An Improved Cluster Delay Line Channel Simulation Method

Abstract

The present invention discloses an improved cluster delay line channel simulation method, comprising: selecting a channel model according to the 3GPP TS 38.901 standard, and inputting scenario parameters and channel scale parameters; generating an ellipsoidal basis sequence #imgabs0# of corresponding length and order, and converting the channel coefficient H[U×S×C×N a The system then divides the coefficients into D groups of equal length and projects them onto the same ellipsoidal basis sequence to obtain reduced-rank basis representations of the D groups of channel coefficients. The reduced-rank basis representations of the channel coefficients are transmitted to a real-time processing unit. In the real-time processing unit, the reduced-rank basis representations are multiplied by the previously generated ellipsoidal basis sequence to obtain recovered channel coefficients. These channel coefficients are then directly convolved with the signal. Compared to traditional wireless channel coefficient generation methods, this method achieves approximately tenfold acceleration, demonstrating high engineering value.

Description

Technical Field

[0001] The present invention relates to the technical field of wireless channel coefficient generation, and in particular to an improved cluster delay line channel simulation method. Background Art

[0002] With the development of wireless communication technology, the iteration cycle of wireless communication equipment is constantly shortening. As an indispensable testing tool, the performance of channel simulators almost determines the development speed of wireless communication equipment.

[0003] Traditional clustered delay line channel simulation methods require offline generation of a large number of channel coefficients H[U×S×C×N] and feed them into the online convolution part for computation. As channel sizes continue to expand, the channel coefficients also increase significantly, and the time required to generate and transmit these coefficients also increases. The excessively long time required by traditional clustered delay line channel simulation methods no longer meets the current requirements of large-scale 5G channel simulation.

[0004] 3GPP TS 38.901, defined by the mobile communications standardization organization 3GPP, defines the channel model and test standards for fifth-generation mobile communications, covering the frequency range from 0.5 GHz to 100 GHz. The protocol abstracts all communication scenarios into 10 types based on the relative positions of the base station and mobile station, the complexity of ambient scatterers, and the presence of a line of sight (LOS). To simplify modeling, it also specifies five clustered delay line (CDL) channel models.

[0005] As a very efficient band-limited sequence, the ellipsoidal basis sequence has been widely used in the field of signal processing. Given a sequence length N and a bandwidth W max , satisfying 0 <W max <1 / 2, and given the order K, a unique set of ellipsoidal basis sequences can be generated, and the frequency f is greatly limited to [-W max ,W max Because the ellipsoidal basis sequence has a very concentrated energy, it can reduce the amount of data of the channel coefficients used as the projection basis of the channel coefficients to express the channel characteristics. In addition, due to some characteristics of the ellipsoidal basis sequence itself, the amount of calculation can be further reduced when calculating the projection coefficients. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to improve the traditional cluster delay line channel simulation method by utilizing the characteristics of ellipsoidal basis sequences to solve the technical problems mentioned in the background technology. The present invention can achieve an acceleration effect of about 20 times, that is, the computing time has an advantage of 1 order of magnitude, the acceleration effect is obvious, and it has high engineering value.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] An improved cluster delay line channel simulation method includes the following steps:

[0009] Step S1: According to the 3GPP TS 38.901 standard, select a channel model and input scenario parameters and channel scale parameters;

[0010] Step S2: Generate an ellipsoidal basis sequence of corresponding length and order according to the channel model and various parameters determined in step S1. and its corresponding eigenvalue λ k and pre-stored on a real-time processing unit;

[0011] Step S3: Update the Doppler frequency shift f of each ray in each cluster in the channel model according to the channel model and channel scale parameters determined in step S1. c,r , the gain term η of each ray in each cluster c,r and the normalized linear power P of each cluster c ;

[0012] Step S4: Project the Doppler term calculated in step S3 onto the ellipsoidal basis sequence generated in step S2, and compare the obtained projection coefficient with the normalized linear power P of each cluster obtained in step S3. c and the gain term η c,r Corresponding multiplication is performed to obtain the reduced rank basis representation coefficient of the final channel coefficient Where C represents the number of clusters;

[0013] Step S5: The reduced rank basis of the channel coefficient obtained in step S4 is expressed as Transmitted to the real-time processing part, and multiplied with the ellipsoidal basis sequence stored there in advance to obtain the recovered channel coefficient H[U×S×C×N];

[0014] Step S6: directly convolve the channel coefficient H[U×S×C×N] obtained in step S5 with the signal;

[0015] Furthermore, the scene parameters include the three-dimensional coordinates of the base station (BS), the three-dimensional coordinates of the terminal (UT), the terminal moving direction and speed v, the carrier frequency f c, Doppler sampling frequency 1 / T a ; Channel scale parameters include the number of single sampling points N, the number of transmitting antennas S, and the number of receiving antennas U, where the number of single sampling points N is an integer multiple of 1024.

[0016] Furthermore, in step S2, the generated ellipsoidal basis sequence The length of each sequence is determined by the number of single sampling points N, and the number of ellipsoidal basis sequences K required is determined by The larger K is, the higher the accuracy is, where N represents the number of single sampling points, and W max represents the normalized maximum Doppler frequency deviation;

[0017] Furthermore, in step S3, the Doppler frequency shift f in the Doppler term c,r refers to the Doppler shift of the rth ray in the cth cluster, and the gain term η c,r Refers to the gain of the rth ray in the cth cluster, where c ranges from 1 to C and r ranges from 1 to R, where C represents the number of clusters and R represents the number of rays in each cluster. In addition, the gain term η c,r Refers to all other terms in the 3GPP TS 38.901 standard channel coefficient generation formula except the Doppler term and the normalized linear power of each cluster, including the directivity pattern and cross-polarization ratio factors, the transmitter phase factor, and the receiver phase factor;

[0018] Furthermore, in the process of generating the channel coefficient, the Doppler sampling frequency used is 1 / T a and the signal sampling frequency f s Equal, that is, 1 / T a =f s , and only the Doppler term is updated within T time, and the Doppler frequency shift f in the Doppler term c,r , the normalized linear power P of each cluster c and the gain term η c,r Their update time T is still determined according to 3GPPTS 38.901 standard;

[0019] Furthermore, within the time interval T, 1 / T a The total number of sampling points is N. a =T / T a , with N sampling points as a group, divided into D=N a / N groups, where the number of sampling points N only needs to be an integer multiple of 1024, and each group of Doppler terms is projected onto the same set of ellipsoid basis sequences. For the dth group of reduced rank basis representation coefficients β c,k (d) The coefficients can be expressed by the first set of reduced-rank bases The calculation formula is shown in (1).

[0020]

[0021] First, the number of sampling points N must be an integer multiple of 1024. Second, if N is too large, the generated ellipsoid basis sequence The length is correspondingly longer. Excessively long sequences require higher computing power and greater storage speed and capacity. Taking all the above factors into consideration, the value range of N is between 1024 and 10240, and is an integer multiple of 1024.

[0022] The beneficial effects of the present invention are:

[0023] Compared with the traditional cluster delay wireless channel coefficient generation method, the present invention can achieve an acceleration effect of about 10-20 times and has high engineering value. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 This is the architecture diagram of the improved cluster delay line channel simulation method;

[0025] Figure 2 A comparison chart of the computation time of the traditional method and the improved method for different channel sizes;

[0026] Figure 3 Error plot of the channel coefficients generated by the improved method relative to those generated by the conventional method. DETAILED DESCRIPTION

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0028] Implementation method as Figure 1 The specific implementation steps are as follows:

[0029] Step S1: According to the 3GPP TS 38.901 standard, select a channel model and input scenario parameters and channel scale parameters;

[0030] Specifically, for the channel model, in this embodiment, 10 conventional channel models defined in the TS 38.901 test standard and 5 simplified channel models such as CDL-A, CDL-B, CDL-C, CDL-D, and CDL-E are provided. Depending on the model, the number of clusters N and the number of sub-paths in each cluster M will also be different; the scenario parameters include the three-dimensional coordinates of the base station (BS), the three-dimensional coordinates of the terminal (UT), the terminal moving direction and speed v (unit: m / s), the carrier frequency f c (Unit: Hz), Doppler sampling frequency 1 / T a (Unit: Hz); the channel scale parameters include the number of single sampling points N, the number of transmitting antennas S, and the number of receiving antennas U, where the number of single sampling points N must be an integer multiple of 1024; enter the required parameters in sequence. It should be noted that the larger the number of sampling points N in each group, the larger the scale of the ellipsoid basis sequence that needs to be generated in advance, and the larger the corresponding storage resources required. On the other hand, the effect of reducing the amount of transmitted data is also more obvious.

[0031] Step S2: Generate an ellipsoidal basis sequence of corresponding length and order according to the channel model and various parameters determined in step S1. and its corresponding eigenvalue λ k and pre-stored on a real-time processing unit;

[0032] Specifically, in this embodiment, according to the parameters given in step S1, the normalized maximum Doppler frequency f can be obtained. d =(f c *v) / (c*f s ), where c represents the speed of light, which is 3×10 8 m / s, that is, W max =f d Based on the number of single sampling points N given in step S1, the minimum value of K can be obtained. Considering the accuracy, K can generally be appropriately enlarged. Then, an ellipsoid basis sequence with a length of N and an order of K is generated.

[0033] Step S3: Update the Doppler frequency shift f in the Doppler term in the channel model according to the channel model and various parameters determined in step S1. c,r , the normalized linear power P of each cluster c and the gain term η c,r ;

[0034] Specifically, the Doppler frequency shift f in the Doppler term described in this embodiment is c,r refers to the Doppler shift of each ray r in each cluster c, and the gain term η c,ris the gain of each ray r in each cluster c, C represents the number of clusters, R represents the number of rays in each cluster, and the gain term η c,r Refers to all other terms except the Doppler term and the normalized linear power of each cluster in the channel coefficient generation formula of the 3GPP TS 38.901 standard, including the directivity pattern and cross-polarization ratio factor, the transmitting end phase factor and the receiving end phase factor. And it is necessary to adjust the Doppler frequency shift f in the Doppler term in the channel model at every T time according to the 3GPP TS 38.901 standard. c,r , the normalized linear power P of each cluster c and the gain term η c,r Update is performed, and the update time T is determined according to the 3GPPTS 38.901 standard.

[0035] Step S4: The Doppler frequency shift f calculated in step S3 is c,r Project it onto the ellipsoid basis sequence generated in step S2, and compare the obtained projection coefficient with the normalized linear power P of each cluster obtained in step S3 c and the gain term η c,r Corresponding multiplication is performed to obtain the reduced rank basis representation coefficient of the final channel coefficient

[0036] Specifically, due to the ellipsoidal basis sequence Its own characteristics, and its corresponding ellipsoid function U k (N, W; f), calculate the Doppler frequency shift f in the Doppler term c,r There is an algorithm to reduce the amount of calculation for the projection, so the calculation formula of the reduced rank basis representation coefficient of the first group of channel coefficients is shown in (2).

[0037]

[0038] Among them U k (N,W;f c,r ) Another approximate calculation formula The calculation formula is shown in (3),

[0039]

[0040] in

[0041]

[0042] f c,r ∈[W0-W max ,W0+W max ], so in this embodiment W0=0, and the positive and negative signs in formula (3) are determined by the inequality group (5).

[0043]

[0044] Through the above process, the reduced rank basis representation coefficients of a set of channel coefficients within time T are obtained According to the 3GPPTS 38.901 standard, the parameters in step S3 need to be updated at each time T, that is, within the time interval T, the Doppler frequency shift f in the Doppler term c,r , the normalized linear power P of each cluster c and the gain term η c,r The Doppler sampling frequency used in step S1 is 1 / T. a , the number of single sampling points is N, then the total number of sampling points in time T is N a =T / T a , with N sampling points as a group, divided into D=N a / N groups, for the dth group, the reduced rank basis represents the coefficient The coefficients can be expressed by the first set of reduced rank basis The calculation formula is shown in (1).

[0045]

[0046] The above steps need to be repeated after the parameters are updated in step S3.

[0047] Step S5: The reduced rank basis of the channel coefficient obtained in step S4 is expressed as Transmitted to the real-time processing part and compared with the ellipsoid base sequence stored there in advance Multiply them to obtain the recovered channel coefficient H[U×S×C×N];

[0048] Specifically, the reduced rank basis representation coefficients of each group of channel coefficients obtained in step S4 need to be transmitted to the real-time processing part and the ellipsoid basis sequence stored there in advance. That is to say, the channel coefficients are restored in groups of 5120 points.

[0049] Step S6: directly convolve the channel coefficient H[U×S×C×n] obtained in step S5 with the signal;

[0050] In order to verify the effectiveness and universality of the present invention, the performance of the improved cluster delay line channel simulation method is compared with the traditional cluster delay line channel simulation method. The channel coefficients with different data amounts are calculated and the average calculation time is tested. Figure 2As shown in the figure. For a single-input single-output channel, the improved clustered delay line channel simulation method generates channel coefficients with the same amount of data, and the calculation time is shorter than that of the traditional clustered delay line channel simulation method, which is about an order of magnitude better. The acceleration effect is more obvious. Since the reduced rank basis representation coefficient algorithm for calculating the channel coefficient is approximate, the error analysis of the improved clustered delay line channel simulation method is also carried out, as shown in the figure. Figure 3 As shown, the errors are all below 1%, which basically meets the requirements of channel simulation.

[0051] Any details not described in detail herein are generally known to those skilled in the art. The preferred embodiments of the present invention have been described in detail above. It should be understood that a person skilled in the art can make numerous modifications and variations based on the concepts of the present invention without inventive effort. Therefore, any technical solution that a person skilled in the art can arrive at through logical analysis, reasoning, or limited experimentation based on the concepts of the present invention and the prior art should be within the scope of protection defined by the claims.

Claims

1. An improved cluster delay line channel simulation method, characterized in that: The steps include: Step S1: According to the 3GPP TS 38.901 standard, select a channel model and input scenario parameters and channel scale parameters; Step S2: Generate an ellipsoidal basis sequence based on the channel model and channel scale parameters determined in step S1. and its eigenvalue λ k and pre-stored on a real-time processing unit; Step S3: Based on the channel model and channel scale parameters determined in step S1, update the Doppler frequency shift f of each ray in each cluster of the channel model. c,r , the gain term η of each ray in each cluster c,r and the normalized linear power P of each cluster c ; Step S4: Project the Doppler terms calculated in step S3 into the ellipsoid basis sequence generated in step S2. The obtained projection coefficient is combined with the normalized linear power P of each cluster obtained in step S3. c and the gain term η c,r Multiply them together to get the reduced rank basis representation coefficients of multiple groups of channel coefficients Where C represents the number of clusters, S is the number of transmitting antennas, U is the number of receiving antennas, and K is the number of ellipsoidal basis sequences; Step S5: The reduced rank basis of the channel coefficient obtained in step S4 is expressed as Transmitted to the real-time processing unit and compared with the ellipsoid base sequence stored in the real-time processing unit in advance Multiply them to obtain the recovered channel coefficient H[U×S×C×N]; Step S6: directly perform a convolution operation on the channel coefficient H[U×S×C×N] obtained in step S5 and the signal.

2. The improved cluster delay line channel simulation method according to claim 1, characterized in that: In step S1, the scene parameters include the base station three-dimensional coordinates, the terminal three-dimensional coordinates, the terminal moving direction and speed, the carrier frequency, the Doppler sampling frequency 1 / T a ; The channel scale parameters include the number of single sampling points N, the number of transmitting antennas S, and the number of receiving antennas U, where the number of single sampling points N is an integer multiple of 1024.

3. The improved cluster delay line channel simulation method according to claim 2, characterized in that: In step S2, the generated ellipsoid basis sequence The length of each column of the sequence is equal to the number of single sampling points N, and the ellipsoidal basis sequence The number K satisfies Under the premise of positive correlation with the channel simulation accuracy, W max Indicates the normalized maximum Doppler frequency deviation.

4. The improved cluster delay line channel simulation method according to claim 3, characterized in that: In step S3, the Doppler frequency shift f in the Doppler term c,r refers to the Doppler shift of the rth ray in the cth cluster, and the gain term η c,r Refers to the gain of the rth ray in the cth cluster, where C is the number of clusters and R is the number of rays per cluster, where c ranges from 1 to c and r ranges from 1 to R.

5. The improved cluster delay line channel simulation method according to claim 4, characterized in that: In step S1, the Doppler sampling frequency used is 1 / T a and the signal sampling frequency f s are equal, and only the Doppler term itself is updated within the time interval T, while the Doppler frequency shift f in the Doppler term is c,r , the normalized linear power P of each cluster c and the gain term η c,r are unchanged, the Doppler frequency shift f c,r , the normalized linear power P of each cluster c and the gain term η c,r The update time T is still determined according to the 3GPP TS 38.901 standard.

6. The improved cluster delay line channel simulation method according to claim 5, characterized in that: In step S3, if within the time interval T, 1 / T a The total number of sampling points is N. a =T / T a , with n sampling points as a group, divided into D=N a / N groups, each group of Doppler terms is projected onto the same set of ellipsoid basis sequences, and the coefficients of the dth group of reduced rank basis are expressed The coefficients are represented by the first set of reduced-rank basis The calculation formula is shown in (1): First, the number of sampling points N must be an integer multiple of 1024. Second, if N is too large, the generated ellipsoid basis sequence The length is correspondingly longer, and overly long sequences require higher computing power, greater storage speed and capacity.

7. The improved cluster delay line channel simulation method according to claim 6, characterized in that: The value of N ranges from 1024 to 10240 and is an integer multiple of 1024.

Images