阅读说明：本技术 融合5g nr的leo卫星多普勒频偏变化率估计方法 (LEO satellite Doppler frequency offset change rate estimation method fused with 5G NR ) 是由 张永亮 景小荣 于 2021-05-12 设计创作，主要内容包括：本发明涉及一种基于第五代(5thGeneration,5G)通信新空口(NewRadio,NR)协议与低轨(LowEarthOrbit,LEO)卫星融合下高动态场景中的多普勒频偏变化率估计方法,属于无线通信技术领域。针对5GNR协议与LEO卫星通信系统所融合的高动态场景下存在的多普勒频偏变化率的问题,本发明基于5GNR协议,利用二阶差分方法实现对多普勒频偏变化率的估计。本发明复杂度较低同时具有良好的性能,能够很好的满足LEO卫星通信系统对于实时性的需求。(The invention relates to a Doppler frequency offset change rate estimation method in a high dynamic scene based on the fusion of a New Radio (NR) protocol of fifth generation (5G) communication and a Low Earth Orbit (LEO) satellite, belonging to the technical field of wireless communication. Aiming at the problem of Doppler frequency offset change rate in a high dynamic scene fused by a 5GNR protocol and an LEO satellite communication system, the Doppler frequency offset change rate estimation method is based on the 5GNR protocol and is realized by using a second-order difference method. The invention has low complexity and good performance, and can well meet the real-time requirement of the LEO satellite communication system.)

1. a LEO satellite Doppler frequency offset change rate estimation method fused with 5GNR is characterized in that: the method comprises the following steps:

and according to a 3GPP 5GNR frame structure, the Doppler frequency offset change rate estimation is realized by utilizing a primary synchronization signal in an SS/PBCH block and combining a second-order difference method.

2. The method for estimating the Doppler frequency shift rate under the fusion of 5GNR and LEO satellites as claimed in claim, wherein the position of the SS/PBCH block obtained in the above steps under different carrier frequencies and different subcarrier spacings is used to estimate the Doppler frequency shift rate by using the primary synchronization signal in the SS/PBCH block in combination with an improved second order difference method. In the protocol of the 5GNR frame structure, Primary Synchronization Signal (PSS), Secondary Synchronization Signal (SSS), and Physical Broadcast Channel (PBCH) (hereinafter referred to as SS/PBCH block). According to the number L of SS/PBCH blocks in 5 cases defined in 5G NR_maxAre all even number, and the ith SS/PBPSS and Lth in CH_maxThe distance between PSSs in the/2 + i SS/PBCH blocks is fixed.

Taking CaseA as an example, the position of the PSS in the 1 st SS/PBCH block on the time domain is on the 2 nd OFDM symbol of the half frame, the position of the PSS in the 3 rd SS/PBCH block on the time domain is on the 16 th OFDM symbol of the half frame, and the interval between the two PSSs is 14 OFDM symbols; the position of the PSS in the 2 nd SS/PBCH block in the time domain is on the 8 th OFDM symbol of the half frame, the position of the PSS in the 4 th SS/PBCH block in the time domain is on the 22 OFDM symbols of the half frame, and the interval between the two primary synchronization signals is 14 OFDM symbols. The corresponding PSS pitch in the corresponding SS/PBCH blocks in the other 4 cases is similar to that of the Case A. Since the base station serving the user is kept unchanged in a half frame, the PSS values in a plurality of SS/PBCH blocks in the corresponding half frame are the same. Thus, a second order difference method may be employed to estimate the rate of change of the doppler frequency offset.

Wherein r (k) represents the PSS received by the receiving end, s (k) represents the PSS transmitted by the transmitting end, epsilon is Doppler frequency offset, gamma is Doppler frequency offset change rate, and theta₀Representing the initial phase, n_kIs gaussian noise. Sampling frequency of f_sIf the time interval between two corresponding PSS is T, the sampling distance D between the two corresponding PSS is f_sT. The expression of PSS in the synchronization block corresponding to the other is:

further obtained by using a second order difference method:

similarly, correlating with sample points k +1 and k + D +1 yields the following expression:

the two above equations are autocorrelation to estimate the doppler shift change rate γ:

and then amplitude angle operation is carried out on the formula to obtain the value of the Doppler frequency offset change rate:

where the arg () function is used to calculate the argument of the complex number. The estimated range of the Doppler frequency offset change rate isWherein N is the number of sampling points of a PSS, D represents the interval between sampling points of the corresponding PSS, which is influenced by the subcarrier interval, and the value range of the arg function is (-pi, pi) when the argument operation is carried out]. The SS/PBCH block exists in a pair in a half frame, and the accumulation operation is carried out on all sampling points of the PSS in the corresponding synchronization block so as to reduce the interference of noise to an estimated value, which is specifically expressed as:

The technical field is as follows:

the invention belongs to the technical field of communication, and particularly relates to an estimation method for Doppler frequency change rate.

Background art:

the 5G New air interface (New Radio, NR) is currently the most advanced wireless mobile communication specification, and mainly meets three application scenarios: enhanced mobile broadband, ultra-reliable low-delay communication and mass machine communication. This specification has become the strongest terrestrial communications infrastructure and digital conduit today with the introduction of advanced technology. Although the 5G NR has special radio frequency technology, the 5G ubiquitous coverage of the world, such as open sea and high altitude, cannot be realized. Satellite communication has become an important component of future global communication networks due to the advantages of easy setup, flexible networking, wide coverage range, no environmental restriction and the like. Among different types of satellite networks, a Low Earth Orbit (LEO) communication satellite has the advantages of small transmission loss, short communication delay, high reliability and the like because the LEO communication satellite runs in a Low Orbit. Therefore, the 5G NR mobile communication can be integrated with the LEO mobile communication system, and the advantages of the two systems are fully utilized, thereby meeting the communication service requirement of the global coverage.

With the intensive research of 5G technology, the international satellite communication community has more and more involved the work of 5G standard formulation of 3 GPP. There are references to: the LEO system can be fused with a ground 5G NR network to form an integrated ubiquitous network pattern, and can provide more reliable consistent service experience for users so as to reduce the network deployment cost of operators. However, there are many differences between the LEO communication system and the 5G NR mobile communication system in terms of deployment environment, channel propagation characteristics, etc., and the integration of the two will face many challenges. On the one hand, the 5G NR still uses an Orthogonal Frequency Division Multiplexing (OFDM) transmission method, and the system performance is very sensitive to Frequency offset, and a small Frequency offset may destroy the orthogonality of subcarriers. On the other hand, there is high-speed relative motion between the LEO satellite and the ground reception, which will generate a fast time-varying doppler frequency offset effect, which will seriously affect the reception performance of the ground terminal.

In the LEO satellite communication system fusing 5G NR, research on doppler frequency offset estimation is less, but frequency offset estimation problem in the OFDM system has been widely researched. However, the research on doppler frequency offset change rate estimation based on the OFDM system is still few. Two algorithms for the rate of change of doppler frequency offset are proposed in the related literature, namely, a Second Order Difference (SOD) algorithm and a Maximum Likelihood (ML) algorithm. The second order difference algorithm is less complex but is also less accurate. The accuracy of the maximum likelihood algorithm is higher, but the algorithm involves optimization of a plurality of variables in the solution process and has a larger search range, so that the complexity is high, and the algorithm is difficult to implement for a real-time LEO satellite communication system.

In the above current research analysis aiming at the fusion of the OFDM and LEO satellite communication systems, the research of the patent is based on the problem of carrier synchronization of the LEO satellite communication system under the 5G NR structure framework. The method further improves the protocol related to the data N in the 5G NR by utilizing the original second-order difference calculation, the original second-order difference method utilizes the content of a Cyclic Prefix (CP) in an OFDM symbol as a copy of the length L behind the data N, and the Doppler change rate is estimated by utilizing the phase difference accumulated by the repeated data. But since the value of the doppler change rate is small, the accumulated phase difference within one OFDM symbol is small. In the protocol of the 5G NR frame structure, Primary Synchronization Signal (PSS) and Secondary Synchronization Signal (SSS) and Physical Broadcast Channel (PBCH) (hereinafter referred to as SS/PBCH block) are transmitted over a certain half frame (5 ms). By estimating the doppler rate of change using the accumulated phase difference of the intervals between the primary synchronization signals in the SS/PBCH block, the noise can be averaged to reduce the influence of the noise, as the number of simultaneous sampling points is increased compared to that of the previous second-order difference method. In actual simulation, under different signal-to-noise ratios, the Mean Square Error (MSE) of the doppler frequency offset change rate is improved to a great extent.

Disclosure of Invention

In view of the above, it is a primary object of the present invention to estimate the rate of change of doppler frequency offset using the primary synchronization signal in the SS/PBCH block in the 5G NR protocol without additional system overhead. The method does not need extra system overhead and has lower complexity, compared with the original second-order difference method, the method keeps the lower complexity and greatly improves the performance, and is more suitable for an LEO satellite communication system with higher real-time requirement.

In order to achieve the purpose, the invention provides the following technical scheme:

a LEO satellite Doppler frequency offset change rate estimation method fused with 5G NR comprises the following steps:

according to the 3GPP 5G NR frame structure, the Doppler frequency offset change rate estimation is realized by using a main synchronization signal in an SS/PBCH block and combining a second-order difference method.

The method of claim for estimating the rate of change of Doppler shift under fusion of 5G NR and LEO satellites, wherein the SS/PBCH blocks obtained in the above steps are carried on different carriersThe wave frequency and the positions of different sub-carrier intervals are further utilized to estimate the value of the Doppler frequency offset change rate by combining the PSS in the SS/PBCH block with a second order difference method. In the protocol of the 5G NR frame structure, Primary Synchronization Signal (PSS), Secondary Synchronization Signal (SSS), and Physical Broadcast Channel (PBCH) (hereinafter referred to as SS/PBCH block). According to the number L of SS/PBCH blocks in 5 cases defined in 5G NR_maxAre even number, and the PSS and the Lth in the ith SS/PBCH_maxThe distance between PSSs in the/2 + i SS/PBCH blocks is fixed.

Taking Case A as an example, the position of the PSS in the 1 st SS/PBCH block on the time domain is on the 2 nd OFDM symbol of the half frame, the position of the PSS in the 3 rd SS/PBCH block on the time domain is on the 16 th OFDM symbol of the half frame, and the interval between the two PSSs is 14 OFDM symbols; the position of the PSS in the 2 nd SS/PBCH block in the time domain is on the 8 th OFDM symbol of the half frame, the position of the PSS in the 4 th SS/PBCH block in the time domain is on the 22 OFDM symbols of the half frame, and the interval between the two primary synchronization signals is 14 OFDM symbols. The corresponding PSS pitch in the corresponding SS/PBCH blocks in the other 4 cases is similar to Case A. Since the base station serving the user is kept unchanged in a half frame, the PSS values in the plurality of SS/PBCH blocks in the corresponding half frame are the same. Thus, a second order difference method may be employed to estimate the rate of change of the doppler frequency offset.

Wherein r (k) represents the PSS received by the receiving end, s (k) represents the PSS transmitted by the transmitting end, epsilon is Doppler frequency offset, gamma is Doppler frequency offset change rate, and theta₀Representing the initial phase, n_kBeing gaussian noise, N represents the number of sample points. Sampling frequency f_sIf the time interval between two corresponding PSS is T, the sampling distance D between the two corresponding PSS is f_sT. Then the expression of PSS in the other corresponding synchronization block is:

in the above formula r_k+DRepresents the PSS received by the corresponding receiving terminal, s (k + D) represents the PSS sent by the sending terminal, epsilon is Doppler frequency offset, gamma is Doppler frequency offset change rate, theta is₀Representing the initial phase, n_k+DIs gaussian noise. By using a mixture of [0012 and [0014 ]]And carrying out difference operation.

Similarly, the sampling points k +1 and k + D +1 are used for correlation, and the following expression is obtained:

and (3) carrying out autocorrelation operation on the formulas [0016] and [0018] to estimate the Doppler frequency offset change rate gamma:

and then, carrying out amplitude angle operation on the formula to obtain an estimated value of the Doppler frequency offset change rate:

the estimated range of the Doppler frequency offset change rate isWherein N is the sampling point number of the PSS, D represents the interval between corresponding PSS sampling points, the interval is influenced by the subcarrier interval, and the value range of the arg function is (-pi, pi) when the argument operation is carried out]. The SS/PBCH block exists in pairs in a half frame, and all sampling points in the corresponding PSS are subjected to accumulation operation,therefore, the interference of noise on the estimated value is reduced, which is specifically expressed as:

FIG. 1 is an abstract drawing showing a specific flow of the patent implementation;

FIG. 2 is a complete OFDM symbol;

figure 3 is a complete SS/PBCH block;

FIG. 4 shows the location of sync blocks at different carrier frequencies with 15KHz subcarrier spacing;

FIG. 5 shows the location of sync blocks at different carrier frequencies with 30KHz subcarrier spacing;

FIG. 6 shows the location of sync blocks at different carrier frequencies with 30KHz subcarrier spacing;

fig. 7 shows a specific method in the Case where the carrier frequency in Case a is less than or equal to 3 GHz.

Detailed Description

The following will explain the implementation flow of this patent in further detail with reference to the attached drawings.

The method mainly aims to estimate the Doppler frequency offset change rate by using the PSS in the SS/PBCH block without extra system expenditure, does not need extra system expenditure and has lower complexity, compared with the original second-order difference method, the method has greatly improved holding performance and is more suitable for an LEO satellite communication system with higher real-time requirement. Meanwhile, in order to clearly recognize the original second-order difference method, the specific flow and principle of the original second-order difference method are given firstly, and the defects of the original second-order difference method are pointed out at the same time.

The following is a specific flow and principle of the original second-order difference method, and points out advantages and disadvantages of the method, which helps us to have a clearer understanding of the method provided by the present patent. Fig. 2 is a schematic diagram of an OFDM structure including a CP, where the length of the OFDM is L, the length of the data portion is N, and the content of the CP is a copy of OFDM symbol data. The signal received by the receiving end of the OFDM system is represented by the following form:

wherein r is_kRepresenting the received signal, s_kRepresenting the transmission signal, epsilon₀Representing the value of the Doppler frequency offset, alpha representing the value of the rate of change of the Doppler frequency offset, theta₀Denotes the magnitude of the initial phase offset, n_kRepresenting noise.

Since the content between CP and OFDM part data is the same in the OFDM system, there is a correlation between its sampling points, and from this correlation, the following expression can be obtained:

wherein k is [0, L-1]

The signal received by the receiving end of the above-mentioned OFDM system, ignoring the interference of noise, is brought into the above-mentioned formula, and the correlation function can be converted into the following form:

in the same way, the autocorrelation function of the k +1 th and k + N +1 th sample points can be expressed in the form:

to obtain an estimate of the Doppler rate of change α, a correlation function P is applied_corr(k) And P_corr(k +1) conjugate multiplication yields the following expression:

the amplitude-angle operation is performed on the above formula to obtain an estimated value alpha of the doppler frequency offset change rate estimation:

the second order difference method can be realized by utilizing four sampling points in the CP, and then the Doppler frequency offset change rate is estimated. Meanwhile, in order to better eliminate the interference of noise, the number of sampling points needs to be increased to average the noise, and all P's are added'_corrAccumulating to obtain a more accurate estimated value of the Doppler frequency offset change rate alpha:

the second order difference method for estimating the doppler frequency offset change rate is described above. The second order difference method is less complex and easier to implement, however, in the estimation of the doppler frequency offset change rate, we need to consider the multipath channel transmission problem. The delay of the multipath channel may cause the last data symbol to "contaminate" the CP of the next data symbol. Assuming that the maximum multipath delay of the equivalent baseband signal is H (where the length of H is smaller than the length of cyclic prefix L), that is, there is multipath interference in the first H data of CP, resulting in an increase in the mean square error of the doppler frequency offset change rate.

The original second order difference method was described and summarized above. The method of claim 1, wherein a frame structure of the 5G NR is introduced, and details of the number of SS/PBCH blocks and their positions in the time domain in different frequency bands and different sub-carrier bandwidths are introduced. Since the method proposed by the patent needs to utilize the primary synchronization signal in the SS/PBCH block, partial 3GPP protocol needs to be introduced to obtain the number of SS/PBCH blocks in different frequency bands and different sub-carrier bandwidths, so as to lay down the next algorithm. In 5G NR each SS/PBCH block consists of 240 consecutive subcarriers in the frequency domain, the subcarriers being numbered in ascending order 0 to 239 within the SS/PBCH block and consisting of 4 OFDM symbols in the time domain, the OFDM symbols being numbered in ascending order 0 to 3 within the SS/PBCH block, the structure of the SS/PBCH block being as shown in fig. 3. The period of the SS/PBCH block of the 5G NR is variable and can be configured to be 5, 10, 20, 40, 80 and 160ms, and the SS/PBCH block is transmitted only on a certain half frame (5ms) in each period. The time positions of the candidate SS/PBCH blocks have A, B, C, D, E five cases according to the difference of the subcarrier spacing of the SS/PBCH blocks.

Case A: the subcarrier spacing of the SS/PBCH block is 15KHz, the 1 st OFDM symbol position indication of the candidate SS/PBCH block is {2,8} +14 xn, for a carrier frequency less than or equal to 3GHz, n ═ 0,1, the SS/PHCH block is transmitted on subframe 0,1 of a certain half-frame, for a total of 4 candidate positions (L _ s) (L _ s/p _ ch _ block is transmitted on subframe 0,1 of a certain half-frame)_max4); for a carrier frequency greater than 3GHz and less than or equal to 6GHz, with n being 0,1,2,3, the SS/PBCH block is transmitted on subframe 0,1,2,3 of a certain half-frame, there are 8 candidate positions (L)_max8). Case a is shown in fig. 4.

Case B: the sub-carrier spacing of the SS/PBCH block is 30KHz, the 1 st OFDM symbol position indication of the candidate SS/PBCH block is {4,8,16,20} +28 xn, for a carrier frequency less than or equal to 3GHz, n is 0, the SS/PHCH block is transmitted on sub-frame 0 of a certain half-frame, for a total of 4 candidate positions (L _ candidate position)_max4); for a carrier frequency greater than 3GHz and less than or equal to 6GHz, n is 0,1, SS/PBCH block transmitted on subframe 0,1,2,3 of a certain field, there are 8 candidate locations (L)_max8). Case B is shown in fig. 5.

Case C: the sub-carrier spacing of the SS/PBCH block is 30KHz, the 1 st OFDM symbol position indication of the candidate SS/PBCH block is {2,8} +14 xn, for a carrier frequency less than or equal to 3GHz, n ═ 0,1, the SS/PHCH block is transmitted on subframe 0 of a certain half-frame, for a total of 4 candidate positions (L_max4); for a carrier frequency greater than 3GHz and less than or equal to 6GHz, with n being 0,1,2,3, SS/PBCH block transmitted on subframe 0,1 of a certain field, there are 8 candidate positions (L)_max8). Case C is shown in fig. 6.

Case D: the subcarrier spacing for the SS/PBCH block is 120KHz, and the 1 st OFDM symbol position indication for the candidate SS/PBCH block is 4,8,16,20 +28 xn,for a carrier frequency greater than 6GHz, with n being 0,1,2,3,5,6,7,8,10,11,12,13,15,16,17,18, an SS/PHCH block transmitted on sub-frames 0,1,2,3 of a half-frame, there are 64 candidate positions (L) in total (L_max＝64)。

And (3) CaseE: the subcarrier spacing of the SS/PBCH block is 120KHz, the 1 st OFDM symbol position indication of the candidate SS/PBCH block is {8, 12,16,20,32,36,40,44} +56 xn, for a carrier frequency greater than 6GHz, n ═ 0,1,2,3,5,6,7,8,10,11,12,13,15,16,17,18, SS/PHCH block transmitted on subframes 0,1,2 of a certain half-frame, for a total of 64 candidate positions (L) in the SS/PBCH block (L1, 2)_max＝64)。

The number of SS/PBCH blocks under different carrier frequencies and different sub-carrier bandwidths and the positions of the SS/PBCH blocks are respectively explained above, so as to lay the foundation for introducing a new second-order difference method next.

The method of claim 1 wherein the second order difference method is performed by using the primary synchronization signal in the SS/PBCH block corresponding to the position of SS/PBCH at different carrier frequencies and different sub-carrier spacings obtained in the above step.

Taking Case a as an example, the position of the primary synchronization signal in the 1 st SS/PBCH block in the time domain is on the 2 nd OFDM symbol of the located half frame, the position of the primary synchronization signal in the 3 rd SS/PBCH block in the time domain is on the 16 th OFDM symbol of the located half frame, and the interval between the two primary synchronizations is 14 OFDM symbols; the position of the main synchronization signal in the 2 nd SS/PBCH block in the time domain is on the 8 th OFDM symbol of the located half frame, the position of the main synchronization signal in the 4 th SS/PBCH block in the time domain is on the 22 OFDM symbols of the located half frame, and the interval between the two main synchronization signals is also 14 OFDM symbols. The same is true for the distances between the primary synchronization signals in the other 4 cases. The base station serving the user is kept unchanged in a half frame, and correspondingly, the values of the primary synchronization signals in the plurality of SSBs in the half frame are the same. Therefore, the Doppler frequency offset change rate can be estimated by carrying out second-order differential multiplication on the corresponding main synchronous signals. Expression for a model of doppler rate of change:

wherein, r (k) represents the PSS received by the receiving end, s (k) represents the PSS transmitted by the transmitting end, epsilon is Doppler frequency offset, gamma is Doppler frequency offset change rate, theta₀Representing the initial phase, n_kIs gaussian noise. Sampling frequency of f_sIf the time interval between two corresponding PSS is T, the sampling distance D between the two corresponding PSS is f_sT. Then the expression of the other corresponding PSS is:

in the above formula r_k+DRepresents the PSS received by the corresponding receiving terminal, s (k + D) represents the PSS sent by the sending terminal, epsilon is Doppler frequency offset, gamma is Doppler frequency offset change rate, theta is₀Representing the initial phase, n_k+DIs gaussian noise. Using the original second order difference method to convert [0060 ]]And [0062]And (3) carrying out correlation operation:

similarly, correlating with sample points k +1 and k + D +1 yields the following expression:

and carrying out correlation operation on the two formulas to estimate the Doppler frequency offset change rate gamma:

and then carrying out amplitude angle operation on the [0068] formula to obtain an estimated value of the Doppler frequency offset change rate:

the estimated range of the Doppler frequency offset change rate isWherein N is the number of sampling points of a PSS, D represents the interval between corresponding PSS sampling points, which is influenced by the subcarrier interval and the carrier frequency band, and the range of the arg function is (-pi, pi) when carrying out argument operation]. The SS/PBCH blocks exist in a pair in a half frame, and all sampling points in the corresponding PSS are subjected to accumulation operation so as to reduce the interference of noise on an estimated value, which is specifically expressed as follows: