Large-scale random access method based on asynchronous enhanced time slot ALOHA

Abstract

The invention provides a large-scale random access method based on asynchronous enhanced time slot ALOHA. The large-scale random access method comprises the following steps: step 1, establishing a system basic framework and a frame structure; 2, signal coding modulation and sending transmission are carried out; step 3, oversampling operation; step 4, iterative soft cancellation signal detection; 5, time slot level interference elimination operation is carried out; step 6, an iterative solution process is carried out; and 7, performing simulation verification and performance evaluation. According to the invention, an oversampling technology is utilized to effectively solve the intra-symbol asynchronization phenomenon between users in an enhanced ALOHA-based random access mechanism. Meanwhile, a corresponding over-sampling signal iteration soft elimination detection algorithm is designed, and the recovery of the over-sampling signal can be effectively realized. According to the scheme provided by the invention, the user asynchronization phenomenon can be effectively resisted, and meanwhile, the access throughput of the system can be improved by utilizing the gain generated by the asynchronization phenomenon.

Description

TECHNICAL FIELD

[0001] Random access technology is the key to efficient information transmission between users and ground base stations (or communication satellites). The present invention proposes a large-scale random access method based on asynchronous reinforcement time slot ALOHA, which can effectively utilize the asynchronous characteristics of random access and improve the access capacity of the system, providing support for large-scale ground communication and satellite communication. The present invention belongs to the field of communication and signal processing. BACKGROUND

[0002] Ultra-large-scale connection is one of the key technologies to support future 6G typical application scenarios, such as Internet of Things, Internet of Vehicles, and unmanned aerial vehicle network. In order to establish the initial connection between user terminals and base stations, random access technology is indispensable. Traditional random access technology is mostly based on ALOHA or time slot ALOHA concept, which lacks a mechanism to deal with large-scale user access collision, resulting in a relatively low user carrying rate. Reinforced time slot ALOHA effectively solves the user collision problem by user copy repeated sending and receiver interference cancellation method, which can significantly improve the user carrying rate of the system.

[0003] In reinforced time slot ALOHA, the synchronization of the receivers between users is a prerequisite for implementing interference cancellation operations. However, receiver synchronization needs to be achieved through precise timing information feedback and other steps, which will cause a large amount of signaling overhead, and also goes against the spirit of random access. Existing asynchronous time slot ALOHA schemes only design and solve time slot-level asynchrony, ignoring the phenomenon of symbol-level asynchrony. In fact, due to the randomness of user location, perfect user synchronization and time slot-level asynchrony are difficult to guarantee at the receiver. Therefore, it is necessary to study the symbol-level asynchronous reinforcement time slot ALOHA scheme.

[0004] In order to solve the problem of symbol-level asynchrony, document [1] uses a matched filter and oversampling technology. It analyzes the channel capacity of symbol-level asynchrony in a Gaussian channel, and proves that symbol-level asynchrony between users can significantly improve the channel capacity of the system compared to symbol-level synchronization. This provides a theoretical basis for studying symbol-level asynchronous reinforcement time slot ALOHA. Document [2] studies the non-orthogonal multiple access technology between users with symbol-level asynchrony, and proposes a related detection algorithm, which confirms the advantages of symbol-level asynchrony and provides a practical basis for the research of the present invention.

[0005] REFERENCES

[0006] [1] S. Verdu, "The capacity region of the symbol-asynchronous Gaussian multiple-access channel," IEEE Transactions on Information Theory, vol. 35, no. 4, pp. 733-751, 1989.

[0007] [2] M. Ganji, X. Zou, and H. Jafarkhani, "Asynchronous Transmission for Multiple Access Channels: Rate-Region Analysis and System Design for Uplink NOMA," IEEE Transactions on Wireless Communications, vol. 20, no. 7, pp. 4364-4378, 2021 SUMMARY

[0008] In order to support the random access of a large number of users in the future 6G high-density scene, the application designs a large-scale random access scheme based on asynchronous reinforcement time slot ALOHA. The application considers the symbol-level asynchronous phenomenon between users existing in the actual random access process, processes the user signal through oversampling technology, and proposes a user signal detection method and an interference cancellation method under asynchronous conditions, which can effectively solve the symbol-level asynchronous problem and improve the user carrying rate of the system. The scheme proposed in the application can effectively realize the random access of users in the ground and star-ground large-scale scenarios. The overall block diagram of the application is shown in Figure 1

[0009] Therefore, the application proposes a large-scale random access method based on asynchronous reinforcement time slot ALOHA, and the steps are as follows:

[0010] Step one: system basic framework and frame structure

[0011] This part introduces the basic framework of the system composition of the application, mainly including the communication scene, the user and the base station architecture, and the access time slot resource.

[0012] Consider a single-cell large-scale communication system composed of a single base station (single satellite) and K users. Use k to represent the user index. The user and the base station use a single antenna for transmission and reception. Consider using the reinforcement time slot ALOHA mechanism for transmission. Each frame includes L time slots, where L represents the number of time slots, and l is used to represent the lth time slot. The transmission load of the system is defined as: ​

[0013]

[0014] The message sequence of user k is defined as b k In each frame, user k randomly selects d k time slots and generates an equal number of message sequence copies to be sent in the selected time slots. For each copy, user k adds a sequence of pointers to other copies to it. For the sake of expression, let the message sequence transmitted by user k in the lth time slot be If time slot l is not selected by user k, then

[0015] Let K l denote the number of users actually sending information in time slot l. If K l = 1, it means that the time slot is a non-contention time slot. If K l > 1, it becomes a contention time slot. Let denote the set of user indices actually sending information in time slot l, and the index of the set is k l , denote the index of the k l th user actually sending information in time slot l.

[0016] Step two: signal encoding, modulation and transmission

[0017] This part mainly introduces the signal transmission link of the user, mainly including encoding, modulation, pulse shaping operation.

[0018] User k performs channel encoding on the message sequence transmitted by it in time slot l, to obtain the encoded sequence, denoted as The sequence is subjected to QPSK modulation to obtain a symbol sequence, denoted as The length of the symbol sequence is N.

[0019] The symbol sequence is subjected to pulse shaping operation. Let ψ(t) denote the pulse shaping function, where t denotes continuous time. The present application adopts a rectangular waveform as the pulse shaping function, i.e. where T s denotes the symbol duration.

[0020] The pulse-shaped signal of user k in the lth time slot is denoted as

[0021]

[0022] where, denotes the k th symbol in ​​This represents the discrete-time index of user k within time slot l.

[0023] All users perform the operation in formula (2) and transmit via antenna. The signal received by the receiver in the l-th time slot is:

[0024]

[0025] in, τ represents the channel coefficient of user k in time slot l. k z represents the delay for user k. l (t) represents the additive white Gaussian noise in time slot l. This represents the power of white noise. This invention primarily focuses on 0 ≤ τ. k <T s The situation.

[0026] Step 3: Oversampling operation

[0027] This section mainly introduces the oversampling operation adopted by the receiver to solve the problem of asynchronous user signals.

[0028] Oversampling and signal detection operations are performed on a single time slot. The following section uses time slot 1 as an example to illustrate the oversampling and signal detection process.

[0029] In this invention, it is assumed that K is the actual amount of information transmitted in time slot l. l The relationship between the latency of each user is as follows: in This represents the delay for the first user in time slot l. This represents the delay for the second user in time slot 1, and so on. Indicates the Kth time slot l l Delay for each user. The CCP exists M l Different delays make Indicates the m-th l A type of delay.

[0030] The receiver for signal r l (t) are aligned with M respectively l Sampling operations are performed for different delays, including alignment delay. The sampling operation yielded the first The symbols are:

[0031]

[0032] in, Indicates alignment τ′m in time slot l l The index of the symbol obtained from the sampling operation, ψ * This represents the conjugate of the pulse shaping function.

[0033] symbol It can be represented as:

[0034]

[0035] in, In time slot l, the kth... l Channel coefficients for each user In time slot l, the kth... l The first symbol sequence sent by the user A symbol, In time slot l, the kth... l Discrete-time index for each user Symbols In the sampling noise, α represents the crossover coefficient, which is defined as:

[0036]

[0037] Where ψ(t) represents the pulse shaping function.

[0038] alignment The vector formed by all the symbols obtained from sampling is represented as Arrange all the delayed sampled signals to obtain Where T represents the matrix transpose operation. Vector y l It can be represented as:

[0039] y l =G l H l x l +z l (7)

[0040] Among them, G l H represents the overall cross coefficient matrix. l Let x represent the overall channel coefficient matrix. l Represents the overall transmitted signal vector.

[0041]

[0042] in, This represents the cross coefficient matrix between the signal of the first user and the first sampled signal within time slot l. Indicates the Kth time slot within time slot l l The cross coefficient matrix between the signal of each user and the first sampled signal. This indicates the signal of the first user within time slot l and the signal of the Mth user. l Cross coefficient matrix between subsampled signals Indicates the Kth time slot within time slot ll The signal of the user and the Mth l Cross coefficient matrix between subsampled signals This represents the channel coefficient of the first user within time slot l. Indicates the Kth time slot within time slot l l The channel coefficients of each user, I N This represents an N×N identity matrix. This represents the symbol sequence of the first user within time slot l. Indicates the Kth time slot within time slot l l A sequence of symbols belonging to each user. l Indicates with y l The corresponding colored noise vector. The k-th time slot within time slot l. l The signal of the mth user and the mth l Cross coefficient matrix between subsampled signals The definition of is:

[0043]

[0044] The parameters within the matrix are calculated using formula (6). Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l Crossover coefficients where the index difference of the subsampled signals is 0 Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l Crossover coefficients with an index difference of 1 between the next sampled signals. Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l The crossover coefficient is N-1 times the index difference of the subsampled signals. Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l Crossover coefficients with a subsampled signal index difference of -1 Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l The crossover coefficient is the crossover coefficient with an index difference of -N+1 between the subsampled signals.

[0045] Step 4: Iterative soft cancellation signal detection

[0046] To recover user message sequences from oversampled signals, this invention proposes an iterative soft cancellation signal detection method. The receiver only solves for K... l ≤K max The time slot, where K maxThe receiver supports solving for the maximum number of users in a time slot. The user information in time slot 1 will be used as an example for explanation.

[0047] First, combine formula (7) to express it as follows:

[0048]

[0049] Among them, y l This indicates the received signal after merging. Let x represent the equivalent channel matrix after merging. l The merged transmitted signal, z l This indicates the noise after launch.

[0050] for It is contained in y l 2M l -1 elements.

[0051] Let set Indicates that it contains

[0052] A set of indices of the elements. and The equivalent channel coefficients between them are make Indicates about A vector consisting of all channel coefficients.

[0053] At the same time each With x l 2K l -1 element is involved. Let the set Indicates included The index of the element in the array. Representing vectors The first in A symbol.

[0054] Iterative soft elimination is a sign-by-sign operation. The following section uses detection... Let's take an example to introduce it. In time slot l, the kth... l The first symbol sent by the user. First, the receiver checks all symbols that satisfy... ( Indicates y l Contains (a set of indexes of elements) (Total 2M) l -1) performs interference cancellation operation to obtain the canceled signal, i.e.:

[0055]

[0056] in, Indicates to The signal after interference cancellation

[0057] Symbols With symbols The channel coefficients between them This indicates that all elements whose index belongs to the set... About Symbol index, This indicates feedback from the previous iteration. of The estimated value and the solution process will be introduced later. This represents residual interference. Let all satisfy... of (Total 2M) l The vector composed of the signals after -1 interference cancellation is Its corresponding residual interference vector is Its corresponding noise vector is

[0058] The receiver combines the interference-cancelled signals to obtain the combined signal:

[0059]

[0060] in, This represents the filter vector used for information merging. H This represents the conjugate transpose. The result of the filter vector is:

[0061]

[0062] in, Indicates about A vector composed of all channel coefficients. E represents the desired operation. Let the covariance matrix of the noise vector have the element in the i-th row and j-th column as follows: in G represents the overall cross coefficient matrix. l The Middle row N(k) l -1)+1 column elements.

[0063] According to the Gaussian approximation formula, the combined signal can be approximately represented as:

[0064]

[0065] in, This represents the statistical mean. Indicates statistical variance. This represents normalized noise that follows a complex Gaussian distribution with a mean of 0 and a variance of 1.

[0066] symbol The log-likelihood ratio (LLR) between the real and imaginary parts is:

[0067]

[0068]

[0069] Here, Re() and Im() represent the operations of finding the real part and the imaginary part, respectively.

[0070] Receiver pair All elements undergo the aforementioned interference cancellation and information merging operations, i.e., formulas (12) to (17). The receiver obtains the encoded sequence. The log-likelihood ratio sequence is denoted as Its input is fed into a decoder to obtain the encoded sequence. The posterior log-likelihood ratio sequence is denoted as...

[0071] The receiver feeds back the element-by-element extrinsic information estimation results to the symbol. Used for the next interference cancellation iteration. The calculation of the element-wise extrinsic information estimation result is divided into two steps: first, the element-wise extrinsic information is calculated. Let... They respectively represent mappings to Bits in the encoded sequences of real and imaginary parts. Regarding symbols. The result of the element-by-element extrinsic information calculation is as follows:

[0072]

[0073] in, Indicates to The signal after interference cancellation Represents about bits Element-by-element external information Represents about bits element-wise external information, () * The expression indicates the conjugate operation, and Var() indicates the variance operation.

[0074] Then the feedback is... about The estimated value is the result of the calculation:

[0075]

[0076] in, This indicates the bit output of the channel decoder. The posterior information, This indicates the bit output of the channel decoder. The posterior information, where 1j represents the imaginary unit.

[0077] Through the iterative steps described above, the receiver recovers user information within each time slot. The receiver can then use verification techniques such as channel coding to determine whether the information recovery is correct.

[0078] Step 5: Time-slot-level interference cancellation operation

[0079] The interference cancellation operation is described below using time slot 1 as an example. The receiver, by successfully resolving the pointer information in the user information, can locate the time slot where other copies of the user's data have been successfully transmitted. Assume there are a total of [number missing] copies in time slot 1. The user who has successfully solved the information. Let κ l This represents the set of indices containing information about users who have successfully solved the problem. Within time slot 1, the receiver regenerates user information based on the information of the solved users and the selected time slot location. Subsequently, the regenerated user information is used for interference cancellation. The interference cancellation operation and its results are as follows:

[0080]

[0081] in, Indicates y l The signal after interference cancellation Indicates the indicator factor, when The value is 1 for the time and 0 for the rest.

[0082] After interference cancellation In the alignment of user k, where, The signal from the set will be fed back to the detector for the next round of iteration. Once the solved user information is eliminated, it will be removed from the set. Eliminate the indexes of solved users in the set. Corresponding K l It has also changed.

[0083] Step Six: Iterative Solution Process

[0084] The receiver repeats the information detection and interference cancellation processes in steps four and five to finally solve for all user information.

[0085] Step 7: Simulation Verification and Performance Evaluation

[0086] Simulations were conducted on the throughput of the proposed system as a function of user load factor under both terrestrial Rayleigh channels and satellite channels. The simulation results reveal that in large-scale communication scenarios, the proposed scheme can effectively combat user asynchrony issues and simultaneously leverage asynchrony to improve user access throughput.

[0087] The advantages and beneficial effects of this invention are as follows: This invention effectively solves the intra-symbol asynchrony phenomenon between users in the enhanced ALOHA random access mechanism by utilizing oversampling technology. Simultaneously, this invention designs a corresponding iterative soft cancellation detection algorithm for oversampled signals, which can effectively recover the oversampled signals. The proposed scheme can effectively combat user asynchrony and simultaneously utilize the gain generated by the asynchrony phenomenon to improve the system's access throughput. [Attached Image Description]

[0088] Figure 1 This is a block diagram of the overall system of the present invention.

[0089] Figure 2 The figure shows the simulation results of the system throughput under the ground Rayleigh channel.

[0090] Figure 3 The figure shows the simulation results of the system throughput under satellite channel conditions.

Detailed Implementation Methods

[0091] To make the objectives, technical solutions, and advantages of this invention clearer and more understandable, the invention will now be described in further detail with reference to the accompanying drawings and technical solutions.

[0092] Step 1: Basic System Framework and Frame Structure

[0093] Consider a large-scale communication system in a single cell consisting of a single base station (single satellite) and K users. The enhanced slotted ALOHA mechanism is used for transmission, with each frame comprising L slots (L = 100), denoted by l.

[0094] The binary message sequence of user k is defined as b k In each frame, user k randomly selects d. k User k generates an equal number of message sequence copies and sends them in the selected time slots. For each copy, user k adds a sequence of pointers to the other copies. For ease of explanation, let the message sequence transmitted by user k in the l-th time slot be _l_. If time slot l is not the time slot selected by user k, then

[0095] Let K l This represents the number of users actually transmitting information in time slot l. If K of time slot l... l =1 indicates that the time slot is a non-contention time slot. If Kl If the value is greater than 1, it becomes a contention slot. Let... Let k represent the set of user indices that actually transmit information in time slot l. l , In time slot l, the kth... l An index of the users who actually sent the information.

[0096] Step Two: Signal Encoding, Modulation, and Transmission

[0097] User k sends a sequence of messages in time slot l. Channel coding is performed, and this part considers using low-density parity-check codes (LDPC) to obtain the encoded sequence, represented as follows: QPSK modulation of the sequence yields a symbol sequence, represented as follows: The length is N.

[0098] This invention performs pulse shaping on a symbol sequence, considering a rectangular pulse waveform as the shaping signal. The pulse-shaped signal of user k in the l-th time slot is shown below. This can be obtained through the pulse shaping operation in formula (2). All users perform the operation in formula (2) and transmit via antenna. The signal received by the receiver in the l-th time slot is given by formula (3).

[0099] Step 3: Oversampling operation

[0100] Oversampling and signal detection operations are performed on a single time slot. In this invention, it is assumed that K of the information actually transmitted in time slot l is... l The relationship between the latency of each user is as follows: in This represents the delay for the first user in time slot l. This represents the delay for the second user in time slot 1, and so on. Indicates the Kth time slot l l Delay for each user. The CCP exists M l Different delays make Indicates the m-th l A type of delay.

[0101] The receiver for signal r l (t) are aligned with M respectively l Sampling operations are performed for different delays to obtain the alignment delay τ′m l The sampling operation yielded the first a symbol The sampling operation is given by formula (4).

[0102] By calculating the cross coefficient in formula (6), the receiver can obtain the direct relationship between the transmitted symbol and the received signal in formula (5). To facilitate signal detection by the receiver, all sampled signals are represented as vectors, and their relationship with the transmitted symbol is obtained, as shown in formula (7). In formula (7), G... l H represents the overall cross coefficient matrix. l Let x represent the overall channel coefficient matrix. l The total transmitted signal vector is represented by the three vectors, which are calculated by formulas (8), (9), and (10).

[0103] Step 4: Iterative soft cancellation signal detection

[0104] The receiver performs iterative soft cancellation signal detection on the received signal to recover the signal transmitted by the user. The signal detection operation is processed separately for each time slot. The receiver only solves for the condition K here. l ≤K max The time slot, where K max The receiver supports the maximum number of users in the time slot to be solved. First, the receiver represents the signal as the combined signal in formula (11).

[0105] Subsequently, the receiver finds y according to the signal relationship expressed in formula (11). l Contains A set of element indices and The equivalent channel coefficients between them are Indicates about A vector composed of all channel coefficients. Simultaneously, the receiver, based on the signal relationships expressed in formula (11), finds x... l Included in A set of indices of elements in

[0106] Based on the above set representation, the receiver performs a symbol-by-symbol iterative soft cancellation operation. In each iteration, the receiver first performs the interference cancellation operation in formula (12) to obtain the canceled signal. Secondly, the receiver uses formula (13) to combine the information according to the minimum mean square error criterion to obtain the combined signal. The filter vector used for information merging in formula (13) It can be obtained from formula (14).

[0107] According to the Gaussian approximation formula, the combined signal can be approximately represented by formula (15). Using the combined signal, the receiver calculates the log-likelihood ratio mapped to the real and imaginary parts of the symbol, using formulas (16) and (17).

[0108] The receiver calculates the encoded sequence using the calculation methods in formulas (12)-(17). The sequence of log-likelihood ratios of all bits in the set is expressed as: The sequence is input into the channel decoder to obtain the encoded sequence. The posterior log-likelihood ratio sequence is denoted as...

[0109] The receiver feeds back the element-by-element extrinsic information estimation results to the symbol. Used for the next interference cancellation iteration. The calculation of the element-wise extrinsic information estimation result is divided into two steps: first, the element-wise extrinsic information is calculated. Let... They respectively represent mappings to Bits in the encoded sequence of real and imaginary parts.

[0110] Using formulas (18) and (19), we obtain the information about bits respectively. Element-wise external information Represents about bits Element-wise external information Finally, the receiver uses formula (20) to feed back the estimated symbol result and performs the next signal detection operation.

[0111] Step 5: Time-slot-level interference cancellation operation

[0112] The receiver, by successfully resolving the pointer information in the user information, can locate the time slot where the user has successfully sent other copies. Assume there are a total of [number missing] copies in time slot l. The user who has successfully solved the information. This represents the set of indices containing information about users who have successfully solved the problem. Within time slot 1, the receiver regenerates user information based on the information of the solved users and the selected time slot location. Subsequently, the regenerated user information is used for time slot-level interference cancellation operations. The time slot-level interference cancellation operation is performed using formula (21).

[0113] After interference cancellation In the alignment of user k, where, The signal from the set will be fed back to the detector for the next round of iteration. Once the solved user information is eliminated, it will be removed from the set. Eliminate the indexes of solved users in the set κ. l Corresponding K l It has also changed.

[0114] Step Six: Iterative Solution Process

[0115] The receiver repeats the information detection and interference cancellation processes in steps four and five to finally solve for all user information.

[0116] Step 7: Simulation Verification and Performance Evaluation

[0117] Simulations were conducted on the throughput of the proposed system as a function of user load factor under both terrestrial Rayleigh channels and satellite channels. The simulation results reveal that in large-scale communication scenarios, the proposed scheme can effectively combat user asynchrony issues and simultaneously leverage asynchrony to improve user access throughput.

[0118] First, consider a terrestrial Rayleigh channel. Assume 100 time slots and a system load ξ ranging from 10% to 200%. Assume a signal-to-noise ratio of 10dB and that users employ standard 5G LDPC coding with a code rate of 0.5. Of all users, 40% send 2 copies and 60% send 3 copies. Assume user latency τ. k In the set {0, 0.1T} s ,…,0.9T s The data is uniformly distributed within the range. When detecting information, the receiver only considers K. l Time slots ≤ 2. Simulation results are as follows: Figure 2 , Figure 3 As shown.

[0119] Depend on Figure 2 It is understood that the proposed solution eliminates the adverse effects of user asynchrony through oversampling technology, and can effectively utilize user asynchrony, thereby improving throughput by 20%-30% compared to the existing enhanced time-slot ALOHA under user synchronization.

[0120] The following simulation considers the throughput of the proposed system as a function of user load rate under satellite channel conditions. A Ricean channel with a Ricean factor of 10 is used to model the satellite channel. The number of time slots L = 100, and the system load ξ ranges from 10% to 250%. The signal-to-noise ratio is 4.85 dB, and users employ standard 5G LDPC coding with a code rate of 0.5. Among all users, 50% send 2 copies, and 50% send 3 copies. Assume the user latency τ... k In the set {0, 0.1T} s ,…,0.9T s The data is uniformly distributed within the range. When detecting information, the receiver only considers K. l Time slots ≤3. Simulation results are shown in the figure.

[0121] Depend on Figure 3As can be seen, under satellite channel conditions, the proposed solution can still eliminate the adverse effects of user asynchrony through oversampling technology. Compared with the existing enhanced slotted ALOHA solution under user synchronization, the proposed solution can improve throughput by 80%.

Claims

1. A large-scale random access method based on asynchronous reinforced slotted ALOHA, characterized in that, The steps are as follows: Step 1: Basic System Framework and Frame Structure The basic framework of the system includes communication scenarios, user and base station architecture, and access time slot resources; In a large-scale communication scenario in a single cell consisting of a single base station and K users, k represents the user index; both users and the base station use a single antenna for transmission and reception; the enhanced time slot ALOHA mechanism is used for transmission, and each frame includes L time slots; Step Two: Signal Encoding, Modulation, and Transmission User k sends a sequence of messages in time slot l. Channel coding is performed to obtain the encoded sequence, represented as QPSK modulation of the sequence yields a symbol sequence, represented as follows: The length is N; Step 3: Oversampling operation Oversampling and signal detection operations are performed on a single time slot. Step 4: Iterative soft cancellation signal detection An iterative soft cancellation signal detection method is proposed; the receiver only solves for K. l ≤K max The time slot, where K max The receiver supports solving for the maximum number of users in a time slot; Step 5: Time-slot-level interference cancellation operation The receiver, by using the pointer information in the successfully solved user information, locates the time slot where the user has successfully sent other copies; assuming there are a total of [number missing] copies in time slot l. The users who have successfully solved the information; let This represents the set of indices containing information about users who have successfully solved the problem; within time slot 1, the receiver regenerates the user information based on the information of the users who have solved the problem and the selected time slot position; subsequently, the regenerated user information is used for interference cancellation operations; Step Six: Iterative Solution Process The receiver repeats the information detection and interference cancellation process in steps four and five to finally solve for all user information. Step 7: Simulation Verification and Performance Evaluation Simulations were performed on throughput as a function of user load factor under both terrestrial Rayleigh channels and satellite channels.

2. The large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 1, characterized in that: In step one, L represents the number of time slots, and l represents the l-th time slot; the system's transmission load is defined as: The message sequence for user k is defined as b k In each frame, user k randomly selects d. k In the l-th time slot, user k generates an equal number of message sequence copies and sends them in the selected time slots. For each copy, user k adds a sequence of pointers to other copies. For ease of description, let the message sequence transmitted by user k in the l-th time slot be (The original text is incomplete and requires further context). If time slot l is not the time slot selected by user k, then 3. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 2, characterized in that: Let K l This represents the number of users actually sending information in time slot l; if K of time slot l l =1 indicates that the time slot is a non-contention time slot; if K l If the value is greater than 1, it becomes a contention slot; let Let k represent the set of user indices that actually transmit information in time slot l. l , In time slot l, the kth... l An index of the users who actually sent the information.

4. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 1, characterized in that: In step two, a pulse shaping operation is performed on the symbol sequence; let ψ(t) denote the pulse shaping function, where t represents continuous time; a rectangular waveform is used as the pulse shaping function, i.e. Among them, T s Indicates the duration of the symbol; The signal after pulse shaping by user k in the l-th time slot is utilized express: in, express The first in A symbol, This represents the discrete-time index of user k within time slot l; All users perform the operation in formula (2) and transmit via antenna; the signal received by the receiver in the l-th time slot is: in, τ represents the channel coefficient of user k in time slot l. k z represents the delay for user k. l (t) represents the additive white Gaussian noise in time slot l. This represents the power of white noise.

5. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 1, characterized in that: In step three, let K be the actual amount of information transmitted in time slot 1. l The relationship between the latency of each user is as follows: in, This represents the delay for the first user in time slot l. This represents the delay for the second user in time slot l. Indicates the Kth time slot l l Delay for individual users; The CCP exists M l Different delays make Indicates the m-th l Delay; The receiver for signal r l (t) are aligned with M respectively l Sampling operations are performed for different delays, including alignment delay. The sampling operation yielded the first The symbols are: in, Indicates alignment in time slot l The index of the symbol obtained from the sampling operation, ψ * This represents the conjugate of the pulse shaping function.

6. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 5, characterized in that: symbol Represented as: in, In time slot l, the kth... l Channel coefficients for each user In time slot l, the kth... l The first symbol sequence sent by the user A symbol, In time slot l, the kth... l Discrete-time index for each user Symbols In the sampling noise, α represents the crossover coefficient, which is defined as: Wherein, ψ(t) represents the pulse shaping function; alignment The vector formed by all the symbols obtained from sampling is represented as Arrange all the delayed sampled signals to obtain Where T represents the matrix transpose operation; vector y l Represented as: y l =G l H l x l +z l (7) Among them, G l H represents the overall cross coefficient matrix. l Let x represent the overall channel coefficient matrix. l This represents the overall transmitted signal vector.

7. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 6, characterized in that: kth time slot l l The signal of the mth user and the mth l Cross coefficient matrix between subsampled signals The definition of is: The parameters within the matrix are calculated using formula (6). Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l Crossover coefficients where the index difference of the subsampled signals is 0 Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l Crossover coefficients with an index difference of 1 between the next sampled signals. Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l The crossover coefficient is N-1 times the index difference of the subsampled signals. Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l Crossover coefficients with a subsampled signal index difference of -1 Indicates the kth time slot within time slot l l The symbol sequence of each user and the m-th time slot l l The crossover coefficient is the crossover coefficient with an index difference of -N+1 between the subsampled signals.

8. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 7, characterized in that: Formula (7) can be combined into the following expression: Among them, y l This indicates the received signal after merging. Let x represent the equivalent channel matrix after merging. l The merged transmitted signal, z l This indicates the noise after launch; for It is contained in y l 2M l -1 elements; Let set Indicates that it contains A collection of the indices of the elements; and The equivalent channel coefficients between them are make Indicates about A vector consisting of all channel coefficients; At the same time each With x l 2K l -1 element is related; let the set Indicates included The index of the element in the array. Representing vectors The first in One symbol; The iterative soft elimination operation is a sign-by-sign operation; where, In time slot l, the kth... l The first symbol sent by the user; first, the receiver checks all symbols that satisfy... of After performing interference cancellation, the signal after cancellation is obtained, i.e.: in, Indicates to The signal after interference cancellation Symbols With symbols The channel coefficients between them This indicates that all elements whose index belongs to the set... About Symbol index, This indicates feedback from the previous iteration. of The estimated value; Indicates residual interference; let all satisfy of The vector formed by the interference cancellation signal is The corresponding residual interference vector is Its corresponding noise vector is The receiver combines the interference-cancelled signals to obtain the combined signal: in, This represents the filter vector used for information merging. H This represents the conjugate transpose; the result of the filter vector is: in, Indicates about A vector composed of all channel coefficients; E represents the expectation operation. Let the covariance matrix of the noise vector have the element in the i-th row and j-th column as follows: in G represents the overall cross coefficient matrix. l The Middle row N(k) l -1)+1 column elements; According to the Gaussian approximation formula, the merged signal can be approximated as follows: in, This represents the statistical mean. Indicates statistical variance. This represents normalized noise that follows a complex Gaussian distribution with a mean of 0 and a variance of 1. symbol The log-likelihood ratio (LLR) between the real and imaginary parts is: Here, Re() and Im() represent the operations of finding the real part and the imaginary part, respectively.

9. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 8, characterized in that: Receiver pair All elements undergo the aforementioned interference cancellation and information merging operations, i.e., formulas (12) to (17); the receiver obtains the encoded sequence. The log-likelihood ratio sequence is denoted as Its input is fed into a decoder to obtain the encoded sequence. The posterior log-likelihood ratio sequence is denoted as... The receiver feeds back the element-by-element extrinsic information estimation results to the symbol. Used for the next interference cancellation iteration; the calculation of the element-wise extrinsic information estimation result is divided into two steps, first calculating the element-wise extrinsic information; let They respectively represent mappings to Bits in the encoded sequences of real and imaginary parts; regarding symbols The result of the element-by-element extrinsic information calculation is as follows: in, Indicates to The signal after interference cancellation Represents about bits Element-by-element external information Represents about bits element-wise external information, () * The expression indicates the conjugate operation, and Var() indicates the variance operation. Feedback to about The estimated value is the result of the calculation: in, This indicates the bit output of the channel decoder. The posterior information, This indicates the bit output of the channel decoder. The posterior information, where 1j represents the imaginary unit; After the above iterative operations, the receiver recovers user information in each time slot; the receiver uses verification techniques such as channel coding to determine whether the information recovery is correct.

10. A large-scale random access method based on asynchronous reinforced slotted ALOHA according to claim 1, characterized in that: In step five, the interference cancellation operation and result are as follows: in, Indicates y l The signal after interference cancellation Indicates the indicator factor, when The value is 1 for the time period and 0 for the rest. After interference cancellation In the alignment of user k, where, A portion of the signal will be fed back to the detector for the next round of iteration; once the solved user information is eliminated, it will be removed from the set. Eliminate the indexes of solved users in the set. Corresponding K l It has also changed.

Images