CDMA transmission method for multiple intelligent reflecting surfaces
阅读说明：本技术 一种针对多智能反射面的码分多址传输方法 (CDMA transmission method for multiple intelligent reflecting surfaces ) 是由 肖悦 马滕 雷霞 于 2021-07-19 设计创作，主要内容包括：本发明属于移动通信技术领域,涉及一种针对多智能反射面的码分多址传输方法。该方案利用智能反射面可变反射系数的特性,为每个反射面配置一个时变的反射系数序列作为扩频码,对多径信号进行扩频,并利用码分多址的原理对不同反射面反射的信号进行标记,从而接收机可根据不同的扩频码对多径信号进行识别,实现多径信号的分离与合并。本发明克服了传统的基于智能反射面的无线通信系统在有多径时延影响下信道容量急剧恶化的问题,并且具有较高的信道估计精度,具备更高的实用性。(The invention belongs to the technical field of mobile communication, and relates to a code division multiple access transmission method for multiple intelligent reflecting surfaces. The scheme utilizes the characteristic of variable reflection coefficient of the intelligent reflection surface, configures a time-varying reflection coefficient sequence for each reflection surface as a spread spectrum code, spreads the frequency of the multipath signals, and marks the signals reflected by different reflection surfaces by utilizing the principle of code division multiple access, so that a receiver can identify the multipath signals according to different spread spectrum codes, and the separation and combination of the multipath signals are realized. The invention overcomes the problem that the channel capacity of the traditional wireless communication system based on the intelligent reflecting surface is rapidly deteriorated under the influence of multipath time delay, has higher channel estimation precision and higher practicability.)
1. A CDMA transmission method for multiple intelligent reflecting surfaces includes transmitting end Tx and receiving end Rx with number of antennas 1, N intelligent reflecting surfaces IRS with number of units K, and features that the antennas are arranged in a same plane and the antennas are arranged in same planeAndrespectively representing Tx-RX, Tx-IRSkAnd IRSk-channel coefficients of Rx, characterized in that said transmission method comprises:
a sending end:
s1, transmitting the baseband signal as
Wherein s ismFor digitally modulating symbols, TsIs the symbol length; Π (t) is a rectangular pulse function defined as
S2, Tx transmits the transmission data frame by frame, and adds the length T before and after each framep＝PTsThe training symbols s of (a) are,whereinIs an integer set;
s3, each IRS sets the following time-varying reflection coefficient vector to identify itself:
in the formulakAdjusting according to the channel information to realize the reconstruction of the reflected wave beam;for spreading to implement CDMA, further denoted as
Wherein theta isk[n]＝θk(nTc) A discrete time series with a period G,Tc＝Tsthe/G is the chip length;
when theta isk[n]When the binary values of 0 and pi are taken,for constructing an m-sequence; in addition, the Zadoff-Chu (ZC) sequence and the DFT sequence are respectively constructed by the following formulas:
tx is synchronized with IRS, and the signal arrival time of Tx-RX link is t0The signal reflected by the l-th cell of the k-th IRS is delayed by tau with respect to the Tx-RX linkklTherefore, after IRS reflection, the signal r (t) reaching the receiving end at time t is
WhereinFor a single sideband power spectral density of N0White gaussian noise of (1);
receiving end:
s4, adopting correlation detection method to identify multipath signal, making maximum multipath time delay not exceed training symbol length TpAnd let s equal to 1 and the transmission start time be 0, so [ t [ [ t ]0,t0+Tp]Reduction of received signals in time intervals to
S41, taking the sample mean value of the received signal as H for the TX-RX chain0Is estimated, i.e.
S42, for the signal reflected by the kth IRS, letAnd detected by sliding correlation, i.e.
Wherein argmax returns the first N local maximum points;
s43, according toTo give H as shown belowklEstimation of (2):
s5, calculating the weighting coefficient of each branch, and carrying out time delay weighting combination on the sampling output, wherein the method specifically comprises the following steps:
s51, calculating the weighting coefficient of each branch, wherein in the maximum ratio combination, the weighting coefficient is proportional to the signal amplitude and is inversely proportional to the trunkPower of interference plus noise, order TX-RX and TX-IRSkl-the weighting factor of the RX link is α0And alphaklFor the scene that the multipath delay tends to 0, there are
In the formula Hk＝∑lHkl(ii) a Estimated value of channel coefficient obtained at S4Replacing the true value H to obtain a corresponding branch weighting coefficient;
in the same way, the relative multipath time delay among different IRSs exceeds one symbol length TsA weighting coefficient of
S52, obtaining the sampling output of each branch, for the m-th transmission symbol SmLet its start time be tmThen TX-RX and TX-IRSklThe sampled outputs of the RX chains are respectively
And
s53, combining the output results as
And the receiving end carries out maximum likelihood judgment on the combined and output signals and restores the original data.
Technical Field
The invention belongs to the technical field of mobile communication, and particularly relates to a code division multiple access transmission method for multiple intelligent reflecting surfaces.
Background
The IRS (Intelligent reflection Surface) has the characteristic of dynamic configuration of radio propagation environment, has a wide application prospect in the aspects of data transmission, network resource optimization, positioning and the like, and is a hot research object of a new generation of wireless communication system. Specifically, the IRS is composed of reflection units that can adjust the amplitude, phase, and even polarization of an incident electromagnetic wave in real time, and a large number of low-cost reflection units are integrated on a plane, thereby intelligently reconstructing a wireless propagation environment. Because of the two-dimensional characteristics of the material, the IRS can be flexibly mounted on the surfaces of buildings such as walls, floors, ceilings and the like, and is well compatible with the existing wireless communication system. Although the IRS has such excellent characteristics, when a plurality of IRS are deployed to assist data transmission, signals from different IRS often have different time delays, which brings a problem of intersymbol interference, thereby affecting communication quality; meanwhile, under the scenes of channel estimation, positioning and the like, information such as time delay of multipath signals and the like also needs to be distinguished and extracted. The current research work of the IRS mainly focuses on the communication performance of a single IRS and the optimization of IRS coefficients, and research aiming at the problems is lacked.
The traditional multipath separation scheme usually adopts spread spectrum-based code division multiple access technology, utilizes the autocorrelation characteristic of spread spectrum code words, separates out multipath through sliding correlation, and further combines multipath signals through delay weighted addition. However, the above conventional scheme cannot be directly applied to a communication system with multiple IRS, and if the transmitter transmits a spread spectrum signal, the receiver can separate out multipath, but cannot match the multipath signal with multiple IRS, which limits its application in the scenarios of channel estimation and positioning.
Disclosure of Invention
The variable time-varying reflection coefficient of the IRS offers the possibility to solve the above-mentioned problems: by using the idea of code division multiple access, different time-varying reflection coefficient sequences are configured for each IRS as spreading codes, so that a user can perform sliding correlation operation by using a received signal and the spreading codes of one IRS, i.e. different IRSs can be distinguished while a multipath signal is separated. The present invention takes advantage of the ability of an IRS to vary the reflection coefficient to separate and combine signals from different IRS.
For better illustration of the present invention, terms and system structures used in the technical solution of the present invention will be described.
BS: base Station, Base Station.
CDMA: code Division Multiple Access.
CRLB: cram er-Rao Lower Bound, Cramer-Lower Bound.
DFT: discrete Fourier Transform, Discrete Fourier Transform.
IRS: the Intelligent Reflecting Surface can dynamically change the electromagnetic property of the Intelligent Reflecting Surface according to different bias voltages, so that the amplitude, the phase and even the polarization mode of a wireless signal are optimized.
MIRS-CDMA: multiple Intelligent Reflecting surface Assisted code division Multiple access.
MSE: mean Squared Error, Mean Squared Error.
SNR: Signal-to-Noise Ratio.
UE: User-End, User side.
FIG. 1 is a schematic diagram of the MIRS-CDMA system of the present invention:
in the MIRS-CDMA system, it is assumed that the number of base stations and user antennas is 1, the number of IRS units and the number of IRS units are N and K, respectively, and each IRS has a time-varying reflection coefficient vector as follows to identify itself:
in the formulakThe reconstruction of the reflected wave beam can be realized by adjusting according to the channel information;for spreading to implement code division multiple access, which may be further denoted as
Wherein theta isk[n]＝θk(nTc)，Is a discrete time sequence of period G, TcFor chip length, pi (t) is a rectangular pulse function defined as
When theta isk[n]When the binary values of 0 and pi are taken,m sequences can be constructed; in addition, the Zadoff-Chu (ZC) sequence and the DFT sequence can be constructed by the following formulas, respectively:
consider a base station transmitting a signal of
Wherein s ismFor digitally modulating symbols, Ts＝GTcThe symbol length indicates that the spectrum of the transmitted signal is expanded to G times after the transmission signal is reflected by the IRS, and therefore G is also called the spreading gain.
The invention assumes independent flat Rayleigh fading channel and uses H0、Andrespectively represent BS-UE, BS-IRSkAnd IRSk-a channel of the UE, whereinhkAnd gkEach element in (1) is independently and identically distributedAndwithout loss of generality, assume that the BS is synchronized with the IRS and that the signal arrival time of the BS-UE link is t0The time delay of the signal reflected by the l-th element of the kth IRS (with respect to the BS-UE link) is τkl. The received signal of the user at time t can be represented as
WhereinFor a single sideband power spectral density of N0White gaussian noise.
The technical scheme adopted by the invention is as follows:
a CDMA transmission method for multiple intelligent reflecting surfaces includes transmitting end Tx and receiving end Rx with number of antennas 1, N intelligent reflecting surfaces IRS with number of units K, and H0、Andrespectively representing Tx-RX, Tx-IRSkAnd IRSk-channel coefficients of Rx, the transmission method comprising:
a sending end:
s1, transmitting the baseband signal as
Wherein s ismFor digitally modulating symbols, TsIs the symbol length; Π (t) is a rectangular pulse function defined as
S2, Tx transmits the transmission data frame by frame, and adds the length of the front and back of each frameFor the receiving end to perform channel estimation, whereinIs an integer set;
the intelligent reflecting surface:
s3, each IRS sets the following time-varying reflection coefficient vector to identify itself:
in the formulakAdjusting according to the channel information to realize the reconstruction of the reflected wave beam;for spreading to implement CDMA, further denoted as
Wherein theta isk[n]＝θk(nTc) A discrete time series with a period G,Tc＝Tsthe/G is the chip length;
when theta isk[n]When the binary values of 0 and pi are taken,can be used for constructing m sequence; in addition, the Zadoff-Chu (ZC) sequence and the DFT sequence are respectively constructed by the following formulas:
tx is synchronized with IRS, and the signal arrival time of Tx-RX link is t0The signal reflected by the l-th cell of the k-th IRS is delayed by tau with respect to the Tx-RX linkklTherefore, after IRS reflection, the signal r (t) arriving at the receiving end at time t is
WhereinFor a single sideband power spectral density of N0White gaussian noise of (1);
receiving end:
s4, adopting correlation detection method to identify multipath signal, making maximum multipath time delay not exceed training symbol length TpAnd let s equal to 1 and the transmission start time be 0, so [ t [ [ t ]0,t0+Tp]Reduction of received signals in time intervals to
S41, taking the sample mean value of the received signal as H for the TX-RX chain0Is estimated, i.e.
S42, for the signal reflected by the kth IRS, letAnd detected by sliding correlation, i.e.
Wherein argmax returns the first N local maximum points;
s43, according toTo give H as shown belowklEstimation of (2):
s5, calculating the weighting coefficient of each branch, and carrying out time delay weighting combination on the sampling output, wherein the method specifically comprises the following steps:
s51, calculating weighting coefficients of each branch, wherein in maximum ratio combination, the weighting coefficients are in direct proportion to the signal amplitude and in inverse proportion to the power of interference plus noise, and enabling TX-RX and TX-IRSkl-the weighting factor of the RX link is α0And alphaklFor the scene that the multipath delay tends to 0, there are
In the formula Hk＝∑lHkl(ii) a Estimated value of channel coefficient obtained at S4Replacing the true value H to obtain a corresponding branch weighting coefficient;
similarly, the relative multipath delay between different IRSs exceeds one symbol lengthDegree TsA weighting coefficient of
S52, obtaining the sampling output of each branch, for the m-th transmission symbol SmLet its start time be tmThen TX-RX and TX-IRSklThe sampled outputs of the RX chains are respectively
And
s53, combining the output results as
And the receiving end carries out maximum likelihood judgment on the signals after the combined output, and the original data can be recovered.
The invention has the beneficial effects that:
the invention provides a code division multiple access transmission scheme aiming at multiple intelligent reflecting surfaces. The scheme utilizes the characteristic of variable reflection coefficient of the intelligent reflection surface, configures a time-varying reflection coefficient sequence for each reflection surface as a spread spectrum code, spreads the frequency of the multipath signals, and marks the signals reflected by different reflection surfaces by utilizing the principle of code division multiple access, so that a receiver can identify the multipath signals according to different spread spectrum codes, and the separation and combination of the multipath signals are realized. The invention overcomes the problem that the channel capacity of the traditional wireless communication system based on the intelligent reflecting surface is rapidly deteriorated under the influence of multipath time delay, has higher channel estimation precision and higher practicability.
Drawings
FIG. 1: the invention provides a schematic diagram of an intelligent reflector CDMA system;
FIG. 2: and the channel estimation precision when the multipath time delay approaches zero.
FIG. 3: and the channel estimation precision when the multipath time delay is not zero.
FIG. 4: the ergodic channel capacity when the multipath time delay tends to zero.
FIG. 5: the multipath delay is greater than the traversal channel capacity for a symbol length.
Detailed Description
Having described the invention in detail in the summary of the invention, the following description, taken in conjunction with the accompanying drawings and simulation examples, illustrates the utility of the invention.
In the simulation examples of fig. 2-5, unless otherwise specified, a sequence with spreading gain of $ G ═ 839$ is used, and the channel attenuation is configured to beSNR is defined as Tc/N0。
Fig. 2 shows the performance curve of MSE versus SNR for the proposed scheme when the multipath delay approaches zero, where 2(a) is compared with the performance of the system using DFT sequences, and the simulation parameters are configured as N-64 and K-4; 2(b) compares the following system performance for K2 and 4, N256; the symbol epsilon in the figure indicates the corresponding CRLB. As can be seen from fig. 2(a), the system employing the ZC sequence has a plateau effect, whereas the performance of the DFT sequence is close to the CRLB. This is because ZC sequences do not satisfy zero mean and zero cross correlation properties, so there is multipath interference; although the DFT sequence satisfies the above properties, its autocorrelation function is not a dirac delta function, and thus cannot be applied to a scenario where multipath delay exists. Meanwhile, the results of fig. 2(a) show that the MSE performance of the system is close to CRLB at low SNR and close to asymptote at high SNR. Further, as can be seen from fig. 2(a) and 2(b), when an imperfect spread spectrum signal is used, increasing the number of reflecting surface elements or the number of reflecting surface blocks increases the estimation error.
Fig. 3 shows that in the presence of multipath delay scenario, the proposed channel estimation method is applied under N-64 and K-2 conditions,theoretical performance bound after normalization, where FIG. 3(b) willThe reduction is-40 dB. Meanwhile, since the DFT sequence is no longer applicable, fig. 3 only shows the simulation of the ZC sequence. It can be seen from the figure that the estimation method provided by the present invention has good estimation accuracy (slightly more than or less than 10%) under high signal-to-noise ratio. In addition, as the average power of the link decreases, the channel estimation accuracy also decreases accordingly.
Fig. 4 shows the ergodic channel capacity curve for a multipath delay of 0, and compares it with the performance of a system without CDMA, where the parameters are set to N-64 and K-4, the spreading gain is set to 839 in fig. 4(a), and to 127 in fig. 4 (b). The results of fig. 4 show that the channel capacity of the scheme proposed by the present invention is significantly improved, and the system performance using ZC sequence as the spreading code is very close to that of an ideal spreading code.
Fig. 5 shows the ergodic channel capacity curve when the multipath delay exceeds one symbol length, and the parameter configuration is in accordance with fig. 4 and compared with the case where K is 2. The results of fig. 5 show that the ergodic capacity increases as the SNR increases, but the growth rate gradually decreases due to multipath residual interference caused by imperfect mean and correlation properties of ZC sequences. Meanwhile, if CDMA is not used, the multipath interference will seriously affect the transmission performance of the system, so that the capacity is close to 0.
Therefore, the intelligent reflector CDMA scheme provided by the invention has higher channel estimation precision, overcomes the problem that the channel capacity of the traditional wireless communication system based on IRS is rapidly deteriorated under the influence of multipath time delay, and has higher practicability.