determined during the optimization phase, thereby optimizing the compensation parameters of the compensation matrix. During the adjustment phase, a balance matrix is ​​determined, thereby adjusting the balance parameters.

2. The method of claim 1, wherein verifying whether the calculated pre-equilibrium matrix solves the optimization problem. ,in Involves transfer matrix, Involving pre-equilibrium matrix and It involves the identity matrix.

3. The method according to claim 2, wherein the transfer matrix describes the communication between the radio communication tester (12) and the device under test (14). MIMO connection (15).

4. The method of claim 1, wherein an initialization phase occurs during which the corresponding parameters of the compensation matrix and the balance matrix are initialized.

5. The method of claim 1, wherein an initial calibration of the compensation matrix is ​​performed to initially calibrate the amplitude values ​​of the compensation parameters.

6. The method of claim 5, wherein the amplitude value of the compensation parameter is determined by taking into account the received power measurement of the reference signal for each branch during the initial calibration.

7. The method of claim 5, wherein during the initial calibration, only one channel (28) is activated at a time, and the reference signal received power measurement of each branch (26) is collected by each branch (26).

8. The method according to claim 1 or 2, wherein the optimal phase value and / or optimal amplitude value of the compensation parameter is determined during the optimization phase.

9. The method of claim 8, wherein the optimal phase value is determined by activating two channels (28) in pairs while deactivating all other channels (28).

10. The method of claim 4 or 5, wherein the optimization phase occurs after the initialization phase and / or the initial calibration.

11. The method according to claim 1 or 2, wherein the adjustment phase occurs after the optimization phase.

12. A test apparatus (10) for testing a device under test (14), the test apparatus (10) comprising a radio communication tester (12) and the device under test (14), wherein the test apparatus (10) is configured to perform the method according to claim 1 or 2.

Images