A low-orbit satellite beamforming method based on channel prediction

Abstract

The present invention discloses a low-orbit satellite beamforming method based on channel prediction, which belongs to the field of wireless communications. The method specifically includes: the low-orbit satellite first sends a pilot sequence to an Internet of Things device through a downlink channel; each Internet of Things device estimates the channel state information based on the received pilot sequence, and then feeds it back to the low-orbit satellite through an uplink channel; the low-orbit satellite predicts the channel state information of the current time slot based on the channel state information of the previous two time slots, and then uses the predicted channel state information to design a transmission beam, and transmits the data signal after beamforming. The present invention provides a simple but effective beamforming method for low-orbit satellites with large-scale Internet of Things device access.

Description

Technical Field

[0001] The present invention relates to the field of wireless communications, and in particular to a low-orbit satellite beamforming method based on channel prediction. Background Art

[0002] Today, the growing number of IoT devices has put forward extremely challenging requirements for future wireless networks, which are expected to provide extremely high throughput, global and seamless coverage, ultra-reliability, low latency and large-scale connectivity. As a key enabler to achieve this ambitious goal, satellite communications can provide continuous and ubiquitous connectivity to areas without adequate Internet access.

[0003] In order to make more active use of limited spectrum, full frequency reuse schemes have been proposed, in which all beams share the same frequency band, thereby improving spectrum efficiency. In this case, advanced signal processing techniques are essential to mitigate inter-beam interference. To this end, precoding techniques from multi-user multiple-input multiple-output MIMO communications have been widely studied in multi-beam satellite systems.

[0004] However, it is well known that the performance of multi-user MIMO precoding depends largely on the quality of the channel state information at the transmitter. Most of the work on precoding for multi-beam satellite communication systems assumes that the transmitter can track the instantaneous channel state information. However, in actual satellite communication systems, due to inherent channel impairments such as large transmission delay and large Doppler effect, it is very difficult to obtain integrated service transmission information. Specifically, in a time division duplex system, the estimated uplink instantaneous channel state information is used for downlink transmission, which may be outdated after the downlink signal reaches the ground IoT device. Similarly, in a frequency division duplex system, the downlink instantaneous channel information is first estimated at each IoT device and then fed back to the satellite, which incurs considerable channel estimation and feedback overhead. Summary of the invention

[0005] The purpose of the present invention is to solve the problems in the prior art such as difficulty in obtaining instantaneous channel state information and large channel estimation overhead, and proposes a low-orbit satellite beamforming method based on channel prediction.

[0006] The specific technical solutions adopted by the present invention are as follows:

[0007] The present invention provides a low-orbit satellite beamforming method based on channel prediction, comprising the following steps:

[0008] 1) A low-orbit satellite equipped with M antennas obtains the downlink channel state estimation information of all IoT devices at the current time t according to a non-orthogonal downlink channel estimation method Where K is the number of IoT devices;

[0009] 2) The low-orbit satellite estimates the downlink channel state based on the information obtained in the past two time slots and Where m is the time slot interval. According to a channel prediction method, the downlink channel state information at the current moment is predicted.

[0010] 3) The low-orbit satellite predicts the downlink channel state information at the current moment based on the information obtained in step 2) According to a beam design method, a corresponding transmission beam w is designed for the IoT device k. k ;

[0011] 4) Based on the transmit beam w k , the low-orbit satellite uses a beamforming method to send the signal s of the kth IoT device k Perform beamforming and then transmit all beamformed signals together;

[0012] 5) After receiving the signal transmitted in step 4), the IoT device decodes the signal it needs.

[0013] Preferably, in step 1), a non-orthogonal downlink channel estimation method is:

[0014] a) The low-orbit satellite transmits a pilot sequence matrix X to all IoT devices, where the transmission power is S and the pilot length is L;

[0015] b) The low-orbit satellite obtains the estimated value of the downlink channel state information between the kth IoT device and the low-orbit satellite based on the pilot sequence matrix x received by the IoT device using the minimum mean square error estimation method

[0016] Preferably, a channel prediction method in step 2) is:

[0017] The low-orbit satellite uses the second-order autoregressive model to obtain the current channel state information based on the channel state estimation information obtained in the past. Where a1=2r d cos(2πf d m) and Both represent model prediction parameters, f d represents the maximum Doppler shift; r d Indicates the peak value of the power spectrum at the maximum Doppler frequency shift, which is generally set to 1. represents the correlation coefficient between the real downlink channel state information and the estimated channel state information, represents the autocorrelation matrix of the channel, represents the Gaussian noise variance, and I represents the identity matrix.

[0018] Preferably, a beam design method in step 3) is:

[0019] a) Initialize the beam in represents the feasible point of the previous iteration, and P is the maximum power of the low-orbit satellite;

[0020] b) Set the signal decoding vector in represents the square of the F-norm; represents the Gaussian noise variance; Indicates the downlink channel state prediction information; the variance coefficient of the channel prediction error Where R k [·] represents the autocorrelation function of channel k; S represents the power of the pilot sequence, and L represents the length of the pilot sequence;

[0021] c) Setting parameters represents the mean square error, where Re{·} represents the real part of a complex number;

[0022] d) By

[0023] make t k I N +S k ≥0, where r k represents the minimum signal-to-interference-noise ratio required, Indicates that the kth IoT device satisfies r k The interruption probability, I N represents the N-dimensional identity matrix, S k ,u k , is an intermediate parameter, γ k and t k is an auxiliary variable, tr(·) represents matrix trace, vec(·) represents matrix column vectorization, and ||·|| represents the 2-norm;

[0024] e) According to step d), use the CVX toolkit to solve w k ;

[0025] f) If the weighted total rate and convergence are achieved, the final w is obtained k , otherwise repeat steps b) to f).

[0026] Preferably, a beam forming method in step 4) is: the low-orbit satellite transmits a beam w k Constructing the total emission signal

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The non-orthogonal downlink channel estimation method proposed in the present invention realizes channel estimation of large-scale Internet of Things devices by using short preamble sequences, effectively solving a series of problems caused by the long preamble sequences. The channel prediction method proposed in the present invention solves the problem of difficulty in obtaining instantaneous channel state information by using past channel state estimation information to predict the current instantaneous channel state information. The beam design method proposed in the present invention has the characteristics of low complexity and strong anti-interference ability. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A block diagram of a low-orbit satellite beamforming method based on channel prediction;

[0030] Figure 2 is a performance comparison of the weighted sum rate and the number of iterations at different maximum powers in the embodiment;

[0031] Figure 3 It is a performance comparison of the weighted sum rate and the maximum power under different numbers of antennas in the embodiment. DETAILED DESCRIPTION

[0032] The present invention is further described and illustrated below in conjunction with the accompanying drawings and specific embodiments. The technical features of each embodiment of the present invention can be combined accordingly without conflicting with each other.

[0033] Example

[0034] The block diagram of the large-scale unlicensed random access method for low-orbit satellite Internet of Things is as follows Figure 1 As shown, the satellite receives the tensor-form leading sequence transmitted by the IoT device, detects the channel state matrix of the IoT device, then determines the activated IoT device according to the channel state matrix and the threshold, and obtains the channel state estimation information of the activated device according to the channel state matrix.

[0035] The present invention provides a low-orbit satellite beamforming method based on channel prediction, which mainly includes the following steps:

[0036] 1) A low-orbit satellite equipped with M antennas obtains the downlink channel state estimation information of all IoT devices at the current time t according to a non-orthogonal downlink channel estimation method Where K is the number of IoT devices;

[0037] 2) The low-orbit satellite estimates the downlink channel state based on the information obtained in the past two time slots and Where m is the time slot interval. According to a channel prediction method, the downlink channel state information at the current moment is predicted.

[0038] 3) The low-orbit satellite predicts the downlink channel state information at the current moment based on the information obtained in step 2) According to a beam design method, a corresponding transmission beam w is designed for the IoT device k. k ;

[0039] 4) Based on the transmit beam w k , the low-orbit satellite uses a beamforming method to send the signal s of the kth IoT device k Perform beamforming and then transmit all beamformed signals together;

[0040] 5) After receiving the signal transmitted in step 4), the IoT device decodes the signal it needs.

[0041] In this embodiment, a non-orthogonal downlink channel estimation method in step 1) is specifically as follows:

[0042] a) The low-orbit satellite transmits a pilot sequence matrix X to all IoT devices, where the transmission power is S and the pilot length is L;

[0043] b) The low-orbit satellite obtains the estimated value of the downlink channel state information between the kth IoT device and the low-orbit satellite based on the pilot sequence matrix x received by the IoT device using the minimum mean square error estimation method

[0044] In this embodiment, a channel prediction method in step 2) is specifically as follows:

[0045] The low-orbit satellite uses the second-order autoregressive model to obtain the current channel state information based on the channel state estimation information obtained in the past. Where a1=2r d cos(2πf d m) and Both represent model prediction parameters, f d represents the maximum Doppler shift; r d Indicates the peak value of the power spectrum at the maximum Doppler frequency shift, which is generally set to 1. represents the correlation coefficient between the real downlink channel state information and the estimated channel state information, represents the autocorrelation matrix of the channel, represents the Gaussian noise variance, and I represents the identity matrix.

[0046] In this embodiment, a beam design method in step 3) is:

[0047] a) Initialize the beam in represents the feasible point of the previous iteration, and P is the maximum power of the low-orbit satellite;

[0048] b) Set the signal decoding vector in represents the square of the F-norm; represents the Gaussian noise variance; Indicates downlink channel state prediction information; represents the variance coefficient of the channel prediction error, where R k [·] represents the autocorrelation function of channel k; S represents the power of the pilot sequence, and L represents the length of the pilot sequence;

[0049] c) Setting parameters represents the mean square error, where Re{·} represents the real part of a complex number;

[0050] d) By

[0051] make t k I N +S k ≥0, where r k represents the minimum signal-to-interference-noise ratio required, Indicates that the kth IoT device satisfies r k The interruption probability, I N represents the N-dimensional identity matrix, S k ,u k , is an intermediate parameter, γ k and t k is an auxiliary variable, tr(·) represents matrix trace, vec(·) represents matrix column vectorization, and ||·|| represents the 2-norm;

[0052] e) According to step d), use the CVX toolkit to solve w k ;

[0053] f) If the weighted total rate and convergence are achieved, the final w is obtained k , otherwise repeat steps b) to f).

[0054] In this embodiment, a beam forming method in step 4) is: the low-orbit satellite transmits a beam w k Constructing the total emission signal

[0055] Computer simulations show that Figure 2 As shown, the present invention proposes a low-orbit satellite beamforming method based on channel prediction. As the number of algorithm iterations increases, the weighted sum rate gradually increases and tends to be stable, so the method has good convergence. In addition, Figure 3 It shows that with the increase of maximum power and number of antennas, the weighted sum rate gradually increases and tends to be stable. Therefore, the low-orbit satellite beamforming method based on channel prediction proposed in the present invention can provide a high-performance data transmission method for low-orbit satellite communication systems.

[0056] The above-described embodiment is only a preferred solution of the present invention, but it is not intended to limit the present invention. A person skilled in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A low-orbit satellite beamforming method based on channel prediction, characterized in that: The steps include: 1) A low-orbit satellite equipped with M antennas obtains the downlink channel state estimation information of all IoT devices at the current time t according to a non-orthogonal downlink channel estimation method k=1,2,…,K; where K is the number of IoT devices; 2) The low-orbit satellite estimates the downlink channel state based on the information obtained in the past two time slots and Where m is the time slot interval. According to a channel prediction method, the downlink channel state information at the current moment is predicted. k=1,2,…,K; 3) The low-orbit satellite predicts the downlink channel state information at the current moment based on the information obtained in step 2) According to a beam design method, a corresponding transmission beam w is designed for the IoT device k. k ; 4) Based on the transmit beam w k , the low-orbit satellite uses a beamforming method to send the signal s of the kth IoT device k Perform beamforming and then transmit all beamformed signals together; 5) After receiving the signal transmitted in step 4), the IoT device decodes the signal it needs; A channel prediction method in step 2) is: The low-orbit satellite uses the second-order autoregressive model to obtain the current channel state information based on the channel state estimation information obtained in the past. Where a1=2r d cos(2πf d m) and Both represent model prediction parameters, f d represents the maximum Doppler shift; r d Indicates the peak value of the power spectrum at the maximum Doppler frequency shift, set to 1; represents the correlation coefficient between the real downlink channel state information and the estimated channel state information, represents the autocorrelation matrix of the channel, represents the Gaussian noise variance, I represents the identity matrix; A beam design method in step 3) is: a) Initialize the beam in represents the feasible point of the previous iteration, and P is the maximum power of the low-orbit satellite; b) Set the signal decoding vector in represents the square of the F-norm; represents the Gaussian noise variance; Indicates the downlink channel state prediction information; the variance coefficient of the channel prediction error Where R k [·] represents the autocorrelation function of channel k; S represents the power of the pilot sequence, and L represents the length of the pilot sequence; c) Setting parameters represents the mean square error, where Re{·} represents the real part of a complex number; d) By make t k I N +S k ≥0, where r k represents the minimum signal-to-interference-noise ratio required, Indicates that the kth IoT device satisfies r k The interruption probability, I N represents the N-dimensional identity matrix, S k ,u k , is an intermediate parameter, γ k and t k is an auxiliary variable, tr(·) represents matrix trace, vec(·) represents matrix column vectorization, and ||·|| represents the 2-norm; e) According to step d), use the CVX toolkit to solve w k ; f) If the weighted total rate and convergence are achieved, the final w is obtained k , otherwise repeat steps b) to f); A beam forming method in step 4) is: the low-orbit satellite transmits a beam w k Constructing the total emission signal 2. A low-orbit satellite beamforming method based on channel prediction according to claim 1, characterized in that: A non-orthogonal downlink channel estimation method in step 1) is: a) The low-orbit satellite transmits a pilot sequence matrix X to all IoT devices, where the transmission power is S and the pilot length is L; b) The low-orbit satellite obtains the estimated value of the downlink channel state information between the kth IoT device and the low-orbit satellite based on the pilot sequence matrix X received by the IoT device using the minimum mean square error estimation method

Images