JP6774455B2 - Radio base station accommodation design methods, equipment and programs - Google Patents

Description

The present invention relates to a radio base station accommodation design method, device and program, and in particular, the load of the mobile management device accommodating each radio base station is distributed by using the number of handovers of the radio terminals accommodated by each radio base station as an index. With respect to methods, devices and programs for designing the accommodation of each radio base station in each mobility control device, as described above.

As shown schematically in FIG. 5, a wireless network represented by LTE includes a plurality of wireless base stations eNB (eNodeB) that dynamically accommodate a large number of wireless terminal MNs, and these plurality of wireless base stations eNB. It is composed of a plurality of movement management devices that are fixedly accommodated and perform movement management of the wireless terminal MN, and an upper network (not shown). An example of the movement management device is a core facility MME (Mobility Management Entity).

Allocation of the radio base station eNB to be accommodated in each movement management device and expansion / reduction of the movement management device are called accommodation design. In the conventional accommodation design, in consideration of the capacity of the movement management device, the movement route of the user terminal, etc., in addition to the manual design, as disclosed in Non-Patent Documents 1 and 2, the combination optimization problem / set division problem -It was done by adding constraints to the set cover problem, incorporating it into integer programming, and solving it with a solver.

Tetsuya Fujie, "Introduction to Formulation by Integer Programming," Operations Research, 57-4 (2012), pp.190-197. S. Kirkpatrick, C. D. Gelatt Jr, M. P. Vecchi, "Optimization by Simulated Annealing," Science 220 (4598), pp.671-680.

NP-hard design is difficult, and when the scale of the problem becomes large, that is, when the number of radio base stations increases, it is very difficult to find the optimum solution, and it takes time to converge the solution. Even if an optimal solution or a quasi-optimal solution is sought, the cost of changing from the current state may be enormous.

An object of the present invention is to solve the above-mentioned conventional technique, to obtain an optimum solution or a quasi-optimal solution in a short time even when the number of radio base stations is large, and to keep the cost of changing from the current state low. To provide accommodation design methods, equipment and programs.

In order to achieve the above object, the present invention has the following configuration in a radio base station accommodation design device for designing accommodation of each base station accommodating a plurality of wireless terminals in each movement management device. There is a feature in.

(1) Means for determining the number of handovers of wireless terminals for each base station pair, and each for each movement management device so that the base station pair with a larger number of handovers is preferentially accommodated in the same movement management device. It was equipped with a means for designing the accommodation allocation of base stations.

(2) The means for designing the accommodation allocation is to design the accommodation allocation that minimizes the number of handovers between base station pairs having different accommodation destinations.

(3) The means for designing the accommodation allocation includes the zero-based design means, and the zero-based design means executes clustering with the reciprocal of the number of handovers of each base station pair as the distance between the base station pairs, and is the same. Base stations classified into groups are now housed in the same mobile management device.

(4) The means for designing the accommodation allocation includes the current base design means, and the current base design means repeatedly evaluates by randomly replacing a part of the base stations housed in different mobile management devices. Changed to execute the selection algorithm.

(5) The randomized algorithm is to replace fringe base stations more frequently than non-fringe base stations.

According to the present invention, the following effects are achieved.

(1) The number of handovers of the wireless terminal is determined for each base station pair, and each base station to each movement management device is accommodated in the same movement management device with higher priority so that the base station pair with the larger number of handovers is accommodated in the same movement management device. Since the accommodation allocation is designed, the number of handovers having a high processing load that straddles the movement management devices can be minimized, and the load of each movement management device can be reduced.

(2) A zero-based design means is provided, clustering is performed with the reciprocal of the number of handovers of each base station pair as the distance between the base station pairs, and the base stations classified into the same group are accommodated in the same mobile management device. Therefore, it is relatively difficult to fall into local optimization, and it becomes easy to realize an innovative accommodation design that is not bound by the current accommodation design. In addition, not only is it easy to handle redesign when adding a movement management device, but it is also possible to converge to a quasi-optimal solution at a relatively high speed under simple constraints.

(3) Since the current base design means was set up and some of the base stations housed in different mobile management devices were randomly replaced and evaluated repeatedly, the change cost was reduced although it was only a gradual improvement. It enables a low accommodation design.

(4) Random replacement of base stations by the current base design means is tried more frequently at fringe base stations than at non-fringe base stations, so it is possible to derive a highly evaluated accommodation design with a small number of replacements. ..

It is a block diagram of the accommodation design system which concerns on one Embodiment of this invention. It is a figure which showed typically the 1st accommodation design method by a zero base design part. It is the figure which showed typically the 2nd accommodation design method by the present-state base design part. It is a figure which showed the identification method of a fringe base station. It is a figure which showed the structure of the LTE network to which this invention is applied.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a configuration of a main part of the accommodation design device / system 1 according to the embodiment of the present invention, and is accommodated in each movement management device in the LTE network described with reference to FIG. Design the allocation of the radio base station eNB.

The accommodation design system 1 of the present invention may be configured by mounting an application (program) that realizes each function described later on a general-purpose computer, or a part of the application is dedicated to hardware or ROM. It may be configured as a machine or a single-purpose machine.

The handover monitoring unit 101 monitors the handover of the wireless base station eNB that occurs when the wireless terminal MN moves and communicates, and totals the handover number N for each base station pair. The aggregated result is stored in the database 102 as a handover history. Based on the number of handovers N obtained for each base station pair, the accommodation design unit 103, in principle, manages each movement so that the number of handovers N between base station pairs having different accommodation destination movement management devices is minimized. Design the allocation of radio base stations to be accommodated in the device.

In this embodiment, the accommodation design unit 103 includes the zero base design unit 103a and the current base design unit 103b. The zero-based design unit 103a performs a semi-optimal accommodation design by performing the accommodation design from scratch without being bound by the current state of accommodation allocation. The current status base design unit 103b repeats the evaluation by partially changing the current status of the accommodation allocation, and gradually improves by adopting the accommodation allocation with the best evaluation under predetermined constraints and conditions.

FIG. 2 is a diagram schematically showing a first accommodation design method by the zero base design unit 103a. The first accommodation design method is based on clustering by unsupervised learning that does not utilize the current mobility management device accommodation design.

In the present embodiment, each radio base station eNB is two-dimensionally configured regardless of the physical distance, with the inverse (or counter) n of the number of handovers N between the base station pairs as the logical distance between the base station pairs. By arranging it on the map [Fig. (A)] and applying an appropriate clustering method such as k-means to this, a base station pair with a larger number of handovers, that is, a base station pair with a closer logical distance. Realize clustering that is classified into the same group [Fig. (B)], and each group is housed in the same movement management device.

According to the first accommodation design method by the zero-based design unit 103a, it is relatively difficult to fall into the local optimum, and it is easy to realize an innovative accommodation design that is not bound by the current movement management device accommodation design. In addition, it is easy to handle redesign when adding a movement management device. Furthermore, simple constraints can be used to converge to a semi-optimal solution at a relatively high speed.

When the capacity of each movement management device, for example, the number of mobile terminals that can be accommodated is different, it is desirable to allocate a group having a larger number of classified wireless terminal MNs to a movement management device having a larger capacity.

FIG. 3 is a diagram schematically showing a second accommodation design method by the current base design unit 103b. The second accommodation design method starts from the current accommodation environment and repeats evaluation by replacing the accommodation destination of a part of the radio base station eNB from the current movement management device with another movement management device. Make gradual improvements.

In this embodiment, a randomized algorithm for local search such as Simulated Annealing method is used to replace the accommodation destinations of one or more radio base stations eNB from the current movement management device accommodation state, and the number of handovers N before and after the replacement is N. The evaluation of the replacement is repeated using the increase / decrease as an index.

In addition, in order to keep the change cost from the current state low, the number of replacements may be added as an evaluation function, and the accommodation allocation with a smaller number of replacements of the radio base station eNB from the current state may be evaluated more highly. Alternatively, if the capacity of each movement management device is different, the accommodation allocation in which more radio base stations are accommodated by the movement management device having a larger capacity may be evaluated more highly.

In addition, the accommodation allocation with a smaller number of overcapacity mobility management devices is evaluated higher, and the accommodation allocation in which the signaling load (handover number) of each mobility management device is more leveled is evaluated higher. Is also good.

Further, paying attention to the signaling in which the handover occurs, each handover may be weighted according to the signaling type. In the present embodiment, each handover is weighted based on the connection, disconnection, bearer setting, bearer release, handover within the movement management device, and handover between movement management devices.

Further, in the present embodiment, when applying the randomized algorithm, priority is set for each radio base station eNB so that the higher priority radio base station eNB is tried to be replaced more frequently. ..

In the present embodiment, attention is paid to the physical position of the radio base station eNB accommodated in each movement management device, and the radio base station eNB (fringe base station) physically located near the boundary of the accommodation range is described. Higher priority over the fringe base station, based on the empirical rule that a handover of the radio terminal MN with another radio base station eNB housed in another adjacent mobile management device is likely to occur. The degree is set.

In identifying fringe base stations, as shown in FIG. 4, a triangle and its circumscribed circle are created with the physical arrangement of each radio base station eNB [Fig. (A)] as a set of points, and which triangle is used. The Delaunay division is completed by repeating the subdivision so that the points are not included in the circumscribed circle [Fig. (B)] [Fig. (C)]. Next, a Voronoi diagram is drawn by connecting the centers of the circumscribed circles of each triangle [Fig. (D)], and if the base station at the apex of the triangle is a different TAC (Tracking Area Code), it can be identified as a fringe base station. ..

As described above, according to the present embodiment, the number of handovers of the wireless terminal MN is determined for each base station pair, and the base station pair having a larger number of handovers is preferentially accommodated in the same movement management device. Since the accommodation allocation is designed, the number of handovers having a high processing load that straddles the movement management devices can be minimized, and the load of each movement management device can be reduced.

Further, the system of the present embodiment includes a zero-based design unit 103a, performs clustering in which the reciprocal of the number of handovers of each base station pair is set as the distance between the base station pairs, and the base stations classified into the same group are the same. Since it is housed in the movement management device, it is relatively difficult to fall into local optimization, and it becomes easy to realize an innovative house design that is not bound by the current house design. In addition, not only is it easy to handle redesign when adding a movement management device, but it is also possible to converge to a quasi-optimal solution at a relatively high speed under simple constraints.

Further, the system of the present embodiment is provided with the current base design unit 103b, and repeats evaluation by randomly exchanging a part of the base stations housed in different movement management devices, so that the improvement is limited to gradual improvement. It enables accommodation design that can keep the change cost low.

Furthermore, since the random replacement of base stations by the current base design unit 103b is tried more frequently in fringe base stations than in other than fringe base stations, it becomes possible to derive a highly evaluated accommodation design with a small number of replacements. ..

eNB (eNodeB) ... wireless base station, MN ... wireless terminal, 101 ... handover monitoring unit, 102 ... database, 103 ... accommodation design department, 103a ... zero base design department, 103b ... current base design department

Claims (11)

In the accommodation design device of a radio base station that designs the accommodation of each base station accommodating a plurality of wireless terminals in each movement management device.
A means for determining the number of wireless terminal handovers for each base station pair,
The radio is provided with a means for designing the accommodation allocation of each base station to each movement management device so that the base station pair having a larger number of handovers is more preferentially accommodated in the same movement management device. Base station accommodation design equipment. The accommodation design device for a radio base station according to claim 1, wherein the means for designing the accommodation allocation is to design an accommodation allocation that minimizes the number of handovers between base station pairs having different accommodation destinations. The means for designing the accommodation allocation comprises zero-based design means.
The zero-based design means performs clustering with the reciprocal of the number of handovers of each base station pair as the distance between the base station pairs, and accommodates the base stations classified in the same group in the same movement management device. The accommodation design device for a radio base station according to claim 1 or 2. The means for designing the accommodation allocation includes the current base design means.
The present-state-based design means is characterized in that it executes a randomized algorithm that repeats evaluation by randomly replacing a part of base stations housed in different mobile management devices at present. The accommodation design device for the radio base station according to any one. The accommodation design device for a radio base station according to claim 4, wherein the randomized algorithm is characterized in that the fringe base station is replaced with a higher frequency than that of a non-fringe base station. The accommodation design device for a radio base station according to claim 4 or 5, wherein the randomized algorithm more highly evaluates an accommodation allocation with a smaller number of base station replacements from the current state. Claim 4 or 5 is characterized in that the capacity of each mobile management device is different, and the randomized algorithm more highly evaluates the accommodation allocation in which more radio base stations are accommodated by the larger capacity mobile management device. The accommodation design device for the radio base station described in. The accommodation design device for a radio base station according to claim 4 or 5, wherein the randomized algorithm more highly evaluates the accommodation allocation in which the load of each movement management device is more leveled. The accommodation design device for a radio base station according to claim 4 or 5, wherein the randomized algorithm more appreciates a smaller number of overcapacity mobility management devices. In the accommodation design method of a wireless base station, which designs accommodation of each base station accommodating a plurality of wireless terminals in each movement management device,
The computer
Determine the number of wireless terminal handovers for each base station pair
Accommodation design of a radio base station, characterized in that the accommodation allocation of each base station to each movement management device is designed so that the base station pair having a larger number of handovers is more preferentially accommodated in the same movement management device. Method. In a radio base station accommodation design program that designs accommodation of each base station accommodating a plurality of wireless terminals in each movement management device.
Determine the number of wireless terminal handovers for each base station pair
A radio base station that causes a computer to design the accommodation allocation of each base station to each movement management device so that the base station pair with a higher number of handovers is more preferentially accommodated in the same movement management device. Containment design program.

Images