WO2021133352A1

WO2021133352A1 - A probability - based distribution method for random access pools

Info

Publication number: WO2021133352A1
Application number: PCT/TR2020/051376
Authority: WO
Inventors: Musa Gökhan KORKUT
Original assignee: Netas Telekomunikasyon AS
Current assignee: Netas Telekomunikasyon AS
Priority date: 2019-12-26
Filing date: 2020-12-24
Publication date: 2021-07-01
Anticipated expiration: 2022-06-26
Also published as: TR201921870A2

Abstract

The invention relates to a method in Long - Term Evolution (LTE) and 5G technologies that reduces the probability of collision of preamble (P) requests made by user equipment (UE) to base stations. The method of the invention initially checks whether the preamble (P) that comes first is from a pool of contention based random access or contention free random access. Afterwards, pool assignment values (Scb,Sncb) are obtained by comparing the preamble (P) requests received during a preamble (P) detection period with the random access pool capacities (cbi,ncbi) separately for both contention based and contention free situations, and by using the collision probability table when necessary. In the last step, the optimal pool distribution is determined by combining the pool assignment values obtained.

Description

DESCRIPTION

A PROBABILITY - BASED DISTRIBUTION METHOD FOR RANDOM ACCESS

POOLS

Technical Field

The invention relates to the optimal arrangement of contention based random access and contention free random access pools in long - term evolution (LTE) and 5G technologies.

The invention specifically relates to a method that reduces the likelihood of collision of initial access requests made by the user equipment to base stations.

Prior Art

In Long Term Evolution technology and 5G, a user equipment must first access in order to receive service from the base station. This initial access of user equipment to the base station is also called ‘preamble’. A continuous broadcast message (with a system information block - 2 in LTE) informs the user equipment in what frequency range, in what length, and what type of first access user equipment will receive. According to this information in the system information block - 2, the first access is arranged and transmitted to the base station. This initial access is of two different types, namely contention based random access or contention free random access, depending on the situation of uninterrupted cell change (handover).

The patent document US20100232318A1 found in the current technique mentions an algorithm that reduces the collision of initial access requests to base stations in long term evolution technology. However, this algorithm does not mention using a probability - based distribution method for random access pools, both contention based random access and contention free random access, and the use of a collision probability table to help with this method.

As a result, due to the above - mentioned drawbacks and the inadequacy of the existing solutions, an improvement in the technical field has been required.

Aim of Invention

The invention is inspired by the existing circumstances and aims to solve the above - mentioned drawbacks. The main aim of the invention is to reach the optimum level by taking measurements from both contention based random access pools and contention free random access pools.

Another aim of the invention is that the overall collision amount can be easily calculated by tabulating the possible collision amounts of initial access.

In order to fulfill the aims described above, the method of the invention first checks whether the initial attempt is from a contention based random access pool or contention free random access pool. Afterwards, pool assignment values are obtained by comparing the first access requests received during a first attempt detection period with the random access pool capacities separately for both contention based random access pools and contention free situations, and by using the possible collision table when necessary. In the last step, the optimal pool distribution is determined by combining the pool assignment values obtained.

The structural and characteristic features and all advantages of the invention outlined in the drawings below and in the detailed description made by referring to these figures will be understood clearly; therefore, the evaluation should be made by taking these figures and detailed explanations into consideration.

Brief Description of the Figures

Figure 1 is a view showing the first interference communication of user equipment to the base station (LTE or 5G).

Figure 2 is a flow chart showing the working principle of the method of the invention.

Reference Numbers

UE: User equipment eNB: LTE base station gNB "generation Node B": 5G base station

SIB - 2: System Information Block - 2 with broadcast message for LTE

P: Preamble 1001. Calculation of the number of contention based random attempts and the number of contention free random attempts during the preamble detection period

1002. Comparison of the number of contention based random access with each pool capacity of contention based random access received during the preamble detection period.

1003. Obtaining potential collision data from the intersection of a collision probability table created by pre - calculating the probability values of each pool capacity for contention based random access greater than or equal to the number of contention based random attempts that come in during a preamble perception period, in cases where the number of contention based random attempts is less than or equal to the pool capacity of the contention based random access.

1004. In cases where the number of contention based random attempts is greater than the contention based random access pool capacity, selection of the maximum probability values in each contention based random access pool capacity column from the collision probability table that is smaller than the contention based random access pool capacity as probable collision data, obtaining absolute collision data from the differences between the number of contention based random attempts and the said contention based random access pool capacities, and the collection of probable collision data and absolute collision data separately for each contention based random access pool capacity, which is smaller than the number of contention based random attempts.

1005. Obtaining a separate contention based random access pool assignment value for each contention based random access pool capacity by dividing the resulting collision data by the number of contention based random attempts that come during a preamble perception period.

1006. Comparison of the number of contention free random access with each pool capacity of contention free random access received during the preamble detection period.

1007. Obtaining potential collision data from the intersection of a collision probability table created by pre - calculating that probability values of each pool capacity for contention free random access greater than or equal to the number of contention free random attempts that come in during a preamble perception period, in cases where the number of contention free random attempts is less than or equal to the pool capacity of the contention free random access.

1008. In cases where the number of contention free random attempts is greater than the contention free random access pool capacity, selection of the maximum probability values in each contention free random access pool capacity column from the collision probability table that is smaller than the contention free random access pool capacity during the preamble perception period as probable collision data, obtaining absolute collision data from the differences between the number of contention free random attempts and the said contention free random access pool capacities, and the collection of probable collision data and precise collision data separately for each contention free random access pool capacity, which is smaller than the number of the contention free random attempts.

1009. Obtaining a separate contention free random access pool assignment value for each contention free random access pool capacity by dividing the resulting collision data by the number of contention free random attempts that come during a preamble perception period

1010. Collection of the contention based and contention free pool assignment values obtained for each random access pool capacity and selecting the minimum of these results

P_cb: The number of contention based random attempts received by the base station during the preamble detection period

P_ncb: The number of contention free random attempts received by the base station during the preamble detection period cb: Contention based random access pool capacity cb_i: i'th contention based random access pool capacity ncb: Contention free random access pool capacity ncb_i: i'th contention free random access pool capacity Detailed Description of the Invention

In this detailed description, the method of the invention is explained only for a better understanding of the subject.

Figure 2 represents a flow chart that shows the working principle of the invention.

Accordingly, the method of the invention basically comprises the following process steps:

• Calculation of the number of contention based random attempts (P_cb) ^and the number of the contention free random attempts ( P_ncb ) during the preamble (P) detection period (1001 )

• Comparison of each contention based random access pool capacity (cb_i) with the number of contention based random access (P_cb) received via the base station (eNB or gNB) during the preamble (P) detection period (1002)

• In cases where the number of contention based random attempts (P_Cb) ^{is less} than or equal to (P_cb ≤ cb_i) the contention based random access pool capacity (cb_i), obtaining probable collision data from the intersection of the number of contention based random attempts (P_cb) during a preamble (P) detection period and the probability values of each contention based random access pool capacity (cb_i) that are greater than or equal to the number of contention based random attempts (P_cb) in a collision probability table (1003)

• In cases where the number of contention based random attempts (P_cb) is greater P_cb > cb_i than the contention based random access pool capacity (cb_i), the selection of the maximum probability values in each contention based random access pool capacity (cb_i) column that is smaller than the contention based random attempt number (P_cb) during a preamble (P) detection period as probable collision data from the collision probability table, obtaining absolute collision data from the differences between the number of contention based random attempts (P_cb) and the said contention based random access pool capacities (cb_i), and the collection of probable collision data and absolute collision data separately for each contention based random access pool capacity (cb_i), which is smaller than the number of contention based random attempts (P_cb) (1004)

• Obtaining a separate contention based random access pool assignment value (S_cb) for each contention based random access pool capacity (cb_i) by dividing the resulting collision data by the number of contention based random attempts (P_cb) that come during a preamble (P) perception period (1005)

• Comparison of each contention free random access pool capacity (: ncb_i ) with the number of contention free random access ( P_ncb ) received via the base station (eNB or gNB) during the preamble (P) detection period (1006)

• In cases where the number of contention free random attempts ( P_ncb ) ^{is less} than or equal to (P_ncb ≤ ncb_i) the contention free random access pool capacity (: ncb_i ), obtaining probable collision data from the intersection of the number of contention free random attempts ( P_ncb ) during a preamble (P) detection period and the probability values of each contention free random access pool capacity (ncb_i) that are greater than or equal to the number of contention free random attempts ( P_ncb ) in a ceollision probability table (1007)

• In cases where the number of the contention free random attempts ( P_ncb ) ^is greater than the contention free random access pool capacity (ncb_i), the selection of the maximum probability values in each contention free random access pool capacity (ncb_i) column that is smaller than the contention free random attempt number ( P_ncb ) during a preamble (P) detection period as probable collision data from the collision probability table, obtaining absolute collision data from the differences between the number of contention free random attempts ( P_ncb ) and the said contention free random access pool capacities (ncb_i), and the collection of probable collision data and absolute collision data separately for each contention free random access pool capacity (ncb_i), which is smaller than the number of contention free random attempts ( P_ncb ) (1008) • Obtaining a separate contention free random access pool assignment value (S_ncb) for each contention free random access pool capacity (ncb_i) by dividing the resulting collision data by the number of contention free random attempts ( P_ncb ) that come during a preamble (P) perception period (1009) · Collecting the contention based and contention free pool assignment values

(S_cb,S_ncb) obtained for each random access pool capacity (cb_i,ncb_i) and selecting the minimum of these results (1010)

The above - mentioned collision probability table for access pools of 64 is as follows:

The purpose of this table is to do fixed mathematical operations to be performed beforehand and determine their values. These determined values can be used directly when necessary. In this way, the user will be freed from excess processing load. Access pool capacities (cb_i,ncb_i) (in the table under the heading “preamble pool”) are available in the first row of the table. In this embodiment, it is given in the range 4 - 64. In the first column, the total number of future preambles (P) during the detection period of a preamble (P) (in the table under the heading "simultaneous preamble") is given. The table is used in the same way for both contention based and contention free random preambles (P). The reason why the maximum value is 64 is that the pool value is the maximum of 64. If this value is changed in different technologies, it can be operated with the same algorithm. The only thing to do will be to update only the value 64 with the new value.

The calculations made in the table assess the probability that, for a given access pool capacity (cb_i,ncb_i) (horizontal axis of the table), a certain number of preambles (P) will collide at least 1 time when a preamble (P) occurs during the detection period (vertical axis of the table). For example, when the access pool capacity is 4, there are 2 ways to calculate the probability of collisions for at least 1 time if 3 preambles (P) arrive during a preamble (P) detection period:

In the first way, the first of the 3 first preambles (P) will never collide, but randomly select 1 out of 4 access pools. For the remaining 2 preambles (P), there is a probability of 3 collisions.

First probability: the first preamble (P) collides (1 / 4 probability), second preamble (P) collides (1 / 4 probability). As a result (1 / 16 probability).

Second probability: the first preamble (P) collides (1 / 4 probability), the second preamble (P) does not collide (3 / 4 probability). As a result (3 / 16 probability).

Third probability: the first preamble (P) does not collide (3 / 4 probability), the second preamble (P) does not collide (2 / 4 probability). As a result (6 / 16 probability).

In the sum of all these probabilities: at least 1 collision is calculated with the probability of (1 / 6) + (3 / 16) + (6 / 16) = (10 / 16) = 0.625. In the second way, the first of the 3 first preambles (P) will never collide, but randomly select 1 out of 4 access pools. For the remaining 2 preambles (P), the probability that they will not collide is unique:

The first preamble does not collide (3 / 4 probability), the second preamble does not collide (2 / 4 probability). As a result (6/ 16 probability). If the universal set = 1 then the result is 1 - (6 / 16) = (10 / 16) = 0.625.

As given in this example, all the probabilities are calculated, and the table given above is created. The reason some values are missing in the table is that a certain number of preambles (P) during the detection time of a preamble (P) (vertical axis) is greater than the pool capacity value (horizontal axis). There is an absolute collision when it is greater, and its calculation is performed differently. The table contains only calculations for possible collisions.

On the other hand, some values are rounded to 1 in the said table. For the values specified in the table, as many parameters can be used as desired after the comma, and during this process, it is also possible to consider taking the values by rounding them.

The working principle of an exemplary embodiment of the method of the invention is as follows:

In this embodiment of the invention, optimal distribution of 64 preambles (P) in both contention based random access pools and contention free random access pools is provided, thus reducing the probability of collision of preamble (P) requests from the user equipment (UE) to the base station (eNB or gNB). A total pool capacity of 64 units (see. for LTE: 3GPP TS 36.211 version 15.7.0, under the heading "5.7.2 Preamble Sequence Generation" in Release 15 document, - for 5G: 3GPP TS 38.211 version 15.7.0, under the heading "6.3.3.1 Sequence generation" in the

Release 15 standards on page 35) is shared between contention based and contention free random access pools as follows: cb + neb = 64 The pool capacity of 64 mentioned in this embodiment of the method of invention is divided into groups of 4 (see. For LTE: 3GPP TS 36.331 version 15.7.0, page 525 of Release 15 with the heading "RACH - ConfigCommon information element," - For 5G: 3GPP TS 38.331 version 15.7.0, page 312 of the Release 15 standard with the heading "RACFI - ConfigCommon") Assuming that there will be no preamble (P) in the first contention based random access pool, and all the preambles (P) will take place in the last contention based random pool, the capacities of the said pools can be represented as [0, 4, 8,..., 64], In short; schematic representation for the ith pool is as follows:

Similarly, capacities of the contention free random access pools [64, 60, 56, ... 0] can be shown as:

In this case, in the above collision probability table, for example; cb₂ value is shown as 4, and ncb₂ value as 60. Since the value of cb₁ is 0, it is not in the table, while ncb₁ value is shown as 64.

The method is basically executed in two separate ways for contention based random access pools and contention free random access pools. Collision data obtained from calculations made in both separate ways is compiled in the last step, and the overall amount of collision is obtained.

First, the number of contention free random attempts ( P_ncb ) and the number of contention based random attempts (P_cb) arriving at the base station (gNB or eNB) during the preamble (P) detection period is calculated (1001 ).

The operations made from the first way start with the 1002th step. A comparison of each contention based random access pool capacity (cb_i) with the number of contention based random access (P_cb) received via the base station (eNB or gNB) during the preamble (P) detection period is made (1002). In cases where the number of contention based random attempts (P_cb) is less than or equal to (P_cb ≤ cb_i) the contention based random access pool capacity (cb_i), the probable collision data from the intersection of the number of contention based random attempts (P_cb) during a preamble (P) detection period and the probability values of each contention based random access pool capacity (cb_i) that are greater than or equal to the number of contention based random attempts (P_cb) in a collision probability table is obtained (1003). In cases where the number of contention based random attempts (P_cb) ^is greater P_cb > cb_i than the contention based random access pool capacity ( cb_i ), the selection of the maximum probability values in each contention based random access pool capacity (cb_i) column that is smaller than the contention based random attempt number (P_cb) during a preamble (P) detection period as probable collision data from the collision probability table is made, and the absolute collision data from the differences between the number of contention based random attempts (P_cb) and the said contention based random access pool capacities (cb_i), and the collection of probable collision data and absolute collision data separately for each contention based random access pool capacity (cb_i), which is smaller than the number of contention based random attempts (P_cb) is made (1004).

As an example of the process in step 1004; The calculations for the value cb_i =4 are as follows: When 7 preambles (P) occur during a preamble (P) detection time, it is obtained as the absolute collision data 7 - 4 = 3. For probable collision data, 4 is selected from the ‘pool capacities' field located on the horizontal axis in the table. The maximum value is also selected as the vertical axis of the table. For this example, this value is 0.90625. In this way, both exact and probable collision data are obtained from all these data. The total collision data is 3 + 0.90625 = 3.90625. Flere, the collision probability of the preambles (P), other than the first preambles (P) that will definitely collide, is found and added to the absolute collision data.

In obtaining possible collision data for step 1003, only the table in which probability calculations are made is used. For example, the calculations for the value cb_i =8 are as follows: Let's assume that the 5 preamble (P) comes if the first attempt (P) is in detection time. In this case, the only thing to do is to check the value from the table. For the data to be obtained, when 8 is selected from the horizontal axis and 5 from the vertical axis, the value obtained is 0.79492. This value is directly the probable collision data. Since incoming preambles (P) during the detection time of a preamble (P) will be smaller than the contention based random access pool capacity (cb_i), the absolute collision part is considered as zero. Because there will be no absolute collision.

Then, a separate contention based random access pool assignment value (S_cb) is obtained for each contention based random access pool capacity (cb_i) by dividing the resulting collision data by the number of contention based random attempts (P_cb) that come during a preamble (P) perception period (1005).

Similar to the operations for contention based random access pools ( cb ) are also performed for contention free random access pools (neb). First of all, a comparison of each contention free random access pool capacity (ncb_i) with the number of contention free random access ( P_ncb ) received via the base station (eNB or gNB) during the preamble (P) detection period is made (1006). As a result, in cases where the number of contention free random attempts ( P_ncb ) is less than ^or equal to (P_ncb ≤ ncb_i) the contention free random access pool capacity (: ncb_i ), probable collision data is obtained from the intersection of the number of contention free random attempts ( P_ncb ) during a preamble (P) detection period and the probability values of each contention free random access pool capacity (ncb_i) that are greater than or equal to the number of contention free random attempts ( P_ncb ) in a collision probability table (1007) Again, as a result, in cases where the number of the contention free random attempts ( P_ncb ) ^is greater than the contention free random access pool capacity (ncb_i), a selection of the maximum probability values is made in each contention free random access pool capacity (ncb_i) column that is smaller than the contention free random attempt number ( P_ncb ) during a preamble (P) detection period as probable collision data from the collision probability table, and absolute collision data from the differences between the number of contention free random attempts ( P_ncb ) and the said contention free random access pool capacities (ncb_i), and probable collision data and absolute collision data is collected separately for each contention free random access pool capacity (ncb_i), which is smaller than the number of contention free random attempts ( P_ncb ) (1008). As an example of the process in step 1008; The calculations for the value ncb_i =4 are as follows: When 7 preambles (P) occur during a preamble (P) detection time, it is obtained as the absolute collision data 7 - 4 = 3. For probable collision data, 4 is selected from the preamble (P) pool heading located on the horizontal axis in the table. The maximum value is also selected as the vertical axis of the table. For this example, this value is 0.90625. In this way, both exact and probable collision data are obtained from all these data. The total collision data is 3 + 0.90625 = 3.90625.

In obtaining probable collision data for step 1007, only the table in which probability calculations are made is used. For example, the calculations for the value ncb_i =8 are as follows: Let's assume that the 5 preamble (P) comes if the first attempt (P) is in detection time. In this case, the only thing to do is to check the value from the table. For the data to be obtained, when 8 is selected from the horizontal axis and 5 from the vertical axis, the value obtained is 0.79492. This value is directly the probable collision data. Since incoming preambles (P) during the detection time of a preamble (P) will be smaller than the contention free random access pool capacity (ncb_i), the absolute collision part is considered as zero. Because there will be no absolute collision.

Then, a separate contention free random access pool assignment value (S_ncb) is obtained for each contention free random access pool capacity (ncb_i) by dividing the resulting collision data by the number of contention free random attempts ( P_ncb ) that come during a preamble (P) perception period (1009).

In the final step of the method, the contention based and contention free pool assignment values (S_cb,S_ncb) obtained for each random access pool capacity ( cb_i , ncb_i) are summed with each other, and the minimum of the results is selected (1010). In this way, the minimum resulting contention based and contention free pool capacities (cb_i,ncb_i) are selected for distribution.

With the method of the invention, measurements are taken in the system that serves the user. Basically, the said measurements cover the time interval up to a preamble (P) detection time. Measurements taken during this period are calculated over a certain period of time, depending on the user request. In this case, pool capacities (cb_i,ncb_i), which give the sum of the minimum pool assignment values (S_cb,S_ncb) obtained at the end of a user - specified period (preferably 1 day), are selected for the distribution.

The pool distribution performed according to this data is then published in the broadcast message. After determining optimal values with the developed method, changes can be made to the system for new values. Continuous broadcasting of new configurations made in this period of change in the LTE system is performed via System Information block - 2 (SIB - 2).

Claims

1. A method that reduces the probability of collisions of preambles (P) for contention based random attempts (P) and contention free random attempts (P) from the user equipment (UE) on cell towers (eNB or gNB) that support long - term evolution or 5G technology, characterized by comprising of the following steps; the use of a table that represents the access pool capacities (cb_i,ncb_i) in the first row, and the total preamble (P) numbers are positioned during the preamble (P) detection period, so as to have the elements as the maximum pool capacity value in the first column, at the intersections of the elements in this row and column, collision probabilities of at least 1 time of a certain number of preambles (P) coming during a preamble (P) detection period for a given access pool capacity (cb_i,ncb_i) are located, and a collision probability table in which no value is located at the intersections of these values, since there will be definite collisions in cases where the first attempt (P) numbers are greater than the pool capacities (cb_i,ncb_i) is used; and,

• Calculation of the number of contention based random attempts (P_cb) and the number of the contention free random attempts ( P_ncb ) during the preamble (P) detection period (1001)

• In cases where the number of contention based random attempts (P_cb) is less than or equal to (P_cb ≤ cb_i) the contention based random access pool capacity (cb_i), obtaining probable collision data from the intersection of the number of contention based random attempts (P_cb) during a preamble (P) detection period and the probability values of each contention based random access pool capacity (cb_i) that are greater than or equal to the number of contention based random attempts (P_cb) in a collision probability table (1003)

• In cases where the number of contention based random attempts (P_cb) is greater (P_cb ≤ cb_i) than the contention based random access pool capacity (cb_i), the selection of the maximum probability values in each contention based random access pool capacity (cb_i) column that is smaller than the contention based random attempt number (P_cb) during a preamble (P) detection period as probable collision data from the collision probability table, obtaining absolute collision data from the differences between the number of contention based random attempts (P_cb) and the said contention based random access pool capacities (cb_i), and the collection of probable collision data and absolute collision data separately for each contention based random access pool capacity (cb_i), which is smaller than the number of contention based random attempts (P_cb) (1004)

• Comparison of each contention free random access pool capacity (ncb_i) with the number of contention free random access {P_ncb) received via the base station (eNB or gNB) during the preamble (P) detection period (1006)

• In cases where the number of contention free random attempts ( P_ncb ) ^is less than or equal to (P_ncb ≤ ncb_i) the contention free random access pool capacity (ncb_i), obtaining probable collision data from the intersection of the number of contention free random attempts ( P_ncb ) during a preamble (P) detection period and the probability values of each contention free random access pool capacity ( ncb_i ) that are greater than or equal to the number of contention free random attempts ( P_ncb ) in a collision probability table (1007)

• In cases where the number of the contention free random attempts ( P_ncb ) ^is greater than the contention free random access pool capacity (ncb_i), the selection of the maximum probability values in each contention free random access pool capacity (ncb_i) column that is smaller than the contention free random attempt number ( P_ncb ) during a preamble (P) detection period as probable collision data from the collision probability table, obtaining absolute collision data from the differences between the number of contention free random attempts ( P_ncb ) and the said contention free random access pool capacities (ncb_i), and the collection of probable collision data and absolute collision data separately for each contention free random access pool capacity (ncb_i), which is smaller than the number of contention free random attempts ( P_ncb ) (1008)

• Obtaining a separate contention free random access pool assignment value (S_ncb) for each contention free random access pool capacity (ncb_i) by dividing the resulting collision data by the number of contention free random attempts ( P_ncb ) that come during a preamble (P) perception period (1009), and

• Collecting the contention based and contention free pool assignment values (S_cb,S_ncb) obtained for each random access pool capacity {cb_i,ncb_i) and selecting the minimum of these results (1010)

2. The method according to Claim 1 , characterized in that; the application of the said process steps for each first attempt (P) detection time that occurs over a predetermined period and the selection of the minimum of the totals of pool assignment values (S_cb,S_ncb) obtained at the end of this predetermined period. The method according to Claim 1 or 2, characterized in that; the selection of pool capacities (cb_i,ncb_i) that give the sum of the minimum pool assignment values (S_cb,S_ncb) obtained for optimal distribution and publishing in the broadcast message.