KR101989373B1 - method for calibrating massive MIMO air-channel simulator - Google Patents

Abstract

The present invention relates to a calibration method for a massive MIMO channel simulator which can correct the magnitude and phase of each input/output signal of the channel simulators. The calibration method for the massive MIMO channel simulator, performed by a calibration server, includes: (a) outputting a reference signal from any UE I/F board of the massive MIMO channel simulator to a calibration kit side; (b) receiving the reference signal at each BS/IF board, delaying the same by a predetermined sample, and outputting the delayed reference signal; (c) capturing a combined reference signal, which is combined while delayed at the BS I/F board, and computing a magnitude and a phase offset (Δp) of each delayed reference signal using a peak of an autocorrelation with the original reference signal; (d) carrying out a calibration after computing a calibration value that keeps the magnitude and phase offset (Δp) of each delayed reference signal the same; and (e) converting the calibration value calculated in (d) into a converted calibration value when a base station is connected to a main port.

Description

METHOD FOR CALIBRATING A MEMBRANE MIMO AIR-CHANNEL SIMULATOR

In particular, the present invention relates to a method of calibrating a channel simulator for a massive MIMO. More particularly, the present invention relates to a method of calibrating a channel simulator for a massive MIMO system, To a method of calibrating a channel simulator for a massive MIMO.

In recent years, the amount of wireless data required by the exponential increase of smart devices has also increased at a very high speed. However, since the usable frequency bandwidth and radio channel capacity are limited, efficient use of existing radio resources (frequency, etc.) It is demanding stronger than ever.

Massive Multiple Input and Multiple Output (MIMO) is a multi-antenna technology that aims to achieve high energy efficiency with a high transmission rate by attaching dozens or more antennas to the base station. If the complete channel vector can be obtained by using the uplink / downlink channel correlation as a result of the first study on the massive MIMO in a TDD (Time Division Multiplexing) system, interference between different users' Since the announcement that users can be served simultaneously by using simple transmit and receive filters canceled, studies on massive MIMO have been actively studied and become one of the hottest fields of interest nowadays.

Accordingly, in an attempt to improve the performance of the LTE-Advanced system, an international standardization organization such as the 3GPP (Third Generation Partnership Project) has proposed a technique of beam forming, in which multiple users simultaneously use the same radio resources, (FD) -MIMO (Full-Dimension (or 3D) Multiple-Input and Multiple-Output) technique that maximizes the wireless channel capacity. Hereinafter, MIMO is used as a comprehensive concept including FD-MIMO as a subset, and it is predicted that such massive MIMO will be adopted as an essential technology in the 5G system which is under development.

Meanwhile, in a multipath communication channel, a line-of-sight (LOS) component (hereinafter, referred to as a " LOS ") component is transmitted between a base station (eNB) And the reflected wave component and the diffraction wave component affect each other at the same time. Since these signals are received not only by the terminal but also by Doppler spreading due to the movement of the terminal, the mobile communication has a poor radio environment compared with the fixed communication.

Generally, rural or suburban environments in which direct waves exist can be explained by the Ricean channel model, and urban areas where the direct path signal is thin and the composite signal by multipath is large can be explained by the Rayleigh model have. There is also a shading effect due to the non-uniformity of the surrounding terrain.

In order to guarantee the performance of the wireless system, it is necessary to verify the performance of the wireless system in addition to the verification through the simulation and analysis, as well as the prototype And a field test. However, real-time channel simulator can be used as a more practical method because field test of the developed wireless system in all environmental conditions is time consuming and costly. This is a system that can simulate virtually any environment that may actually occur in a wireless channel.

The applicant of the present invention has proposed a technique for efficiently applying bi-directional path loss and bidirectional real-time fading to all paths (P * Q) between P (> 2 integers) The channel simulator has been patented and has been patented under the registration number 1286023. While continuing the related research, the channel simulator for massive MIMO has recently been developed. A 'massive MIMO base station' is defined as a base station having 10 or more antennas, and a 'massive MIMO channel simulator' is defined as a simulator capable of connecting at least one massive MIMO base station.

On the other hand, since the channel simulator for a massive MIMO is based on the beam forming concept, it is not necessary to precisely correct the signal after measuring it in real time, It is important.

However, in the conventional large capacity channel simulator, only the calibration operation is performed to maintain the same signal size between channels, and therefore, when the same is applied to the channel simulator for the mass MIMO, stable performance can not be guaranteed at all .

Prior Art 1: 10-1286023 (Patent Title: CHANNEL SIMULATOR) Prior Art 2: 10-1606354 (Patent Title: Calibration Method of Channel Simulator) Prior Art 3: 10-2017-0077671 Patent Application (Name of invention: Control method of channel simulator)

SUMMARY OF THE INVENTION The present invention is conceived to solve the above-described problems, and it is an object of the present invention to provide a channel simulator for real time channel generation by connecting a massive MIMO base station having a plurality of antennas and a terminal, And a method of calibrating a channel simulator for a massive MIMO.

According to a first aspect of the present invention, there is provided a mobile communication system including a BS I / F board for performing RF signal processing for a base station having 10 or more antennas and a terminal having at least one antenna, A main body having a board and an LP board for performing fading processing for each channel between the boards; A calibration kit and a calibration server for externally connecting the BS I / F board and the UE I / F board are provided. The BS I / F board and the UE I / F board are provided with a main port connected to the base station and the terminal, F unit and a UE I / F board are connected to the calibration kit side, and the UE I / F is connected to the calibration I / F board in the channel simulator for massive MIMO, (A) outputting a reference signal from the board to the calibration kit side; (B) receiving the reference signal from each BS I / F board and delaying the reference signal by a predetermined number of samples; (C) calculating the size and phase offset (? _P ) of each delayed reference signal by the peak value of the autocorrelation value with the original reference signal after capturing the delayed combined reference signal on the UE I / F board; step; (D) performing a calibration after calculating a calibration value that keeps each of the magnitude and phase offsets (? _P ) the same, and connecting the calibration value calculated in the step (d) And (e) a step of calibrating the channel simulator for a massive MIMO.

According to a second aspect of the present invention, there is provided a radio communication system including a BS I / F board and a UE I / F board each performing RF signal processing for a base station having ten or more antennas and a terminal having one or more antennas, A main body having an LP board for performing fading processing for each channel; A calibration kit and a calibration server for externally connecting the BS I / F board and the UE I / F board are provided. The BS I / F board and the UE I / F board are provided with a main port connected to the base station and the terminal, F unit and a UE I / F board are connected to the calibration kit side, and the UE I / F is connected to the calibration I / F board in the channel simulator for massive MIMO, (H) outputting a reference signal from the board to the LP board side; (I) receiving the reference signal from each BS I / F board and delaying the reference signal by a predetermined number of samples; (J) calculating the magnitude and phase offset (Δ _p ) of each delayed reference signal by the peak value of the autocorrelation value with the original reference signal after capturing the delayed combined reference signal on the UE I / F board, step; (K) performing calibration after calculating a calibration value that keeps each of the magnitude and phase offsets (? _P ) the same, and connecting the calibration value calculated in step (k) And a step (l) of calibrating the channel simulator for a massive MIMO.

In the above-described first and second features, each channel of the LP board is simply connected without performing fading processing.

In the first and second aspects described above, the conversion correction value is determined by the size and phase offset (? _C ) by the RF cable between each antenna of the base station and the BS I / F board, the size and phase The offset (Δ _bu ) and the magnitude and phase offset (Δ _mt ) between the main port and the test port are performed, and the respective offsets Δ _c , Δ _bu and Δ _mt are measured and stored in the calibration server.

In the first and second features described above, the calculation of the magnitude and phase offset (? _P ) of each of the delayed reference signals is performed with the fractional delay offset of each of the delayed reference signals compensated.

In the first and second aspects described above, the reference signal is a Zadoff-add signal.

과 같이 정의된다.In the first and second aspects described above, the reference signal is a modulator dop-boost signal x _u (n) whose bandwidth is limited by the used bandwidth of the communication system,

.

가 BS I/F 보드의 i번째 테스트 포트를 거쳐서 UE I/F 보드에서 캡쳐되면

가 되고, 여기에서

및

는 각각 원래의 변형 자도프-추 신호에 대한 시간 지연 성분과 위상 및 크기 오프셋 성분이며, 상기 자기 상관 값의 피크치는 상관계수 함수인In the above-described first and second aspects, the modulated Zadoff-tripled signal outputted from the UE I /

Is captured on the UE I / F board through the i-th test port of the BS I / F board

, And here

And

Is a time delay component and a phase and magnitude offset component, respectively, for the original demodulator dop-add signal, and the peak value of the autocorrelation value is a correlation coefficient function

의 최대치를 검색하여

로 추정한다.

The maximum value of

Respectively.

In the first and second features described above, the input reflection coefficient of the base station signal input through the main port of the BS I / F board, the input port of the test port input terminal and the calibration port of the BS I / F board to the calibration kit, So as to remain the same.

According to the method of calibrating a channel simulator for massive MIMO of the present invention, a channel simulator can be connected to a base station employing a massive MIMO antenna and a terminal, It is possible to give high reliability to the logging and analysis results by correcting the signals so that the magnitudes and phases of the signals of the respective channels are equal beforehand.

1 is a general system configuration diagram of a channel simulator for a massive MIMO to which the calibration method of the present invention can be applied.
2 is an internal functional block diagram of the BS I / F board in FIG.
FIG. 3 is an internal configuration view of the calibration kit in FIG. 1; FIG.
4 is a flowchart illustrating a method of calibrating a channel simulator for massive MIMO of the present invention.
FIGS. 5A and 5B are graphs showing time-domain and frequency domain demodulator-adder signals used in the calibration method of the channel simulator for massive MIMO of the present invention, respectively.
6A and 6B are graphs showing a capture signal of the sum of the modulator-doped-add signals output from 64 antennas of the base station in time domain and frequency domain, respectively.
FIG. 7 is a graph showing autocorrelation characteristics of the sum of modulator dop-add signals received from 64 base station antennas; FIG.

Hereinafter, a preferred embodiment of a method of calibrating a channel simulator for massive MIMO of the present invention will be described in detail with reference to the accompanying drawings.

1 is a general system configuration diagram of a channel simulator for a massive MIMO to which the calibration method of the present invention can be applied. As shown in FIG. 1, a channel simulator for a massive MIMO system to which the calibration method of the present invention can be applied includes a real-time fading processing for a plurality of channel signals connecting the massive MIMO base station 600 and the terminal 700, A link processor board 110 for performing fading and fast fading processing, a base station interface board 110 for connecting the base station 600 and the LP board 110, A UE I / F board (hereinafter simply referred to as a 'UE I / F board') 120 that connects the LP board 110 and the terminal 700 The signal of the main body 100 for each channel without connecting the simulator main body 100, the base station 600, and the terminal 700, To measure the size and phase of A calibration kit 200 used to support the user and a test scenario desired by the user such as the position or distance of the base station 600 and the terminal 700 or the direction of each antenna can be set based on GUI (Graphic User Interface) A scenario server (S-server) 300 for controlling the LP board 110 by calculating channel coefficients according to a test scenario set by the user through a test manager (TM) 300 and a test manager 300, And a calibration server (Cal server or C-server) 500 for controlling the calibration and verification of the channel simulator.

1 illustrates a channel simulator for 64 * 32-mass MIMO in which one massive MIMO base station 600 having 64 antennas and 16 terminals 700 having two antennas are connected. In this case, when 16 BS I / F boards 120 are constituted and 16 UE I / F boards 130 are constituted, 16 base station antennas are connected to each BS I / F board 120, One terminal 700 may be connected to the I / F board 130. The main port (MP) connected to the base station and the terminal per channel and the test port connected to the calibration kit 200 are connected to the BS I / F board 120 and the UE I / F board 130, (Test Port) TP. Switching between these ports can be performed collectively under the control of the calibration server 500. Meanwhile, since the test manager 300 and the calibration server 500 operate at different times at all times, they can be mounted on the same PC.

In FIG. 1, a channel simulator, that is, a 64 * 32 channel simulator in which 16 terminals 700 are connected to one massive MIMO base station 600 is illustrated, but two base stations having 32 antennas may be connected. Furthermore, it is also possible to expand channels that can be generated by 128 * 64, etc. Hereinafter, one BS I / F board 120 corresponds to one base station antenna port, and one UE I / F board corresponds to one terminal antenna port.

FIG. 2 is an internal functional block diagram of the BS I / F board in FIG. 1, only for one uplink channel and downlink channel pair for convenience. 2, the BS I / F board 120 includes an RF switch rfs for switching the signal path for the downlink channel to the main port MP or the test port TP, An RF duplexer rfd for supporting bidirectional communication on the downlink and uplink, an up / down converter 128 for downconverting the RF signal to convert the RF signal into a baseband signal (hereinafter simply referred to as BB signal) A gain adjuster 127d for adjusting the gain of the signal, preferably a variable attenuator or a variable amplifier, an ADC (Analog to Digital Converter) 126d for converting the analog BB signal into a digital BB signal, (Amp./Phase) calibrator 124d for calibrating the magnitude and phase of the signal and a signal generator (SG) and a signal analyzer ; A signal generating / analyzing unit 122d and a photoelectric An optical transceiver, for performing a ring or inverse transformation, for example, be achieved by having the SFP I / F (Small Form Factor Pluggable Interface) (121d) one by one.

Next, for the uplink channel, the SFP I / F 121u, the signal generating / analyzing unit 122u, the size / phase correcting unit 124u, the fractional delay compensating unit 125u, A DAC (Digital to Analog Converter) 126u for converting the signal into an analog BB signal, a gain adjuster 127u for adjusting the gain of the signal, which is composed of a variable attenuator or a variable amplifier, an up- 128u, the RF duplexer rfd, and the RF switch rfs sequentially.

In the above-described configuration, the signal generator SG of the signal generating / analyzing unit 122d generates a demodulator dop-add signal as described later, and the signal analyzer SA includes a sample buffer, And measures the magnitude and phase offset between each antenna signal after capturing the combined signal into a digital sample.

Although not shown, the configuration of the downlink channel and the uplink channel pair of each UE I / F board 130 may be the same as that of the BS IF board 120 only in different directions. In the drawing, the part indicated by the dot-dash line represents a configuration implemented by an FPGA (Field Programmable Gate Array).

3 is an internal configuration diagram of the calibration kit in FIG. As shown in FIG. 3, the calibration kit 200 according to the present invention may include first and second signal distributors 210 and 220 combined in mutually different directions. In this case, the first signal distributor 210 is composed of, for example, a 64 * 1 signal distributor to combine the RF signals of the test port TP for each channel (antenna) of the BS I / F board 120, 2 to the BS I / F board 120 by dividing the RF signals received from the second signal distributor 220 into signals for each channel.

Similarly, the second signal distributor 220 may include a 1 * 32 signal distributor to combine the RF signals of the test port TP for each channel of the UE I / F board 130, Or separates the RF signals received from the first signal distributor 210 for each channel and transmits the RF signals to the UE I / F board 130.

MP.BS # 1, MP.BS # 64, MP.UE # 1 and MP.UE # 32 are respectively provided in the BS I / F board and the UE I / F board of the simulator main body 100 TP # BS # 1, TP.BS # 64, TP.UE # 1 and TP.UE # 32 represent the first and 64th main ports (MP) and first and 32.sup.th main ports The first and 64th test ports TP and the first and 32th test ports TP provided in the BS I / F board and the UE I / F board of the simulator main body 100, respectively.

On the other hand,? _C ,? _Bu ,? _Mt and? _P can be defined as shown in Table 1 below.

Item Contents Remarks Δ _c An offset of the RF cable between each antenna of the base station and the BS I / F board of the main body Correction is possible only by linking with base station.
Measure by tapping the cable into the NA (Network Analyzer). _Mt The offset between the main port and the test port by the connector and RF switch for each antenna of the base station Preset and store the offset value in advance at the time of production or change of H / W. Δ _bu For each antenna of the base station, the internal cable of the calibration kit, the connector, the offset by the first and second signal distributors Preset and store the offset value in advance at the time of production or change of H / W. Δ _p Size / phase offset for each TP as measured by the actual calibration operation.

4 is a flowchart for explaining the calibration method of the channel simulator for massive MIMO of the present invention, and may be performed under the control of the calibration server 500. [ Further, it will be described that the calibration operation for the downlink channel is performed first.

As described above, in the calibration server 500 of the present invention, Δ _c , Δ _bu (n) measured by an external network analyzer for each base station antenna, And Δ _mt there value is stored, for example, the available carrier frequency band to 10㎒ within 20㎒ unit frequency interval and the available gain or attenuator in a range Δ _c, Δ and Δ _{bu mt} measured values for the amplifier of (Step S10).

In this state, as shown in Fig. 4, in step S20, various parameters are set after switching the channel simulator to the calibration mode. For example, a calibration sequence or a sample delay interval setting for the downlink channel or the uplink channel . In this process, the calibration server 500 controls the channel server 500 so that the channel coefficient for each path of the LP board 110 is 1, that is, the simple connection mode in which no fading is applied through the scenario server 400 do.

Next, in step S30, the SG of the signal generating / analyzing unit of the UE I / F board 130 generates a reference signal, preferably a modulator dop-add signal, and outputs it to the calibration kit 200 continuously.

In this regard, in the calibration method of the present invention, as a kind of a Constant Amplitude Zero Auto-Correlation (CAZAC) signal for simultaneously measuring the size / phase offset for a plurality of downlink channels (paths), Peak-to-Average Power Ratio ) And a Zadoff-Chu signal (hereinafter, simply referred to as a ZC signal) having excellent auto-correlation characteristics are used as reference signals. However, in consideration of power efficiency, It is preferable to use the Zadoff-add signal.

The modified ZC signal used in the method of the present invention is defined as Equation 1 below and can be generated for 10 ms periodically as N _ZC samples at a sampling frequency (Fs) of 30.72 MHz when targeting LTE do.

Generally, the number of samples (N _ZC ) of a ZC signal is defined as a prime number. However, in the method of the present invention, the number of samples is set to a power of 2 (2 ⁿ ) 14), for example, 16384 (= 2 ¹⁴ ) which is close to 16381, which is a prime number, to simplify implementation.

5A and 5B are graphs showing a modified ZC signal used in the method of calibrating a channel simulator for a massive MIMO system according to the present invention in a time domain and a frequency domain, respectively, and a complex BB I / Q Quad-phase signals. As shown in FIG. 5A, the modified ZC signal shows the CAZAC characteristic in the time domain. As shown in FIG. 5B, the band is limited to the LTE frequency bandwidth of 9 * 2 MHz in the frequency domain and shows a flat spectrum have.

Referring back to FIG. 4, in step S40, each BS I / F board 120 outputs a reference signal delayed by a predetermined sample, for example, 13 samples, that is, a modified ZC signal. In this regard, the modified ZC signals output from the UE I / F board 130 are distributed to 64 signals via the first signal distributor 210 of the calibration kit 200, Each BS I / F board 120 delays the input deformed ZC signal by 13 samples and outputs it. Accordingly, the modified ZC signal output from the 64th BS I / F board 120 is delayed by 819 (= 13 * 63) samples (64 * 13 samples in the case of delay from the first) to the first modified ZC signal do. The signal output from the UE I / F board 130 moves along each channel path of the LP board 110 and is input to the UE I / F board 130 while being combined. In step S50, The sub SA captures the combined BB signal and calculates the autocorrelation between the original reference signal and the modified ZC signal.

In this regard, if a pure ZC signal having no bandwidth is used, delaying arbitrary samples is not a problem. However, since the method of the present invention uses a modified ZC signal having a limited bandwidth, , And this delay value can be obtained by precomputation.

6A and 6B are graphs showing a capture signal of a sum of modified ZC signals output from 64 antennas of a base station in a time domain and a frequency domain, respectively. As shown in FIG. 6A, the PAPR increases in the sum signal of the 64 time-delayed deformed ZC signals, and it can be seen that the spectrum is not a flat spectrum in the frequency domain, as shown in FIG. 6B.

FIG. 7 is a graph showing the autocorrelation characteristics of the sum of modified ZC signals received from 64 base station antennas. According to this graph, peak detection in 64 base station antennas is performed at a predetermined time delay, for example, You can measure the antenna's sampling timing offset and size / phase offset.

Returning to Fig. 4, in step S60, an integer multiple and a fractional multiple sample position are searched for by detecting the peak value of the autocorrelation value as described above. In this regard, even if all the BS I / F boards 120 use the same clock, since the start timing of each ADC 126d changes due to a difference in logic or cable length, it is inevitably a fractional (fractional sampling) Timing offsets are generated and must be compensated for accurate calibration.

To this end, in step S70, a fractional timing offset compensation filter coefficient (C _ff ) for each TP is calculated, and this compensation filter coefficient (C _ff ) can be performed by, for example, an interpolation method through oversampling.

Next, in step S80 the measured magnitude / phase offset (Δ _p) of each TP, and the step S90 again verified by applying the respective TP-specific compensation filter coefficient (C _ff) and magnitude / phase offset (Δ _p) the calibration value to carry out, such a correction value, for example, be obtained by each TP specific size / phase offset (Δ _p) the first manner from the normalized based on the size / phase offset (Δ _p) of the first TP state taking the reciprocal .

Next, in step S90, it is determined whether the size offset of all the TPs is within 0.35 dB, and the phase offset is within 3 degrees, for example, based on the size offset of the first TP. When an If it is determined that any one at any that do not satisfy the reference value of the magnitude and phase offset at step S100 is satisfied all the while repeating the steps S10 or less, the process advances to step S110 to Δ _c, Δ _bu and Δ _mt And calculates a final calibration value for each MP and configures a fractional delay calibration unit 125d and a size / phase calibration unit 124d of the BS I / F board 120 based on the calculated final calibration value.

가 교정 키트(200) 및 BS I/F 보드(120)의 i번째 TP 포트를 거쳐서 UE I/F 보드(130)의 SA에 수신되면

가 된다. 즉, i번째 TP 포트의 신호에는 원 신호의 시간 지연 성분

와 위상/크기 오프셋 성분

가 포함되어 있다. 따라서, 아래의 수학식 2와 같은 상관계수 함수

의 최대치를 검색하여

을 추정한다.The following is summarized. First, the transformed ZC signal generated at the SG of the UE I / F board 130

Is received by the SA of the UE I / F board 130 via the i < th > TP port of the calibration kit 200 and the BS I / F board 120

. That is, the signal of the i-th TP port includes the time delay component

And a phase / magnitude offset component

. Therefore, the correlation coefficient function < RTI ID = 0.0 >

The maximum value of

.

=1이 되어서 단계 S100에서 규정한 기준치를 만족시키게 된다. 도 4에의 각 단계에서의 계산 및 측정은 캘리브레이션 서버에 의해 수행될 수 있다.Finally, multiply this value by a reciprocal to complete the calibration operation. Theoretically,

= 1 and the reference value defined in step S100 is satisfied. Calculation and measurement at each step in Fig. 4 can be performed by the calibration server.

On the other hand, the calibration operation for the uplink channel is performed when the modified ZC signal generated in the SA of the UE I / F board 130 is input to the BS I / F board 120 via the LP board 110, Port), and is then captured by the SA of the UE I / F board 130 through the calibration kit 200 and then performed through the same process as shown in FIG.

The calibration operation is performed on the downlink channel input from all the UE I / F boards 130 through the calibration kits 200 to all the BS I / F boards 120, And the uplink channel input to all the BS I / F boards 120 through the base station 110.

During the calibration process, a base station signal input terminal through the main port of the BS I / F board 120 of the mass-MIMO channel simulator, a main port of the BS I / F board 120 with respect to the calibration kit 200, And the calibration kit 200, the magnitude and the phase of the deformed ZC signal may be changed due to the reflection of the input signal. Therefore, it may be necessary to further correct the magnitude and phase of the distorted ZC signal.

In consideration of this, in the method of the present invention, the reflection coefficient S11C of the base station signal input through the main port of the BS I / F board 120, the reflection coefficient S11C of the BS I / F board 120, The input reflection coefficient (S11M and S11T) of the test port and the input reflection coefficient of the calibration kit 200 are measured and stored in advance through a network analyzer in a state where all the equipments are configured. And can be calibrated through a magnitude / phase calibration unit 124d such that the reflection coefficients become equal.

The calibration method of the present invention can be performed for a frequency range of 10 MHz to 20 MHz within the usable carrier frequency band and for a gain range of an available attenuator or amplifier. Also, it can be performed at the time of replacement of each board or verification of the calibration state of the simulator.

While the preferred embodiments of the method of calibrating a channel simulator for massive MIMO of the present invention have been described in detail with reference to the accompanying drawings, it is to be understood that various changes and modifications may be made without departing from the scope of the present invention . Accordingly, the scope of the present invention should be determined by the following claims. For example, in the above-described embodiment, LTE is taken as an example, but the present invention can be applied to future 5G technology within the scope of maintaining the technical idea.

In addition, terms such as 'board' and 'kit' are arbitrarily borrowed for the sake of logical or functional explanatory convenience and should not be used to limit the scope of the rights, It may be explained separately in units.

For example, in the above-described embodiment, four base station antennas are connected to one BS I / F board and two antennas of a terminal are connected to one UE I / F board. However, One base station antenna may be connected to one BS I / F board or all 64 base station antennas may be connected to one BS I / F board.

100: a simulator main body, 110: a link processor (LP) board,
120: a base station interface (BS I / F) board,
121d, 121u: SFP I / F, 122d, 122u: signal generating / analyzing unit,
124d, 124u: size / phase calibration unit, 125d, 125u: fractional delay calibration unit,
126d, ADC, 126u: DAC,
127d, 127u: gain adjustment unit, 128: up / down converter,
130: terminal interface (BS I / F) board,
200: calibration kit, 210: first signal distributor,
220: second signal distributor, 300: test manager,
400: scenario server, 500: calibration server,
600: Massive MIMO base station, 700: Terminal,
MP: Main port, TP test port

Claims (9)

A BS I / F board and a UE I / F board each performing RF signal processing for a base station having 10 or more antennas and a terminal having one or more antennas, and an LP board for performing fading processing for each channel between these boards ; A calibration kit and a calibration server for externally connecting the BS I / F board and the UE I / F board are provided. The BS I / F board and the UE I / F board are provided with a main port connected to the base station and the terminal, And a test port connected to the kit, wherein the test port is performed under the control of a calibration server in a channel simulator for a massive MIMO,
(A) outputting a reference signal from an arbitrary UE I / F board to a calibration kit side while the BS I / F board and the UE I / F board are connected to the calibration kit side;
(B) receiving the reference signal from each BS I / F board and delaying the reference signal by a predetermined number of samples;
(C) calculating the size and phase offset (? _P ) of each delayed reference signal by the peak value of the autocorrelation value with the original reference signal after capturing the delayed combined reference signal on the UE I / F board; step;
(D) performing a calibration after calculating a calibration value that keeps the respective magnitude and phase offsets (? _P ) the same, and
And converting the calibrated value calculated in step (d) into a converted calibration value when the base station is connected to the main port, and finally calibrating the calibrated value, and (e) calibrating the calibrated value for the massive MIMO channel simulator. A BS I / F board and a UE I / F board each performing RF signal processing for a base station having 10 or more antennas and a terminal having one or more antennas, and an LP board for performing fading processing for each channel between these boards ; A calibration kit and a calibration server for externally connecting the BS I / F board and the UE I / F board are provided. The BS I / F board and the UE I / F board are provided with a main port connected to the base station and the terminal, And a test port connected to the kit, wherein the test port is performed under the control of a calibration server in a channel simulator for a massive MIMO,
(H) outputting a reference signal from an arbitrary UE I / F board to the LP board side while the BS I / F board and the UE I / F board are connected to the calibration kit side;
(I) receiving the reference signal from each BS I / F board and delaying the reference signal by a predetermined number of samples;
(J) calculating the magnitude and phase offset (Δ _p ) of each delayed reference signal by the peak value of the autocorrelation value with the original reference signal after capturing the delayed combined reference signal on the UE I / F board, step;
(K) performing a calibration after calculating a calibration value that keeps the respective magnitude and phase offset (? _P ) the same, and
Converting the calibration value calculated in the step (k) into a conversion calibration value when the base station is connected to the main port, and finally calibrating the calibrated value; and (l) calibrating the calibration value for the massive MIMO channel simulator. The method according to claim 1 or 2,
Wherein each channel of the LP board is simply connected without performing fading processing. The method of claim 3,
The conversion correction value is determined by the size and phase offset (? _C ) of the RF cable between each antenna of the base station and the BS I / F board, the size and phase offset (? _Bu ) _Lt ; RTI ID = 0.0 > ( _mt ), < / RTI >
_Wherein the respective offsets [Delta] _c , [Delta] _bu and [Delta] _mt are previously measured and stored in a calibration server. The method of claim 4,
Wherein the calculation of the magnitude and phase offset (? _P ) of each delayed reference signal is performed in a state of compensating for a fractional delay offset of each of the delayed reference signals. The method of claim 4,
Wherein the reference signal is a Zadoff-add signal. The method of claim 4,
The reference signal is a modulator dop-boost signal x _u (n) whose bandwidth is limited by the bandwidth used by the communication system,

The method of claim 1, wherein the channel simulator is a channel simulator. The method of claim 7,
The modulated zero-add signal output from the UE I / F board

Is captured on the UE I / F board through the i-th test port of the BS I / F board

Lt; / RTI &
From here

And

Are the time delay components and the phase and magnitude offset components for the original modulator dope-add signal, respectively,
The peak value of the autocorrelation value is a correlation coefficient function

The maximum value of

And estimating the channel simulator for the MIMO channel. The method of claim 4,
The step of calibration is added so that the input reflection coefficient at the input port of the base station signal through the main port of the BS I / F board, the main port of the BS I / F board for the calibration kit, the input port of the test port and the input of the calibration kit remain the same. Wherein the channel simulator includes a plurality of channel simulators.

Images