Calibration method for parallel multi-channel wireless channel measurement and system for the same

Abstract

The present invention discloses a calibration method for parallel multi-channel wireless channel measurement and a calibration system thereof. The transmitting end and the receiving end are disconnected, and are respectively connected to the calibration receiving channel and the calibration transmitting channel, and the calibration receiving channel/calibration transmitting channel cooperates with a measurement channel in air interface measurement, to calibrate a channel response characteristic of the transmitting end and the receiving end. By means of the present invention, a current channel response characteristic between multiple channels can be online supervised in real time, so as to ensure that a measurement error because of an impact of mutual interference between multiple channels can be avoided in the channel measurement process.

Claims

1. A calibration method for parallel multi-channel wireless channel measurement, wherein a calibration receiving channel is added at a transmitting end and a calibration transmitting channel is added at a receiving end, and the method comprising the following steps: disconnecting the transmitting end and the receiving end, connecting the transmitting end with the calibration receiving channel, connecting the receiving end with the calibration transmitting channel, and recording channel response characteristic of the transmitting end and the receiving end with cooperation of the calibration receiving channel and the calibration transmitting channel in air interface measurement, the receiving end acquires a peak value by using a correlation between orthogonal sequences, and transforms the peak value to a frequency domain by using FFT, to measure a frequency response matrix of channels.

2. The calibration method according to claim 1, wherein: in a calibration process, the transmitting end parallel transmits multi-channel data, and the receiving end parallel receives multi-channel data in a code division multiplexing manner.

3. The calibration method according to claim 2, further comprising, Before the step of disconnecting the transmitting end and the receiving end, directly connecting the transmitting end and the receiving end, disconnecting the calibration receiving channel and the calibration transmitting channel, and testing a response characteristic of a passive device.

4. The calibration method according to claim 1, wherein: the receiving end selects a j.sup.th receiving channel, generates a local sequence of an i.sup.th transmitting channel of the transmitting end, performs a correlation operation on a receiving sequence of the j.sup.th receiving channel and the local sequence of the i.sup.th transmitting channel to acquire the correlation peak value, wherein both i and j are positive integers; then transforms the correlation peak value to the frequency domain by using FFT, to acquire a frequency domain channel response characteristic; and acquires channel responses of all channels as the frequency response matrix.

5. The calibration method according to claim 1, wherein: calibrating the channel response characteristic of the transmitting end and the receiving end with a measurement transmitting channel and a measurement receiving channel, when the measurement transmitting channel in a silent period, the calibration transmitting channel transmits a narrow pulse signal for circularly testing, and transmission of the narrow pulse signal satisfies a time delay requirement of the measurement transmitting channel and the measurement receiving channel.

6. The calibration method according to claim 5, wherein: when the transmitting end transmits a signal, the calibration transmitting channel is closed; after transmission of the signal, the calibration transmitting channel begins to transmit the narrow pulse signal.

7. The calibration method according to claim 1, wherein: at the transmitting end and the receiving end, a pulse per second is used as an initial synchronization trigger signal source; and a periodic trigger signal is generated with the initial synchronization trigger signal used as a reference signal, and is then transmitted to the calibration transmitting channels and the calibration receiving channels.

8. A calibration system for parallel multi-channel wireless channel measurement comprising: a transmitting end including a combiner and a calibration receiving channel, the calibration receiving channel selectively connecting to an output end of the combiner; and a receiving end including a splitter and a calibration transmitting channel, the calibration transmitting channel selectively connecting to an input end of the splitter, in a condition that the transmitting end disconnect with the receiving end, the calibration receiving channel connects to the output end of the combiner at the transmitting end and the calibration transmitting channel connects to the input end of the splitter at the receiving end; in a condition that the transmitting end connects with the receiving end, the calibration receiving channel disconnects to the output end of the combiner at the transmitting end and the calibration transmitting channel disconnects to the input end of the splitter at the receiving end.

9. The calibration system according to claim 8, wherein: the transmitting end further includes a plurality of splitters, whose output ends connect with an input end of the combiner at the transmitting end; the receiving end further includes a plurality of combiners, whose input ends connect to an output end of the splitter at the receiving end.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The present disclosure will become more fully understood from the detailed description given herein below for illustration only, and therefore are not limitative of the present disclosure, and wherein:

(2) FIG. 1 is a schematic diagram of a calibration solution for a parallel multi-channel wireless channel in the prior art:

(3) FIG. 2 is a flowchart showing that a transmitting end performs channel calibration by using a neighboring channel relationship in the prior art;

(4) FIG. 3 is a flowchart showing that a receiving end performs channel calibration by using a neighboring channel relationship in the prior art;

(5) FIG. 4 is a block diagram showing that eight antenna ports transmit/receive data in parallel according to an embodiment of the present invention;

(6) FIG. 5 is a logic block diagram of a calibration method for parallel multi-channel wireless channel measurement according to the present invention;

(7) FIG. 6 is a flowchart of the calibration method for parallel multi-channel wireless channel measurement according to the present invention;

(8) FIG. 7(A) to FIG. 7(C) are schematic diagrams of a connection relationship between eight transmit paths to eight receive paths according to the present invention;

(9) FIG. 8 is a flowchart of calculating and calibrating a channel frequency response according to the present invention;

(10) FIG. 9 is a diagram of an operating time sequence between a measurement transmitting channel and a calibration transmitting channel according to the present invention; and

(11) FIG. 10 (A) to FIG. 10 (C) are respectively schematic diagrams of pairwise multi-path time delay power spectrum of 64 groups of receive/transmit antenna pairs in cases of using raw data, using calibration data, and using calibration data compensation.

DETAILED DESCRIPTION

(12) Technical content of the present invention is described in detail below with reference to accompanying drawings and specific embodiments.

(13) FIG. 4 is a block diagram showing that eight antenna ports transmit/receive data in parallel according to an embodiment of the present invention. Antennas TX1 through TX8 transmit data simultaneously, and antennas RX 1, RX 2 . . . RX 8 simultaneously receives the data transmitted by TX1 through TX8. In this way, it is not needed to repetitively switch, and a frequency response matrix of transmitting channels and receiving channels could be simultaneously acquired by transmitting once. Compared with a conventional calibration method, calibration data is simultaneously transmitted and received without switching in the present invention, so as to resolve a time jitter problem caused by frequently switching and thus improve measurement precision.

(14) On the other hand, measurement error is inevitable to some extent. Multi-channel concurrent testing signals interfere with each other. But, the channel interference in the measurement process cannot be obtained because the interference such as a phase jitter is time-varying for clock asynchronization and non-idealization of radio frequency device. Therefore, it is necessary to supervise the measurement error in real time.

(15) To implement a real-time supervision function, it is meaningful to resolve a time-varying response correlation between parallel multi-channels caused by the asynchronization between the radio frequency device and the clock, so as to obtain a channel response difference between N transmitting channels or N receiving channels in the measurement process. Therefore, a calibration receiving channel is added at a transmitting end and a calibration transmitting channel is added at the receiving end in the present invention. By using these two calibration channels to record channel responses of N transmitting channels and N receiving channels at the current moment, a channel response relationship between TX2 through TX8 and TX1 and a channel response relationship between RX 2 through RX 8 and RX 1 can be calculated. In data post processing, the measurement error caused by the multi-channels responses is eliminated. This is described in detail below with reference to a specific embodiment.

(16) FIG. 5 is a logic block diagram of a calibration method for wireless parallel multi-channel measurement according to the present invention. In the present calibration method, a traditional time division multiplexing switch is not used. Instead, a calibration receiving channel is added at the transmitting end and a calibration transmitting channel is added at the receiving end, for online supervising a channel response characteristic between multiple channels at a current moment in real time. Thus, it ensures that a measurement error caused by interference between multiple channels can be avoided in the channel measurement process.

(17) In the embodiment shown in FIG. 5, a plurality of 12 splitters and a calibration receiving channel are added at the transmitting end, and a plurality of 21 combiners and a calibration transmitting channel are added at the receiving end. A main function of the 12 splitter in the measurement process is to simultaneously split a transmitting signal into two signals and respectively transmit the two signals to an antenna port and the calibration receiving channel. A main function of the 21 combiner is to ensure during measurement that the receiving end can receive a measurement signal that is output by the antenna port and a calibration measurement signal that is output by the calibration transmitting channel. These two calibration channels implements real-time supervising channel response characteristics of the transmitting end and the receiving end.

(18) Considering that in an actual measurement process, the transmitting channel and the receiving channel of a measurement signal locate far away, a conventional solution in which the transmitting end and the receiving end are directly connected by using a radio frequency cable is not used. Instead, a method for adding calibration channels is used in the present invention to implement real-time supervision and calibration, so as to ensure that the calibration would not affected by a physical distance. Output signals of the transmitting channels (that is, the transmitting end) pass through a plurality of 12 splitters and then are fed into an N1 combiner. The output end (point a in FIG. 5) of the N1 combiner is connected to the input end (point b in FIG. 5) of a 1N splitter. Output ends of the 1N splitter are connected to input ends of a plurality of 21 combiners, and output ends of the 21 combiners are connected to the receiving channels (that is, the receiving end). The calibration receiving channel is led out from the output end of the N1 combiner (point a in FIG. 5), and the calibration transmitting channel is led out from the input end of the 1N splitter (point b in FIG. 5).

(19) FIG. 6 is a flowchart of the calibration method for wireless parallel multi-channel measurement according to the present invention. In the calibration process, the point a and the point b are directly connected by using a radio frequency cable, the calibration receiving channel and the calibration transmitting channel are disconnected, and a response characteristic of passive devices (for example, a connection cable, a splitter and a combiner) is tested, to obtain a calibration channel response matrix H(f).sub.88. In an air interface measurement process, the point a and the point b, without connecting each other, are respectively connected to the calibration receiving channel and the calibration transmit channel. Through the calibration receiving channel and the calibration transmitting channel as well as with a measurement channel, a channel response characteristic of the transmitting end and the receiving end is recorded to be calibrated. A specific description is as follows:

(20) First, a frequency response characteristic of the added passive device is measured by using a vector network analyzer (VNA). In the embodiment shown in FIG. 7 (A), a frequency response of the passive device from eight transmit paths to eight receive paths is marked as S(f):

(21) $S\left(f\right)=\left[\begin{array}{ccccc}{S}_{11}\left(f\right)& & & & S_{18}\left(f\right)\\ & S_{22}\left(f\right)& & & \\ & & S_{33}\left(f\right)& & \\ & & & & \\ S_{81}\left(f\right)& & & & S_{88}\left(f\right)\end{array}\right]$

(22) In the embodiment shown in FIG. 7 (B), a frequency response characteristic from eight transmit paths to the calibration receiving channel is marked as S(f):
S(f)=[S.sub.11(f)S.sub.21(f) . . . S.sub.81(f)]

(23) In the embodiment shown in FIG. 7 (C), a frequency response from the calibration transmitting channel to eight receive paths is marked as S(f):
S(f)=[S.sub.11(f)S.sub.12(f) . . . S.sub.18(f)]

(24) Secondly, at a calibration moment t.sub.0, the transmitting end and the receiving end are connected, and the calibration transmitting channel and the calibration receiving channel are closed. The transmitting end simultaneously transmits eight RF signals, and the receiving end processes to obtain 8 groups of signal sequences orthogonal to each other. Then, in subsequent processing, a frequency response characteristic of a measurement device itself is H.sup.t.sup.0(f), where H.sup.t.sup.0(f) is an 88 matrix:

(25) $H^{t_0}\left(f\right)=\left[\begin{array}{ccc}{H}_{11}^{t_0}\left(f\right)& \mathrm{.Math.}& H_{18}^{t_0}\left(f\right)\\ \mathrm{.Math.}& & \mathrm{.Math.}\\ H_{81}^{f_0}\left(f\right)& \mathrm{.Math.}& H_{88}^{t_0}\left(f\right)\end{array}\right]$

(26) H.sub.ij(f) denotes a channel response from the i.sup.th transmit port to the j.sup.th receive port, and S.sub.q(f) denotes a response characteristic of a combiner and a splitter through which the signal passes. Therefore, there is:

(27) $H_{\mathrm{ij}}^{t_0}\left(f\right)=H_{\mathrm{TXi}}^{t_0}\left(f\right).Math.S_{\mathrm{ij}}\left(f\right).Math.H_{\mathrm{RXj}}^{t_0}\left(f\right)$$i=1,\mathrm{.Math.},8;j=1,\mathrm{.Math.},8$

(28) Finally, at a measurement moment t.sub.1, the radio frequency cable between the transmitting end and the receiving end is disconnected, a signal from the transmitting end of the N channels passes through a splitter, and turns to 2 signals. One of the signals passes via an antenna port to the receiving end, and the other passes through an 81 combiner to the calibration receiving channel. In this way, at the same time of measuring the channel response of an air interface from the transmitting end to the receiving end, a time-varying response result of N measurement channels between the receiving end and the transmitting end may be tested by using the calibration receiving channel/calibration transmitting channel. Thus, the calibration receiving channel processes and acquires a time-varying channel response of the N transmitting channels. A calibration response of the i.sup.th transmitting channel is denoted by H.sub.i.sub.TX.sup.t.sup.1(f):
H.sub.i.sub.TX.sup.t.sup.1(f)=H.sub.TXi.sup.t.sup.1(f).Math.S.sub.i1(f).Math.H.sub.mtRX.sup.t.sup.1(f) i=1, . . . ,8

(29) The receiving end receives via the antennas radio frequency data sent by the transmitting end. In a silent period of the transmitting end, the receiving end receives a circular testing signal sent by the calibration transmitting channel through the 18 combiner. After the receiving end processes the signal sent by the calibration transmitting channel, a time-varying channel response of the N receiving channels may be calculated, and a calibration response of the j.sup.th receiving channel is denoted by H.sub.j.sub.RX.sup.t.sup.1(f):
H.sub.j.sub.RX.sup.t.sup.1(f)=H.sub.asTX.sup.t.sup.1(f).Math.S.sub.1j(f).Math.H.sub.RXj.sup.t.sup.1(f) j=1, . . . ,8

(30) It may be seen from the foregoing formula that, a calibration response may be a product of a transmit response, a receive response, and a frequency response of a passive device. Furthermore, a relationship between the transmit channels may be obtained by a ratio of the calibration responses of the second through the eighth transmitting channels to the calibration response of the first transmitting channel:

(31) $\frac{H_{i_{\mathrm{TX}}}^{t_1}}{H_{1_{\mathrm{TX}}}^{t_1}}\left(f\right)=\frac{H_{\mathrm{TXi}}^{t_1}\left(f\right).Math.S_{i1}^{}\left(f\right).Math.H_{\mathrm{calRX}}^{t_1}\left(f\right)}{H_{\mathrm{TX}1}^{t_1}\left(f\right).Math.S_{11}^{}\left(f\right).Math.H_{\mathrm{calRX}}^{t_1}\left(f\right)}=\frac{H_{\mathrm{TXi}}^{t_1}\left(f\right)S_{i1}^{}\left(f\right)}{H_{\mathrm{TX}1}^{t_1}\left(f\right)S_{11}^{}\left(f\right)}$$i=2,\mathrm{.Math.},8$

(32) Likewise, a relationship between receiving channels may be obtained by a ratio of the calibration response of the second through the eighth receiving channels to the calibration response of the first receiving channel:

(33) $\frac{H_{j_{\mathrm{RX}}}^{t_1}}{H_{1_{\mathrm{RX}}}^{t_1}}\left(f\right)=\frac{H_{\mathrm{calTX}}^{t_1}\left(f\right).Math.S_{1j}^{}\left(f\right).Math.H_{\mathrm{RXj}}^{t_1}\left(f\right)}{H_{\mathrm{calTX}}^{t_1}\left(f\right).Math.S_{11}^{}\left(f\right).Math.H_{\mathrm{RX}1}^{t_1}\left(f\right)}=\frac{H_{\mathrm{RXj}}^{t_1}\left(f\right)S_{1j}^{}\left(f\right)}{H_{\mathrm{RX}1}^{t_1}\left(f\right)S_{11}^{}\left(f\right)}$$j=2,\mathrm{.Math.},8$

(34) Based on a hypothesis that the channel response of the first receiving channel and the first transmitting channel is kept unchanged at the measurement time and calibration time, that is, H.sub.ll.sup.t.sup.0(f)=H.sub.ll.sup.t.sup.1(f),

(35) the calibration response can be calculated by the following formula:

(36) $H_{\mathrm{ij}}^{t_1}\left(f\right)=H_{11}^{t_0}\left(f\right).Math.\frac{H_{i_{\mathrm{TX}}}^{t_1}}{H_{1_{\mathrm{TX}}}^{t_1}}\left(f\right).Math.\frac{H_{j_{\mathrm{RX}}}^{t_1}}{H_{1_{\mathrm{RX}}}^{t_1}}\left(f\right)$

(37) In an embodiment of the present invention, a process of calculating and calibrating the channel frequency response is shown in FIG. 8. The transmitting end/receiving end transmits/receives concurrent multi-channel data in a code division multiplexing manner, and the transmitting ports/receiving ports are individually distinguished by using a group of orthogonal sequences orthogonal to each other. After receiving data, the calibration transmitting channel and the calibration receiving channel perform subsequent processing. In detail, when a response result of a channel is processed in a receiving process, depending on data amount and hardware performance, it is selective to simultaneously perform streaming (that is, consecutively transmitting data from a memory or transmitting data to a memory) and processing or to perform processing after the streaming. In the subsequent processing, the receiving end receives data transferred by a combiner and performs streaming simultaneously. Then, the receiving end selects the j.sup.th receive channel, generates a local sequence sent by the i.sup.th channel of the transmitting end, and performs a correlation operation on a receiving sequence of the j.sup.th channel and the local sequence of the i.sup.th channel to obtain a correlation peak value. Subsequently, the receiving end converts the correlation peak value to a frequency domain by using an FFT operation, to obtain a frequency domain channel response characteristic. Accordingly, the sequence of the i.sup.th transmitting channel is circularly processed, and the j.sup.th receiving channel is circularly processed, so as to obtain a channel response result of 64 channels, where both i and j are positive integers. After acquiring the peak value by using the correlation of the orthogonal sequences, the receiving end converts the peak value to the frequency domain by using FFT, thereby measuring the frequency response matrix of the channels.

(38) $H\left(f\right)=\left[\begin{array}{ccc}{H}_{11}\left(f\right)& \mathrm{.Math.}& H_{18}\left(f\right)\\ \mathrm{.Math.}& & \mathrm{.Math.}\\ H_{81}\left(f\right)& \mathrm{.Math.}& H_{88}\left(f\right)\end{array}\right]$

(39) To ensure that the receiving end can correctly receive a signal, a time point at which the calibration transmitting channel transmits data needs to be distinguished from a time point at which the measurement transmitting channel transmits data. In an embodiment of the present invention, when the measurement transmitting channel is in a silent period, the calibration transmitting channel transmits a narrow pulse signal for circularly testing. Moreover, transmission of the narrow pulse signal needs to satisfy a time delay requirement of the measurement transmitting channel and the measurement receiving channel, so as to ensure that aliasing does not occur between a PN (pseudo random noise) sequence received by the receiving end and the narrow pulse. When the transmitting end transmits a signal (for example, a PN sequence), the calibration transmitting channel is closed; after transmission of the signal is ended, the calibration transmitting channel begins to transmit the narrow pulse signal. An illustrative time sequence diagram is shown in FIG. 9.

(40) Additionally, it is necessary to realized strict synchronization between transmitting channels, between receiving channels, and between the receiving channels and the transmitting channels. At first, in the present invention, a GPS receiver (or Beidou terminal) is used at either of the receiving end and the transmitting end. The OPS receiver (or Beidou terminal) is used to provide a 1 pps (pulse per second) as an initial synchronization trigger signal source. Secondly, the initial synchronization trigger signal is used as a reference signal, and transmitted to an AWG (arbitrary waveform generator) at the receiving end or the transmitting end to generate a periodic trigger signal. The periodic trigger signal is transmitted by using a PXI (PCI extensions for instrumentation) bus, and transmitted to the transmitting or receiving channels, to trigger signal transmission of an FPGA. Additionally, clocks at the receiving end and the transmitting end need to be strictly synchronous in period and phase. Therefore, in the present invention, two rubidium clocks are separately used at the receiving end and the transmitting end. The two rubidium clocks are aligned for a period of time, phase jitters of the two rubidium clocks can be controlled at a level of p seconds. 10 Mhz reference signals from the clocks are transmitted to the receiving end and the transmitting end, and channel reference clocks at the receiving end and the transmitting end share these two 10 Mhz reference signals, thereby precisely synchronizing clocks of the receiving channels and the transmitting channels.

(41) FIG. 10 (A) to FIG. 10 (C) respectively show pairwise multi-path time delay power spectrum of 64 groups of receive/transmit antenna pairs in cases of using raw data, calibration data, and calibration data compensation. Impacts of a wireless measurement channel and an RF radio frequency response on a time delay power spectrum may be seen from an image of raw data. The calibration data also has a time delay power spectrum same as that of the raw data, and it may be seen from a calibrated pattern that all power spectrums are already aligned. It indicates that at a post-processing stage, the calibration data effectively compensates for a response difference of an RF channel, and a time delay of these 64 groups of data decreases from 3.0277 ns (before calibration) to 0.048 ns (after calibration).

(42) Compared with the prior art, in the calibration method for parallel multi-channel wireless channel measurement and the system for the same provided in the present invention, without a conventional time division multiplexing switch, multi-channel data is transmitted and received parallel and concurrently. Thereby, the present invention better satisfies a channel measurement requirement in a high-speed scenario while shortening a calibration time. Moreover, the calibration transmitting channel and the calibration receiving channel are respectively added at the receiving end and the transmitting end, so that the current channel response characteristic between multiple channels can be on-line supervised in real time, so as to ensure that a measurement error caused by an impact of mutual interference between the multiple channels can be avoided in the channel measurement process. In the measurement process, response results between channels at the transmitting end and the receiving end may be separately obtained, thereby eliminating a time-varying correlation between the radio frequency channels to more precisely measure the response result of the wireless channel.

(43) The calibration method for parallel multi-channel wireless channel measurement and the system for the same provided in the present invention are described in detail above. Any apparent modification made to the present invention by persons of ordinary skill in the art without departing from the essence of the present invention constitutes violation on patent rights of the present invention, and the persons should bear corresponding legal liabilities.