Method for simulating wireless channel Doppler diffusion and multipath time delay

Abstract

The invention relates to a method for simulating Doppler diffusion and multipath time delay of a wireless channel, belongs to the field of wireless communication channel simulation, and is used for simulating Doppler diffusion and multipath time delay of a Rayleigh channel. According to the method, the maximum number of sine waves of a channel simulator is expanded through a time division multiplexing Cordic calculation module, a limited number of sine waves are superposed to generate a channel envelope, the channel sampling rate is improved through a CIC interpolation filter, fractional time delay in a wireless channel is simulated through a Farrow structure filter based on Lagrange interpolation, and the frequency of the wireless channel is improved. Therefore, the multi-path number is flexibly expanded, the resource consumption is reduced, the channel simulation precision and the sampling rate are improved, and the multi-path fractional time delay is flexibly simulated. Compared with an ROM table look-up method, the method has the advantages of being low in resource consumption, high in channel simulation precision, easy to implement and deploy and the like, and is suitable for the fields of wireless communication channel simulators, signal processing and the like.

Description

Technical Field

[0001] The invention belongs to the field of wireless communication channel simulation, and relates to a method for simulating Doppler diffusion and multipath delay of a wireless channel. Background Art

[0002] 1 Wireless Channel Characteristics

[0003] In wireless communication, when a signal propagates from a transmitter to a receiver, it will experience multiple reflections and finally reach the receiver from different incident angles. When there are many reflection paths, the incident angles θ of each sub-path of the same indivisible path (i.e., the propagation delay difference of each path is extremely small) obey the uniform distribution of [0,2π]. After the signal passes through each sub-path, a Doppler spread spectrum with a specific shape will be generated. When there is a direct path (LOS), the envelope statistics of the signal received by the receiver obeys the Rice distribution. When there is no direct path (NLOS), the envelope statistics of the signal received by the receiver obeys the Rayleigh distribution. When simulating the Rice channel, the signal can be superimposed with the direct signal after passing through the Rayleigh channel, thereby achieving the effect of simulating the Rice channel.

[0004] When the carrier frequency is f c , when the receiving terminal moves at a speed v, the Doppler frequency deviation f generated by the signal d As follows:

[0005]

[0006] , where θ n is the incident angle of the nth path, and c is the speed of light constant. Due to the influence of the Doppler effect, the Doppler frequency shift deviation generated by the LOS path and the NLOS path received by the receiving end. The continuous Doppler change caused by the superposition of the LOS path and the NLOS path makes the single signal power spectrum of the transmitter program a continuous Doppler spectrum. The distribution of the incident angle θ of each sub-path and the frequency selective attenuation caused by the multipath delay determine the shape of the Doppler diffusion spectrum, which is typically the Jakes power spectrum and the Gaussian power spectrum.

[0007] 2 Channel model and channel simulation method

[0008] In a wireless time-varying channel, the received signal is the delayed amount of the transmitted baseband signal x(t) multiplied by a complex random process, that is,

[0009]

[0010] in

[0011]

[0012] Therefore, when simulating wireless channels, it is necessary to simulate complex random processes and signal transmission delays. There are currently two common methods for simulating channels, namely the sine wave superposition method (SOS, Sum of Sinusoid) and the shaping filter method. The shaping filter method generates Gaussian noise in the time domain, and then passes the Gaussian noise through a filter of a specific shape to synthesize colored Gaussian white noise, whose envelope obeys the Rayleigh distribution. Compared with the shaping filter method, the sine wave superposition method is currently used more. The principle is to simulate the Gaussian random process by superposing finite weighted harmonic functions with equidistant frequencies and random phases, that is,

[0013]

[0014] In the above expression, c i,n is the discrete Doppler coefficient, f i,n is the discrete Doppler frequency, θ i,n is a random phase. Δf in discrete Doppler frequency i =f i,n -f i-1,n As a differential factor, satisfying the Δf→0 condition, in order to cover all Doppler frequency ranges, N is required i →∞, that is, an infinite number of sine waves need to be generated, but this is impossible for hardware implementation. Therefore, it is necessary to use finite harmonics to superimpose analog channels while ensuring performance.

[0015] In wireless communications, common signal envelope distributions such as Lognormal distribution, Rician distribution, and Rayliegh distribution can all be generated using the sinusoidal wave superposition method.

[0016] 1) For Lognormal distribution, the simulation model is as follows:

[0017]

[0018] Where μ is the mean of the lognormal distribution, c n is the Doppler coefficient, f n is the discrete Doppler frequency, θ n is a random phase.

[0019] 2) For Rayleigh distribution, the model is as follows:

[0020]

[0021] In the above formula, N is the number of sine waves, f max is the maximum Doppler frequency deviation, and α is the arrival angle of the subpath.

[0022] 3) For the Rice distribution, its envelope is a simple superposition of the Rayleigh distribution on the direct path.

[0023] There are many methods for calculating the parameters of the above three common channel models. In the present invention, the Rayleigh channel parameter calculation model proposed by Wang Chengxiang is taken as an example to illustrate the present invention.

[0024] 3 Wireless channel parameter calculation model

[0025] Wang Chengxiang et al. proposed a channel parameter calculation model based on sine wave superposition to generate uncorrelated ray fading waveforms. It is now widely integrated into a variety of mathematical tools for simulating wireless channels. The complex envelope expression of Rayleigh fading is shown in (3), k The amplitude statistics of (t)| obey Rayleigh distribution:

[0026]

[0027] When using the sine wave superposition method to simulate the Rayleigh channel, The expression form is as follows: The amplitude statistics of

[0028]

[0029] k is the kth tap, and K is the total number of taps.

[0030] N k is the number of sinusoids used for SOS by the kth tap, indicating the number of indivisible subpaths in one tap.

[0031] is the discrete Doppler frequency in the nth subpath in the kth tap, that is, the frequency of the sine wave.

[0032] is the phase component in the nth subpath of the kth tap, which obeys the uniform distribution of (0,2π].

[0033] t is the fading process time.

[0034] The calculation method of discrete Doppler frequency is:

[0035]

[0036] in is the rotation angle factor, expressed as

[0037]

[0038] From the above formula, we can know that each indivisible sub-path channel can be represented by To express it, all the sub-paths of a single tap are superimposed to obtain The envelopes of its real and imaginary parts are approximately Gaussian distributed, so Z k The complex envelope of (t) obeys the Rayleigh distribution, and its frequency spectrum appears as a Doppler spread spectrum with a specific shape.

[0039] From the above analysis, we can see that the key to simulating Rayleigh channels is to generate sine waves with a specified frequency range and different phases. If the number of sine waves is too small, the randomness of the superimposed data will be poor or the performance will be poor, which will lead to poor simulation of the Rayleigh channel. Therefore, in the simulation process, it is necessary to ensure that the number of sine waves used for SOS is large enough.

[0040] Problems with 4-channel simulation

[0041] In FPGA, there are two main methods for generating sine waves, namely the Cordic real-time calculation method and the lookup table method.

[0042] The generation method of the Cordic real-time calculation method is as follows: first, the step phase is obtained according to the required frequency, the phase is accumulated under the control of the clock, and the phase accumulation result is input into the Cordic calculation module to calculate the amplitude corresponding to the phase.

[0043] The table lookup method is to calculate the amplitude of the sine wave in one cycle in advance and store it in a ROM table. By controlling the starting address and step length of the ROM table, a sine wave with a specific phase and frequency is realized. Generally, it is necessary to store half a cycle (0, π) or a quarter cycle. The amplitude of .

[0044] Compared with the table lookup method, the Cordic real-time calculation method has high accuracy. In the process of Cordic calculation, a large number of registers will be used and multiple shift iterations will be performed, which will consume a certain amount of time and space resources. The table lookup method requires a large amount of space to store the sine wave amplitude. When generating a sine wave, it only needs to query the address and read the data, and the time cycle is shorter than that of Cordic.

[0045] In the channel simulation process, it is necessary to generate a large number of sine waves at a high sampling rate. The channel simulator is sensitive to resources and has a large tolerance for time periods. Therefore, in the implementation process, it is necessary to use as few space resources as possible to generate as many sine waves as possible. When using FPGA devices for channel simulation and generating a sine wave curve with a precision of 16 bits, a quarter cycle is stored. The LUT resources consumed by the lookup table are about 2000, and the LUT resources consumed by the pipeline Cordic are about 1000. When a higher-precision sine wave curve is required, the resources required for the lookup table will increase exponentially, while the Cordic method only needs to add a small number of intermediate iteration registers. Therefore, when generating a large number of high-precision sine wave curves at a high sampling rate at the same time, Cordic consumes fewer resources. Due to the characteristics of FPGA, when a project layout and routing is completed, the internal circuit is fixed. If you still want to change the number of sine waves, you need to re-synthesize and layout and routing, and then deploy it to the FPGA chip. This process will consume a lot of time and is not flexible enough.

[0046] Based on this, the present invention proposes a Doppler spread simulation based on Cordic and cascaded integral comb filter interpolation, and a multipath delay simulation method of a Farrow structure filter based on Lagrange interpolation. The method can expand the number of multipaths, and resource consumption does not increase with the increase in the number of sine waves. Data of multiple superimposed sine waves can be quickly generated. During the multipath delay simulation, the filter coefficient is fixed, and the multipath delay effect can be flexibly and effectively achieved. Summary of the invention

[0047] In view of this, the object of the present invention is to provide a method for simulating Doppler spread and multipath delay of wireless channels. When performing channel simulation, the number of sine waves used for Doppler synthesis can be expanded by parameter configuration. This expansion method does not increase FPGA resource consumption and can flexibly and effectively achieve the multipath delay effect of wireless channels.

[0048] In order to achieve the above object, the present invention provides the following technical solutions:

[0049] A method for simulating Doppler spread and multipath delay of a wireless channel, the method comprising the following steps:

[0050] S1: Configure channel parameters: carrier frequency is f c , the terminal moving speed is v, the number of delay taps is K, and the tap delay is τ k , the number of sine waves N used for SOS k , the channel is a Rayleigh channel, and the channel sampling rate is f s , the number of sine waves is expanded by a multiple of M;

[0051] S2: PS side calculates the maximum Doppler frequency f according to the following formulas: mak , rotation angle factor

[0052]

[0053] c represents the speed of light constant; i = 0 represents the in-line component, i = 1 represents the orthogonal component; k is the kth tap, K is the total number of taps; N k is the number of sinusoids used for SOS in the kth tap, indicating the number of indivisible subpaths in a tap;

[0054] Assume that for the kth tap, the discrete Doppler frequencies of the real and imaginary parts of the kth tap are generated by the following formula:

[0055]

[0056] Indicates that the discrete Doppler frequency is the discrete Doppler frequency of the nth subpath of the real part of the Rayleigh channel in the kth tap;

[0057] Indicates that the discrete Doppler frequency is the discrete Doppler frequency of the nth subpath of the imaginary part of the Rayleigh channel in the kth tap;

[0058] Generate 2 sets of random phases at the same time The size of each group is N k , the random phase θ follows a uniform distribution in [0,2π];

[0059]

[0060] is the phase component in the nth subpath of the kth tap, which obeys the uniform distribution of (0,2π]; t represents the fading process time; Represents the frequency and starting phase of a single sine wave respectively. For a single tap, 2N k Sine waves are used to synthesize Rayleigh channels;

[0061] S3: Channel envelope generation, which is realized by the sine wave superposition method.

[0062] For the real part The generation process is as follows:

[0063] Imaginary part The generation process is the same as the real part;

[0064] S3.1: Sine wave superposition, by controlling the expansion multiple M, the pipeline Cordic is time-division multiplexed and the generated sine wave data is superimposed;

[0065] Several Cordic computing modules are preset in the FPGA chip. The operating clock frequency of Cordic is Af s , a single Cordic can be f sA sine wave curve is generated at a sampling rate of fs. When the number of sine waves is expanded by M times, a single Cordic can generate A sine wave curves at a sampling rate of fs s A sampling rate of / M generates MA sine wave curves;

[0066] The multiplexed output results of Cordic are superimposed to obtain the sum of the output results of MA sinusoidal wave curves;

[0067] Sum the superposition values ​​of all Cordics to get the result of one channel sampling;

[0068] That is, according to Perform superposition and summation, that is, sum all the data generated by Cordic to obtain the sampling result of the channel real envelope;

[0069] When Cordic is not multiplexed, the channel sampling interval T = 1 / f s , the sampling rate is f s ;

[0070] When Cordic is time-division multiplexed M times, the channel sampling interval T = M / f s , the sampling rate is f s / M;

[0071] S3.2: CIC filter interpolation. Since the time division multiplexing of Cordic reduces the channel sampling rate, the channel sampling rate is increased by interpolation.

[0072] First, the channel sampling result is input into the third-order comb filter with an input frequency of f s / M;

[0073] Insert M-1 zero values ​​between two consecutive output results of the third-order comb filter, and input the interpolated sequence into the third-order integral filter. The output frequency of the third-order integral filter is f s ;

[0074] S4: The output of the third-order CIC filter is the real and imaginary parts of the Rayleigh channel The envelope of the channel, including the real envelope of the channel synthesized by superposition of sine waves Virtual Envelope Complex Envelope The amplitude.

[0075] S5: Multipath fractional delay of baseband signal is realized by Lagrange interpolation and Farrow structure filter;

[0076] The baseband signal sampling sequence is Y, and the filter interpolation polynomial is designed as:

[0077] P(x)=X·C·Y,V=C·Y

[0078] P(x)=X·V

[0079] P(x)=((((v n x+v n-1 )x+v n-2 )x+v n-3 )x+...+v1)x+v0

[0080] Wherein, the coefficient C is calculated based on the horizontal coordinate of the continuous channel sampling sequence Y;

[0081] For the kth tap, the input coefficient x=τ of the control interpolation polynomial P(x) k , the output of the filter is y(t-τ k ).

[0082] Furthermore, in S5, the Lagrangian interpolation filter adopts a Farrow structure.

[0083] The beneficial effects of the present invention are:

[0084] (1) Through parameter configuration, the number of sine waves used for Doppler synthesis can be easily expanded, thereby increasing the number of multipaths. This expansion method does not increase FPGA resource consumption. The number of sine waves required for channel simulation can be configured after FPGA chip layout and routing, which improves the flexibility of channel simulation.

[0085] (2) Compared with the table lookup method, the present invention adopts a method combining the Cordic calculation method and the CIC interpolation method to effectively reduce the FPGA resource consumption. By time-division multiplexing the pipeline Cordic, the large amount of physical space resources required by the table lookup method is avoided, while the CIC interpolation uses a third-order comb filter and an integral filter for interpolation, avoiding the problem of resource consumption increasing with the increase of the interpolation multiple.

[0086] (3) The present invention can generate more and more accurate sine waves by time-division multiplexing the Cordic calculation module and the CIC filter interpolation, thereby improving the accuracy of channel simulation. The generated channel envelope has statistical characteristics that are closer to the real environment, such as the amplitude characteristics of the Rayleigh distribution and the Doppler diffusion spectrum of a specific shape.

[0087] (4) Through the Lagrange interpolation filter, the baseband signal sampling results can be fractionally delayed to simulate the multipath delay of the channel. The tap coefficient of the interpolation filter is fixed, and the multipath delay can be flexibly adjusted as needed to meet the needs of different channel simulation scenarios.

[0088] (5) The method of the present invention is easy to implement and deploy on FPGA chips. Cordic, CIC filters and Lagrange interpolation are all mature algorithms that can be easily integrated into FPGA design. In addition, through parameter configuration, the parameters of channel simulation, such as the number of multipaths, sampling rate, etc., can be easily adjusted, which is convenient for users to customize according to actual needs.

[0089] Other advantages, objectives and features of the present invention will be described in the following description to some extent, and to some extent, will be obvious to those skilled in the art based on the following examination and study, or can be taught from the practice of the present invention. The objectives and other advantages of the present invention can be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0090] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below in conjunction with the accompanying drawings, wherein:

[0091] Figure 1 It is a schematic diagram of multipath propagation;

[0092] Figure 2 This is a schematic diagram of Rayleigh channel complex envelope simulation;

[0093] Figure 3 This is a schematic diagram of Cordic module reuse;

[0094] Figure 4 This is the overall block diagram of Cordic multiplexing and CIC interpolation;

[0095] Figure 5 It is the structure diagram of Farrow structure filter;

[0096] Figure 6 This is the CIC double interpolation effect diagram;

[0097] Figure 7 This is the Lagrangian fractional delay effect diagram;

[0098] Figure 8 is the statistical histogram of the imaginary and real envelope amplitudes of the Rayleigh channel;

[0099] Fig. 9 is the statistical histogram of the complex envelope amplitude of the Rayleigh channel. DETAILED DESCRIPTION

[0100] The following describes the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0101] Among them, the drawings are only used for illustrative explanations, and they only represent schematic diagrams rather than actual pictures, and should not be understood as limitations on the present invention. In order to better illustrate the embodiments of the present invention, some parts of the drawings may be omitted, enlarged or reduced, and do not represent the size of actual products. For those skilled in the art, it is understandable that some well-known structures and their descriptions in the drawings may be omitted.

[0102] The same or similar numbers in the drawings of the embodiments of the present invention correspond to the same or similar parts; in the description of the present invention, it should be understood that if the terms "upper", "lower", "left", "right", "front", "rear", etc. indicate the orientation or position relationship, they are based on the orientation or position relationship shown in the drawings, which is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the terms describing the position relationship in the drawings are only used for illustrative purposes and cannot be understood as limiting the present invention. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to specific circumstances.

[0103] Configure the channel parameters on the PS side: the carrier frequency is f c , the terminal moving speed is v, the number of delay taps is K, and the tap delay is τ k , the number of sine waves N used for SOS k , the channel is a Rayleigh channel, and the channel sampling rate is f s , the number of sine waves is expanded by a multiple of M;

[0104] The discrete Doppler frequency is calculated by the above configuration and random phase As well as the multiplexing times of the Cordic module and the step length of each sine wave (also called frequency control word), the amplitude of each sine wave is superimposed to obtain a sampling result of the analog channel.

[0105] Due to the time-division multiplexing of Cordic, the channel sampling rate is reduced. In order to compensate for the sampling rate, the channel sampling results are interpolated through the CIC interpolation filter to achieve the effect of increasing the number of sine wave curves and the channel sampling rate;

[0106] The baseband signal sampling result is simulated by the fractional delay through the Lagrangian interpolation filter of the Farrow structure. The tap coefficient of the interpolation filter is fixed and will not be affected by the input delay coefficient τ k changes with the changes.

[0107] In order to describe the present invention more clearly, the principles of the CIC interpolation filter, the Lagrange interpolation method and the Farrow structure filter are now explained.

[0108] 1CIC interpolation filter

[0109] The CIC interpolation filter consists of a comb filter, a zero insertion module, and an integral filter. The comb filter is mainly used to suppress low-frequency noise and prevent aliasing. It attenuates specific frequencies by forming zero points in the frequency domain. The interpolation module inserts zero values ​​into the output sequence of the comb filter to increase the signal sampling rate. However, since the zero insertion operation will introduce image components, the high-frequency image components will be suppressed in the subsequent integral filter. The integral filter is a low-pass filter. This module will enhance the low-pass signal and suppress the high-frequency components.

[0110] Since the discrete sine wave frequency simulated by the channel simulator is very low relative to the channel sampling rate, it is distributed near the low frequency in the spectrum, and its bandwidth is extremely narrow. Therefore, its image component has little effect on the signal itself after being suppressed by the integral filter.

[0111] 2 Lagrange interpolation method

[0112] The Lagrangian method generates a fitting curve that passes through several known points. For example, several known points

[0113] (x0,y0),(x1,y1),(x2,y3),...,(x n ,y n )

[0114] A fitting curve can be generated from these points

[0115] P(x)=l0(x)y(0)+l1(x)y(1)+l2(x)y(2)+...+l n (x)y(n)

[0116] In the above formula, l k (x), k = 0, 1, 2, ... n is called the basis function of the interpolation function. The basis function is determined by the horizontal coordinate of the current known point and the expression is as follows:

[0117]

[0118] Rewrite P(x) into matrix form:

[0119]

[0120] Will l k (x)Expand:

[0121]

[0122] Rewritten in matrix form:

[0123]

[0124] Then P(x) can be expressed as:

[0125]

[0126] Let P(x) = X·C·Y, V = C·Y, then P(x) = X·V

[0127]

[0128] Then P(x)=x n v n +x n-1 v n-1 +x n-2 v n-2 +...xv1+v0

[0129] 3Farrow structure filter

[0130] Rewrite P(x) in the above formula as:

[0131] P(x)=((((v n x+v n-1 )x+v n-2 )x+v n-3 )x+...+v 1) )x+v0

[0132] By reasonably controlling the input coefficient x = τ k , fractional delay can be achieved for the input baseband signal sequence.

[0133] In order to better describe the invention, specific embodiments are now described.

[0134] S1: Configure channel parameters on the PS side: carrier frequency is f c , the terminal moving speed is v, the number of delay taps is K, and the tap delay is τ k , the number of sine waves N used for SOS k , the channel is a Rayleigh channel, and the channel sampling rate is f s, the number of sine waves is expanded by a multiple of M;

[0135] S2: PS side calculates the maximum Doppler frequency f according to the following formulas: max , rotation angle factor

[0136]

[0137] For a single delay tap (taking the kth tap as an example),

[0138] The discrete Doppler frequencies of the real and imaginary parts of the kth tap are generated by

[0139]

[0140] Indicates that the discrete Doppler frequency is the discrete Doppler frequency of the nth subpath of the real part of the Rayleigh channel in the kth tap;

[0141] Indicates that the discrete Doppler frequency is the discrete Doppler frequency of the nth subpath of the imaginary part of the Rayleigh channel in the kth tap;

[0142] Generate 2 sets of random phases at the same time The size of each group is N k , the random phase θ follows a uniform distribution in [0,2π];

[0143]

[0144]

[0145] are the frequency and starting phase of a single sine wave. For a single tap, 2N k Sine waves are used to synthesize Rayleigh channels.

[0146] S3: Channel envelope generation, according to Figure 2 As shown, the channel envelope is realized by the sine wave superposition method.

[0147] For the real part The generation process is as follows (the imaginary part The generation process is the same as the real part):

[0148] S3.1: Sine wave superposition, according to Figure 3 and Figure 4 As shown, by controlling the expansion multiple M, the Cordic is time-division multiplexed and the generated sine wave amplitude values ​​are superimposed;

[0149] Several Cordic computing modules are preset in the FPGA chip. The operating clock frequency of Cordic is Af s , a single Cordic can be f s A sine wave curve is generated at a sampling rate of;

[0150] When the number of sine waves configured is N k <N max When the sampling rate is f, there is no need to interpolate the channel sampling results through the interpolation filter. The several Cordic modules preset inside the chip are sufficient to s Under the condition of k Sine wave.

[0151] When the number of sine waves is expanded by M times, a single Cordic can s A sampling rate of / M generates MA sine wave curves;

[0152] The multiplexed output results of Cordic are superimposed to obtain the sum of the output results of MA sinusoidal wave curves;

[0153] Sum the superposition values ​​of all Cordics to get the result of one channel sampling;

[0154] That is, according to Perform superposition and summation, that is, sum the data generated from all Cordics to obtain the sampling result of the channel real envelope;

[0155] When Cordic is not multiplexed, the channel sampling interval T = 1 / f s , the sampling rate is f s ;

[0156] When Cordic is time-division multiplexed M times, the channel sampling interval T = M / f s , the sampling rate is f s / M;

[0157] S3.2: CIC filter interpolation. Since the time division multiplexing of Cordic reduces the channel sampling rate, the channel sampling rate is increased by interpolation.

[0158] First, the channel sampling result is input into the third-order comb filter with an input frequency of f s / M;

[0159] There are M-1 0 values ​​between two consecutive output results of the third-order comb filter. The sequence after the 0 value insertion is input into the third-order integral filter. The output frequency of the third-order integral filter is f s ;

[0160] The effect of using this interpolation method to double the channel sampling results is as follows: Figure 6 As shown;

[0161] S4: The output of the third-order CIC filter is the real and imaginary parts of the Rayleigh channel The envelope of the channel, including the real envelope of the channel synthesized by superposition of sine waves Virtual Envelope Complex Envelope The amplitude.

[0162] S5: Multipath fractional delay of baseband signal is realized by Lagrange interpolation and Farrow structure filter;

[0163] The baseband signal sampling sequence is Y, and the filter interpolation polynomial is designed as:

[0164] P(x)=X·C·Y,V=C·Y

[0165] P(x)=X·V

[0166] P(x)=((((v n x+v n-1 )x+v n-2 )x+v n-3 )x+...+v1)x+v0

[0167] The coefficient C is calculated based on the horizontal coordinate of the continuous channel sampling sequence Y. In this embodiment, the coefficient C is calculated based on 6 sampling points x = -3, -2, -1, 0, 1, 2, 3. The last column of the coefficient matrix C′ is

[0168] C′0=[0,0,0,1,0,0,0]

[0169] For the kth tap, the input coefficient x=τ of the control interpolation polynomial P(x) k , the output of the filter is y(t-τ k ).

[0170] The fractional delay effect is as follows Figure 7 As shown;

[0171] Figure 8 is a statistical histogram of the imaginary and real envelope amplitudes of the Rayleigh channel generated by the channel simulation method; Fig. 9 It is the statistical histogram of the complex envelope amplitude of the Rayleigh channel generated by the channel simulation method.

[0172] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solution, which should be included in the scope of the claims of the present invention.

Claims

1. A method for simulating Doppler spread and multipath delay of a wireless channel, characterized in that: The method comprises the following steps: S1: Configure channel parameters: carrier frequency is f c , the terminal moving speed is v, the number of delay taps is K, and the tap delay is τ k , the number of sine waves N used for SOS k , the channel is a Rayleigh channel, and the channel sampling rate is f s , the number of sine waves is expanded by a multiple of M; S2: The PS side calculates the maximum Doppler frequency f according to the following formulas: max , rotation angle factor c represents the speed of light constant; i = 0 represents the in-line component, i = 1 represents the orthogonal component; k is the kth tap, K is the total number of taps; N k is the number of sinusoids used for SOS in the kth tap, indicating the number of indivisible subpaths in a tap; Assume that for the kth tap, the discrete Doppler frequencies of the real and imaginary parts of the kth tap are generated by the following formula: Indicates that the discrete Doppler frequency is the discrete Doppler frequency of the nth subpath of the real part of the Rayleigh channel in the kth tap; Indicates that the discrete Doppler frequency is the discrete Doppler frequency of the nth subpath of the imaginary part of the Rayleigh channel in the kth tap; Generate 2 sets of random phases at the same time The size of each group is N k , the random phase θ follows a uniform distribution in [0,2π]; is the phase component in the θth subpath in the kth tap, which obeys the uniform distribution of (0,2π]; t represents the fading process time; Represents the frequency and starting phase of a single sine wave respectively. For a single tap, 2N k Sine waves are used to synthesize Rayleigh channels; S3: Channel envelope generation, which is realized by the sine wave superposition method. For the real part The generation process is as follows: Imaginary part The generation process is the same as the real part; S3.1: Sine wave superposition, by controlling the expansion multiple M, the pipeline Cordic is time-division multiplexed and the generated sine wave data is superimposed; Several Cordic computing modules are preset in the FPGA chip. The operating clock frequency of Cordic is Af s , a single Cordic can be f s The sampling rate generates A sine wave curves. When the number of sine waves is expanded by M times, a single Cordic can generate A sine wave curves at f s A sampling rate of / M generates MA sine wave curves; The multiplexed output results of Cordic are superimposed to obtain the sum of the output results of MA sinusoidal wave curves; Sum the superposition values ​​of all Cordics to get the result of one channel sampling; That is, according to Perform superposition and summation, that is, sum the data generated from all Cordics to obtain the sampling result of the channel real envelope; When Cordic is not multiplexed, the channel sampling interval T = 1 / f s , the sampling rate is f s ; When Cordic is time-division multiplexed M times, the channel sampling interval T = M / f s , the sampling rate is f s / M; S3.2: CIC filter interpolation. Since the time division multiplexing of Cordic reduces the channel sampling rate, the channel sampling rate is increased by interpolation. First, the channel sampling result is input into the third-order comb filter with an input frequency of f s / M; There are M-1 0 values ​​between two consecutive output results of the third-order comb filter. The sequence after the 0 value insertion is input into the third-order integral filter. The output frequency of the third-order integral filter is f s ; S4: The output of the third-order CIC filter is the real and imaginary parts of the Rayleigh channel The envelope of the channel, including the real envelope of the channel synthesized by superposition of sine waves Virtual Envelope Complex Envelope the magnitude of S5: Multipath fractional delay of baseband signal is realized by Lagrange interpolation and Farrow structure filter; The baseband signal sampling sequence is Y, and the filter interpolation polynomial is designed as: P(x)=X·C·Y,V=C·Y P(x)=X·V P(x)=((((v n x+v n-1 )x+v n-2 )x+v n-3 )x+...+v1)x+v0 Wherein, the coefficient C is calculated based on the horizontal coordinate of the continuous channel sampling sequence Y; For the kth delay tap, the input coefficient x=τ of the control interpolation polynomial P(x) k , the output of the filter is y(t-τ k ).

2. The method for simulating Doppler spread and multipath delay of a wireless channel according to claim 1, characterized in that: In S3.1, when the number of configured sine waves is less than or equal to the preset number of basic sine waves, the Cordic module is not multiplexed; when the number of configured sine waves is greater than the preset number of basic sine waves, the Cordic is multiplexed according to the configured expansion multiple.

3. The method for simulating Doppler spread and multipath delay of a wireless channel according to claim 1, characterized in that: In S3.2, the interpolation filter is a CIC filter.

4. The method for simulating Doppler spread and multipath delay of a wireless channel according to claim 1, characterized in that: In S5, the Lagrangian interpolation filter adopts a Farrow structure.

Images