CN110912621B

CN110912621B - Communication system simulation method and device

Info

Publication number: CN110912621B
Application number: CN201911055716.0A
Authority: CN
Inventors: 牛凯; 董超; 李元杰
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2019-10-31
Filing date: 2019-10-31
Publication date: 2020-10-30
Anticipated expiration: 2039-10-31
Also published as: CN110912621A

Abstract

The embodiment of the invention provides a communication system simulation method and a device, wherein the method comprises the following steps: the simulation sending end obtains a baseband sampling rate and sends a digital baseband signal; acquiring a channel coefficient by the simulation channel under the condition of a baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from the frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, and converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal; and performing down-sampling on the convolution signal received by the simulation receiving end to obtain a digital baseband signal. By intercepting the baseband low-pass channel response from the frequency domain response as the equivalent frequency domain response and using the equivalent frequency domain response in the time domain, the data volume participating in convolution in the time domain is reduced, the redundant information of the whole communication system simulation process is reduced, and the complexity of the whole communication system simulation process is reduced.

Description

Communication system simulation method and device

Technical Field

The present invention relates to the field of wireless communication technologies, and in particular, to a communication system simulation method and apparatus.

Background

A channel is a generic term of a media medium between a transmitting end and a receiving end in a communication system, and determines the rate of information transmission and the quality of communication, and thus, the channel characteristics directly affect the performance of the communication system. However, since the communication system itself has the characteristics of high cost, complex structure, and difficult modification, the communication system needs to be simulated by simulation software before designing the communication system, that is, the communication system simulation, and the channel simulation becomes one of the important links for realizing the communication system simulation.

In order to describe the channel as accurately as possible, in the related art, when the simulation of the communication system is realized through simulation software, the signal is sampled in an oversampling manner, wherein the simulation software can be used for simulating a transmitting end, a simulation transmitting filter, a simulation channel, a simulation receiving filter and a simulation receiving end. The simulation method of the general communication system mainly comprises the following steps:

firstly, a simulation sending end obtains a channel coefficient from a Tapped Delay Line (TDL) model, generates a digital baseband signal by using a random number and sends the digital baseband signal; and performing zero-filling interpolation on the frequency spectrum of the digital baseband signal by adopting an oversampling mode, such as performing zero-filling interpolation on the frequency spectrum of the digital baseband signal, and increasing the baseband sampling rate of the digital baseband signal to be the same as the sampling rate of channel time domain response in the simulation channel to obtain an oversampling signal. Generally, the sampling rate of the channel time domain response is generally higher than that of the digital baseband signal, and the baseband sampling rate of the digital baseband signal can be the same as that of the channel time domain response in the simulation channel by using oversampling;

the TDL model describes the channel impulse response of multipath in an ideal state, each tap corresponds to a path, and the TDL model can obtain a channel coefficient.

Secondly, filtering a high-frequency signal in the over-sampled signal by the simulation sending filter to obtain a first filtered signal, and sending the first filtered signal to a simulation channel;

thirdly, in the simulation channel, convolving the first filtered signal with the channel time domain response to obtain and output a convolution signal;

fourthly, filtering the high-frequency signal in the convolution signal by the simulation receiving filter to obtain a second filtered signal, and sending the second filtered signal to the simulation receiving end;

and fifthly, the simulation receiving end down-samples the second filtered signal, namely extracts partial signals to obtain a digital baseband signal, wherein the 'first' in the first filtered signal and the 'second' in the second filtered signal are used for distinguishing the two filtered signals and are not limited in sequence.

In the transmission process of the digital baseband signal in the simulation of the communication system, the actual transmitting end simulated by the simulation transmitting end is limited by the factory sampling rate, so that the sampling rate of the digital baseband signal is lower. In order to match the sampling rate of the channel time domain response of the artificial channel, the convolution process in the artificial channel is implemented, and therefore, the digital baseband signal needs to be subjected to oversampling processing. And the digital baseband signal finally obtained at the simulation receiving end is the required signal obtained by down-sampling, so that other signals in the second filtered signal all belong to redundant information. Since the redundant information participates in the whole communication system simulation process, as the oversampling order increases, not only the redundant information of the whole communication system simulation process also increases, but also the complexity of the whole communication system simulation process also increases exponentially.

Disclosure of Invention

The embodiment of the invention aims to provide a communication system simulation method and a communication system simulation device, which are used for solving the technical problems that in the prior art, as redundant information participates in the whole communication system simulation process, the redundant information of the whole communication system simulation process is increased along with the increase of an oversampling order, and the complexity of the whole communication system simulation process is also increased in an exponential order. The specific technical scheme is as follows:

in a first aspect, the present invention provides a communication system simulation method, including:

the simulation sending end obtains a baseband sampling rate and sends a digital baseband signal;

acquiring a channel coefficient by the simulation channel under the condition of a baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from the frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; obtaining a convolution signal by the convolution baseband time domain response and the digital baseband signal;

and the simulation receiving end receives the convolution signal and takes the convolution signal after down sampling as a digital baseband signal.

In a second aspect, the present invention provides another communication system simulation method, including:

acquiring a digital baseband signal, wherein the digital baseband signal carries a baseband sampling rate;

acquiring a channel coefficient under the condition of the baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from a frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal to obtain a convolved signal;

and performing down-sampling on the convolution signal, and taking the down-sampled convolution signal as a digital baseband signal.

In a third aspect, the present invention provides a communication system simulation apparatus, including:

the simulation sending end is used for acquiring a baseband sampling rate and sending a digital baseband signal;

the simulation channel is used for acquiring a channel coefficient under the condition of a baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from the frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal;

and the simulation receiving end is used for receiving the convolution signal as a digital baseband signal.

In a fourth aspect, the present invention provides another communication system simulation apparatus, including:

the signal acquisition module is used for acquiring a digital baseband signal, and the digital baseband signal carries a baseband sampling rate;

the signal convolution module is used for acquiring a channel coefficient under the condition of the baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from a frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal to obtain a convolved signal;

and the down-sampling module is used for down-sampling the convolution signal and taking the down-sampled convolution signal as a digital baseband signal.

The embodiment of the invention provides a simulation method and a simulation device of a communication system, which intercept a baseband low-pass channel response from a frequency domain response as an equivalent frequency domain response and use the equivalent frequency domain response in a time domain, so that in the time domain, the data volume participating in convolution is reduced, the redundant information of the whole simulation process of the communication system is reduced, and the complexity of the simulation process of the whole communication system is reduced. Of course, not all of the advantages described above need to be achieved at the same time in the practice of any one product or method of the invention.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.

Fig. 1 is a schematic flowchart of a communication system simulation method according to an embodiment of the present invention;

fig. 2 is a schematic flow chart of another communication system simulation method according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of a communication system simulation apparatus according to an embodiment of the present invention;

fig. 4 is a schematic structural diagram of another communication system simulation apparatus according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The frequency domain and the time domain of the embodiment of the present invention are first described below.

The embodiment of the invention analyzes the conversion of the oversampling method from a time domain to a frequency domain. The transmission sequence is N long, the oversampling multiplying factor is q, sampling is not aliasing, and the receiving and transmitting filter is an ideal sinc waveform.

And performing oversampling, such as performing zero-filling interpolation on a frequency spectrum of the transmission signal, increasing the baseband sampling rate of the transmission signal to be the same as the sampling rate of channel time-domain response in the simulation channel, and performing convolution on the transmission signal and the channel time-domain response with the same sampling rate to obtain an oversampled signal.

Meanwhile, the transmission process of the over-sampling at the transmitting end is equivalent to zero-filling interpolation of the frequency spectrum of the transmission signal, and the discrete frequency response of the transmission signal is expanded to qN point. Meanwhile, a Discrete Fourier Transform (DFT) of qN points is performed on the channel time domain response of the integral sampling to obtain a Discrete frequency response of the transmission signal. The multiplication of the transmitted signal and the channel time domain response in the frequency domain is equivalent to the time domain convolution of the oversampling method.

The down-sampling process of receiving filtering at the receiving end: and windowing and truncating the frequency response output by the channel to N points, and correspondingly performing Inverse Discrete Fourier Transform (IDFT) conversion on the frequency response output by the channel, namely obtaining a received signal by an oversampling method. The above process is the frequency domain equivalent of the oversampling approach.

Based on the above description of the oversampling method, the inventors found that due to the frequency domain truncation, i.e., the existence of the ideal low pass, (q-1) the information of N points is redundant, this redundancy ratio becomes more significant as q is larger, therefore, aiming at the problem that the redundant information participates in the whole simulation process of the communication system, as the oversampling order is increased, not only the redundant information of the whole simulation process of the communication system is increased, in addition, the complexity of the whole simulation process of the communication system is also the problem of increasing index level, the embodiment of the invention provides a simulation method and a simulation device of the communication system, by intercepting the baseband low-pass channel response from the frequency domain response as the equivalent frequency domain response, and the equivalent frequency domain response is used in the time domain, so that the data volume participating in convolution is reduced in the time domain, redundant information of the whole communication system simulation process is reduced, and the complexity of the whole communication system simulation process is reduced.

The following is a continued description of a communication system simulation method provided in the embodiment of the present invention.

As shown in fig. 1, a communication system simulation method provided in an embodiment of the present invention may include the following steps:

step 110, the simulation transmitting end obtains a baseband sampling rate and transmits a digital baseband signal.

The digital baseband signal is a signal sent by the simulation sending end. The baseband sampling rate may be, but is not limited to, set according to user requirements. The baseband sampling rate adopts the sampling rate input by the sending end and the receiving end. The baseband sampling rate may be an oversampling multiple.

Step 120, acquiring a channel coefficient by the simulation channel under the condition of a baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from the frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time-domain response with the digital baseband signal.

The above-mentioned emulation channels may include, but are not limited to: a single Input single Output system-Tapped Delay Line (SISO-TDL) channel, and a Multiple Input Multiple Output Cluster Delay Line (MIMO-CDL) channel.

The channel coefficients represent the negative gain, time delay, Angle of Departure (AOD), Angle of Arrival (AOA) of the path. The channel coefficients may include, but are not limited to: channel tap coefficients. However, the general tapped delay line model does not directly provide the channel coefficient, and therefore the channel coefficient may be determined by generating the channel coefficient through the tapped average power, power distribution, doppler shift, and angle spread in the tapped delay line model.

The above tap delay model may include, but is not limited to:

a 5G New Radio Tapped-Delay Line (5G NR TDL) model, a 5G New Radio Cluster Delay Line (5G NR CDL) model, and a 4G Long Term Evolution (4G Long Term Evolution, 4G LTE) channel model, where the 4G LTE channel model includes three common models: an Extended Pedestrian channel model (EPA for short), an Extended vehicle channel model (EVA for short), and an Extended typical urban channel model (ETU for short).

The predetermined truncation window is used to obtain an equivalent frequency domain response, where h (k) represents the equivalent frequency domain response. The preset truncation window in the embodiment of the invention can be set according to the requirements of a user. Wherein the content of the first and second substances,

where k represents the subcarrier number of the frequency domain, L is the total number of tap coefficients, n is the number of elements, a_nFor the nth element of the vector a,

d_nis the nth element of the vector d, k is the frequency domain subcarrier number, j is the imaginary unit, F is the transform coefficient matrix, a is the tap coefficient vector, a ═ a₀，a₁，…，a_L-1]D is a decimal time delay vector, and d is T/T_s＝[d₀，d₁，…，d_L-1]＝[p₀/q，p₁/q，…，p_L-1/q]Q is oversampling factor, p_nIs the corresponding integer sample position after oversampling. To facilitate the definition of parameters in the formula, the bold parameters represent a matrix or vector.

In step 120, the equivalent frequency domain response may be converted into the time domain by using the following formula to obtain the baseband time domain response:

wherein h (n) is the baseband time domain response, K is the subcarrier number, K is the total number of the frequency domain subcarriers, and h (K) is the frequency domain correspondence.

In this step 120, any possible manner may be adopted, based on the channel coefficient, to intercept the baseband low-pass channel response from the frequency domain response by presetting the truncation window length, as an equivalent frequency domain response:

in one possible implementation, in a first step, the fractional delay of a channel tap coefficient in each channel coefficient is determined according to the baseband sampling rate.

The channel tap coefficients may include, but are not limited to: tap time delay and tap average power.

And secondly, determining a linear transformation matrix by using the preset truncation window length and the decimal time delay, wherein the preset truncation window length is the rounding result of the maximum decimal time delay in the decimal time delay.

The linear transformation matrix is obtained by cutting from a standard Discrete Fourier Transform (DFT) matrix, the row dimension of the linear transformation matrix is truncated, the column dimension of the linear transformation matrix is decimal, and the standard DFT matrix is a transformation matrix with the time dimension being an integer.

And thirdly, converting the original baseband signal response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain the equivalent frequency domain response.

Specific applications of the above possible implementations are as follows:

for example, one: when the simulation channel is a SISO-TDL channel model, the SISO-TDL channel model gives a tap time delay t ═ t in a channel tap coefficient₀，t₁，…，t_L-1],t₀Tap time delay, t, with serial number 0₁Tap time delay with sequence number 1, t_L-1Is a tap time delay with serial number L-1, L is the total number of taps, and the average power of the taps is P ═ P₀，P₁，…，P_L-1]，P₀Tap average power, P, of sequence number 0₁Tap average power, P, of number 1_L-1Is the tap average power with the serial number of L-1, and the baseband sampling period is T_s，T_sWhere S is to be distinguished from T in the two-dimensional channel tap coefficient matrix A, and the maximum Doppler shift is f_maxFor the maximum, the statistical property of the channel is generalized stationary uncorrelated Scattering (WSSUS).

Step 1, acquiring a channel coefficient by the simulation channel under the condition of the baseband sampling rate, and further including:

11, according to the sine sum method, the following formula is adopted

Wherein the content of the first and second substances,

calculating a channel tap coefficient matrix;

a_l(t) is the channel response on the first path at time t, l is the channel multipath number, t is time, P_lIs the power of the first diameter,

is a_l(t), R represents the in-phase component, j is in imaginary units,

is a_l(t), I represents the orthogonal component,

is a_lIn-phase or quadrature component of (t), N_lThe number of sinusoids used for the first path to fit to generate the channel response, N_lIs a preset constant, R/I represents R or I, when R or I component of the parameter is calculated specifically, the superscript of the parameter is embodied as R or I,

is the frequency of the doppler frequency and is,

in order to be a random initial phase,

is a [0,2 π ]]Is uniformly distributed with respect to the random number,

is the phase of the frequency function of the nth sine wave component on the ith path, n is the serial number of the superposed sine wave,

is the phase of the frequency function of the 0 th sine wave component in the l-th path, and l,0 is the nth sine wave component in the l-th path.

Average power per path is given by

Term determination, in-phase component

And the orthogonal component

Respectively by corresponding N_lAn initial phase of

At a frequency of

And the sinusoidal signals are superposed. The above implementation is an implementation of a rayleigh channel, and for a rice channel with a line of sight (LOS) path, the channel tap coefficient matrix needs to be modified according to a rice factor K as follows:

wherein, a'₀(t) is the 0 th path channel response after power adjustment, K is the Rice K factor (Rician K-factor), f_LOS,0Doppler frequency, θ, of LOS path_LoS,0Is the initial phase, θ, of the LOS path_LOS,0Is a [0,2 π ]]Uniformly distributed random numbers.

Due to time-varying nature of the channel, in_nSymbol time t ═ nT_sWhere the channel coefficients are a vector of functions with respect to time a [ nT_s]＝[a₀(nT_s)，a₁(nT_s)，…，a_l(nT_s)]Each element of this vector is a function a with respect to time_l(t) at a sampling instant nT_sThe values of (b) are, for example: a is_l[nT_s]As a function a at time n_l(t) at a sampling instant nT_sValue of (c), each set of channel coefficients corresponds to a transmit symbol x_n. Also, since the sampling is uniform, a can be adjusted_l[nT_s]Abbreviated as a_l[n]. Combining each set of channel coefficient function vectors a [ n ]]And forming a two-dimensional channel tap coefficient matrix: a ═ a [0 [ ]]，a[1]，···，a[N-1]]^TT is the transpose, and the dimension of the two-dimensional channel tap coefficient matrix is N × L. The vector a is a row vector or a column vector in the matrix A, wherein N is the sampling frequency and the number of transmitted signal samples; l is the total number of tap coefficients, which is also the number of multipaths. Then, the decimal time delay of the channel tap coefficient is added, so that each line of A can be independently transformed, and an equivalent frequency domain response matrix is obtained.

Step 2, according to the baseband sampling rate, determining the decimal time delay of the channel tap coefficient in each channel coefficient, and determining a linear transformation matrix by using the preset truncation window length and the decimal time delay, further comprising:

step 21, acquiring decimal time delay d as T/T_s＝[d₀，d₁，…，d_L-1],t＝[t₀，t₁，…，t_L-1]I.e. d_n＝P_n/q，P_nThe tap average power with the sequence number n and q is the oversampling multiplying factor. The oversampling ratio q is sometimes too large, for example, the oversampling ratio q is larger than the representation range of the 32-bit binary integer, which makes the following steps difficult to calculate. As can be seen from the characteristics of the sinc sampling function, the value at a position close to the main value is closer to the main value, so the tap delay t can be operated by the following formula, and a reasonable oversampling multiple q is selected:

where < · > represents rounding, and the processed fractional delay d is an approximation to the tap delay t below the oversampling factor q.

Step 22, determining the preset truncation window length W, which may further include: rounding-up the maximum fractional delay of the fractional delays, i.e.

Wherein the content of the first and second substances,

the result of rounding up is shown to be,

is the rounding-up result of the maximum decimal time delay.

Step 23, obtaining a preset cut-off window

Wherein for each a n]All have a n with each]A corresponding set of fractional time delays d, each a [ n ]]The oversampling magnification is q.

Step 24, calculating a linear transformation matrix according to the preset truncation window w, namely

F_k,l＝{exp(-j2πw_kd_l)}

Wherein, F_k,lIs an element of the k-th row/column of the matrix F, w_kK element of w, d_lThe i-th element of d, i is the serial number of the element.

Step 3, converting the original baseband signal response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficients to obtain an equivalent frequency domain response, which further includes:

step 31, the original baseband signal response in the time domain at the nth symbol time is obtained by the following linear transformation to obtain the equivalent frequency domain response

Wherein the content of the first and second substances,

for equivalent frequency domain response, F is the standard fourier transform matrix.

Step 4, converting the equivalent frequency domain response into the time domain to obtain a baseband time domain response, which further includes:

by adopting a formula, the method adopts the following steps,

wherein h [ n ] is the baseband time domain response, n represents the serial number of the baseband time domain response, K is the serial number of the subcarrier, and K is the total number of the frequency domain subcarriers.

To pair

The inverse discrete Fourier transform is performed one by one and is recorded as

Obtaining a baseband time domain response, namely:

and circularly executing the step 3 until the baseband time domain responses of all the symbols are obtained, wherein the baseband time domain responses are equivalent to the equivalent frequency domain responses and can also be called as equivalent time domain channel responses. The baseband time domain responses of all the symbols are combined to obtain an equivalent two-dimensional channel coefficient matrix, namely

H＝[h[0]，h[1]，···，h[N-1]]^TH differs from a in that the sampling rate in the two multipath delay dimensions is different, the former being the low-pass equivalent of the latter, and N representing the total number of baseband time-domain responses.

Example two: when the simulation channel is a MIMO-CDL channel model, the MIMO-CDL channel model is different from the SISO-TDL channel model in that a configuration of an array antenna is added, the dimension of a channel response is increased, and a direction angle parameter and an angle spreading characteristic are added, so that generation modes of tap coefficients are also different. However, the steps of obtaining the equivalent time domain channel response by the MIMO-CDL channel model are the same as the steps of obtaining the equivalent time domain channel response by the SISO-TDL channel model.

SISO-TDL channel refers to a channel constructed by a single-transmitting single-receiving tap delay line model

Wherein, the MIMO-CDL channel model gives a base band sampling period T_sThe cluster time delay t ═ t₀，t₁，…，t_L-1],t_iThe representation i represents the time delay of the ith path, and the cluster average power is P ═ P₀，P₁，…，P_L-1]，P_iRepresents the power of the ith cluster, i is from 0 to L-1, and the departure angle of the azimuth angle corresponding to each cluster is

The AOD of the ith path is expressed, and the arrival angle of the azimuth angle is phi_AOAAzimuthal Angle Of Departure (AOD), azimuthal angle Of Departure (AOA), AOD being distinguished from AOA by Φ_AODAnd phi_AOAThe angle at the zenith departure is theta_ZODThe zenith angle is theta_ZOAZOD used to differentiate Θ from ZOA_ZODAnd Θ_ZOAZOD denotes the apex angle of departure, ZOA denotes the apex angle of arrival, and the root mean square angle spread characteristic is c_ASD，c_ASA，c_ZSD，c_ZSAWherein c is_ASDFor angular root mean square angular spread characteristics away from azimuth, c_ASAFor the root mean square angular spread characteristic of the angle to the zenith angle, c_ZSDFor angular root mean square angular spread characteristics away from zenith angles, c_ZSAThe root mean square angular spread characteristic of the angle to the zenith angle. The relative moving speed of the antenna is v, and the relative moving speed of the antenna is a three-dimensional vector. The spatial gain characteristic of the receiving array antenna and the spatial gain characteristic of the transmitting array antenna are respectively G_rx,u(ψ)，G_tx,s(psi), u is the index of the receiving antenna, s is the index of the transmitting antenna, rx, u is the u-th receiving antenna, rx, s is the s-th transmitting antenna, rx is the subscript to indicate the receiving, tx, s is the s-th transmitting antenna, and tx is the subscript to indicate the transmitting.

and 11, acquiring a channel tap coefficient matrix A of the MIMO-CDL channel model as a 4-dimensional channel coefficient matrix, wherein the first dimension is a sending symbol dimension, the second dimension is a cluster dimension of a channel, the third dimension is a sending antenna dimension, and the fourth dimension is a receiving antenna dimension. And, each cluster is formed by the superposition of M sub-clusters. The "first" of the first channel tap coefficients and the "second" of the second channel tap coefficients are for distinguishing the two channel tap coefficients, and are not limited to the order. In the MIMO-CDL channel model, paths are represented in clusters, and one path is a cluster. The following steps 12, 13 and 14 can be adopted to obtain the channel tap coefficient matrix a of the MIMO-CDL channel model.

Step 12, under the non-line-of-sight condition, for each cluster of channels, determining a first channel tap coefficient of the channel of the cluster from the nth transmission symbol, the ith cluster, the s-th transmission antenna to the u-th receiving antenna:

wherein l is a cluster, M is any sub-cluster, u is the serial number of any receiving antenna, s is the serial number of any transmitting antenna, M is the total cluster number of the sub-clusters, and Non-Line of Sight (NLOS) is also a corner mark, which indicates that n is the serial number of any transmitting symbol under the condition of NLOS,

first channel tap coefficient, P, for the channel at the nth transmitted symbol, the l cluster, the s-th transmitting antenna to the u-th receiving antenna_lFor the power of the l-th path, θ represents the horizontal polarization angle, φ represents the vertical polarization angle, G represents the directional gain of the antenna at a certain polarization angle, G_rx，u，θRepresenting the gain G in the horizontal polarization angle direction of the u-th receiving antenna_rx，u，θ(θ_l，m，ZOA，φ_l，m，AOA) Representing a functional relationship between the angular parameters, meaning according to theta_l，m，ZOA，φ_l，m，AOACan determine a G_rx，u，θValue of (a), theta_l，m，ZOAZOA-derived horizontal polarization angle, φ, of the mth sub-cluster representing the ith cluster_l，m，AOAThe m-th sub-cluster representing the l-cluster is derived from the AOA as the vertical polarization angle, G_rx，u，Φ(θ_l，m，ZOA，φ_l，m，AOA) According to theta_l，m，ZOA，φ_l，m，AOACan determine a G_rx，u，ΦValue of (a), phi_l，m，AODThe AOD-derived vertical polarization angle of the mth sub-cluster representing the ith cluster, rx, u, phi is the vertical polarization angle of the uth receiving antenna rx, psi_l，mIs the random initial phase of the mth sub-cluster of the lth cluster, obeys [ - π, π]Random numbers uniformly distributed, j being an imaginary unit, G_tx，s，θRepresenting the gain in the horizontal polarization angle direction of the s-th transmitting antenna, G_tx，s，ΦRepresenting the gain in the direction of the vertical angle of oscillation of the s-th transmitting antenna, G_tx，s，θ(θ_l，m，ZOD，φ_l，m，AOD) Represents according to theta_l，m，ZOD，φ_l，m，AODCan determine a G_tx，s，θThe value of the one or more of,

is the transpose of the receiving direction vector of the mth sub-cluster in the ith cluster, rx, l, m is the receiving direction of the mth sub-cluster in the ith cluster, tx, l, m is the transmitting direction of the mth sub-cluster in the ith cluster,

t in (1) is transposition, b_rx，uOffset vector representing the u-th receiving array element, b_tx，sAn offset vector representing the s-th transmit element, each transmit or receive antenna being referred to as an element,_vis the relative moving speed of the antenna, λ₀Is the center frequency wavelength.

Taking ZOA as an example, due to the existence of angular spread in the cluster, the following formula is adopted to obtain the angular parameter of the mth sub-cluster in the lth cluster:

θ_l，m，ZOA＝θ_l，ZOA+c_ZSAα_m(formula two)

Wherein, theta_l，ZOAIs a statistical mean of the horizontal polarization angles of the first cluster, alpha_mWherein the mth sub-cluster base angle offset.

The other three angle parameters, namely ZOD, AOA and AOD can be obtained by adopting the formula, and the specific parameters of the ZOD, the AOA and the AOD are different.

The following formula is adopted to obtain the transmission direction vector (which is the physical meaning of the r vector) of the mth sub-cluster in the ith cluster, namely

Wherein r is_tx，l，mThe transmission direction vector for the mth sub-cluster inside the lth cluster can be obtained by the departure angle ZOD and AOD using the above formula. And accordingly, r_rx，l，mThe receiving direction vector of the mth sub-cluster inside the lth cluster can be obtained by using the above formula through the angle of arrival ZOA and the AOA.

The above formula is to calculate the first channel tap coefficient of each cluster under NLOS propagation conditions.

Step 13, under the line-of-sight condition, determining the channel tap coefficient of the first cluster (i.e. the 0 th cluster) based on the time delay caused by the three-dimensional space distance between the receiving antenna and the transmitting antenna, wherein the second channel tap coefficient from the nth transmitting antenna to the mth receiving antenna in the 0 th cluster is as follows:

wherein the content of the first and second substances,

for the 0 th cluster under LOS propagation, the nth sampling point nT from the s th transmitting antenna to the u th receiving antenna_sThe channel response (or channel tap coefficients),

for the transposition of the received direction vector under LOS propagation,

for transposing the transmitted direction vector under LOS propagation, d_3DIs the three-dimensional space distance between the receiving antenna and the transmitting antenna, 3D is three-dimensional, wherein the initial phase is 0, and the three-dimensional space distance D between the receiving antenna and the transmitting antenna is added_3DThe resulting time delay.

Step 14, adjusting the proportion between the power of the 0 th path and the power of the rest paths under the condition of line of sight according to a preset power adjustment proportion;

wherein, the rest paths are the rest paths except the 0 th path in a link;

adjusting the ratio between the power of the 0 th path and the power of the rest paths under the condition of the sight distance by the following formula:

wherein K is the Rice factor.

P_LOSRepresenting the total power of LOS propagation. P_NLOSRepresenting the total power of NLOS propagation. The ratio of the two is the power ratio of the part of the channel that propagates through LOS to NLOS. LOS stands for line-of-sight propagation and NLOS stands for non-line-of-sight propagation.

And step 15, obtaining a 4-dimensional channel tap coefficient matrix by using the first channel tap coefficient, the second channel tap coefficient and the ratio of the power of the 0 th path to the power of the rest paths, wherein the 4-dimensional channel tap coefficient matrix comprises: transmitting symbols, clusters of channels, receiving antennas and transmitting antennas, each cluster of channels being formed by M sub-clusters, i.e.

Through the formula I, the formula IV and the formula V, the 4-dimensional channel coefficient matrix A is transformed from the time domain to the frequency domain to obtain equivalent frequency domain response, and the equivalent frequency domain response is equivalent to the baseband time domain response, so that the equivalent frequency domain response is called as equivalent frequency domain response.

And step 16, dividing the 4-dimensional channel tap coefficient matrix into a plurality of sub-matrixes formed by the chain from the s-th transmitting antenna to the u-th receiving antenna. Wherein, the 4-dimensional channel tap coefficient matrix A is [ A [0, 0], A [0, 1], A [ s, u ], [ O.

Step 17, for a plurality of paths of a link in a plurality of submatrices, obtaining a multipath response vector of each signal, namely

Each sub-matrix A [ s, u ]]＝[a^s,u[0]，a^s,u[1]，···，a^s,u[N-1]]^TIs formed by a multipath response vector a of each symbol^s,u[n]Is formed by a process of, among others,

[n]as a discrete delta impulse function, t_lIs the delay of the ith cluster. Thus a^s,u[n]The multipath response vector of the nth transmission symbol from the s-th transmitting antenna to the u-th receiving antenna, wherein the value range of N is 0 to N-1.

Based on the channel coefficient, intercepting the baseband low-pass channel response from the frequency domain response by a preset truncation window length, as an equivalent frequency domain response, which further includes:

step 21, for the multipath response vector a^s,u[n]And transforming the multipath response vectors one by one to obtain equivalent frequency domain response.

The step of obtaining the decimal time delay by the MIMO-CDL channel model is the same as the step of obtaining the decimal time delay by the SISO-TDL channel model, and a formula is adopted

And (6) calculating.

Step 23, obtaining a preset cut-off window

At step 24, a linear transformation matrix is calculated based on the preset truncation window w, i.e.

F_k,l＝{exp(-j2πw_kd_l)}

and step 31, obtaining an equivalent frequency response corresponding to the nth transmission symbol from the s-th transmitting antenna to the u-th receiving antenna through the following linear transformation:

by adopting a formula, the method adopts the following steps,

to pair

Obtaining a baseband time domain response, namely:

and circularly executing the step 3 until the baseband time domain responses of all the symbols on the links of all the transmitting antennas and the receiving antennas are obtained, wherein the baseband time domain responses are equivalent to equivalent frequency domain responses and can also be called equivalent time domain channel responses. The equivalent time domain channel responses on the single link are combined to obtain an equivalent two-dimensional channel coefficient matrix, i.e.

A[s,u]＝h^s,u[0]，h^s,u[1]，···，h^s,u[N-1]^T，h^s,u[n]And the baseband time domain response of the nth transmission symbol from the nth transmission antenna to the nth receiving antenna.

Further, the equivalent time domain channel responses of all links are combined to obtain a channel tap coefficient matrix a ═ a [0, 0], a [0, 1], ·, a [ s, u ], ·. Likewise, H differs from a in that the sampling rates in the two multipath delay dimensions are different, the former being the low-pass equivalent of the latter.

Step 130, the simulation receiving end receives the convolution signal as a digital baseband signal.

In the embodiment of the invention, the baseband low-pass channel response is intercepted from the frequency domain response and is used as the equivalent frequency domain response, and the equivalent frequency domain response is used in the time domain, so that the data volume participating in convolution is reduced in the time domain, the redundant information of the whole communication system simulation process is reduced, and the complexity of the whole communication system simulation process is reduced. Compared with the traditional communication system simulation, the communication system simulation method provided by the embodiment of the invention realizes the same precision as the traditional communication system simulation, and has lower complexity. In addition, in the simulation method of the communication system according to the embodiment of the present invention, the simulation of the channel in the frequency domain of the transform domain is implemented, compared with the simulation of the communication system in the prior art that is implemented in the time domain and requires a filter and a convolution, the simulation method of the communication system according to the embodiment of the present invention has a low implementation complexity.

The communication system simulation method of the embodiment of the invention is a specific application example of the channel simulation realization of the actual communication system.

For example, one: the tap delay model is taken as a 5G NR TDL model for explanation.

Third Generation Partnership project technical Specification (3 GPP TS for short) 38.901 is a channel model Specification for 5G NR. The section 7.7.2 in the 3gpp ts.38.901 channel model specification gives 5 typical SISO-TDL parametric models, where TDL-a/B/C is a typical parametric model in a non-viewing path propagation environment, and TDL-D/E is a typical parametric model in a viewing path propagation environment. Taking the implementation of the TDL-D channel as an example without loss of generality, a simulation implementation of the 5G NR TDL channel model is given.

TABLE 13 GPP TS38.901 TDL-D channel model basic parameter settings

In table 1, the Delay of each path is given in the form of Normalized Delay, and the actual Delay value in seconds can be generally obtained from the Delay Scaling standard Scaling of Delay of section 7.7.3 in the 3GPP ts.38.901 channel model specification according to a specific scenario. And, the power and fading distribution of each path are given accordingly. In addition, since TDL-D is the Rice channel under the visual path propagation, the Rice factor K is also given in the parameter Table 1. The parameters in the parameter table 1 correspond to the tap time delay t of the SISO-TDL channel model, the average tap power is P and the rice factor K, and the total tap coefficient L is 13. And, the maximum Doppler shift is f_maxDetermined by the actual scene setting. The baseband sampling period is T_sAre set according to a specific system configuration.

Firstly, the simulating channel obtains the channel coefficient under the condition of the baseband sampling rate, and further comprises:

according to the formula

Calculating a 'baseband low-pass channel response of each path'_l(t) and in digital systems at integer multiples of the baseband sampling period, i.e., t ═ nT_sTo a'_l(t) sampling to obtain a channel response vector a [ nT ]_s]＝a[n]＝[a′₀(nT_s)，a′₁(nT_s)，···，a′_L-1(nT_s)]. Combining each set of channel coefficients to form a two-dimensional channel tap coefficient matrix: a ═ a [0 [ ]]，a[1]，···，a[N-1]]^T。

Secondly, according to the baseband sampling rate, determining the fractional time delay of the channel tap coefficient in each channel coefficient, and determining a linear transformation matrix by using the preset truncation window length and the fractional time delay, further comprising:

according to

When decimal fraction is obtainedD, prolonging; and according to a formula

Determining the length W of a preset truncation window, and acquiring the preset truncation window W; and, according to F_k,l＝{exp(-j2πw_kd_l) And (6) calculating a linear transformation matrix.

Thirdly, converting the original baseband signal response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain an equivalent frequency domain response, which further comprises:

and adding time delay through frequency domain transformation. According to the formula

To a [ n ]]Discrete Fourier transform one by one and according to formula

Performing inverse discrete Fourier transform to obtain equivalent frequency domain response

And an equivalent time domain response h [ n ]]. According to the sampling position n, circularly executing the step of converting the original baseband signal response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain the equivalent frequency domain response until all symbols, namely the channel response vector a [ n ] of the sampling point]Are all transformed into an equivalent time domain response h n]And obtaining the final equivalent two-dimensional channel coefficient matrix H.

Example two: the tap delay model is taken as a 5G NR CDL model for explanation.

The 3GPP TS38.901 also gives channel model settings in MIMO scenarios. Corresponding to the TDL model, subsection 7.7.1 in the 3GPP TS38.901 channel model specification gives 5 typical MIMO-CDL parametric models, where CDL-a/B/C is a typical parametric model in a non-path-of-view propagation environment, and CDL-D/E is a typical parametric model in a path-of-view propagation environment. Similarly, taking the implementation of the CDL-D channel as an example, a simulated implementation of the 5G NR CDL channel model is given.

TABLE 23 GPP TS38.901 CDL-D channel model basic parameter settings

In Table 2, C_ASDAngle spread coefficient, C, for horizontal departure angle_ASAAngle expansion coefficient for horizontal angle of arrival, C_ZSDExpansion coefficient of vertical departure angle, C_ZSAFor the spreading factor of the vertical departure angle, the Delay of each cluster is given in the form of Normalized Delay, and generally, according to a specific scenario, from the Delay Scaling standard Scaling of Delay of 7.7.3 subsections in the 3GPP TS38.901 channel model specification, an actual value of Delay in seconds is obtained, so that the cluster Delay of the MIMO-CDL model is obtained as t. Moreover, the average power of each cluster is given by the corresponding cluster power parameter P in Table 2, and the four angle information of each cluster is AOD (automatic optical inspection), i.e. phi_AODAOA being phi_AOAZOD is theta_ZODZOA is theta_ZOAAnd the root mean square angle spread characteristic corresponding to each angle information is c_ASD，c_ASA，c_ZSD，c_ZSAAlso given. The K parameter of CDL-D is given in the parameter table for setting the scene in subsection 7.2, and the parameter table for setting the scene in subsection 7.2 in the specification of 3GPP TS38.901 channel model also gives the typical value of the relative moving speed, so as to determine the relative moving speed of the antenna as v, and XPR is the power ratio of two polarization directions.

Sections 7.7.1, 7.7.3 and 7.2 in the 3GPP TS38.901 channel model specification all refer to the corresponding chapter numbers in the 3GPP TS38.901 standard.

Moreover, the CDL model is set up in relation to the geometric characteristics of the receiving antenna and the transmitting antenna and the specific spatial relationship between them. Wherein the spatial gain characteristic of the receiving array antenna and the spatial gain characteristic of the transmitting array antenna are G_rx,u(ψ)，G_tx,s(psi) and a position vector b_rx，u,b_tx，uDetermined by the geometric characteristics of the antenna, the receiving direction vector is r_rx,l,mAnd a transmit direction vector of r_tx,l,mHe-ShiThe three-dimensional space distance between the receiving antenna and the transmitting antenna under radial propagation is d_3DIs determined by the spatial relationship between the antennas. The baseband sampling period is T_sAnd the central wavelength of the carrier is lambda₀As determined by the particular system configuration. Thereby, the basic setting of the channel is completed.

according to the formula one

Under the condition of non-line-of-sight (NLOS), aiming at the channel of each cluster, calculating the first channel tap coefficient of the channel of the cluster from the nth transmission symbol, the ith cluster, the s-th transmission antenna to the u-th receiving antenna

And according to the formula four

Under the condition of line-of-sight (LOS), calculating a second channel tap coefficient from the nth transmitting symbol, the 0 th cluster, the s-th transmitting antenna to the u-th receiving antenna for the channel tap coefficient of the 0 th path on the basis of time delay caused by the three-dimensional space distance between the receiving antenna and the transmitting antenna;

according to the formula five

Adjusting the power distribution between the 0 th path and the rest paths under the LOS condition so as to obtain a 4-dimensional channel coefficient matrix A;

the 4-dimensional channel tap coefficient matrix A is equal to [ A [0, 0]]，A[0，1]，···，A[s,u],···]Is divided into a plurality of sub-matrixes A [ s, u ] formed by the chain from the s-th transmitting antenna to the u-th receiving antenna]＝[a^s,u[0]，a^s,u[1]，···，a^s,u[N-1]]^T。

Each sub-matrix A [ s, u ]]＝[a^s,u[0]，a^s,u[1]，···，a^s,u[N-1]]^TDividing into multipath response vector a corresponding to each symbol^s,u[n]。

according to

Acquiring decimal time delay d; and according to a formula

To a^s,u[n]Discrete Fourier transform one by one and according to formula

Performing inverse discrete Fourier transform to obtain equivalent frequency response

And equivalent time domain response h^s,u[n]. Traversing all the links among the s-th transmitting antennas of the u-th receiving antenna and setting the sampling position as n, circularly executing the steps of converting the original baseband signal response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain the equivalent frequency domain response until the equivalent frequency domain response is obtainedThe channel response vector a of all symbols, i.e. sampling points, on the link from transmitting to receiving antenna^s,u[n]Are all transformed into an equivalent time domain response h^s,u[n]And obtaining the final equivalent two-dimensional channel coefficient matrix H.

Example three: the tap delay model is taken as a 4G LTE channel model for example.

3GPP TS36.104 specifies three typical channel parameters for a 4G LTE system under multipath propagation conditions: EPA, EVA and ETU. All three channel models belong to SISO-TDL models, so that the simulation can be carried out by the implementation method of the SISO-TDL channel. In addition, the three channel model setting modes are that under the NLOS propagation condition, multipath time delay, power and maximum Doppler frequency shift are given, and only specific values are different. Therefore, an EPA is taken as an example to introduce a channel simulation method in a 4G LTE multipath propagation scenario. The following table is the multipath parameter settings for the EPA channel:

TABLE 3 multipath parameter settings for EPA channels

In table 3, the delay per path is given by the multi-tap delay process tap delay, corresponding to the tap delay being t, the power value per path is given by the Relative power, corresponding to the tap average power being P, and the number of multipaths is L-7. Maximum Doppler shift f_maxDoppler shift is a typical spectrum at 5 Hz. The baseband sampling period is T_sAre set according to a specific system configuration.

according to the formula

Calculating the baseband low-pass channel response a of each path_l(t) and in digital systems at integer multiples of the baseband sampling period, i.e., t ═ nT_sTo a, a_ll (t) sampling to obtain a channel response vector a [ nT ]_s]＝a[n]＝[a₀(nT_s)，a₁(nT_s)，···，a_L-1(nT_s)]。

Combining the channel responses at each sampling time to form a two-dimensional channel tap coefficient matrix: a ═ a [0 [ ]]，a[1]，···，a[N-1]]^T。

according to

Acquiring decimal time delay d; and according to a formula

To a [ n ]]Discrete Fourier transform one by one and according to formula

And an equivalent time domain response h [ n ]]. According to the sampling position n, circularly executing the step of converting the original baseband signal response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain the equivalent frequency domain responseVector of channel responses a [ n ] to all symbols, i.e. sample points]Are all transformed into an equivalent time domain response h n]And obtaining the final equivalent two-dimensional channel coefficient matrix H. The implementation of the ETU and EVA channels is consistent with the above steps.

The following proceeds to describe another communication system simulation method provided by the embodiment of the present invention.

The other communication system simulation method provided by the embodiment of the invention can be applied to electronic equipment, and further the electronic equipment can be a server, a computer and the like.

As shown in fig. 2, another communication system simulation method provided in the embodiment of the present invention includes:

step 210, obtaining a digital baseband signal, wherein the digital baseband signal carries a baseband sampling rate;

step 220, acquiring a channel coefficient under the condition of the baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from a frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal to obtain a convolved signal;

and step 230, performing down-sampling on the convolution signal, and taking the down-sampled convolution signal as a digital baseband signal.

The following continues to describe a communication system simulation apparatus according to an embodiment of the present invention.

As shown in fig. 3, an embodiment of the present invention further provides a communication system simulation apparatus, including:

the simulation transmitting terminal 31 is configured to obtain a baseband sampling rate and transmit a digital baseband signal;

the simulation channel 32 is used for acquiring a channel coefficient under the condition of a baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from the frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal;

and the simulation receiving end 33 is configured to receive the convolution signal and use the downsampled convolution signal as a digital baseband signal.

Further, the emulation channel is configured to:

according to the baseband sampling rate, determining the decimal time delay of a channel tap coefficient in each channel coefficient;

determining a linear transformation matrix by using the length of a preset truncation window and the decimal time delay, wherein the length of the preset truncation window is the rounding result of the maximum decimal time delay in the decimal time delay;

and converting the original baseband channel response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain the equivalent frequency domain response.

Further, the emulation channel is configured to:

the formula is adopted:

so as to obtain the equivalent frequency domain response,

wherein k represents the subcarrier number of the frequency domain,

l is the total number of tap coefficients, n is the number of elements, a_nIs the nth element of the vector a, d_nIs the nth element of the vector d, k is the unit of imaginary number of the frequency domain subcarrier number j, F is the transformation coefficient matrix, a is the tap coefficient vector, a ═ a₀，a₁，…，a_L-1]D is a decimal time delay vector, and d is T/T_s＝[d₀，d₁，…，d_L-1]＝[p₀/q，p₁/q，…，p_L-1/q]Q is oversampling factor, p_nIs the corresponding integer sample position after oversampling.

Further, an embodiment of the present invention provides a communication system simulation apparatus, further including: an obtaining module, configured to obtain, when the simulation channel is a multiple-input multiple-output clustered delay line channel model, an original baseband channel response in a time domain by using the following steps:

under the condition of non-line-of-sight, aiming at the channel of each cluster, determining a first channel tap coefficient of the channel of the cluster from the nth transmission symbol, the ith cluster, the s-th transmission antenna to the u-th receiving antenna;

under the line-of-sight condition, determining a second channel tap coefficient from an nth transmitting symbol, an l cluster, an u transmitting antenna to an s receiving antenna for the channel tap coefficient of the 0 th path based on time delay caused by a three-dimensional space distance between the receiving antenna and the transmitting antenna;

adjusting the proportion between the power of the 0 th path and the power of the rest paths under the condition of sight distance according to a preset power adjustment proportion;

obtaining a 4-dimensional channel tap coefficient matrix by using the first channel tap coefficient, the second channel tap coefficient and the adjusted ratio between the power of the 0 th path and the power of the rest paths under the line-of-sight condition, wherein the 4-dimensional channel tap coefficient matrix comprises: transmitting symbols, clusters of channels, receiving antennas and transmitting antennas, wherein each cluster of channels is formed by M sub-clusters;

dividing a 4-dimensional channel tap coefficient matrix into a plurality of sub-matrixes formed by an s-th transmitting antenna to an u-th receiving antenna link;

for a plurality of paths of one link in a plurality of sub-matrixes, obtaining a multi-path response vector of each signal;

an emulation channel to:

and intercepting the baseband low-pass channel response from the frequency domain response by the preset truncation window length on the basis of the channel coefficient for the original baseband channel response in each time domain in the multipath response vector to serve as the equivalent frequency domain response.

The following continues to describe another communication system simulation apparatus provided in the embodiment of the present invention.

As shown in fig. 4, a communication system simulation apparatus provided in an embodiment of the present invention includes:

a signal obtaining module 41, configured to obtain a digital baseband signal, where the digital baseband signal carries a baseband sampling rate;

a signal convolution module 42, configured to obtain a channel coefficient under the condition of the baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from a frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal to obtain a convolved signal;

and a down-sampling module 43, configured to down-sample the convolution signal, and use the down-sampled convolution signal as a digital baseband signal.

It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, as for the apparatus embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

The above description is only for the preferred embodiment of the present invention, and is not intended to limit the scope of the present invention. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for communication system simulation, comprising:

acquiring a channel coefficient by the simulation channel under the condition of the baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from a frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal to obtain a convolved signal;

the simulation receiving end receives the convolution signal and takes the convolution signal after down sampling as a digital baseband signal;

the intercepting of the baseband low-pass channel response from the frequency domain response through a preset truncation window length based on the channel coefficient includes, as an equivalent frequency domain response:

determining decimal time delay of a channel tap coefficient in each channel coefficient according to the baseband sampling rate;

determining a linear transformation matrix by using the preset truncation window length and the decimal time delay, wherein the preset truncation window length is an integer result of the maximum decimal time delay in the decimal time delay;

converting the original baseband channel response in the time domain into the frequency domain by using the linear transformation matrix and the channel tap coefficient to obtain equivalent frequency domain response;

converting the original baseband channel response in the time domain into the frequency domain by using the linear transformation matrix to obtain an equivalent frequency domain response, wherein the method comprises the following steps:

the formula is adopted:

so as to obtain the equivalent frequency domain response,

wherein k represents the subcarrier number of the frequency domain,

l is the total number of tap coefficients, n is the number of elements, a_nIs the nth element of the vector a, d_nIs the nth element of the vector d, k is the frequency domain subcarrier number, j is the imaginary unit, F is the transform coefficient matrix, a is the tap coefficient vector, a ═ a₀，a₁，…，a_L-1]Is a decimal time delay vector, d is T/T_s＝[d₀，d₁，…，d_L-1]＝[p₀/q，p₁/q，…，p_L-1/q]Q is oversampling factor, p_nIs the corresponding integer sample position after oversampling.

2. The method of claim 1, wherein when the simulated channel is a multiple-input multiple-output (mimo) clustered delay line channel, the method comprises the following steps:

under the line-of-sight condition, determining a second channel tap coefficient from an nth transmitting symbol, an ith cluster, an s-th transmitting antenna to an u-th receiving antenna for the channel tap coefficient of the 0 th path based on time delay caused by the three-dimensional space distance between the receiving antenna and the transmitting antenna;

adjusting the proportion between the power of the 0 th path and the power of the rest paths under the line-of-sight condition according to a preset power adjustment proportion, wherein the rest paths are the rest paths except the 0 th path in a link;

obtaining a 4-dimensional channel tap coefficient matrix by using a first channel tap coefficient, a second channel tap coefficient and a ratio of the power of the 0 th path to the power of the rest paths under the line-of-sight condition after adjustment, wherein the 4-dimensional channel tap coefficient matrix comprises: transmitting symbols, clusters of channels, receiving antennas and transmitting antennas, wherein the cluster of channels of each cluster is formed by M sub-clusters;

dividing the 4-dimensional channel tap coefficient matrix into a plurality of sub-matrixes formed by the links from the s-th transmitting antenna to the u-th receiving antenna;

and intercepting the baseband low-pass channel response from the frequency domain response through the preset truncation window length on the basis of the channel coefficient as the equivalent frequency domain response for the original baseband channel response in each time domain in the multipath response vector.

3. A method for communication system simulation, comprising:

down-sampling the convolution signal, and taking the down-sampled convolution signal as a digital baseband signal;

the formula is adopted:

so as to obtain the equivalent frequency domain response,

wherein k represents the subcarrier number of the frequency domain,

l is a tap systemTotal number of numbers, n being the number of the elements, a_nIs the nth element of the vector a, d_nIs the nth element of the vector d, k is the frequency domain subcarrier number, j is the imaginary unit, F is the transform coefficient matrix, a is the tap coefficient vector, a ═ a₀，a₁，…，a_L-1]Is a decimal time delay vector, d is T/T_s＝[d₀，d₁，…，d_L-1]＝[p₀/q，p₁/q，…，p_L-1/q]Q is oversampling factor, p_nIs the corresponding integer sample position after oversampling.

4. A communication system simulation apparatus, comprising:

the simulation channel is used for acquiring a channel coefficient under the condition of the baseband sampling rate; based on the channel coefficient, intercepting a baseband low-pass channel response from a frequency domain response through a preset truncation window length to serve as an equivalent frequency domain response, wherein the frequency domain response is generated by converting an original baseband channel response in a time domain into a frequency domain; converting the equivalent frequency domain response into a time domain to obtain a baseband time domain response; convolving the baseband time domain response with the digital baseband signal;

the simulation receiving end is used for receiving the convolution signal as a digital baseband signal;

wherein the emulation channel is configured to:

the simulation channel is specifically configured to:

the formula is adopted:

so as to obtain the equivalent frequency domain response,

wherein k represents the subcarrier number of the frequency domain,

l is the total number of tap coefficients, n is the number of elements, a_nIs the nth element of the vector a, d_nIs the nth element of the vector d, k is the frequency domain subcarrier number, j is the imaginary unit, F is the transform coefficient matrix, a is the tap coefficient vector, a ═ a₀，a₁，…，a_L-1]D is a decimal time delay vector, and d is T/T_s＝[d₀，d₁，…，d_L-1]＝[p₀/q，p₁/q，…，p_L-1/q]Q is oversampling factor, p_nIs the corresponding integer sample position after oversampling.

5. The apparatus of claim 4, wherein the apparatus further comprises: an obtaining module, configured to obtain, when the simulation channel is a multiple-input multiple-output clustered delay line channel, an original baseband channel response in a time domain by using the following steps:

obtaining a 4-dimensional channel tap coefficient matrix by using a first channel tap coefficient, a second channel tap coefficient and a ratio of the power of the 0 th path to the power of the rest paths under the line-of-sight condition, wherein the 4-dimensional channel tap coefficient matrix comprises: transmitting symbols, clusters of channels, receiving antennas and transmitting antennas, wherein the cluster of channels of each cluster is formed by M sub-clusters;

the emulation channel is configured to:

6. A communication system simulation apparatus, comprising:

the down-sampling module is used for down-sampling the convolution signal and taking the down-sampled convolution signal as a digital baseband signal;

wherein the signal convolution module is configured to:

the signal convolution module is specifically configured to:

the formula is adopted:

so as to obtain the equivalent frequency domain response,

wherein k represents the subcarrier number of the frequency domain,

l is the total number of tap coefficients, n is the number of elements, a_nIs the nth element of the vector a, d_nIs the nth element of the vector d, k is the frequency domain subcarrier number, j is the imaginary unit, F is the transform coefficient matrix, a is the tap coefficient vector, a ═ a₀，a₁，...，a_L-1]D is a decimal time delay vector, and d is T/T_s＝[d₀，d₁，…，d_L-1]＝[p₀/q，p₁/q，…，p_L-1/q]Q is oversampling factor, p_nIs the corresponding integer sample position after oversampling.