Frequency offset estimation method and device for non-ground network, and base station

Abstract

The embodiment of the invention provides a frequency offset estimation method and device for a non-ground network and a base station, which are applied to a base station of which part or all facilities are arranged on a satellite or a base station of which the facilities are arranged on the ground, and the method comprises the following steps: the base station receives a format signal sent by terminal equipment through a physical uplink control channel, according to the format signal, executing the following frequency offset compensation operations: performing integer frequency offset estimation, and executing corresponding frequency offset compensation; and / or performing decimal frequency offset coarse estimation, and executing corresponding frequency offset compensation; and if the number of symbols configured by the format signal does not reach a preset standard or a frequency hopping function in a time slot is not activated, decimal frequency offset fine estimation is further carried out, and corresponding frequency offset compensation is executed. The base station provided by the invention can gradually and accurately estimate and compensate the frequency offset on different levels, and finally realizes correct demodulation of the signal.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a frequency offset estimation method, device and base station for a non-terrestrial network. Background Art

[0002] The Physical Uplink Control Channel (PUCCH) is a channel for the terminal to feedback uplink control information to the base station. In 5G, PUCCH has multiple transmission formats, and different formats have different transmission requirements and carry different uplink control information. During the initial access and connection state of the terminal, HARQ (Hybrid Automatic Repeat reQuest) information feedback and SR (Scheduling Request) information will be sent according to the business situation. This information can be transmitted on the PUCCH channel. The specific format to use (such as Format 0, Format 1, Format 2, Format 3 and Format 4) is determined by the base station.

[0003] If the base station fails to correctly demodulate the HARQ information or SR information transmitted by the terminal, it will cause the base station to make an incorrect judgment, which is inconsistent with the expected behavior of the terminal. Repeated occurrences of such incorrect judgments may cause the base station or terminal to believe that the current wireless link is unreliable, thereby triggering other processes such as re-access. For fast-running low-orbit satellites, due to the short coverage time, the cost of incorrect judgments caused by demodulation failures is relatively high. Summary of the invention

[0004] In order to solve the problem of how a base station can quickly and accurately estimate frequency offset, the present invention provides a frequency offset estimation method, device and base station for a non-terrestrial network, so as to enable the base station to correctly demodulate information transmitted by a terminal.

[0005] In a first aspect, the present invention provides a frequency offset estimation method for a non-terrestrial network, which is applied to a base station in which some or all facilities are set on a satellite or a base station set on the ground, and the method comprises:

[0006] The base station receives the format signal sent by the terminal device through the physical uplink control channel, and performs the following frequency offset compensation operations according to the format signal:

[0007] Perform integer frequency offset estimation and perform corresponding frequency offset compensation; and / or

[0008] Make a rough estimate of the fractional frequency offset and perform corresponding frequency offset compensation;

[0009] If the number of symbols configured in the format signal does not reach a preset standard, or the frequency hopping function is not activated in the time slot, further fractional frequency offset fine estimation is performed, and corresponding frequency offset compensation is executed.

[0010] In a second aspect, the present invention further provides a frequency offset estimation device for a non-terrestrial network, which is applied to a base station where some or all facilities are located on a satellite or a base station located on the ground, and the device comprises:

[0011] The primary frequency offset compensation module receives the format signal sent by the terminal device through the physical uplink control channel by the base station, and performs the following frequency offset compensation operations according to the format signal:

[0012] Perform integer frequency offset estimation and perform corresponding frequency offset compensation; and / or

[0013] Make a rough estimate of the fractional frequency offset and perform corresponding frequency offset compensation;

[0014] The advanced frequency offset compensation module is used to further perform fractional frequency offset fine estimation and perform corresponding frequency offset compensation when the number of symbols configured in the format signal does not reach a preset standard or the frequency hopping function in the time slot is not activated.

[0015] In a third aspect, the present invention further provides a base station, which is used to execute the steps of the frequency offset estimation method for a non-terrestrial network as described in any one of the first aspects.

[0016] The frequency offset estimation method, device and base station of a non-terrestrial network provided by an embodiment of the present invention, the method first comprises that the base station receives a format signal from a terminal device through a physical uplink control channel. Then, the base station performs integer frequency offset estimation. This means that the base station will estimate the frequency offset value as an integer multiple of the carrier frequency, and perform corresponding frequency offset compensation. Following the integer frequency offset estimation, the base station performs a coarse estimation of the fractional frequency offset, using an adjustment step size that is smaller than the adjustment step size of the integer frequency offset. If the number of symbols configured by the format signal does not meet the preset standard, or the frequency hopping function is not activated within the time slot, the base station will further perform a fine estimation of the fractional frequency offset. This is a more precise adjustment step, using a very small adjustment step size to finely adjust the frequency to ensure the highest accuracy of frequency offset compensation.

[0017] Therefore, the base station of the present invention can gradually and accurately estimate and compensate for the frequency offset at different levels, and finally realize the correct demodulation of the signal. It not only improves the accuracy of frequency offset adjustment, but also optimizes the processing speed, ensuring that the base station can effectively cope with various signal conditions, thereby improving the communication quality and the overall performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1a This is a schematic diagram of a transparent transmission mode;

[0020] Figure 1b It is a schematic diagram of a regeneration mode;

[0021] Figure 1c is a schematic diagram of another regeneration mode;

[0022] Figure 2 It is a schematic diagram of a non-terrestrial network;

[0023] Figure 3 is a schematic diagram of offset subcarriers;

[0024] Figure 4 is a flow chart of a frequency offset estimation method for a non-terrestrial network;

[0025] Figure 5a is a flow chart of a frequency offset estimation method in a transparent transmission mode;

[0026] Figure 5b is a flow chart of a frequency offset estimation method for a regeneration mode;

[0027] Figure 6 It is a schematic diagram of the communication between the terminal device and the base station;

[0028] Figure 7a It is a schematic diagram of frequency hopping in non-time slot;

[0029] Figure 7b It is a schematic diagram of frequency hopping within a time slot;

[0030] Figure 8a It is a schematic diagram of integer frequency offset estimation;

[0031] Figure 8b It is a schematic diagram of receiving data without moving in a specific case;

[0032] Fig. 9 It is a lookup table for rough estimation of fractional frequency bias;

[0033] Fig.10 It is a schematic diagram of the fractional multiple frequency estimation;

[0034] Fig.11 It is a structural diagram of a frequency offset estimation device for a non-terrestrial network. DETAILED DESCRIPTION

[0035] In the embodiments of the present invention, the term "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the associated objects before and after are in an "or" relationship. In the embodiments of the present invention, the term "plurality" refers to two or more than two, and other quantifiers are similar.

[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions 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 creative work are within the scope of protection of the present invention.

[0037] 3GPP (3rd Generation Partnership Project) defines different types of wireless communication services for different application scenarios and requirements in different application fields. Among them, non-terrestrial network (NTN) is one of the directions of protocol evolution of 3GPP at present. According to the definition of 3GPP, NTN communication scenarios can be divided into transparent transmission mode and regeneration mode. In the "transparent transmission mode" (Transparent satellitebased NG-RAN architecture), the base station is located on the ground and forwards the signal through the satellite; while in the "regenerative mode" (Regenerative satellite based NG-RAN architecture), the base station is partially or completely set on the satellite, so that the signal can be processed directly on the satellite. The main difference between these two modes is the use of satellites and the location of signal processing. However, no matter which mode of networking is used for NTN communication, satellites are required to forward or process signals. For example, in the transparent transmission mode, the base station is located on the ground, and the role of the satellite is mainly to forward the signal received from the ground base station to the user or other base stations. The satellite does not process the signal in this process, but is only responsible for signal relay. In regenerative mode, the satellite not only forwards the signal, but also processes the signal. The satellite in this mode has more complex functions and then forwards the signal.

[0038] In the Long Term Evolution (LTE) standard, the physical uplink control channel (PUCCH) is designed to transmit uplink control information (UCI), including HARQ-ACK (Hybrid Automatic Repeat Request Acknowledgement), Scheduling Request (SR) and Channel State Information (CSI). The format of PUCCH in the LTE system is slightly different from that of 5G NR (New Radio), and multiple formats are defined for different purposes and functional requirements, such as Format 0, Format 1, Format 2, Format 3 and Format 4. Different formats have different transmission requirements and carry different uplink control information. Among them, PRACH, PDSCH, PUSCH and PUCCH are designated physical channels for different types of data transmission. PRACH (Physical Random Access Channel) is used for user equipment to initiate a random access process to access the network. PDSCH (Physical Downlink Shared Channel) is used to transmit downlink data to multiple user equipment. PUSCH (Physical Uplink Shared Channel) is used to transmit uplink data from user equipment. PUCCH (Physical Uplink Control Channel) is used to transmit uplink control information.

[0039] Please refer to Figure 1a , Figure 1a It is a schematic diagram of the transparent transmission mode. The user equipment (UE) communicates with the nearest next-generation base station (gNB) through the new radio user plane (NRUu). The gNB receives these signals and sends the data to the satellite. The satellite receives the signal from the ground and forwards the signal back to the non-terrestrial network gateway (NTN Gateway) on the ground through transparent transmission. The NTNGateway receives the signal forwarded by the satellite and further forwards it to the 5G core network (5GCN) through the NG (next generation) interface. The 5G core network processes the signal and sends the data to the final data network through the N6 interface. The N6 interface refers to the interface connecting the 5G core network and the data network (DN).

[0040] Please refer to Figure 1b , Figure 1bIt is a schematic diagram of the regeneration mode. The user equipment (UE) communicates with the next generation base station (gNB) located on the satellite through the new radio user plane (NRUu). The gNB receives these signals and uses NG over SRI technology for service routing, and then sends the data to the non-terrestrial network gateway (NTN Gateway). The NTN Gateway receives the signal forwarded from the gNB and transmits it to the 5G core network (5G CN) through the NG interface. The 5G core network further processes the signal, including security checks and routing decisions, and sends the data to the final data network through the N6 interface. Among them, NG over SRI refers to the technology of service routing on the NG interface through SRI (Service Routing Identifier).

[0041] Please refer to Figure 1c , Figure 1c This is a schematic diagram of another regeneration mode. The user equipment (UE) communicates with the distribution unit (gNB-DU) of the next-generation base station located on the satellite via the new radio user plane (NR Uu). After receiving and processing the signal, the gNB-DU sends the control information and user data to the non-terrestrial network gateway (NTN Gateway) via the F1 interface. In this process, the F1 over SRI technology is used, that is, efficient service routing is performed via the service routing identifier (SRI) on the F1 interface. After forwarding by the NTN Gateway, the signal is sent to the central unit (gNB-CU) of the next-generation base station. The gNB-CU further processes the signal and then transmits the processed signal to the 5G core network (5GCN) via the NG interface. After completing the necessary signal processing, the 5G core network sends the data to the final data network via the N6 interface.

[0042] Please refer to Figure 2 , Figure 2This is a schematic diagram of a non-terrestrial network. Satellites can be divided into low earth orbiting (LEO), medium earth orbiting (MEO) and geostationary earth orbiting (GEO) according to their orbits. Since LEO and MEO satellites move relative to the earth, their speed relative to the earth's surface increases as their orbital altitude decreases, resulting in a larger Doppler frequency deviation. This frequency deviation has become a major challenge in non-terrestrial network (NTN) communications. Taking a LEO satellite as an example, assuming that it is operating in an orbit at an altitude of 600 kilometers, when the system operating frequency is 2GHz, the maximum Doppler frequency deviation can reach ±48KHz; and when the system frequency is increased to 20GHz, the maximum Doppler frequency deviation can increase to ±480KHz. If the Doppler frequency deviation in NTN communication exceeds the subcarrier spacing of the system, the wireless signal cannot be correctly demodulated without frequency deviation estimation and compensation. Please refer to Figure 3 , Figure 3 This is a schematic diagram of offset subcarriers. In the transmitted signal frequency domain, the signal is transmitted in a specific frequency domain. Due to factors such as the Doppler effect, the received signal frequency domain changes, resulting in the actual received signal position being different from the expected position. This offset causes the received signal to move to an adjacent subcarrier, which is called subcarrier offset. When this offset occurs, the signal that was supposed to be received falls into a no-signal area, resulting in the inability to correctly demodulate the signal.

[0043] Therefore, in order to solve the problem of how a base station can quickly and accurately estimate frequency offset, the present invention provides a frequency offset estimation method, device and base station for a non-terrestrial network, so that the base station can correctly demodulate information transmitted by a terminal.

[0044] Please refer to Figure 4 , Figure 4 A method for estimating frequency offset of a non-terrestrial network is applied to a base station in which some or all facilities are set on a satellite or a base station set on the ground, and the method comprises:

[0045] S410, the base station receives a format signal sent by the terminal device through a physical uplink control channel, and performs the following frequency offset compensation operation according to the format signal:

[0046] Perform integer frequency offset estimation and perform corresponding frequency offset compensation; and / or

[0047] Make a rough estimate of the fractional frequency offset and perform corresponding frequency offset compensation;

[0048] S420: If the number of symbols configured in the format signal does not reach a preset standard, or the frequency hopping function in the time slot is not activated, further fine estimation of the fractional frequency offset is performed, and corresponding frequency offset compensation is executed.

[0049] Specifically, under certain configuration conditions, such as when the number of symbols is small and the frequency hopping function is not enabled, the signal is more susceptible to frequency deviation, which may result in information loss or errors. By performing detailed frequency deviation compensation, these effects can be effectively reduced to ensure that the signal can be correctly demodulated in each transmission time slot. The purpose of step S420 is to ensure that the signal can be accurately demodulated even under undesirable configuration conditions (such as when the number of symbols does not meet the preset standard and the frequency hopping function is not activated in the time slot) by performing fine estimation of fractional frequency deviation and corresponding frequency deviation compensation. This fine frequency deviation processing helps to improve the quality of the signal and the accuracy of demodulation, thereby optimizing the performance of the entire communication system. Exemplarily, the preset standard is that the format signal contains at least four symbols.

[0050] In some embodiments, the format signal may be a PUCCH Format 1 signal specified by the Long Term Evolution standard.

[0051] Specifically, PUCCH Format 1 includes demodulation reference signal (DMRS) symbols and data symbols (DATA), which are arranged at intervals in the time domain. According to the 3GPP protocol, this format signal can use 4 to 14 symbols in the time domain and use one resource block (RB) in the frequency domain. In addition, PUCCH Format 1 supports frequency hopping within a time slot, which can be turned on or off as needed. When turned on, time domain resources are configured in accordance with 38.211Table 6.3.2.4.1-1 (this content is located in the technical standard document of 3GPP), and 2 segments of data symbols and DMRS symbols are allocated, while in the frequency domain, one resource block is occupied at each end of the frequency hopping area.

[0052] based on Figure 4 The frequency deviation estimation method of the non-terrestrial network shown in FIG. 1 is for the NTN communication scenario using non-synchronous satellites. The transparent transmission mode (such as Figure 5a as shown) and regeneration mode (as Figure 5b How does the base station estimate the frequency offset and compensate for it?

[0053] In transparent transmission mode networking, the base station needs to compensate for the Doppler frequency deviation on the feeder link, such as Figure 2The magnitude of the frequency offset shown depends mainly on the satellite altitude and trajectory. When the base station receives PUCCH Format 1 data, the data not only has the Doppler frequency offset caused by the feeder link, but also has the residual frequency offset after the UE pre-compensates the frequency offset. After the two are superimposed, it is impossible to determine the frequency offset range of the actual received data. For example, if the feeder link has a 28KHz frequency offset and the residual frequency offset is 2KHz, the received data has a 30KHz frequency offset. At this time, if the system uses a subcarrier spacing of 30KHz, a frequency offset of an integer multiple of the subcarrier occurs. However, if the residual frequency offset is -2KHz, the received frequency offset is 26KHz, and no frequency offset of an integer multiple of the subcarrier is sent.

[0054] Please refer to Figure 5a , Figure 5a A method for estimating frequency offset in a non-terrestrial network, the method comprising:

[0055] S5a1, the base station receives the format signal sent by the terminal equipment through the physical uplink control channel.

[0056] S5a2, the base station estimates the integer multiple frequency offset of the received signal.

[0057] S5a3, if it is detected that there is an integer frequency offset, the base station performs corresponding frequency offset compensation on the subcarrier according to the estimation result.

[0058] S5a4: If no integer frequency offset is detected, the base station does not need to perform a subcarrier shift operation, that is, the signal has no offset at the integer frequency, and there is no need to perform integer frequency offset compensation, and S5a5 is executed.

[0059] S5a5, the base station performs a coarse estimation of the fractional frequency offset, and performs corresponding frequency offset compensation according to the coarse estimation result.

[0060] The purpose of this step is that even if there is no integer frequency offset, the signal may still have a small frequency offset, and this small deviation may also affect the signal quality and demodulation effect, so it is necessary to perform a rough estimation of the fractional frequency offset first.

[0061] S5a6, detecting whether the number of symbols configured by the format signal meets a preset condition, that is, the number of symbols configured by the format signal does not reach a preset standard, or the frequency hopping function in the time slot is not activated.

[0062] If the preset condition is not met, that is, the number of symbols configured by the format signal does not reach the preset standard, or the frequency hopping function is not activated in the time slot, then S5a7 is executed;

[0063] If the preset condition is met, that is, the number of symbols configured by the format signal reaches the preset standard, and the intra-time slot frequency hopping function is activated, S5a8 is executed.

[0064] S5a7, the base station performs a fine estimation of the fractional frequency offset and performs corresponding frequency offset compensation.

[0065] S5a8, end the operation.

[0066] For the regeneration mode, when the base station receives PUCCH Format 1 data, it needs to determine whether integer frequency offset estimation is required according to the receiving conditions (ie, condition one, condition two, and condition three).

[0067] Please refer to Figure 5b , Figure 5b A method for estimating frequency offset in a non-terrestrial network, the method comprising:

[0068] S5b1, the base station receives the format signal sent by the terminal equipment through the physical uplink control channel.

[0069] S5b2, the base station determines whether the format signal meets the preset conditions.

[0070] The preconditions include condition 1, condition 2 and condition 3:

[0071] Condition 1: The bearer content is the HARQ feedback of MSG4 in the initial access process;

[0072] Condition 2: No uplink scheduling is performed within the preset time interval;

[0073] Condition 3: The last uplink scheduling and demodulation failed.

[0074] If the format signal satisfies any one of the above conditions 1, 2 and 3, step S5b3 is executed.

[0075] If the format signal does not satisfy any of the above conditions 1, 2 and 3, step S5b6 is executed.

[0076] S5b3, the base station estimates the integer multiple frequency offset of the received signal.

[0077] S5b4, if it is detected that there is an integer frequency offset, the base station performs corresponding frequency offset compensation on the subcarrier according to the estimation result.

[0078] S5b5: If no integer frequency offset is detected, the base station does not need to perform a subcarrier shift operation, that is, the signal has no offset at the integer frequency, and there is no need to perform integer frequency offset compensation, and S5b6 is executed.

[0079] S5b6, the base station performs a rough estimation of the fractional frequency offset and performs corresponding frequency offset compensation according to the rough estimation result.

[0080] The purpose of this step is that even if there is no integer frequency offset, the signal may still have a small frequency offset, and this small deviation may also affect the signal quality and demodulation effect, so it is necessary to perform a rough estimation of the fractional frequency offset first.

[0081] S5b7, detecting whether the number of symbols configured by the format signal meets a preset condition, that is, the number of symbols configured by the format signal does not reach a preset standard, or the frequency hopping function in the time slot is not activated.

[0082] If the preset condition is not met, that is, the number of symbols configured by the format signal does not reach the preset standard, or the frequency hopping function is not activated in the time slot, then S5b8 is executed;

[0083] If the preset condition is met, that is, the number of symbols configured by the format signal reaches the preset standard and the frequency hopping function within the time slot is activated, S5b9 is executed.

[0084] S5b8, the base station performs a fine estimation of the fractional frequency offset and performs corresponding frequency offset compensation.

[0085] S5b9, end the operation.

[0086] Please refer to Figure 6 , Figure 6 This is a schematic diagram of the communication between the terminal device and the base station. The process is as follows:

[0087] MSG 1 (first message): sent by the UE through the uplink PRACH channel.

[0088] Carrier content: Random Access Preamble.

[0089] Description: The user equipment (UE) initiates the random access procedure and requests network access by sending a random access preamble.

[0090] MSG 2 (second message): gNB responds via downlink PDSCH channel.

[0091] Carrier content: Random Access Response.

[0092] Description: After receiving the random access preamble, the base station (gNB) sends a random access response through the downlink PDSCH channel, which includes time alignment and uplink resource allocation information.

[0093] MSG 3 (the third message): sent by the UE using the allocated uplink PUSCH channel.

[0094] Carrier content: Scheduled Transmission.

[0095] Description: According to the instruction of gNB, UE sends its connection establishment request or initial data on the allocated uplink PUSCH channel.

[0096] MSG 4 (the fourth message): sent by gNB through the downlink PDSCH channel.

[0097] Carrying content: Contention Resolution.

[0098] Description: The gNB sends a contention resolution message to ensure that no other device sends the same preamble during the random access process.

[0099] At the same time, the UE responds through the uplink PUCCH channel.

[0100] Bearer content: HARQ (Hybrid Automatic Repeat reQuest) feedback response to MSG 4, which may include a scheduling request (SR).

[0101] Description: In response to MSG 4 from the gNB, the UE sends HARQ feedback to confirm the reception status and may include a Scheduling Request (SR) to further ensure the reliability and continuity of communications.

[0102] It can be seen that during the initial access and connection state of the terminal, the terminal needs to demodulate the fourth message (MSG4) of the downlink shared channel (PDSCH) and send HARQ feedback through the physical uplink control channel (PUCCH). If the message is demodulated correctly, the terminal sends an acknowledgment signal (ACK); if the demodulation fails, a negative acknowledgement signal (NACK) is sent. At the same time, the terminal may send a scheduling request (SR) to request uplink data transmission resources. These operations are performed using PUCCH Format 0 or Format 1. If the base station demodulates the HARQ or SR information incorrectly, it may cause an incorrect network response, such as mistakenly believing that the terminal has not received downlink data or does not need uplink resources, thereby affecting the reliability and stability of the network. Frequent errors may trigger re-access or other network correction processes.

[0103] The following is a detailed description of the preset conditions (condition 1, condition 2 and condition 3) in the above step S5b2.

[0104] Condition 1: The bearer content is the HARQ feedback of MSG4 in the initial access process.

[0105] Specifically, during the initial access process, whether the initial frequency deviation caused by the UE's own crystal oscillator is properly corrected and the access process in the NTN communication network has a longer delay than that in the terrestrial NR network, the data after the UE pre-compensates the frequency deviation cannot determine the approximate range of the received data frequency deviation when the base station receives it. If the HARQ feedback of MSG4 carried by PUCCH Format 1 is not detected by the base station, the base station will perform retransmission of MSG4, which will not only waste resources but also seriously increase the time consumption of the UE access network process; so when condition 1 is met, the base station needs to use integer frequency deviation estimation.

[0106] Condition 2: No uplink scheduling is performed within the preset time interval.

[0107] Exemplarily, the step of determining the preset time interval includes: obtaining the maximum Doppler change rate and the subcarrier spacing, and dividing the subcarrier spacing by the maximum Doppler change rate to obtain a first time interval; obtaining a scheduling request period and a maximum number of transmissions, and multiplying the scheduling request period by the maximum number of transmissions minus one to obtain a second time interval; and using the smaller value of the first time interval and the second time interval as the preset time interval.

[0108] For example, the determination of the preset time interval T should be related to the maximum Doppler change rate calculated by the system parameters and the scheduling request information (SR) of the system configuration. The specific method is: first, the first time interval T1 can be determined according to the maximum Doppler change rate calculated by the system parameters and the subcarrier spacing set by the system, T1 = subcarrier spacing / maximum Doppler change rate; secondly, the second time interval T2 is determined according to the SR period and the maximum number of transmissions set by the system, T2 = SR transmission period * (maximum number of transmissions - 1); finally, the time interval T is determined, T = min (T1, T2). For example, if the maximum Doppler change rate calculated by the system parameters is 2KHz / s and the system subcarrier spacing is 30KHz, then T1 = 15 seconds; the SR scheduling period set by the system is 40ms, and the maximum number of scheduling times is 6, then T2 = 0.2 seconds; based on the minimum value of T1 and T2, it can be determined that T is 0.2 seconds; that is, during the 0.2 second UE uplink scheduling, it is impossible to determine whether the UE's own maintenance frequency has a frequency offset, so at this time the base station cannot determine whether the current UE's frequency offset pre-compensation value is reasonable. In order to ensure that the HARQ or SR information carried by PUCCH Format 1 is correctly demodulated, the base station needs to use integer frequency offset estimation.

[0109] Condition 3: The last uplink scheduling and demodulation failed.

[0110] For the UE after access, the base station should record the demodulation results of the uplink channel PUSCH or PUCCH. For example, the demodulation results of PUSCH or PUCCH received in time slot n need to be recorded as demodulation history values, and PUCCHFormat 1 needs to be received in time slot n+m (M>m>0), and there is no PUSCH and PUCCH scheduling between time slot n and time slot n+m-1. At this time, the reception of time slot n+m needs to query the recorded demodulation history value to see whether the demodulation is successful or failed. If the recorded result is a demodulation failure, the base station cannot determine whether it is caused by inaccurate frequency offset estimation, so the base station needs to use integer multiple frequency offset estimation to ensure that the frequency offset estimation will not be wrong due to the limited estimation range.

[0111] Based on condition 1, condition 2 and condition 3, it is determined whether integer frequency offset estimation is required. If the format signal satisfies any one of the above conditions 1, condition 2 and condition 3, integer frequency offset estimation is performed.

[0112] For the fractional frequency offset rough estimation, since it has less calculation and can guarantee the frequency offset estimation range within one subcarrier, the fractional frequency offset rough estimation needs to be done in any case; since both integer frequency offset estimation and fractional frequency offset rough estimation are calculated in the frequency domain of the received data, the fractional frequency offset fine estimation selects the algorithm of frequency domain estimation, which can reduce the amount of calculation caused by time-frequency conversion, and even if the fractional frequency offset fine estimation cannot be used in some scenarios, most of the frequency offset has been compensated for in the previous frequency offset estimation process. The conditions for using fractional frequency offset estimation are as the preset conditions of step S5b7, that is, the number of symbols configured by the format signal does not meet the preset standard, or the frequency hopping function is not activated in the time slot.

[0113] Please refer to Figure 7a , Figure 7b , Figure 7a This is a schematic diagram of frequency hopping in non-time slots. Figure 7b It is a schematic diagram of frequency hopping within a time slot. Figure 7a For PUCCH Format 1 with 4 symbols, white indicates the DMRS symbol occupancy portion, and gray indicates the DATA symbol occupancy portion. In the frequency domain, a resource block consists of 12 subcarriers, and each subcarrier corresponds to a resource element. When PUCCH Format 1 is configured with 4 symbols, Figure 7bAs shown, white represents the part occupied by PUCCH DMRS symbols, gray represents the part occupied by PUCCH Data symbols, and dotted lines represent the non-PUCCH part. At this time, the data is divided into 2 groups, each group has only 1 PUCCH DMRS symbol, and is transmitted using different frequencies. Since at least 2 PUCCH DMRS symbols are required for frequency offset estimation in the frequency domain. Therefore, if this condition is met, that is, the number of symbols configured by the format signal reaches the preset standard, and the frequency hopping function within the time slot is activated, for example, 4 symbols are configured and the frequency hopping function within the time slot is turned on, then the fractional frequency deviation cannot be finely estimated. If the condition is not met, that is, the number of symbols configured by the format signal does not reach the preset standard, or the frequency hopping function within the time slot is not activated, for example, 4 symbols are not configured and the frequency hopping function within the time slot is turned off, then the fractional frequency deviation can be finely estimated and compensated, making the overall frequency offset estimation algorithm more refined.

[0114] In some embodiments, the frequency offset estimation method for a non-terrestrial network proposed in the present invention includes the following steps: assuming that the frequency offset is xy times the subcarrier size (x, y are both integers), firstly using an integer multiple frequency offset estimation algorithm to estimate the integer part of the frequency offset. Then, using the inherent properties of the base sequence, the fractional part is quickly calculated through the fractional multiple frequency bias rough estimation method. Finally, the pilot method is used to accurately estimate the fractional frequency offset, and the estimated residual frequency offset is combined To improve the overall frequency offset estimation range. If the pilot method (i.e. fractional frequency offset precise estimation) cannot be used, Most of the frequency deviation can be eliminated by this method.

[0115] For example: Assume that the frequency offset is 4.7 times the size of the subcarrier, where the integer part is 4 and the fractional part is 0.7. First, use the integer frequency offset estimation algorithm to determine the integer part of the frequency offset. That is, the algorithm identifies that the integer part of the frequency offset is 4 subcarriers. Then, the fractional frequency offset rough estimation method is applied. The fractional part is quickly calculated using the inherent properties of the signal base sequence. If the fractional part obtained by the rough estimate of the fractional frequency offset is 0.6. Here, d (i.e., the accuracy threshold) is assumed to be 0.1. Since the difference between the actual fractional part 0.7 and the fractional frequency offset rough estimate 0.6 is less than d (0.1), the pilot method (i.e., fractional frequency offset fine estimation) can be directly used for precise estimation. Specifically, the residual frequency offset 0.1 (0.7-0.6) is calculated, and the estimated frequency offset is finally corrected to 4.7 (integer part 4 + fractional part 0.6 of the rough estimate + residual 0.1 of the fine estimate).

[0116] Therefore, through the three steps of integer frequency offset estimation, fractional frequency offset coarse estimation and fractional frequency offset fine estimation, the system can accurately estimate and compensate for the total frequency offset, ensure the correct demodulation of the signal and the accurate transmission of the data, and can be applied to communication environments with large and complex frequency offsets, such as satellite communications or communications in high-speed mobile environments. Integer frequency offset estimation is used to adjust the frequency deviation caused by changes in format signal transmission, and the adjustment value is an integer multiple of the carrier frequency. Fractional frequency offset coarse estimation uses a step size that is less than an integer multiple of the carrier frequency, which is less than the range of fine frequency offset adjustment. Fractional frequency offset coarse estimation uses a non-integer multiple of a preset step size to adjust the frequency.

[0117] The following specifically describes integer frequency offset estimation, fractional frequency offset coarse estimation, and fractional frequency offset fine estimation.

[0118] (1) Integer frequency offset estimation:

[0119] In some embodiments, in the above step S410, the steps of performing integer frequency offset estimation and performing corresponding frequency offset compensation include:

[0120] S411a, performing a fast Fourier transform on the received demodulation reference signal, converting it from the time domain to the frequency domain, so as to generate a corresponding frequency domain signal.

[0121] Specifically, the calculation process of integer frequency offset estimation is performed in the frequency domain of the received data. In order to estimate the integer frequency offset of the received signal, the first received DMRS symbol can be used to perform frequency domain similarity calculation.

[0122] S412a, performing integer subcarrier offset processing on the frequency domain signal, and performing frequency domain similarity calculation on the frequency domain signal after each offset and the local reference signal of the base station to determine a similarity value corresponding to each offset.

[0123] Specifically, the received DMRS symbol data is first offset by an integer multiple of subcarriers in the frequency domain, and similarity calculation is performed between the data after each offset and the local reference signal (ie, local pilot). The number of similarity calculations required is determined according to the maximum Doppler frequency offset range preset by the system parameters.

[0124] The Doppler frequency deviation range is calculated based on the satellite's altitude and speed (calculated by Newton's second law) and the preset maximum cosθ value (determined by the cell location and satellite orbit) to calculate the maximum Doppler frequency deviation range ±F. Figure 2 As shown in the figure, for a LEO satellite altitude of 600 km, a system frequency of 2 GHz, and a maximum cosθ value, the maximum range of the Doppler frequency deviation can be calculated to be ±48 kHz, while the subcarrier spacing is 30 kHz.

[0125] S413a, among all the offsets, selecting the offset with the highest similarity as the estimation result of the integer frequency offset, and performing corresponding frequency offset compensation on the frequency domain signal according to the estimation result.

[0126] Please refer to Figure 8a , Figure 8a This is a schematic diagram of integer frequency offset estimation. Assuming the time domain signal is: x(n), the corresponding frequency domain signal is: X(k) = DFT(x(n)), when there is a frequency offset δ: Where δ is the normalized frequency deviation. When δ is an integer: the frequency domain data index is directly shifted by δ, and the corresponding time domain data rotates by δ. When δ is a negative integer, according to Figure 8a The example can be shifted in the negative direction.

[0127] For example, considering that the subcarrier spacing in the above example is 30KHz, a total of 4 shift operations are required to determine the positive and negative directions. The received data is shifted by 0, 1, and 2 subcarriers respectively, where 0 represents the original received data (no subcarrier shift operation is performed), 1 represents the original received data is shifted by one subcarrier, and 2 represents the original received data is shifted by two subcarriers; 3 shifts correspond to 3 groups of data, and the frequency domain similarity between the data after each shift and the local pilot is calculated.

[0128] Similarity calculation:

[0129] Among them, L is the sequence used in the calculation, y i is the received sequence after the i-th shift, k is the k-th element in the sequence, conj is the conjugate calculation, and x(k) is the local pilot sequence.

[0130] Integer frequency offset estimation: i = argmax s(i), where argmax is the index corresponding to the maximum value.

[0131] For example, the similarities corresponding to the frequency domain similarity calculations with the local pilot are 0.7, 0.99, and 0.6, respectively. The offset with the highest similarity is selected as the estimation result of the integer frequency offset. Finally, according to the estimation result, all received data (including DMRS symbols and Data symbols) of the frequency domain signal (such as PUCCH Format 1 signal) are subjected to corresponding subcarrier offset operations to complete the estimation and compensation of the frequency offset. For example, if the highest similarity is 1 offset, then the integer frequency offset result is 1 subcarrier offset.

[0132] In some embodiments, in the above step S410, the step of performing integer frequency offset estimation and performing corresponding frequency offset compensation further includes:

[0133] S414a, if the frequency domain signal cannot be subcarrier offset in a certain direction under certain circumstances, the position of the local reference signal of the base station is adjusted to perform a corresponding offset; after completing the offset operation, the frequency domain similarity calculation is performed again on the frequency domain signal after each offset and the adjusted local reference signal.

[0134] That is to say, it is a special case of integer frequency offset estimation. For example, if an integer offset in the negative direction is required, but the starting index of the received sequence is already 0 and cannot be moved to the left, you can choose not to move the received data, but to shift the local pilot and then calculate the frequency domain similarity.

[0135] Please refer to Figure 8b , Figure 8b This is a schematic diagram of receiving data without moving in a specific case. However, in special cases, such as Figure 8a As shown in the figure, if you want to calculate a negative integer value, and the starting received sequence index is already 0, you cannot move to the left (negative direction of 0) as a whole, such as Figure 8b As shown, there is no sequence in the negative direction of 0; in this case, the received data can be shifted without moving, and the frequency domain similarity can be calculated after the local pilot is shifted. After a shift of -1, the effective alignment data length changes from 7 to 6, that is, L in the similarity calculation formula changes to 6. Therefore, when the starting received sequence index is already 0, the value of the negative integer index cannot be calculated by directly moving the received data. Therefore, the local pilot can be shifted to simulate the effect of moving the received data to the left, thereby calculating the similarity in the frequency domain. This avoids directly moving the received data and meets the need to calculate the negative integer index value.

[0136] (2) Rough estimate of fractional frequency:

[0137] In some embodiments, in the above step 410, the steps of performing a coarse estimation of the fractional frequency offset and performing corresponding frequency offset compensation include:

[0138] S411b, performing a fast Fourier transform on the received demodulation reference signal, converting it from the time domain to the frequency domain, so as to generate a corresponding frequency domain signal.

[0139] S412b, in the frequency domain signal, identifying pairs of elements satisfying the requirement that adjacent frequency domain values ​​are opposite numbers of each other, and performing an addition operation on the pairs of elements to generate a new frequency domain data set.

[0140] S413b, analyzing the signs of the real and imaginary parts of each data point in the new frequency domain data set and their fluctuation range, and estimating the frequency deviation range based on this information.

[0141] S414b, calculating the theoretical frequency according to the preset fractional frequency deviation value and the estimated frequency deviation range, and calculating the actual frequency according to one or more pairs of elements in the selected new frequency domain data set that satisfy the requirement that adjacent frequency domain values ​​are opposite to each other.

[0142] S415b, comparing the similarity between the real part and the imaginary part of the actual frequency and all theoretical frequencies, and selecting the theoretical frequency closest to the actual frequency, so as to perform corresponding frequency offset compensation on the frequency domain signal according to the selected theoretical frequency.

[0143] The entire calculation process of the above steps S411b to S415b is performed in the frequency domain of the received data. The inherent property of the base sequence of PUCCH Format 1 (including DMRS and Data) can be used, that is, adjacent data are opposite to each other. For example, adjacent data in the base sequence (such as x(i) and x(j)) have a relationship of opposite numbers, that is, x(i) = -x(j), j = i+1. According to the formula, directly adding two adjacent frequency domain values ​​that meet the conditions can eliminate most of the inter-carrier interference, so that the common rotation term accounts for a large proportion of the calculation result. At this time, according to the sign of the real and imaginary parts of the calculation results (such as 4 positive and negative combinations) and the fluctuation range, the range of the frequency deviation can be roughly determined, and the roughly determined frequency deviation is used as the result of the rough estimation of the decimal frequency deviation and the data is compensated for the rough estimation of the frequency deviation. Among them, the specific formula derivation is as follows:

[0144]

[0145] is the frequency deviation value relative to SCS (normalized result);

[0146]

[0147] Let X(m) be the frequency domain of x(n); let the sequence length be N;

[0148]

[0149] Let the above formula:

[0150]

[0151] Let: X(i) = -X(j), i+1 = j;

[0152]

[0153]

[0154] The mean value of z(n) is 0, and multiple operations similar to the above formula can reduce the influence of noise; finally:

[0155]

[0156] Please refer to Fig. 9 , Fig. 9 It is a lookup table for rough estimation of fractional frequency bias. Since the base sequence of PUCCH Format 1 is generated according to the 3GPP protocol 38.211 Table 5.2.2.2-2, it can be seen that no matter which row of the table is taken, Calculation, there is a relationship of X(i) = -X(j), i+1 = j (this formula relationship is referred to as condition G). The specific operation is: Assume that the decimal multiple frequency deviation in the above derivation formula The value ranges from -0.9 to 0.9 with a step size of 0.1, and ≠0, according to the formula of the final result in the above derivation formula, different The corresponding different results are calculated on the right side of the formula to obtain the result A(n), where n corresponds to each The index of When n=1, When n=2, and so on; by taking all or one or several groups of element pairs with condition G in the received sequence (X(i) and X(j) count as one element pair), the result B is calculated according to the left side of the equal sign in the above final result formula. If multiple groups are taken, the calculation result needs to be averaged. At this time, compare the signs and sizes of the real and imaginary parts of the result A(n) and the result B. When the result B is closest to the value of the result A(m), the index m corresponds to This is the rough estimated frequency deviation result.

[0157] Therefore, by utilizing the inherent properties of the PUCCH Format 1 base sequence, especially the relationship that adjacent data are inverse numbers, frequency offset estimation can be performed in the frequency domain. By formula derivation and calculation, most of the inter-carrier interference can be eliminated, and the range of frequency offset can be roughly determined based on the sign and fluctuation range of the real and imaginary parts of the calculation results. The corresponding result A(n) and the result B in the received sequence can determine the coarse estimated frequency offset result.

[0158] (3) Fractional frequency bias estimation

[0159] In some embodiments, in the above step S420, the steps of performing fine estimation of fractional frequency offset and performing corresponding frequency offset compensation include:

[0160] S421, performing a fast Fourier transform on the received demodulation reference signal, converting it from the time domain to the frequency domain, so as to generate a corresponding frequency domain signal.

[0161] S422, extracting each symbol from the frequency domain signal, and determining whether the number of symbols exceeds a preset threshold or the frequency hopping function is not activated in the time slot.

[0162] S423, if the number of symbols exceeds the preset threshold or the frequency hopping function is not activated in the time slot, the frequency offset is calculated based on the symbols of the two demodulation reference signals, and the channel estimation results of the same frequency domain resource positions of different symbols are conjugate multiplied and the average value is taken. The frequency offset value can be calculated based on the average value, and then frequency offset compensation is performed on the frequency domain signal.

[0163] In the above steps S421 to S423, the calculation process is processed in the frequency domain of the received data. If there are more than 1 DMRS symbols in the configuration of PUCCH Format1, the conventional method of estimating the frequency offset by the pilot is used to calculate the data ratio factor according to the subcarrier spacing and the number of 2 DMRS symbol spacings used and the length of the cyclic prefix (CP) and data (DATA) symbols of the system to calculate the frequency offset size, and finally estimate the frequency offset according to the data and compensate it.

[0164] Exemplarily, the calculation formula is as follows:

[0165]

[0166] FO(Hz)=δ×SCS.

[0167] δ is the deviation normalized to the subcarrier spacing SCS, N re is the total number of subcarriers involved in the calculation, H1(k) is the channel estimation result of the k-th subcarrier of the first symbol used, H2(k) is the channel estimation result of the k-th subcarrier of the second symbol used, and FO is the estimated offset frequency, in Hz;

[0168] The following formula is the frequency offset range that can be estimated by fractional frequency deviation fine estimation:

[0169]

[0170] Among them, N DMRs is the distance between two DMRS symbols used for estimation, unit = number of symbols, β is the proportion of the data part in the entire OFDM symbol, calculated as follows:

[0171]

[0172] In the above formula, N μ For the data part, is the CP (cyclic prefix) length in symbols.

[0173] Please refer to Fig.10 , Fig.10It is a schematic diagram of fractional frequency bias estimation. In NR, a symbol consists of two parts: CP and Data. Different μ values ​​will affect the subcarrier spacing of the system, thus affecting N in the above formula. μ and For the value of , please refer to section 5.3.1 of 3GPP protocol 38.211 for specific calculation.

[0174] In some embodiments, in the above step S420, the step of performing fine estimation of the fractional frequency offset and performing corresponding frequency offset compensation further includes:

[0175] S425: If the number of symbols extracted from the frequency domain signal does not exceed the preset threshold and the frequency hopping function is activated, the frequency offset estimation and frequency offset compensation operations are not performed.

[0176] For example, if there is only one DMRS symbol, no processing is performed.

[0177] The frequency offset estimation device for a non-terrestrial network provided in the present application is described below. The frequency offset estimation device for a non-terrestrial network described below and the frequency offset estimation method for a non-terrestrial network described above can refer to each other.

[0178] Please refer to Fig.11 , Fig.11 A frequency offset estimation device 1100 for a non-terrestrial network is applied to a base station with some or all facilities located on a satellite or a base station located on the ground, and includes a primary frequency offset compensation module 1110 and an advanced frequency offset compensation module 1120 .

[0179] Exemplarily, the primary frequency offset compensation module 1110 receives a format signal sent by a terminal device through a physical uplink control channel by the base station, and performs the following frequency offset compensation operations according to the format signal:

[0180] Perform integer frequency offset estimation and perform corresponding frequency offset compensation; and / or

[0181] Make a rough estimate of the fractional frequency offset and perform corresponding frequency offset compensation;

[0182] Exemplarily, the advanced frequency offset compensation module 1120 is used to further perform fractional frequency offset fine estimation and corresponding frequency offset compensation when the number of symbols configured in the format signal does not reach a preset standard or the frequency hopping function in the time slot is not activated.

[0183] In some embodiments, the present application further provides a base station, which is used to execute the steps of the frequency offset estimation method for a non-terrestrial network described in the above embodiments.

[0184] It should be noted here that the above-mentioned device and base station provided in the embodiment of the present invention can implement all the method steps implemented in the above-mentioned method embodiment, and can achieve the same technical effect. The parts and beneficial effects that are the same as the method embodiment in this embodiment will not be described in detail here.

[0185] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A frequency offset estimation method for a non-terrestrial network, characterized in that: Applicable to a base station in which some or all facilities are arranged on a satellite or a base station arranged on the ground, the method comprises: The base station receives the format signal sent by the terminal device through the physical uplink control channel, and performs the following frequency offset compensation operations according to the format signal: Perform integer frequency offset estimation and perform corresponding frequency offset compensation; and / or Make a rough estimate of the fractional frequency offset and perform corresponding frequency offset compensation; If the number of symbols configured in the format signal does not reach a preset standard, or the frequency hopping function is not activated in the time slot, further fractional frequency offset fine estimation is performed, and corresponding frequency offset compensation is executed.

2. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: For a base station in which some or all facilities are arranged on a satellite, the method further comprises: The base station determines whether the format signal satisfies any of the following conditions: Condition 1: The bearer content is the HARQ feedback of MSG4 in the initial access process; Condition 2: No uplink scheduling is performed within the preset time interval; Condition 3: The last uplink scheduling and demodulation failed; If the format signal meets any of the above conditions, an integer frequency offset compensation operation is performed.

3. The frequency offset estimation method for a non-terrestrial network according to claim 2, characterized in that: The step of determining the preset time interval includes: Acquire a maximum Doppler change rate and a subcarrier spacing, and divide the subcarrier spacing by the maximum Doppler change rate to obtain a first time interval; Obtaining a scheduling request period and a maximum number of transmissions, and multiplying the scheduling request period by the maximum number of transmissions minus one to obtain a second time interval; The smaller value between the first time interval and the second time interval is used as the preset time interval.

4. The method for estimating frequency offset of a non-terrestrial network according to claim 1, characterized in that: The format signal is a PUCCH Format 1 signal specified by the Long Term Evolution standard.

5. The frequency offset estimation method for a non-terrestrial network according to claim 1 or 2, characterized in that: The format signal includes a demodulation reference signal symbol and a data symbol, and the steps of performing integer frequency offset estimation and performing corresponding frequency offset compensation include: Performing a fast Fourier transform on the received demodulation reference signal to convert it from the time domain to the frequency domain to generate a corresponding frequency domain signal; Performing integer multiple subcarrier offset processing on the frequency domain signal, and performing frequency domain similarity calculation on the frequency domain signal after each offset and the local reference signal of the base station to determine the similarity value corresponding to each offset; Among all the offsets, the offset with the highest similarity is selected as the estimation result of the integer frequency offset, and according to the estimation result, corresponding frequency offset compensation is performed on the frequency domain signal.

6. The frequency offset estimation method for a non-terrestrial network according to claim 5, characterized in that: The steps of estimating the integer frequency offset and performing corresponding frequency offset compensation also include: If the frequency domain signal cannot be subcarrier-shifted in a certain direction under certain circumstances, the position of the local reference signal of the base station is adjusted to perform a corresponding shift; After the shift operation is completed, the frequency domain similarity calculation is performed again on the frequency domain signal after each shift and the adjusted local reference signal.

7. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: The format signal includes a demodulation reference signal symbol and a data symbol, and the steps of performing a coarse estimation of a fractional frequency offset and performing corresponding frequency offset compensation include: Performing a fast Fourier transform on the received demodulation reference signal to convert it from the time domain to the frequency domain to generate a corresponding frequency domain signal; In the frequency domain signal, identifying pairs of elements satisfying the requirement that adjacent frequency domain values ​​are opposite to each other, and performing an addition operation on the pairs of elements to generate a new frequency domain data set; Analyze the signs of the real and imaginary parts of each data point in the new frequency domain data set and their fluctuation range, and estimate the frequency deviation range based on this information; The theoretical frequency is calculated according to the preset fractional frequency deviation value and the estimated frequency deviation range, and the actual frequency is calculated according to one or more pairs of elements in the selected new frequency domain data set that satisfy the requirement that adjacent frequency domain values ​​are opposite to each other; The real part and the imaginary part of the actual frequency are compared with the similarity of all theoretical frequencies, and the theoretical frequency closest to the actual frequency is selected, so as to perform corresponding frequency offset compensation on the frequency domain signal according to the selected theoretical frequency.

8. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: The format signal includes a demodulation reference signal symbol and a data symbol, and the steps of performing fine estimation of fractional frequency offset and performing corresponding frequency offset compensation include: Performing a fast Fourier transform on the received demodulation reference signal to convert it from the time domain to the frequency domain to generate a corresponding frequency domain signal; Extracting each symbol from the frequency domain signal and determining whether the number of symbols exceeds a preset threshold or the frequency hopping function is not activated in the time slot; If the number of symbols exceeds the preset threshold or the frequency hopping function is not activated within the time slot, the frequency offset is calculated based on the symbols of the two demodulation reference signals, and the channel estimation results of the same frequency domain resource positions of different symbols are conjugate multiplied and the average value is taken. The frequency offset value can be calculated based on the average value, and then frequency offset compensation is performed on the frequency domain signal.

9. The method for estimating frequency offset of a non-terrestrial network according to claim 8, characterized in that: The steps of finely estimating the fractional frequency offset and performing corresponding frequency offset compensation also include: If the number of symbols extracted from the frequency domain signal does not exceed a preset threshold and the frequency hopping function is activated, the frequency offset estimation and frequency offset compensation operations are not performed.

10. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: The preset standard is that the format signal contains at least four symbols.

11. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: Integer frequency deviation estimation is used to adjust the frequency deviation caused by changes in format signal transmission, and the adjustment value is an integer multiple of the carrier frequency.

12. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: The fractional frequency offset coarse estimation uses a step size that is smaller than an integer multiple of the carrier frequency, which is smaller than the range of fine frequency offset adjustment.

13. The frequency offset estimation method for a non-terrestrial network according to claim 1, characterized in that: Fractional frequency bias coarse estimation uses non-integer multiples of the preset step size to adjust the frequency.

14. A frequency deviation estimation device for a non-terrestrial network, characterized in that: Applicable to a base station where part or all of the facilities are located on a satellite or a base station located on the ground, the device comprises: The primary frequency offset compensation module receives the format signal sent by the terminal device through the physical uplink control channel by the base station, and performs the following frequency offset compensation operations according to the format signal: Perform integer frequency offset estimation and perform corresponding frequency offset compensation; and / or Make a rough estimate of the fractional frequency offset and perform corresponding frequency offset compensation; The advanced frequency offset compensation module is used to further perform fractional frequency offset fine estimation and perform corresponding frequency offset compensation when the number of symbols configured in the format signal does not reach a preset standard or the frequency hopping function in the time slot is not activated.

15. A base station, characterized in that: The base station is used to execute the steps of the frequency offset estimation method for a non-terrestrial network according to any one of claims 1 to 13.

Images