A large-scale terminal random access method based on conflict estimation

Abstract

The present invention is a large-scale terminal random access method based on conflict estimation, which solves the problem of multi-user overload access under limited preamble resources and is suitable for the scenario of sporadic burst transmission of small data packets by a large number of machine-type devices. The specific access method includes the following steps: first, designing a large-scale random access protocol with low signaling overhead and controllable user conflicts; then, proposing a multi-preamble combined random access competition scheme; finally, designing an efficient channel resource allocation scheme based on conflict estimation. The present invention can reduce large-scale terminal random access conflicts without increasing the number of orthogonal preambles, improve the access success probability of users, and enhance the multiple access capability of cellular networks.

Description

Technical Field

[0001] The invention belongs to the field of communication technology, and in particular relates to a large-scale terminal random access method based on conflict estimation. Background Art

[0002] With the vigorous development of mobile Internet, the number of wireless terminal devices and mobile data traffic are growing rapidly. Massive Machine Type Communication (mMTC) is one of the three main scenarios of the fifth generation of cellular communication. Its goal is to solve the connection and communication problems of large-scale devices. It is currently widely used in smart medical, smart home, smart manufacturing and smart transportation. Unlike human-to-human (H2H) communication, mMTC communication presents new characteristics: 1) the uplink data packet is shorter; 2) the data traffic of each device is sporadic and bursty, and only a small number of devices are active at any given time and frequency; 3) the number of devices is huge and the types of devices are numerous. These new features bring new challenges to wireless access technology. For example, the control signaling overhead will become huge when using traditional authorization-based multiple access technology, and limited pilot resources determine that it is impossible to allocate orthogonal preamble sequences to each device. Therefore, how to cope with the access of large-scale terminals with high spectrum efficiency and low signaling overhead, especially to support the burst transmission of sporadic small data packets of a large number of machine-type devices, has become an important problem to be solved in future wireless networks.

[0003] Multiple Input Multiple Output (MIMO) technology provides a solution for high-capacity, high-reliability wireless communications and has become one of the mainstream technologies in the current large-scale access field. The random access technology of traditional cellular networks requires the allocation of independent, orthogonal, dedicated time-frequency preambles to each active terminal. The limited number of preamble codes limits the scale of multiple access. With the support of large-scale MIMO technology, different active terminals can share preamble resources in different spaces. The expanded preamble space can greatly reduce preamble collisions during random access and achieve near-perfect preamble detection. However, how to solve the multiple access of large-scale terminals, especially MTC terminals, based on MIMO technology is still a difficult problem at present. Summary of the invention

[0004] In order to solve the problem of multi-user overload access under limited preamble resources and improve the utilization efficiency of wireless resources, the present invention proposes a large-scale terminal random access scheme based on conflict estimation, which is suitable for the scenario of sporadic burst transmission of small data packets by a large number of machine-type devices. The present invention can reduce large-scale terminal random access conflicts without increasing the number of orthogonal preambles, improve the user's access success probability, and enhance the multiple access capability of the cellular network.

[0005] In order to achieve the above object, the present invention is achieved through the following technical solutions:

[0006] The present invention is a large-scale terminal random access method based on conflict estimation, which is characterized in solving the problem of multi-user overload access under limited preamble (pilot) resources to improve the utilization efficiency of wireless resources. The method comprises the following steps:

[0007] Step 1: To address the problems of high signaling overhead and high user conflict rate of the traditional four-step random access protocol, a three-step random access protocol is designed to reduce the random access signaling overhead;

[0008] Step 2: Propose a multi-preamble combined random access competition scheme, design a corresponding preamble signal detection algorithm to detect the user's combined preamble, reduce large-scale terminal random access conflicts, and improve the user's access success probability;

[0009] Step 3: Estimate the number of competing conflicting users, that is, the number of users using the same combined preamble, and design a channel resource allocation scheme based on conflict estimation to maximize the utilization of wireless channels and improve the multiple access capability of cellular networks.

[0010] A further improvement of the present invention is that in step 1, the three-step random access protocol is described as follows:

[0011] Step 1-1: The active user randomly selects multiple preamble codes from the orthogonal preamble, that is, the pilot set, to form a combined preamble, transmits the combined preamble in the access contention time slot of the uplink channel, and performs access contention;

[0012] Step 1-2: The base station receives and detects the combined preamble sent by the active user, determines whether the user's contention is successful based on the detected preamble, and then designs a resource allocation scheme to allocate physical channels to the successful user, and informs the user of the contention result and available physical channels through broadcasting;

[0013] Step 1-3: According to the base station broadcast, users who have successfully competed or basically successfully competed transmit data on the designated physical channel. Users whose physical channels are not designated are considered to have failed access. Users who have failed to compete wait for a period of time before making the next access attempt.

[0014] A further improvement of the present invention is that: the multi-preamble combined random access competition scheme described in step 2 is specifically: each uplink access competition phase of the user is divided into multiple time slots; each active user randomly selects a preamble code from an orthogonal preamble (i.e., pilot) set in each time slot to send; all preamble codes sent by each user in each access competition constitute a combined preamble; an access conflict occurs if and only if the combined preambles of two or more active users participating in the competition are equal to or more than half the same; otherwise, the user competition is successful.

[0015] The access contention scheme is specifically described and analyzed as follows:

[0016] Suppose there is a base station with M antennas in the center of the cellular cell, and N active terminals (users) send information to the base station at the same time, and all terminals use the time slot transmission structure. There are K orthogonal preamble symbols (preamble codes) of length J in the system. Let s k represents the kth leading symbol, k∈{1, 2, ..., K}, K≤J; the matrix composed of all K orthogonal leading symbols is called the leading symbol matrix, which is expressed as S=[s1, ..., s K ] T ∈C K×J , where C K×J represents a complex field with K rows and J columns, and SS H =I K , where S H represents the Hermitian matrix of the leading symbol matrix S, I K represents the K-row and K-column identity matrix.

[0017] Each active user randomly selects L (L ≥ 1) preamble symbols from K preamble symbols during the access contention phase and transmits them sequentially in L consecutive time slots of the uplink random access channel, L ≤ K. If the preamble sequence number selected by user n, n∈{1, 2, ..., N} in time slot l during the access contention phase is c n，l , c n，l ∈{1, 2, ..., K}, then the leader selection vector of user n in time slot l is expressed as Function 1 [x] Indicates that when condition x is met, 1 [x] =1, otherwise 1 [x] = 0. The leader selection matrix of all users in time slot l is expressed as A l =[a 1，l , ..., a N，l ] T .

[0018] A further improvement of the present invention is that in step 2, the pilot signal detection algorithm is described as follows:

[0019] Step 2-1: The base station receives a pilot signal sent by an active user;

[0020] Let the pilot signal sent by the active user received by the base station in time slot l be Y l ∈C M×J , which is expressed as

[0021] Y l =HP l +N l

[0022] Where H = [h1, ..., h N ]∈C M×N represents the uplink channel response matrix from N users to the base station, C M×N represents a complex field with M rows and N columns, h n =[h n，1 ,h n，2 , ..., h n，M ] T (n=1, 2, ..., N) represents the response coefficient vector of the uplink access channel between user n and the base station, h n，m is the channel response coefficient between user n and base station antenna m. The channel response coefficient matrix satisfies the formula H H H≈MI, where H H represents the Hermitian matrix of matrix H, I is the N-dimensional unit matrix, P l =[p 1，l , ..., p n，l , ..., p N，l ] T ∈C N×J represents the lth pilot symbol matrix sent by all N users, where C N×J represents a complex field with N rows and J columns, p n，l T is the vector p n，l The transposed vector of represents the lth pilot symbol transmitted by user n, N l is the channel additive white Gaussian noise matrix, in is the average noise power, channel additive white Gaussian noise N l The distribution characteristics of can be obtained through channel estimation;

[0023] Step 2-2: The base station operates the received preamble signal and the preamble symbol matrix to obtain the user's preamble selection matrix;

[0024] Let the pilot signal received by the base station in time slot l be Y l , Y l The operation is performed with the leading symbol matrix S, and the result is expressed in B l ∈CM×K Indicates that

[0025] B l =Y l S H =HP l S H +N l S H

[0026] Let A l =P l S H =[a 1，l , ..., a N，l ] T , where a n，l T ∈C 1×K represents the leader selection vector of user n in time slot l, if a n，l T The kth column element of is 1, indicating that the leading symbol selected by user n in time slot l, that is, the lth pilot symbol transmitted by user n, is s k , since user n chooses to transmit only one preamble in any time slot, vector a n，l T There is only one element that is 1, and all other elements are 0;

[0027] Let W l =N l S H , B l It can be simplified as

[0028] B l =HA l +W l

[0029] Given that the leading symbol matrix S is a fixed parameter and its energy is normalized, that is, SS H =I K , W l and N l have the same distribution, i.e. Among them I M is the M-dimensional unit matrix,

[0030] By B l =HA l +W l Get the leader selection matrix A l Estimated value of for

[0031]

[0032] Taking into account the estimated value The elements of may not be integers, so rounding is used to The elements of the integer are taken to get the final estimate of the leader selection matrix of the user in time slot l.

[0033] Step 2-3: Determine the combined leading used by each user;

[0034] According to the method of step 2-1 and step 2-2, all L time slots are traversed to obtain the estimated values ​​of the user leader selection matrix of all L time slots. The combined pilot scheme of all competing users can be obtained;

[0035] Step 2-4: Find the pilot signal and the number of conflicting users where contention conflicts occur;

[0036] The pilot selection matrix estimate A * Indicated as A * =[a1, ..., a N ] T ,in represents the estimated value of the pilot combination selected by user n in all L time slots, and the estimated value of the pilot selection matrix of the nth user and the estimated value of the pilot selection matrix of the n′th user Perform correlation operation, n≠n′, and obtain the pilot conflict pattern d n，n′ ,Right now

[0037]

[0038] d n，n′ The value of represents the number of time slots in which users n, n′ select the same pilot in all L time slots, that is, the number of pilots that send access conflicts. The estimated values ​​of all user pilot selection matrices are correlated pairwise to obtain the pilot conflict pattern matrix D = [d1, ..., d N ],d n =[d n，1 , d n，2 , ..., d n，N ] T It indicates the number of the same pilots selected by user n and other users in the same time slot.

[0039] A further improvement of the present invention is that in step 3, the channel resource allocation based on conflict estimation includes the following steps:

[0040] Step 3-1: Classify competing users into three categories according to the pilot conflict situation obtained in step 2-4:

[0041] Category A is a user without contention conflict, i.e., the user that sends the only combined preamble, and uses E A Represents the set of users of category A;

[0042] Category B is a weak contention conflict user, that is, less than half of the pilots in the combined preamble sent conflict with other users, and E B represents the set of users of category B;

[0043] Category C is a strong contention conflict user, that is, equal to or more than half of the pilots in the combined preamble sent conflict with other users, and E C represents the set of users of category C;

[0044] Step 3-2: Give different categories of users different resource allocation priorities:

[0045] Class A users are users who have successfully competed for access and are given the highest priority in resource allocation;

[0046] Class B users are basically successful users in access competition and are given the second highest priority in resource allocation;

[0047] Class C users are users who have failed in access competition and are given the lowest priority in resource allocation;

[0048] Step 3-3: assigning data channels to competing users according to resource allocation priorities;

[0049] The base station first allocates data channels to Class A users with the highest priority. After all Class A users have obtained data channels, it continues to allocate data channels to Class B users until all Class A and Class B users in the system have obtained transmission resources or all channels have been allocated. Class C users fail in this competition and try to access again after a delay.

[0050] The base station notifies all users of the competition results and channel allocation status through the downlink broadcast channel, and the users perform data transmission or wait according to the base station broadcast.

[0051] The beneficial effects of the present invention are:

[0052] 1. The present invention designs a large-scale random access protocol with low signaling overhead and controllable user conflicts, which reduces the signaling overhead compared with the traditional four-step random access protocol.

[0053] 2. The present invention proposes a multi-preamble combined random access competition scheme, designs a corresponding preamble signal detection algorithm to detect the user's combined preamble, reduces large-scale terminal random access conflicts, and improves the user's access success probability; in traditional authorization-based random access, each device only selects one orthogonal preamble in the preamble transmission stage. The scheme adopted by the present invention greatly reduces the user's access conflict probability.

[0054] 3. The present invention designs a channel resource allocation scheme based on conflict estimation, divides users into different sets, and assigns different resource allocation priorities to users of different categories, so as to maximize the utilization of wireless channels and improve the multiple access capability of cellular networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a network model diagram in an embodiment of the present invention.

[0056] Figure 2 It is a schematic diagram of the transmission protocol in an embodiment of the present invention.

[0057] Figure 3 It is a flowchart of the implementation of the method in the embodiment of the present invention.

[0058] Figure 4 4 is a diagram showing simulation results of user access probability in an embodiment of the present invention. DETAILED DESCRIPTION

[0059] The following will disclose the embodiments of the present invention with drawings. For the purpose of clear description, many practical details will be described together in the following description. However, it should be understood that these practical details should not be used to limit the present invention. That is to say, in some embodiments of the present invention, these practical details are not necessary.

[0060] like Figure 1-3 As shown, the present invention is a large-scale terminal random access method based on conflict estimation, and the specific implementation steps of the method are as follows:

[0061] Step 1: Design a three-step random access protocol, specifically:

[0062] 1a) The active user randomly selects multiple preamble codes from the orthogonal preamble (pilot) set to form a combined preamble, transmits the combined preamble in the access contention time slot of the uplink channel, and performs access contention;

[0063] 1b) The base station receives and detects the combined preamble sent by the active user, determines whether the user's contention is successful based on the detected preamble, and then designs a resource allocation scheme to allocate physical channels to the successful user, and informs the user of the contention result and available physical channels through broadcasting;

[0064] 1c) According to the base station broadcast, users who have successfully competed or basically successfully competed transmit data on the designated physical channel, and users who have failed to compete wait for a period of time before making the next access attempt. Users who have not been assigned a physical channel are considered to have failed access.

[0065] Step 2: Propose a multi-preamble combined random access contention scheme and design a corresponding preamble signal detection algorithm to detect the user's combined preamble.

[0066] The multi-preamble combination random access contention scheme specifically includes:

[0067] Each uplink access contention phase of the user is divided into multiple time slots; each active user randomly selects a preamble code from the orthogonal preamble, i.e., the pilot set, to send in each time slot; all preamble codes sent by each user in each access contention constitute a combined preamble; an access conflict occurs if and only if the combined preambles of two or more active users participating in the contention are equal to or more than half the same; otherwise, the user competition is successful.

[0068] The specific access competition scheme is described and analyzed as follows:

[0069] Suppose there is a base station with M antennas in the center of the cellular cell, and N active terminals, i.e. users, send information to the base station at the same time. All terminals use a time slot transmission structure. There are K orthogonal preamble symbols, i.e. preamble codes, of length J in the system. Let s k represents the kth leading symbol, k∈{1, 2, ..., K}, K≤J; the matrix composed of all K orthogonal leading symbols is called the leading symbol matrix, which is expressed as S=[s1, ..., s K ] T ∈C K×J , where C K×J represents a complex field with K rows and J columns, and SS H =I K , where S H represents the Hermitian matrix of the leading symbol matrix S, I K represents the K-row and K-column identity matrix.

[0070] Each active user randomly selects L preamble symbols from K preamble symbols during the access contention phase and transmits them sequentially in L consecutive time slots of the uplink random access channel, L≤K. If the preamble sequence number selected by user n, n∈{1, 2, ..., N} in time slot l during the access contention phase is c n，l , c n，l ∈{1, 2, ..., K}, then the leader selection vector of user n in time slot l is expressed as Function 1 [x] Indicates that when condition x is met, 1 [x] =1, otherwise 1 [x] = 0, the leader selection matrix of all users in time slot l is expressed as A l =[a 1，l , ..., a N，l ] T .

[0071] The following are some examples:

[0072] Assume that there are 3 active users in the cell and the pilot set is {s1, ..., s4}. During access contention, each user randomly selects 3 pilots from the pilot set and transmits them in 3 time slots in sequence. Assume that the combined pilots selected by the 3 users are {s1, s2, s3}, {s2, s3, s4}, and {s1, s3, s4}, respectively. Then the pilot selection matrices A1, A2, and A3 of the users in the 3 time slots are:

[0073]

[0074] In addition, the combined preamble sequence number vector sent by user n in all L time slots is defined as c n =[c n，1 , ..., c n，L ]; the combined leading symbol sent by user n is The number of combined leading schemes that the user can select is The probability of each combined leader being selected by each user is For example, when K=4, L=2, the number of combined leading elements that the user can select is The combined leading schemes are {s1, s2}, {s1, s3}, {s1, s4}, {s2, s3}, {s2, s4}, {s3, s4}. The probability of each combined leading being selected by each user is

[0075] The preamble signal detection algorithm detects the user's combined preamble and specifically includes the following steps:

[0076] Step 2-1: The base station receives a pilot signal sent by an active user;

[0077] Let the pilot signal sent by the active user received by the base station in time slot l be Y l ∈C M×J , which is expressed as

[0078] Y l =HP l +N l (2)

[0079] Where H = [h1, ..., h N ]∈C M×N represents the uplink channel response matrix from N users to the base station, C M×N represents a complex field with M rows and N columns, h n =[h n，1 ,h n，2 , ..., h n，M ] T (n=1, 2, ..., N) represents the response coefficient vector of the uplink access channel between user n and the base station, h n，mis the channel response coefficient between user n and base station antenna m. The channel response coefficient matrix satisfies the formula H H H≈MI, where H H represents the Hermitian matrix of matrix H, I is the N-dimensional unit matrix, P l =[p 1，l , ..., p n，l , ..., p N，l ] T ∈C N×J represents the lth pilot symbol matrix sent by all N users, where C N×J represents a complex field with N rows and J columns, p n，l T is the vector p n，l The transposed vector of represents the lth pilot symbol transmitted by user n, N l is the channel additive white Gaussian noise matrix, in is the average noise power, channel additive white Gaussian noise N l The distribution characteristics of can be obtained through channel estimation;

[0080] Step 2-2: The base station operates the received preamble signal and the preamble symbol matrix to obtain the user's preamble selection matrix;

[0081] Let the pilot signal received by the base station in time slot l be Y l , Y l The operation is performed with the leading symbol matrix S, and the result is expressed in B l ∈C M×K Indicates that

[0082] B l =Y l S H =HP l S H +N l S H (3)

[0083] Let A l =P l S H =[a 1，l , ..., a N，l ] T , where a n，l T ∈C 1×K represents the leader selection vector of user n in time slot l, if a n，l T The kth column element of is 1, indicating that the leading symbol selected by user n in time slot l, that is, the lth pilot symbol transmitted by user n, is s k, since user n chooses to transmit only one preamble in any time slot, vector a n，l T There is only one element that is 1, and all other elements are 0;

[0084] Let W l =N l S H , B l It can be simplified as

[0085] B l =HA l +W l (4)

[0086] Given that the leading symbol matrix S is a fixed parameter and its energy is normalized, that is, SS H =I K , W l and N l have the same distribution, i.e. Among them I M is the M-dimensional unit matrix,

[0087] The leader selection matrix A is obtained from formula (4): l Estimated value of for

[0088]

[0089] Taking into account the estimated value The elements of may not be integers, so rounding is used to The elements of the integer are taken to obtain the final estimate of the leader selection matrix of the user in time slot l.

[0090] Step 2-3: Determine the combined leading used by each user;

[0091] According to the method of step 2-1 and step 2-2, all L time slots are traversed to obtain the estimated values ​​of the user leader selection matrix of all L time slots. The combined pilot scheme of all competing users can be obtained;

[0092] Step 2-4: Find the pilot signal and the number of conflicting users where contention conflicts occur;

[0093] The pilot selection matrix estimate A * Represented as A * =[a1, ..., a N ] T ,in represents the estimated value of the pilot combination selected by user n in all L time slots, and the estimated value of the pilot selection matrix of the nth user and the estimated value of the pilot selection matrix of the n′th user Perform correlation operation, n≠n′, and obtain the pilot conflict pattern d n，n′ ,Right now

[0094]

[0095] d n，n′ The value of represents the number of time slots in which users n, n′ select the same pilot in all L time slots, that is, the number of pilots that send access conflicts. The estimated values ​​of all user pilot selection matrices are correlated pairwise to obtain the pilot conflict pattern matrix D = [d1, ..., d N ],d n =[d n，1 , d n，2 , ..., d n，N ] T It indicates the number of the same pilots selected by user n and other users in the same time slot.

[0096] The following are some examples:

[0097] Considering that the contention access process includes 3 time slots, the estimated values ​​of the pilot selection matrices of the three users are correlated with each other, and the pilot conflict pattern matrix D is obtained as

[0098]

[0099] The first row and second column element D of the matrix D 1，2 = 0, indicating that user 1 and user 2 select different pilots in all time slots; the first row and third column element D of matrix D 1，3 =1, indicating that user 1 and user 3 have selected the same pilot in one time slot; the second row and third column element D of matrix D 2，3 =2, indicating that user 2 and user 3 have selected the same pilot in 2 time slots.

[0100] Step 3: Estimate the number of competing conflicting users, that is, the number of users using the same combined preamble, and design a channel resource allocation scheme based on conflict estimation to maximize the utilization of wireless channels and improve the multiple access capability of the cellular network.

[0101] Channel resource allocation based on conflict estimation includes the following steps:

[0102] Step 3-1: Classify competing users into three categories according to the pilot conflict situation obtained in step 2-4:

[0103] Category A is a user without contention conflict, i.e., the user that sends the only combined preamble, and uses E A Represents the set of users of category A;

[0104] Category B is a weak contention conflict user, that is, less than half of the pilots in the combined preamble sent conflict with other users, and E B represents the set of users of category B;

[0105] Category C is a strong contention conflict user, that is, equal to or more than half of the pilots in the combined preamble sent conflict with other users, and E C represents the set of users of category C;

[0106] Step 3-2: Give different categories of users different resource allocation priorities:

[0107] Class A users are users who have successfully competed for access and are given the highest priority in resource allocation;

[0108] Class B users are basically successful users in access competition and are given the second highest priority in resource allocation;

[0109] Class C users are users who have failed in access competition and are given the lowest priority in resource allocation;

[0110] Step 3-3: Assign data channels to competing users based on resource allocation priority. The base station first allocates data channels to Class A users with the highest priority. After all Class A users have obtained data channels, it continues to allocate data channels to Class B users until all Class A and Class B users in the system have obtained transmission resources or all channels have been allocated. Class C users fail in this competition and try to access again after a delay.

[0111] The base station notifies all users of the competition results and channel allocation status through the downlink broadcast channel, and the users perform data transmission or wait according to the base station broadcast.

[0112] In order to prove the effect of the present invention, the present invention is further verified by the following simulation.

[0113] 1. Experimental scenario:

[0114] There is one base station in a cell, equipped with multiple antennas. An orthogonal channel matrix is ​​randomly generated according to the distance from the user to the base station. The number of single-antenna users is in [0,100], the number of base station antennas is 128, the number of orthogonal preambles is 10, and it is assumed that each user selects 3 different preamble codes for transmission.

[0115] 2. Experimental content and results:

[0116] Figure 4 is the simulation result, the horizontal axis is the number of active users, and the vertical axis is the user access probability. Single-slot is a traditional authorization-based random access. Each device only selects an orthogonal preamble in the preamble transmission phase, which makes the user's access conflict probability higher.

[0117] The present invention can reduce random access conflicts of large-scale terminals without increasing the number of orthogonal preambles, increase the access success probability of users, and enhance the multiple access capability of cellular networks.

[0118] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent substitution, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.

Claims

1. A large-scale terminal random access method based on conflict estimation, characterized in that: Suppose there is a base station with M antennas in the center of the cellular cell, and N active terminals, i.e. users, send information to the base station at the same time. All terminals use a time slot transmission structure. There are K orthogonal preamble symbols, i.e. preamble codes, of length J in the system. Let s k represents the kth leading symbol, k∈{1, 2, ..., K], K≤J; the matrix composed of all K orthogonal leading symbols is called the leading symbol matrix, which is expressed as S=[s1, ..., s K ] T ∈C K×J , where C K×J represents a complex field with K rows and J columns, and SS H =I K , where S H represents the Hermitian matrix of the leading symbol matrix S, I K represents a K-row K-column identity matrix, and the large-scale terminal random access method comprises the following steps: Step 1: Propose a multi-preamble combination random access competition scheme and design a corresponding preamble signal detection algorithm to detect the user's combined preamble. The multi-preamble combined random access contention scheme comprises the following steps: Step 1a: Divide each uplink access contention phase of the user into multiple time slots; Step 1b: Each active user randomly selects a preamble code from the orthogonal preamble set to send in each time slot; Step 1c: All preambles sent by each user in each access contention constitute a combined preamble. Access conflict occurs if and only if the combined preambles of two or more active users participating in the contention are equal to or more than half the same. Otherwise, the user contention succeeds. The pilot signal detection algorithm comprises: Step 1-1: The base station receives a pilot signal sent by an active user; Step 1-2: The base station operates the received preamble signal with the preamble symbol matrix to obtain the user's preamble selection matrix; Step 1-3: Determine the combined leading used by each user; Step 1-4: Find the pilot signal and the number of conflicting users where contention conflicts occur; Step 2: Design a channel resource allocation scheme based on conflict estimation, including the following steps: Step 2-1: Classify competing users into three categories according to the pilot conflict situation obtained in step 1-4: Category A is a user without contention conflict, i.e., the user that sends the only combined preamble, and uses E A Represents the set of users of category A; Category B is a weak contention conflict user, that is, less than half of the pilots in the combined preamble sent conflict with other users, and E B represents the set of users of category B; Category C is a strong contention conflict user, that is, equal to or more than half of the pilots in the combined preamble sent conflict with other users, and E C represents the set of users of category C; Step 2-2: Give different categories of users different resource allocation priorities: Class A users are users who have successfully competed for access and are given the highest priority in resource allocation; Class B users are basically successful users in access competition and are given the second highest priority in resource allocation; Class C users are users who have failed in access competition and are given the lowest priority in resource allocation; Step 2-3: assigning data channels to competing users according to resource allocation priorities; Step 2-4: The base station notifies all users of the competition results and channel allocation status through the downlink broadcast channel, and the users perform data transmission or wait according to the base station broadcast.

2. The method for large-scale terminal random access based on conflict estimation according to claim 1, characterized in that: In step 1-1, Let the pilot signal sent by the active user received by the base station in time slot l be Y l ∈C M×J , which is expressed as Y l =HP l +N l Where H = [h1,…,h N ]∈C M×N represents the uplink channel response matrix from N users to the base station, C M×N represents a complex field with M rows and N columns, h n =[h n,1 ,h n,2 ,…,h n,M ] T (n=1,2,…,N) represents the response coefficient vector of the uplink access channel between user n and the base station, h n,m is the channel response coefficient between user n and base station antenna m. The channel response coefficient matrix satisfies the formula H H H≈MI, where H H represents the Hermitian matrix of matrix H, I is the N-dimensional unit matrix, P l =[p 1,l ,…,p n,l ,…,p N,l ] T ∈C N×J represents the lth pilot symbol matrix sent by all N users, where C N×J represents a complex field with N rows and J columns, p n,l T is the vector p n,l The transposed vector of represents the lth pilot symbol transmitted by user n, N l is the channel additive white Gaussian noise matrix, in is the average noise power, channel additive white Gaussian noise N l The distribution characteristics of are obtained through channel estimation; In step 1-2, let the pilot signal received by the base station in time slot 1 be Y l , Y l and the leading symbol matrix S = [s1,…,s K ] T ∈C K×J Perform the operation and use B as the result. l ∈C M×K Indicates that B l =Y l S H =HP l S H +N l S H Let A l =P l S H =[a 1,l ,…,a N,l ] T , where a n,l T ∈C 1×K represents the leader selection vector of user n in time slot l, if a n,l T The kth column element of is 1, indicating that the leading symbol selected by user n in time slot l (i.e., the lth pilot symbol transmitted by user n) is s k , since user n chooses to transmit only one preamble in any time slot, vector a n,l T There is only one element that is 1, and all other elements are 0; Let W l =N l S H , B l Simplified expression: B l =HA l +W l Given that the leading symbol matrix S is a fixed parameter and its energy is normalized, that is, SS H =I K , W l and N l have the same distribution, i.e. Among them I M is the M-dimensional unit matrix, By B l =HA l +W l Get the leader selection matrix A l Estimated value of for Taking into account the estimated value The elements of may not be integers, so rounding is used to The elements of the integer are taken to get the final estimate of the leader selection matrix of the user in time slot l. According to the method of step 1-1 and step 1-2, all L time slots are traversed to obtain the estimated values ​​of the user leader selection matrix of all L time slots. The combined pilot scheme of all competing users can be obtained; The pilot selection matrix estimate A * Represented as A * =[a1,…,a N ] T ,in represents the estimated value of the pilot combination selected by user n in all L time slots, and the estimated value of the pilot selection matrix of the nth user and the estimated value of the pilot selection matrix of the n′th user Perform correlation operation, n≠n′, and obtain the pilot conflict pattern d n,n′ ,Right now d n,n′ The value of represents the number of time slots in which users n, n′ select the same pilot in all L time slots, that is, the number of pilots that send access conflicts. The estimated values ​​of all user pilot selection matrices are correlated pairwise to obtain the pilot conflict pattern matrix D = [d1,…,d N ],d n =[d n,1 ,d n,2 ,…,d n,N ] T It indicates the number of the same pilots selected by user n and other users in the same time slot.

3. The method for large-scale terminal random access based on conflict estimation according to claim 1, characterized in that: The step 2-3 assigns data channels to competing users according to the resource allocation priority as follows: the base station first allocates data channels to the highest priority Class A users, and after all Class A users have obtained data channels, it continues to allocate data channels to Class B users until all Class A users and Class B users in the system have obtained transmission resources or the channels are allocated. Class C users fail in this competition and try to access again after a delay.

Images