Uplink and downlink channel reconstruction method and system for duplex system air interface performance test

Abstract

The invention discloses an uplink and downlink channel reconstruction method and system for a duplex system air interface performance test, and the method achieves the selection of an uplink probe antenna and a downlink probe antenna through a double wolf pack hunting optimization algorithm based on population interaction, avoids the cost caused by the complex switching of the uplink probe antenna and the downlink probe antenna, and achieves the precise and efficient reconstruction of a bidirectional channel. The switch array with the functions of switching the uplink and the downlink of the probe antenna and activating the selection of the probe antenna is added, the switch array can be used for regulating and controlling the on-off condition of the uplink and the downlink of the probe antenna on a probe wall, and meanwhile selective activation of the probe antenna is carried out, so that simultaneous reconstruction of an uplink channel and a downlink channel is realized.

Description

Technical Field

[0001] The invention belongs to the field of wireless communications, and in particular relates to an uplink and downlink channel reconstruction method and system for air interface performance testing of a duplex system. Background Art

[0002] In the development of wireless communication equipment, it is crucial to test the performance indicators of the device under test in a real channel environment. Due to the constant changes in the outdoor environment, the reliability and repeatability of the test are difficult to guarantee, and the cost of outdoor testing is high. In order to solve the above problems, the industry and academia are committed to studying how to complete indoor testing in a controllable laboratory environment. In the past, the industry usually used the conduction test method to transmit the test signal to the device under test through a physical connection. Such testing is simple to operate and low-cost. However, in the future, mobile communication equipment will be highly integrated and will no longer provide a radio frequency interface for conduction testing. Artificially manufactured interfaces will affect the performance of the entire communication system. Therefore, conduction testing is no longer suitable for future mobile communication equipment performance testing.

[0003] In response to the above problems, Over-the-Air (OTA) testing technology has gradually come into the public eye and has been recognized by many equipment manufacturers and operators. The so-called Over-the-Air testing aims to create a controllable and repeatable testing environment in the laboratory without the need for cable connection, thereby reconstructing the channel in the anechoic chamber to complete the performance evaluation of the equipment under test. At present, the multi-probe microwave anechoic chamber method is one of the mainstream terminal performance testing methods and has been identified as a standard testing method by the Cellular Telecommunications Industry Association (CTIA), the 3rd Generation Partnership Project (3GPP) and the China Communication Standards Association (CCSA).

[0004] The multi-probe microwave anechoic chamber method is currently the mainstream method for air interface performance testing. The baseband signal generated by the base station or base station simulator is transmitted to the channel fading simulator through an RF cable, and convolved with the pre-defined fading channel in the channel fading simulator to generate a channel impulse response, which is then radiated through multiple probes to create the required radiation propagation environment in the anechoic chamber test domain.

[0005] At present, channel reconstruction based on the multi-probe microwave darkroom method usually adopts Pre-Faded Signal Synthesis (PFS) technology and Plane Wave Synthesis (PWS) technology. Among them, PFS technology reconstructs the target channel from the perspective of the entire cluster. Each channel cluster is jointly reconstructed by multiple probe antennas. The basic synthesis object is a single cluster, while PWS technology focuses on reconstructing each sub-path within the cluster. It synthesizes a single plane wave with a specific arrival angle, Doppler frequency shift and amplitude. Unlike PFS technology, which only needs to calibrate amplitude information, PWS technology needs to complete dual calibration of amplitude and phase information before testing, which is more complex. Therefore, PFS technology is widely used in actual testing.

[0006] With the development of the fifth-generation (5G) wireless communication system, the 5G millimeter wave system mainly adopts time division duplex technology, that is, transmitting uplink signals and downlink signals in the same frequency range in different time slots. The current multi-probe microwave darkroom OTA test setting only considers one-way channel simulation. In order to support time division duplex testing, the multi-probe microwave darkroom needs to use the switch to switch the uplink and downlink probe antennas through the uplink and downlink link switching trigger signal provided by the base station. This requires the base station manufacturer to specifically provide a switching trigger signal, which greatly increases the complexity and cost of the test.

[0007] Due to cost and hardware limitations, a sector-shaped probe wall structure is used in a multi-probe microwave anechoic chamber, and only a small number of probe antennas can be selected for activation at the same time. Therefore, it is necessary to consider the probe antenna selection problem in the sector-shaped probe wall structure. The existing probe antenna selection algorithm is mainly designed for unidirectional channel simulation, ignoring the distinction between uplink probe antennas and downlink probe antennas, and cannot be applied to the simultaneous reconstruction of uplink and downlink channels.

[0008] For example, the patent application document with Chinese patent application number 202310927452.3 discloses a dual-probe wall test system for RIS air interface testing and a test method thereof, wherein the test system includes a microwave darkroom, a transmitting end probe wall, a receiving end probe wall, a RIS panel to be tested, an electrically controlled test turntable, an amplitude modulation and phase modulation network, a port switcher, a vector network analyzer and a main control terminal; the test method includes step 1, the main control terminal controls the amplitude modulation and phase modulation network to perform amplitude and phase control on the input Hertz signal according to the amplitude and phase control test code book and feeds it to the output end, step 2, the main control terminal controls the RIS panel to be tested to perform phase adjustment according to the phase test code book, step 3, the main control terminal controls the electrically controlled test turntable to rotate to a specified test angle, and the like.

[0009] This prior art has the above-mentioned technical problems.

[0010] Based on the above technical problems existing in the prior art, the present invention proposes an uplink and downlink channel reconstruction method and system for air interface performance testing of a duplex system. Summary of the invention

[0011] The purpose of the present invention is to provide an uplink and downlink channel reconstruction method and system for air interface performance testing of a duplex system in view of the deficiencies of the prior art.

[0012] The present invention adopts the following technical solution:

[0013] On the one hand, a method for reconstructing uplink and downlink channels for air interface performance testing of a duplex system is provided, comprising:

[0014] Step 1, establishing a target uplink channel spatial correlation matrix and a downlink channel spatial correlation matrix to be reconstructed;

[0015] Step 2: reconstruct the target uplink and downlink channels, determine the reconstruction accuracy error function according to the target reconstruction parameters, calculate the probe antenna weights through the convex optimization algorithm, and then calculate the error function corresponding to each population individual;

[0016] Step 3: Optimize and iterate the initial probe antenna index matrix to obtain a set of uplink and downlink probe antennas with the lowest reconstruction accuracy error.

[0017] Furthermore, in step 2, reconstructing the target uplink and downlink channels includes:

[0018] Taking spatial correlation ρ as the channel reconstruction index, the optimization objective is expressed as:

[0019]

[0020] where Φ UL , Φ DL are the position vector matrices of the OTA probe antennas selected for uplink channel and downlink channel reconstruction, Φ is the position vector matrix of all candidate probe antennas, This means that due to the coupling effect between antennas and the physical size limitation of the probe antenna, the uplink probe antenna and the downlink probe antenna cannot share the same probe antenna position. is the weight of the selected K uplink probe antennas, is the weight of the selected K downlink probe antennas, They represent the kth uplink probe antenna weight and the kth downlink probe antenna weight, respectively, and w UL 、w DL satisfy as well as

[0021] ρ UL , DLThey represent the uplink target spatial correlation and the downlink target spatial correlation, respectively, and are defined as:

[0022]

[0023] in is a matrix composed of M row vectors of the antenna positions of the device under test, Ω is the space vector corresponding to the beam arrival angle, P UL (Ω) represents the uplink angular power spectrum, P DL (Ω) represents the downlink angle power spectrum, all satisfying

[0024] They represent the uplink reconstruction spatial correlation and the downlink reconstruction spatial correlation respectively, and they are:

[0025]

[0026] Furthermore, in step 2, the probe antenna index matrix is ​​initialized according to the total number of probe antennas N in the multi-probe microwave darkroom, the number of selected probe antennas 2K, and the population size L For each individual G i Randomly divide the antenna index of the uplink probe into equal numbers and downlink probe antenna index in And G i The probe antenna indexes in the same group are not repeated, so the initial double wolf pack G is obtained. UL (1) G DL (1).

[0027] Furthermore, in step 2, each individual Corresponding to a set of probe antenna position vectors Φ UL / DL , according to the determined reconstruction error function, set the fitness function f(·):

[0028]

[0029] Determined by convex optimization algorithm The probe weight w at the corresponding index in UL / DL , calculate each individual The corresponding fitness function Arrange them in ascending order and select the three individuals with the smallest fitness As the three alpha wolves in the wolf pack.

[0030] Furthermore, in step 3, the initial probe antenna index matrix is ​​optimized and iterated by the double wolf pack hunting optimization algorithm based on population interaction. The other wolves in the wolf pack hunt according to the position of the head wolf. When the number of iterations is t, Traverse and get GGWO (t+1), the process is described as follows:

[0031]

[0032] A and C are used to control the hunting method and range of wolves, and are defined as follows:

[0033] A=2·a(t)·r 1 -a(t)……(11),

[0034] C=2·r 2 ……(12),

[0035] a(t) decreases linearly from 2 to 0 during the iteration, r 1 and r 2 is a random number in [0,1];

[0036] For the individual G in the traversal i (t), G i Different neighbors of (t) and random individuals in the population G(t) will move to G i (t) Learn and optimize, except for the In addition, individual optimization will generate a new individual To do this, first calculate G i (t) G i (t) and G j The Euclidean distance R between (t)∈G(t) i (t), R i ′(t), then individual G i The neighborhood N of (t) i (t) = {X j (t)|R i ′(t)≤R i (t)}, and then perform multi-neighborhood learning of the population to obtain

[0037]

[0038] Among them, G n (t) is N i (t) is a randomly selected individual, G r (t) is a randomly selected individual from G(t), r 3 is a random number in [0,1];

[0039] Compare and Select the individual with the lowest fitness and use it as G i (t+1), the traversal of i ends here;

[0040] For G i (t+1), calculate And sort them to find the wolf head G with the smallest fitness α (t+1), and then use the partial matching crossover algorithm to complete the probe index set and The exchange operation between them is performed, and a greedy strategy is adopted to select the one with the smallest fitness before and after the exchange as the new individual G′ α (t+1) put into G i Replace G in (t+1) α (t+1), one iteration is completed.

[0041] Furthermore, in step 3, the leader wolves of different wolf packs exchange information on the location of the hunting target. This information exchange behavior can be mapped to a random exchange between uplink and downlink probe index sets, thereby selecting a probe set that can more accurately reconstruct the uplink and downlink channels.

[0042] Furthermore, step 3 includes the step of setting an upper limit T of the number of iterations and a fitness threshold ε.

[0043] Furthermore, in step 1:

[0044] The uplink and downlink channels to be reconstructed are composed of multiple paths or clusters composed of paths;

[0045] Select different reconstruction target parameters according to different channel characteristics;

[0046] At the same time, the reconstruction accuracy of the channel is measured by the sum of the Euclidean distances between the uplink and downlink targets and the reconstruction parameters of the reconstructed channel.

[0047] On the other hand, a system for air interface performance testing of a duplex system is provided, comprising:

[0048] A test device, comprising an anechoic chamber, a probe wall disposed in the anechoic chamber, and a probe antenna disposed on the probe wall, the test device being used to shield external signal interference;

[0049] A switch array is connected to the test device, wherein the switch array has the functions of switching the uplink and downlink of the probe antenna and activating the probe antenna selection, and is used to adjust the on / off status of the uplink and downlink of the probe antenna on the probe wall;

[0050] A user equipment simulator, used to generate a test signal to feed into the channel simulator and simultaneously receive a downlink signal;

[0051] The channel simulator is used to create and simulate a multipath fading environment. It generates a wireless channel impulse response and convolves it with the user equipment simulator's transmit signal to generate a fading signal that is fed into the probe antenna selected by the switch array and radiated to the test area to complete the channel reconstruction.

[0052] Furthermore, the probe antenna is connected to the switch array through a wire to complete the probe antenna selection.

[0053] The beneficial effects of the present invention are:

[0054] 1. The uplink and downlink channel reconstruction method for duplex system air interface performance testing of the present invention solves the uplink and downlink channel reconstruction problem, realizes the selection of uplink and downlink probe antennas through a double wolf pack hunting optimization algorithm based on population interaction, avoids the cost caused by complex switching of uplink and downlink probe antennas, and realizes accurate and efficient reconstruction of bidirectional channels;

[0055] 2. The system for air interface performance testing of a duplex system described in the present invention, based on the original darkroom structure, adds a switch array with the functions of switching the uplink and downlink links of the probe antenna and activating the probe antenna selection. The switch array can be used to regulate the on / off status of the uplink and downlink links of the probe antenna on the probe wall, and selectively activate the probe antenna to achieve simultaneous reconstruction of the uplink and downlink channels. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flow chart of the uplink and downlink channel reconstruction method for duplex system air interface performance test according to an embodiment of the present invention;

[0057] Figure 2 It is a flowchart of a specific algorithm included in the method described in the embodiment of the present invention;

[0058] Figure 3 The figure is a schematic diagram of the structure of the system for air interface performance testing of a duplex system according to an embodiment of the present invention.

[0059] In the figure, 1-test device, 11-anechoic chamber, 12-probe wall, 13-probe antenna. DETAILED DESCRIPTION

[0060] In order to more clearly understand the above-mentioned purposes, features and advantages of the present invention, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0061] Example

[0062] The uplink and downlink channel reconstruction method for duplex system air interface performance test is based on the spatial characteristics of the channel to be reconstructed, and the gray wolf algorithm is improved by using the population interaction strategy to form a double wolf hunting optimization algorithm. K non-overlapping probe antennas 13 are selected for the uplink channel and the downlink channel on the probe wall 12 with N probe antennas 13 in the anechoic chamber 11, and the power weight on each probe antenna 13 is optimized by using the convex optimization algorithm to complete the high-precision reconstruction of the target uplink and downlink channels. Figure 1 As shown, the method includes:

[0063] Step 1: Based on the system model under the required test scenario, establish the target uplink and downlink channel model to be reconstructed, and select the target reconstruction parameters according to different channel reconstruction criteria;

[0064] Step 2, reconstruct the target uplink and downlink channels, initialize the probe antenna 13 index matrix according to the total number of probe antennas 13 in the anechoic chamber 11, the number of selected probe antennas 13 and the population size, and randomly divide them into equal numbers of uplink probe antennas 13 and downlink probe antennas 13, determine the reconstruction accuracy error function according to the target reconstruction parameters, calculate the weight of the probe antenna 13 by a convex optimization algorithm, and calculate the error function corresponding to each individual in the population accordingly;

[0065] Step 3, based on the above calculation results, the initial probe antenna 13 index matrix is ​​optimized and iterated through the double wolf pack hunting optimization algorithm based on population interaction, and finally a set of uplink and downlink probe antennas 13 with the lowest reconstruction accuracy error is obtained, which is used to achieve simultaneous accurate reconstruction of the uplink and downlink channels.

[0066] In step 1, the uplink and downlink channels to be reconstructed are composed of multiple paths or clusters of paths. Different reconstruction target parameters are selected according to different channel characteristics. At the same time, the reconstruction accuracy of the channel is measured by the sum of the Euclidean distances of the reconstruction parameters between the uplink and downlink targets and the reconstructed channels.

[0067] Exemplarily, a PFS channel reconstruction method is used, such as Figure 2 As shown, by measuring the reconstruction accuracy through the uplink and downlink channel spatial correlation errors, the specific application process of the method and the solution method of the optimization problem are given;

[0068] Taking spatial correlation ρ as the channel reconstruction index, the optimization objective can be expressed as:

[0069]

[0070] where Φ UL , Φ DL are the position vector matrices of the OTA probe antennas 13 selected for uplink channel and downlink channel reconstruction, respectively, Φ is the position vector matrix of all candidate probe antennas 13, This means that due to the coupling effect between antennas and the physical size limitation of the probe antenna 13, the uplink probe antenna 13 and the downlink probe antenna 13 cannot share the same probe antenna 13 position. is the weight of the selected uplink and downlink K probe antennas 13, satisfying as well as ρ UL , DL They represent the uplink target spatial correlation and the downlink target spatial correlation respectively, and are specifically defined as:

[0071]

[0072] in is a matrix composed of M row vectors of the antenna positions of the device under test, Ω is the space vector corresponding to the beam arrival angle, P UL (·) is the uplink angular power spectrum, P DL (·) is the downlink angle power spectrum, which satisfies ∮P(Ω)dΩ=1;

[0073] They represent the uplink reconstruction spatial correlation and the downlink reconstruction spatial correlation respectively, and they are:

[0074]

[0075] In order to achieve the highest reconstruction accuracy, the optimization objective is UL , Φ DL The probe antenna 13 is selected by a double wolf pack hunting algorithm based on population interaction;

[0076] Consider that a sector-shaped probe wall 12 is placed in the anechoic chamber 11, on which N probe antennas 13 are arranged, all of which are located on a spherical surface, and the distance between the probe antenna 13 on the probe wall 12 and the center of the antenna array of the device under test is R, the probe wall 12 covers 120° in the horizontal direction and 60° in the vertical direction, and each probe antenna 13 is equidistantly distributed in the azimuth and vertical angle directions;

[0077] First, the probe antenna 13 index matrix is ​​initialized according to the total number N of probe antennas 13 in the anechoic chamber 11, the number 2K of selected probe antennas 13, and the size L of the wolf pack. And each individual G i Randomly divided into equal parts uplink probe antenna 13 index and downlink probe antenna 13 index in And G i The probe antenna 13 indexes in the same area do not overlap, so as to obtain the initial double wolf pack G UL (1) G DL (1) Each individual Corresponding to a set of probe antenna 13 position vector Φ UL / DL , according to the determined reconstruction error function, set the fitness function f(·):

[0078]

[0079] Determined by convex optimization algorithm The probe weight w at the corresponding index in UL / DL , calculate each individual The corresponding fitness function Arrange them in ascending order and select the three individuals with the smallest fitness As the three alpha wolves in the wolf pack;

[0080] Then, it enters the iteration, and the other wolves in the wolf pack move according to the position of the head wolf to hunt. When the number of iterations is t, Traverse and get G GWO (t+1), the process is described as follows:

[0081]

[0082] A and C are used to control the hunting methods and ranges of wolves, and are defined as follows:

[0083] A=2·a(t)·r 1 -a(t),

[0084] C=2·r 2 ,

[0085] a(t) decreases linearly from 2 to 0 during the iteration, r 1 and r 2 is a random number in [0,1];

[0086] For the individual G in the traversal i (t), G i Different neighbors of (t) and random individuals in the population G(t) will move to G i (t) Learn and optimize; except for the above In addition, individual optimization will also generate a new individual To do this, first calculate G i (t) G i (t) and G j The Euclidean distance R between (t)∈G(t) i (t), R i ′(t), individual G i The neighborhood N of (t) i (t) = {X j (t)|Ri ′(t)≤R i (t)}, and then perform multi-neighborhood learning of the population to obtain

[0087]

[0088] Among them, G n (t) is N i (t) is a randomly selected individual, G r (t) is a randomly selected individual from G(t), r 3 is a random number in [0,1];

[0089] Compare and Select the individual with the lowest fitness and use it as G i (t+1), the traversal of i ends here;

[0090] Since there is a certain correlation in the spatial correlation of the uplink and downlink channels, it is mapped to wolf hunting in nature, that is, the alpha wolves of different wolf packs will obtain information about two preys at the same time during the search for prey. Therefore, in the iterative process, the alpha wolves of different wolf packs exchange information about preys, which has a positive incentive for the wolf pack to surround and hunt the corresponding prey. This behavior is mapped to a random exchange between the uplink and downlink probe index sets, so as to select a probe set that can more accurately reconstruct the uplink and downlink channels;

[0091] For G i (t+1), calculate And sort them to find the wolf head G with the smallest fitness α (t+1), and then use the partially matched crossover (PMX) algorithm to complete the probe index set and The exchange operation between them is performed, and a greedy strategy is adopted to select the one with the smallest fitness before and after the exchange as the new individual G′ α (t+1) put into G i Replace G in (t+1) α (t+1), one iteration is completed;

[0092] The upper limit T of the number of iterations and the fitness threshold ε are set. As the number of iterations increases, the simulated spatial correlation of the uplink and downlink channels generated by the selected probe antenna 13 will become closer and closer to the spatial correlation of the target uplink and downlink channels. Different numbers of iterations and fitness thresholds are set according to different target channels to obtain the optimal distribution position of the uplink and downlink probe antennas 13. At the same time, the weight of the probe antenna 13 is obtained through the convex optimization algorithm, thereby completing the accurate and efficient reconstruction of the target uplink and downlink channels.

[0093] In order to implement the above-mentioned uplink and downlink channel reconstruction method, the embodiment of the present application proposes a method as follows: Figure 3 The system for air interface performance testing shown in the figure includes a test device 1 consisting of an anechoic chamber 11, a probe wall 12 and a probe antenna 13, a switch array, a channel simulator and a user equipment simulator. Specifically, the switch array has the functions of switching the uplink and downlink of the probe antenna and selecting the activation of the probe antenna. The switch array can be used to adjust the uplink and downlink connection and disconnection of the probe antenna on the probe wall 12, and selectively activate the probe antenna 13 at the same time by selecting;

[0094] Taking the uplink communication link as an example, the user simulator first generates a test signal, which is fed to the channel simulator. Then, the channel simulator can generate a multipath channel environment, including fast fading such as path delay and Doppler spread. Furthermore, the channel pulse response generated by the channel simulator is convolved with the test signal. In order to simulate the spatial characteristics of the target channel on the side, the fading signal generated by the channel simulator is mapped to the probe antenna 13 selected by the switch array, thereby synthesizing the radiation signal in the test domain.

[0095] The present invention is not limited by the above embodiments. The above embodiments and descriptions are only for explaining the principle of the present invention. Without departing from the spirit and scope of the present invention, the present invention may be subject to various changes and improvements, which are within the scope of the present invention to be protected. The scope of protection of the present invention is defined by the appended claims.

Claims

1. A method for reconstructing uplink and downlink channels for air interface performance testing of a duplex system, characterized in that: include: Step 1, establishing a target uplink channel spatial correlation matrix and a downlink channel spatial correlation matrix to be reconstructed; Step 2, reconstruct the target uplink and downlink channels, determine the reconstruction accuracy error function according to the target reconstruction parameters, calculate the probe antenna weight, and then calculate the error function corresponding to each population individual; Step 3: Optimize and iterate the initial probe antenna index matrix to obtain a set of uplink and downlink probe antennas with the lowest reconstruction accuracy error.

2. The uplink and downlink channel reconstruction method for duplex system air interface performance testing according to claim 1 is characterized in that: In step 2, reconstructing the target uplink and downlink channels includes: Taking spatial correlation ρ as the channel reconstruction index, the optimization objective is expressed as: where Φ UL , Φ DL are the position vector matrices of the OTA probe antennas selected for uplink channel and downlink channel reconstruction, Φ is the position vector matrix of all candidate probe antennas, This means that due to the coupling effect between antennas and the physical size limitation of the probe antenna, the uplink probe antenna and the downlink probe antenna cannot share the same probe antenna position. is the weight of the selected K uplink probe antennas, is the weight of the selected K downlink probe antennas, They represent the kth uplink probe antenna weight and the kth downlink probe antenna weight, respectively, and w UL 、w DL satisfy as well as ρ UL , DL They represent the uplink target spatial correlation and the downlink target spatial correlation, respectively, and are defined as: in is a matrix composed of M row vectors of the antenna positions of the device under test, Ω is the space vector corresponding to the beam arrival angle, P UL (·) is the uplink angular power spectrum, P DL (·) is the downlink angle power spectrum, which satisfies ∮P(Ω)dΩ=1; They represent the uplink reconstruction spatial correlation and the downlink reconstruction spatial correlation respectively, and they are:

3. The uplink and downlink channel reconstruction method for duplex system air interface performance testing according to claim 2 is characterized in that: In step 2, the probe antenna index matrix is ​​initialized according to the total number of probe antennas N in the multi-probe microwave darkroom, the number of selected probe antennas 2K, and the population size L For each individual G i Randomly divide the antenna index of the uplink probe into equal numbers and downlink probe antenna index in And G i The probe antenna indexes in the same group are not repeated, so the initial double wolf pack G is obtained. UL (1) G DL (1).

4. The uplink and downlink channel reconstruction method for duplex system air interface performance testing according to claim 1 or 3, characterized in that: In step 2, each individual Corresponding to a set of probe antenna position vectors Φ UL / DL , according to the determined reconstruction error function, set the fitness function f(·): Determined by convex optimization algorithm The probe weight w at the corresponding index in UL / DL , calculate each individual The corresponding fitness function Arrange them in ascending order and select the three individuals with the smallest fitness As the three alpha wolves in the wolf pack.

5. The uplink and downlink channel reconstruction method for duplex system air interface performance testing according to claim 4 is characterized in that: In step 3, the initial probe antenna index matrix is ​​optimized and iterated by the double wolf pack hunting optimization algorithm based on population interaction. The other wolves in the wolf pack hunt according to the position of the head wolf. When the number of iterations is t, Traverse and get G GWO (t+1), the process is described as follows: A and C are used to control the hunting methods and ranges of wolves, and are defined as follows: A=2·a(t)·r1-a(t)......(11), C＝2·r2……(12)， a(t) decreases linearly from 2 to 0 during the iteration, r1 and r2 are random numbers in [0,1]; For the individual G in the traversal i (t), G i Different neighbors of (t) and random individuals in the population G(t) will move to G i (t) Learn and optimize, except for the In addition, individual optimization will generate a new individual To do this, first calculate G i (t) G i (t) and G j The Euclidean distance R between (t)∈G(t) i (t), R i ′(t), then individual G i The neighborhood N of (t) i (t) = {X j (t)|R i ′(t)≤R i (t)}, and then perform multi-neighborhood learning of the population to obtain Among them, G n (t) is N i (t) is a randomly selected individual, G r (t) is a randomly selected individual from G(t), r3 is a random number in [0,1]; Compare and Select the individual with the lowest fitness and use it as G i (t+1), the traversal of i ends here; against calculate And sort them to find the wolf head G with the smallest fitness α (t+1), and then use the partial matching crossover algorithm to complete the probe index set and The exchange operation between them is performed, and a greedy strategy is adopted to select the individual with the smallest fitness before and after the exchange as the new individual G′ α (t+1) put into G i Replace G in (t+1) α (t+1), one iteration is completed.

6. The uplink and downlink channel reconstruction method for duplex system air interface performance test according to claim 4, characterized in that: In step 3, the leader wolves of different wolf packs exchange information about the location of the hunting target. This information exchange behavior can be mapped to a random exchange between the uplink and downlink probe index sets, thereby selecting a probe set that can more accurately reconstruct the uplink and downlink channels.

7. The uplink and downlink channel reconstruction method for duplex system air interface performance test according to claim 4, characterized in that: Step 3 includes the steps of setting the upper limit T of the number of iterations and the fitness threshold ε.

8. The uplink and downlink channel reconstruction method for duplex system air interface performance test according to claim 1, characterized in that: In step 1: The uplink and downlink channels to be reconstructed are composed of multiple paths or clusters composed of paths; Select different reconstruction target parameters according to different channel characteristics; At the same time, the reconstruction accuracy of the channel is measured by the sum of the Euclidean distances between the uplink and downlink targets and the reconstruction parameters of the reconstructed channel.

9. A system for air interface performance testing of a duplex system according to any one of claims 1 to 8, characterized in that: include: A test device, comprising an anechoic chamber, a probe wall disposed in the anechoic chamber, and a probe antenna disposed on the probe wall, the test device being used to shield external signal interference; A switch array is connected to the test device, wherein the switch array has the functions of switching the uplink and downlink of the probe antenna and activating the probe antenna selection, and is used to adjust the on / off status of the uplink and downlink of the probe antenna on the probe wall; A user equipment simulator, used to generate a test signal to feed into the channel simulator and simultaneously receive a downlink signal; The channel simulator is used to create and simulate a multipath fading environment. It generates a wireless channel impulse response and convolves it with the user equipment simulator's transmit signal to generate a fading signal that is fed into the probe antenna selected by the switch array and radiated to the test area to complete the channel reconstruction.

10. The system for duplex system air interface performance testing according to claim 9, characterized in that: The probe antenna is connected to the switch array through a wire to complete the probe antenna selection.

Images