CN106850045B

CN106850045B - Self-adaptive over-subscription LEO satellite system channel allocation method

Info

Publication number: CN106850045B
Application number: CN201710033738.1A
Authority: CN
Inventors: 冉静; 郭佳妮; 刘元安; 胡鹤飞; 袁东明; 高锦春
Original assignee: Beijing University of Posts and Telecommunications
Current assignee: Beijing University of Posts and Telecommunications
Priority date: 2017-01-17
Filing date: 2017-01-17
Publication date: 2020-04-24
Anticipated expiration: 2037-01-17
Also published as: CN106850045A

Abstract

The invention discloses a self-adaptive channel allocation method for an over-subscription LEO satellite system, belonging to the field of mobile communication; the method specifically comprises the following steps: firstly, establishing a mobility model of an LEO satellite system, and setting conditions of a channel allocation strategy; for beam cell C_iInitially setting the total number S of occupied virtual channels to be 0; then, the call request for a certain high priority user accesses beam cell C_iJudging whether the call request is a new call request or a switching call request, respectively carrying out self-adaptive excess reservation channel allocation processing aiming at different requests, and determining whether the new call request or the switching call request is accessed; finally, beam cell C_iProcessing the terminated call, and transferring the reservation request on any virtual channel to the occupied available channel; has the advantages that: the method can ensure that a smaller new call blocking rate is obtained under the condition that the switching call drop rate of the high-priority user is close to zero, and the overall service quality of the system is improved.

Description

Self-adaptive over-subscription LEO satellite system channel allocation method

Technical Field

The invention belongs to the field of mobile communication, and particularly relates to a self-adaptive channel allocation method for an over-subscription LEO satellite system.

Background

Compared with the prior art, the ground mobile communication system cannot achieve global accessibility due to geographic factors and economic factors, and the satellite mobile communication system is not influenced by geographic positions and distances and can provide global seamless coverage communication at any time; and the LEO satellite system can provide high-quality instant mobile communication service with the advantages of low transmission power, low transmission delay, high-efficiency bandwidth utilization rate and the like.

The iridium satellite system, as a typical LEO satellite system, has 6 orbital planes, each of which has 11 satellites, as shown in fig. 1, the radiation range of any one satellite on the earth surface is a circular area, which is called a satellite footprint; for efficient frequency division multiplexing, a satellite footprint is divided into a large number of smaller regions, called beam cells.

However, LEO satellites operate at high speed relative to the surface of the earth, resulting in the possibility that a user terminal may experience many satellites or beam cells during a call connection, frequently handing off. Among them, beam cell handover is particularly frequent. Because of the scarce resources in the satellite system, the user's call connection will be interrupted if the destination cell to which the user is about to be handed over has no free channel. This is more unacceptable to the user than if the new call is blocked. It is therefore important to give higher priority to handover calls during channel allocation. However, if too much channel resources are allocated for the handover call, the blocking rate of the new call will be too high.

In addition, users in LEO satellite systems also have different ratings, which are mainly classified into two categories: high priority users and normal users. High priority users, such as business users, are more sensitive to delays and have higher quality of service requirements. The requirements of the common users on time delay and service quality are flexible. In the channel allocation process, emphasis should be placed on meeting the needs of high priority users. However, if too many channels are allocated to the high priority users, the service quality of the ordinary users is seriously reduced.

In summary, for the LEO satellite system, two problems need to be considered simultaneously when designing the channel allocation strategy: i.e. how to make a reasonable channel allocation between the handover call and the new call, and how to make a fair channel allocation between the high priority users and the normal users.

In view of the above problems, the strategies adopted in the prior art include the following:

1) a Guaranteed Handover (GH) strategy;

when a call is made (New call or handover call) to cell C_iThe GH strategy tries cell C_iNext cell C_i+1Reserving channels for high priority users, and if the free channels in the system are all locked, then a reservation request is sent to wait for a free channel. Since the channel reservation occurs before the actual handover of the high priority user, the success of the handover call is actually guaranteed.

The Guaranteed Handover (GH) strategy is based on the following assumptions:

the system employs a fixed channel allocation scheme (FCA), i.e. each cell has a fixed number of channels C. In addition, queued reservation requests are prioritized over new calls (both for high priority users and normal users) and handover calls for normal users. An incoming call can only be accessed when there are free channels in both of the first two cells it will experience, where any one cell has no free channels, and the call will be blocked at the start.

The channel reservation method adopted in the GH strategy is called a channel locking mechanism. This reservation method is very conservative and selfish, and when a channel is locked, other users than the specific user cannot use the channel even if the specific user does not occupy the channel or has a considerable distance from the cell. Therefore, although the GH policy can ensure that the handover drop rate of the high-priority user is zero, the GH policy is at the cost of a higher new call blocking rate, which results in a very low overall service quality of the system.

2) A time-based channel reservation (TCRA) policy:

unlike the GH policy, in the TCRA policy, the channel is locked only for the usage time expected by the user. The key to the TCRA strategy is to determine the cell that the user will visit while calculating the necessary channel lock time. The strategy is used for more accurately reserving channels for the system to access more users, and meanwhile, the satisfaction degree of waiting for accessing the users is enhanced.

Compared with the GH strategy, the TCRA strategy effectively reduces the time for unnecessary locking of channel resources in a cell, and obtains a lower new call blocking rate and higher overall service quality of a system while ensuring that the handover blocking rate is zero.

Although the TCRA policy effectively reduces the instances of call blocking, it does not take into account call termination, i.e. the call terminal occurs before the reserved channel is used, and during the channel reservation process using the TCRA policy, a reservation request is issued at call setup and at each handover brake; in both cases, the channel will be locked for a specified time interval; however, the call may be terminated at any time and the reserved channel at the next visited cell will be cancelled. Thus, the channel is meaninglessly locked, and handling the problem will enable the system to access more users while achieving a higher overall quality of service for the system.

3) An oversubscription TCRA (TCRA-O) strategy

The unique feature of the TCRA-O strategy is the introduction of the concept of overbooking, where in order to obtain maximum resource utilization, the system will admit channel reservation requests that are greater than its capability, and in which part is assumed to be cancelled, the TCRA-O strategy accepts more reservation requests to make up for the waste of resources due to call termination.

The main implementation procedure of the TCRA-O strategy is as follows:

assuming a given cell C_iThere are a number C of channels available, which will receive the channel in its own cell C_iNew call request or last cell C_i-1The handover request of (2). For new calls, cell C_iIs considered to be the source cell; for handover calls, cell C_iIs considered a handover cell. The TCRA-O is an extended strategy based on TCRA, whose requirements for channel reservation time are consistent with the TCRA strategy, i.e. each reservation request can be accurately predicted for the lock time of the channel resource. The TCRA-O strategy considers that in each cell, in addition to the real channel resources, there are a certain number of virtual channel resources, the number being S. Thus, the access or denial of a reservation request will depend on the availability of C + S channels rather than C channels. The cell will accept additional reservations for S channels in anticipation of at least S reservation cancellations.

Compared with the GH strategy and the TCRA strategy, the strategy effectively reduces the new call blocking rate of the high-priority user, however, the new call blocking rate is reduced at the cost of sacrificing the handover call performance of the high-priority user; the call drop rate of the high-priority users is not zero any more, and is influenced by the number S of the virtual channels.

However, for the TCRA-O strategy, if the number S of virtual channels is set too large, the handover failure rate of the high-priority user will be increased, which is unacceptable for the user. If the number S of the virtual channels is set to be too small, it cannot be effectively ensured that the high-priority user acquires a low new call blocking rate. How to obtain the optimal number S of virtual channels to achieve the optimal utilization of channel resources is a key to the TCRA-O policy to obtain better system quality of service.

At present, the strategy adopts a fixed virtual channel number for simulation, and sets the virtual channel number S to be 10% of the real channel number C in the cell, although a certain effect is obtained, the virtual channel number cannot be adaptively adjusted according to the user condition in the system to obtain the optimal system service quality.

In order to adaptively obtain the optimal number S of virtual channels, the following two factors related to the ue need to be considered:

1. location distribution of user terminals

The location distribution of the user terminal determines the switching time, and for the user at the cell switching boundary, the switching is about to occur, and the virtual channel is allocated to the user, which may cause the switching failure and affect the service quality. For users far away from the cell switching boundary, it is more practical to allocate virtual channels for the users to perform switching reservation, and simultaneously, the new calls can be prevented from being blocked. Therefore, it is necessary to consider the setting of adjusting the number of virtual channels for the difference in the location of the user terminal.

2. Traffic ratio for high priority users

Virtual channels in the system serve only channel reservations for high priority user handover calls; if the traffic of the high-priority users in the system is low, the setting of properly reducing the number of virtual channels can be considered to reduce the probability of switching failure; if the traffic of the high priority users in the system is high, it can be considered to increase the number of virtual channels appropriately to reduce the new call blocking rate.

It should be noted that, by combining the above two factors, more reasonable virtual channel number setting can be performed, and it is ensured that a smaller new call blocking rate is obtained to obtain the best system service quality under the condition that the handover call drop rate is close to zero.

Disclosure of Invention

The invention provides a self-adaptive channel allocation method for an over-subscription LEO satellite system, aiming at solving the problem that the best system service quality cannot be obtained because the number of virtual channels cannot be self-adaptively adjusted according to the user condition in a system in a TCRA-O strategy.

The method comprises the following specific steps:

step one, establishing a mobility model of an LEO satellite system, and setting conditions of a channel allocation strategy;

modeling each beam cell into a continuous rectangular cell, which is called a one-dimensional mobility model for short; allocating C number of available channels in each beam cell;

based on the mobility model, when channel allocation strategy research is carried out, the following assumptions are adopted:

1) the generation location of the new call in each beam cell follows a uniform distribution.

2) The average call origination rate is independent of the number of ongoing calls in each beam cell.

The arrival process of the new call and the handover call obeys Poisson distribution, and the new call and the handover call are independent from each other; the call arrival processes of the different features are also independent of each other.

3) The call duration follows an exponential distribution.

4) The channel duration of each beam cell follows an exponential distribution.

Step two, aiming at the beam cell C_iSetting the beam cell C at the initial time_iThe total number S of occupied virtual channels of (1) is 0;

step three, accessing beam cell C aiming at call request of certain high-priority user_iJudging whether the call request is a new call request or a switching call request, if the call request is the new call request, entering a fourth step, and if not, entering a fifth step;

step four, beam cell C_iCarrying out self-adaptive excess booking channel distribution processing on a new call request of a high-priority user, and determining whether to access the new call request;

the method comprises the following specific steps:

step 401, the system receives the beam cell C after the new call request of the high priority user is accessed_iFirst, the source cell C is determined_iIf yes, go to step 402; otherwise, blocking the new call request;

step 402, determining handover cell C_i+1If there is a free channel, if so, requesting handover cell C for the new call_i+1Performs channel reservation in the source cell C_iThe call request is accessed. Otherwise, go to step 403;

step 403, based on the time threshold t_THDividing each beam cell into two areas A and B, and calculating the position distribution of the new call request user;

t_TH*V_trkis the distance of the user from the beam cell switching boundary; v_trkRepresents the sub-satellite rate;

region a is a region far from the cell handover boundary; region B is the region of the cell switching boundary where virtual channel reservations are invalid;

step 404, determining whether the location distribution of the new call requesting subscriber is the source beam cell C_iIf yes, go to step 405; otherwise, whether the position distribution of the new call request user is a source beam cell C_iArea B of (2), block the new call request;

step 405, judge handover cell C_i+1Whether the total number S of occupied medium virtual channels is less than the maximum number S of virtual channels_maxIf yes, go to step 406; otherwise, blocking the new call request;

S_max＝[(αk+β)*C]

k is the proportion of the high-priority user traffic, k is more than or equal to 0 and less than or equal to 1, α is a scale factor for controlling the proportion and is a constant;

step 406, requesting cell C to be handed off for the new call_i+1A virtual channel is reserved, the total occupied number S of the virtual channel is added with 1, and the new call request is sent to the source cell C_iAnd (6) accessing.

Step five, wave beam cell C_iCarrying out adaptive over-booking channel allocation processing on the call switching request of the high-priority user, and determining whether to switch in the call;

the method comprises the following specific steps:

step 501, switching call request to beam cell C for high priority user_iJudging whether the high priority user reaches the beam cell C_iIf yes, go to step 502; otherwise, continue to wait until reaching handover cell C_iThe handover boundary of (1);

step 502, determining that the high priority user call switching request is in switching cell C_iIf yes, switching the call switching request to drop call; otherwise, go to step 503;

step 503, judge the next switching cell C_i+1If there is a free channel, then switching the request for the call in the next switching cell C_i+1Performs channel reservation in the switching cell C_iThe switching is successful; otherwise, go to step 504;

step 504, judge the next switching cell C_i+1Whether the total number S of occupied medium virtual channels is less than the maximum number S of virtual channels_maxIf yes, go to step 505; otherwise, switching the call switching request to drop the call;

step 505, the call handover request is sent to the next handover cell C_i+1A virtual channel is reserved, the total number S of the virtual channels is added with 1, and the call switching request is sent to a switching cell C_iThe handover is successful.

Step six, wave beam cell C_iFor terminatingThe call is processed, and the reservation request on any virtual channel is transferred to the occupied available channel.

Whenever there is a call termination, the system will transfer the call in the virtual channel to the available channel and release the virtual channel, and the total number of occupied virtual channels S is reduced by 1.

The invention has the beneficial effects that:

1) a self-adaptive channel allocation method for an over-subscription LEO satellite system introduces a position distribution factor (represented as a time threshold) of a user terminal, and adjusts the setting of the number of virtual channels in a TCRA-O strategy; the method avoids the switching failure caused by reserving invalid virtual channels for the users at the cell switching boundary, and meanwhile, the virtual channels are allocated for the users far away from the cell switching boundary in a meaningful way, so that the new call is prevented from being blocked.

2) The method introduces the traffic proportion factor of a high-priority user to adjust the setting of the number of virtual channels in a TCRA-O strategy, and properly reduces the setting of the number of the virtual channels to reduce the probability of switching failure under the condition that the traffic proportion of the high-priority user in the system is lower; otherwise, the number of virtual channels is increased appropriately to reduce the new call blocking rate.

3) The self-adaptive excess booking LEO satellite system channel allocation method is characterized in that self-adaptive excess booking channel reservation is carried out based on mutual restriction of a position distribution factor of a user terminal and a flow rate ratio of a high-priority user, so that a small new call blocking rate can be obtained under the condition that the switching call drop rate of the high-priority user is close to zero, and the integral service quality of the system is improved.

Drawings

FIG. 1 is a network coverage diagram of an exemplary LEO satellite system of the present invention;

FIG. 2 is a one-dimensional mobility model diagram in the LEO satellite system of the present invention;

FIG. 3 is a time threshold t of the present invention_THDividing the area of the cell into a situation graph;

FIG. 4 is a flow chart of the adaptive oversubscription LEO satellite system channel allocation method of the present invention;

FIG. 5 is a flow chart of the new call admission for a high priority subscriber of the present invention;

fig. 6 is a flow chart of the handover call access of the high priority user of the present invention.

DETAILED DESCRIPTION OF EMBODIMENT (S) OF INVENTION

The following describes in detail a specific embodiment of the present invention with reference to the drawings.

In the existing communication, the influence of the position distribution of a user terminal and the flow rate ratio of a high-priority user on the selection of the optimal virtual channel number is not considered, and along with the change of the user condition in a system, the improper setting of the virtual channel number can cause that the switching failure rate is too high or the new call blocking rate is not obviously reduced, so that the expected performance of the TCRA-O strategy cannot be achieved. The invention improves the existing self-adaptive TCRA-O strategy, and provides a self-adaptive over-subscription LEO satellite system channel allocation method, which is an ATCR-O strategy for short.

The ATCR-O policy is a major improvement over the TCRA-O policy in the handling of high priority users. For the high priority users, as shown in fig. 4, the specific steps of the implementation process of channel allocation are as follows:

considering the high-rate operation of the relative LEO satellite, the movement and the earth rotation of the user can be ignored, and each beam cell is modeled into a continuous rectangular cell, which is called a one-dimensional mobility model for short. As shown in fig. 2, when a handover occurs, the destination cell is the next cell in the opposite direction of the satellite motion. The rectangular cell length L is typically equal to the diameter of the beam cell, typically 425 km. V_trkRepresenting the sub-satellite velocity, typically 26600 km/h. The time for a user to traverse a complete cell, denoted T_max：

Based on the mobility model, the following classical assumptions are generally adopted when channel allocation strategy research is performed:

1) the generation positions of the new calls in each beam cell are subjected to uniform distribution.

The arrival process of the new call and the handover call obeys Poisson distribution, and the new call and the handover call are independent from each other; the call arrival procedures of different features (user terminal type, service type) are also independent of each other.

3) The call duration follows an exponential distribution.

4) The channel duration of each beam cell follows an exponential distribution.

In the established LEO system model, it is assumed that each beam cell is allocated a number C of channels. The ATCR-O policy keeps the common user's treatment consistent with the TCRA-O policy. That is, the non-priority management is performed for the call establishment and the handover of the common user: if there is a free channel in the system, the call is accessed; otherwise the new call will be blocked or the handover call will be terminated.

and (3) new call access: after receiving a new call request of a user, the system firstly judges a source cell C_iWhether there is a free channel. If not, blocking it, if yes, then judging switching cell C_i+1If there is a free channel, then switching cell C for the call_i+1Performs channel reservation in the source cell C_iThe call is accessed. Otherwise, the call is adaptedThe specific process of the excess reservation channel reservation is as follows: based first on a time threshold t_THJudging the position distribution of the new calling user; if the new calling user is located in the source cell C_iBlocking the area B; if the new calling user is located in the source cell C_iIf the area A is smaller than the area A, judging whether the total number S of the virtual channels occupied in the current cell is smaller than the maximum number S of the virtual channels_max(ii) a If S is less than S_maxThen cell C is handed over for the call_i+1Reserving a virtual channel, adding 1 to the total number S occupied by the virtual channel, and accessing the call in the source cell; if S is greater than or equal to S_maxThe call is blocked to avoid the handoff drop rate increase caused by excessive virtual channel reservation.

As shown in fig. 5, the specific steps are as follows:

as shown in fig. 3, the time for the user to traverse a complete cell is T_maxTherefore, t_TH*V_trkIs the distance of the user from the beam cell switching boundary; v_trkRepresents the sub-satellite rate;

region a is a region far from the cell handover boundary; the area B is an area of a cell switching boundary, the virtual channel reservation in the area is invalid, and a high-speed user in the area B will be subjected to switching soon, and performing virtual channel reservation for the user will likely cause the user to fail in switching.

Step 404Judging whether the position distribution of the new call request user is a source beam cell C_iIf yes, go to step 405; otherwise, whether the position distribution of the new call request user is a source beam cell C_iArea B of (2), block the new call request;

the conditions of the channel allocation strategy indicate: the generation location of the new call in the source cell is subject to a uniform distribution.

S_max＝[(αk+β)*C]

k is the ratio of the high priority user traffic, k is more than or equal to 0 and less than or equal to 1, α is a scale factor for controlling the proportion, is a constant, and is selected according to an empirical value.

With the increase of the traffic ratio of the high-priority users, the virtual channel resources allowed to be used in the handover cell increase, and more access opportunities are provided for new calls of the high-priority users.

Step 406, requesting cell C to be handed off for the new call_i+1A virtual channel is reserved, the total number of occupied virtual channels is added with 1, and the new call request is sent to the source cell C_iAnd (6) accessing.

and (3) call switching: the system receiving it switches to C_iAfter the request of the cell, firstly, the user is waited to arrive at C_iThe handover boundary of the cell is then determined to be at C_iWhether the reserved channel in the cell is a virtual channel or not; if the virtual channel is still present, the handover is dropped. If not, judging the next cell C_i+1If there is a free channel, then at C_i+1Channel reservation in a cell for a handover call for the call and allowing the call to be handed over to C_iAnd (4) a cell. If not, the call is switched toAdaptive oversubscription channel reservation.

It should be noted that for the call handover request, the corresponding subscriber is far from the next cell C_i+1All the time of the switching boundary is T_max(ii) a Then for the call handover user, it is always located at C_iArea a of the cell.

As shown in fig. 6, the specific steps are as follows:

step 501, switching call request to beam cell C for high priority user_iJudging whether the call switching request of the high priority user reaches the beam cell C_iIf yes, go to step 502; otherwise, continue to wait until reaching handover cell C_iThe handover boundary of (1);

if so, the user has not waited for an available channel when actually performing the handover. Otherwise, the user waits until the actual channel is available before actually making the handover.

step 504, judge the next switching cell C_i+1Whether the total number S of occupied medium virtual channels is less than the maximum number S of virtual channels_maxIf yes, go to step 505, perform adaptive over-subscription channel reservation for the call handover request; otherwise, switching the call switching request to drop the call;

Step six, wave beam cell C_iThe terminated call is processed to determine how the channels occupied by the call are to be allocated.

As with the TCRA-O strategy, whenever a call is terminated, the system will transfer the call in the virtual channel to the available channel and release the virtual channel, with the total number of virtual channel occupancies S reduced by 1.

The invention adaptively adjusts the setting of the number of virtual channels in the TCRA-O strategy according to two factors, namely the position distribution of the user terminal and the flow ratio of the high-priority user, processes the new call and the switching call process of the high-priority user and obtains the optimal system service quality.

Claims

1. A self-adaptive channel allocation method for an over-subscription LEO satellite system is characterized by comprising the following specific steps:

the method specifically comprises the following steps:

1) the generation positions of the new calls in each beam cell are subjected to uniform distribution;

2) the average call origination rate is independent of the number of ongoing calls in each beam cell;

the arrival process of the new call and the handover call obeys Poisson distribution, and the new call and the handover call are independent from each other; the call arrival processes of different characteristics are mutually independent;

3) the call duration follows an exponential distribution;

the channel duration of each beam cell follows exponential distribution;

step three, call request for a certain high priority user is receivedBeam-in cell C_iJudging whether the call request is a new call request or a switching call request, if the call request is the new call request, entering a fourth step, and if not, entering a fifth step;

the method comprises the following specific steps:

step 402, determining handover cell C_i+1If there is a free channel, if so, requesting handover cell C for the new call_i+1Performs channel reservation in the source cell C_iAccessing the call request; otherwise, go to step 403;

S_max＝[(αk+β)*C]

step 406, requesting cell C to be handed off for the new call_i+1A virtual channel is reserved, the total occupied number S of the virtual channel is added with 1, and the new call request is sent to the source cell C_iAccessing;

step six, wave beam cell C_iProcessing the terminated call, and transferring the reservation request on any virtual channel to the occupied available channel;

2. The method of claim 1, wherein said step five comprises the steps of:

step 504, judge the next switching cell C_i+1Whether the total number S of occupied medium virtual channels is less than the maximum number S of virtual channels_maxIf yes, go to step 505;otherwise, switching the call switching request to drop the call;