KR101529131B1 - Method for switching dynamic base stations for green cellular networks - Google Patents

Abstract

The present invention relates to a method for switching a dynamic base station (BS) to reduce energy consumption in a wireless cellular network. More specifically, to solve a problem of minimizing general energy related to a BS switching which is known to require a high computational complexity and a big signaling overhead, provided is the method for switching the dynamic base station for the green cellular network to reduce the energy consumption in the wireless cellular network, by using a switching-on/off based energy save (SWES) algorithm which is introduced with a new concept of a network-impact and is operated with a low computational complexity in a dispersive method, and turning off the BS which has a minimum impact on the network, one by one, considering an additional load increase of adjacent BSs.

Description

TECHNICAL FIELD [0001] The present invention relates to a dynamic base station switching method for a green cellular network,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dynamic base station (BS) switching method for reducing energy consumption in a wireless cellular network and, more particularly, to a method and apparatus for high computational complexity and signaling The present invention relates to a method for solving a general energy minimization problem related to BS switching known to require overhead.

To this end, the present invention proposes a switching-on / off based energy save (SWES) algorithm that can be practically implemented so as to be able to be distributed in a distributed manner with low computational complexity I would like to.

In recent years, with the popularization of smart phones, the data usage of mobile communication network has surged, so the energy consumption of the mobile communication industry has been increasing. As a result, the carbon footprint of the mobile communication industry is steadily increasing.

Here, the term "carbon footprint" means the total amount of greenhouse gas generated by an individual or a group directly or indirectly. If a product made of carbon is used without directly producing carbon dioxide, it can be said to generate carbon dioxide indirectly , And in particular in the field of information and communication technology (ICT), it is estimated to generate about 2% of the world's carbon dioxide emissions (cf. References 1 to 5).

Also, between 2007 and 2020, the carbon footprint of the entire ICT sector is expected to increase more than double, while the carbon footprint of cellular networks is expected to nearly triple over the same period.

In addition, as the awareness of the potential adverse effects on the environment caused by the emission of carbon dioxide and the consumption of non-renewable energy sources has increased, the demand for the development of more efficient energy systems in all industries And this is no exception in a wireless communication system.

Moreover, the development of an efficient energy system is important from an economic point of view of the operation of a cellular network system, because electricity costs account for a major part of the operational expenditure of the cellular network system. For example, in 2013, The cost of power consumed to operate the network is estimated to reach $ 22 billion (see reference 6).

Here, in order to reduce the power consumption of the cellular network, it is desirable to reduce the power consumption at the base station (BS) occupying 60 to 80% of the total power consumption (refer to Reference 7 and Reference 8) .

That is, in order to reduce the power consumption of the cellular network, it is desirable to reduce the power consumed in the entire network by turning off some of the under-utilized base stations during periods of low traffic. It is desirable to provide a new switching on / off based dynamic base station operation method for potential energy savings and a cellular network system configured to reduce power consumption using such a method. A device or a method is not provided.

[references]

1. E. Oh and B. Krishnamachari, "Energy savings through dynamic base station switching in cellular wireless access networks ", in Proc. 2010 IEEE GLOBECOM.

2. "An inefficient truth", Global Action Plan, Tech. Rep., Dec. 2007. Available: http://www.globalactionplan.org.uk/

3. Gartner, "Green IT: the new industry shock wave", in 2007 Presentation Symp./ITXPO Conf.

4. J. Malmodin, A. Moberg, D. Lunden, G. Finnveden, and N. Lovehagen, "Greenhouse gas emissions and operational electricity use in the ICT and entertainment & media sectors", J. Industrial Ecology, vol. 14, no. 5, pp. 770-790, Oct. 2010.

5. A. Fehske, G. Fettweis, J. Malmodin, and G. Biczok, "The global footprint of mobile communications: the ecological and economic perspective", IEEE Commun. Mag., Vol. 49, no. 8, pp. 55-62, Aug. 2011.

6. "Mobile network energy OPEX to rise dramatically to $ 22 billion in 2013", Cellular News, July 3, 2008.

7. K. Son, H. Kim, Y. Yi, and B. Krishnamachari, "Toward Energy-efficient Operation of Base Stations in Cellular Wireless Networks", in A Book Chapter of Green Communications: Theoretical Fundamentals, Algorithms, and Applications. CRC Press, Taylor & Francis, 2012.

8. M. A. Marsan, L. Chiaraviglio, D. Ciullo, and M. Meo, "Optimal energy savings in cellular access networks ", in Proc. 2009 IEEE GreenComm.

9. "Optimizing performance and efficiency of PAs in wireless base stations", white paper, Texas Instruments, Feb. 2009.

10. "Sustainable energy use in mobile communications", white paper, Ericsson, Aug. 2007.

11. A. J. Fehske, F. Richter, and G. P. Fettweis, "Energy efficiency improvements through micro sites in cellular mobile radio networks ", in Proc. 2009 IEEE GreenComm.

12. P. Rost and G. Fettweis, "Green communications in cellular networks with fixed relay nodes ", 2010 Cooperative Cellular Wireless Netw.

13. K. Son, E. Oh, and B. Krishnamachari, "Energy-aware hierarchical cell configuration: from deployment to operation", in Proc. 2011 IEEE INFOCOM Workshop Green Commun. and Net.

14. D. Willkomm, S. Machiraju, J. Bolot, and A. Wolisz, "Primary users in cellular networks: a large-scale measurement study ", in Proc. 2008 IEEE DySPAN.

15. L. Correia, D. Zeller, O. Blume, D. Ferling, Y. Jading, I. Godor, G. Auer, and L. Van Der Perre, "Challenges and enabling technologies for energy aware mobile radio networks" IEEE Commun. Mag., Vol. 48, no. 11, pp. 66-72, Nov. 2010.

16. L. Chiaraviglio, D. Ciullo, M. Meo, and M. A. Marsan, "Energy-aware UMTS access networks ", in Proc. 2008 IEEE WPMC.

17. M. A. Marsan and M. Meo, "Energy efficient management of two cellular access networks ", in Proc. 2009 ACM GreenMetrics.

18. K. Son and B. Krishnamachari, "Speedbalance: speed-scaling-aware optimal load balancing for green cellular networks", in Proc. 2012 IEEE INFOCOM.

19. J. Kwak, K. Son, Y. Yi, and S. Chong, "Impact of spatio-temporal power sharing policies on cellular network greening ", in Proc. 2011 WiOpt, pp. 167-174.

20. E. Oh, B. Krishnamachari, X. Liu, and Z. Niu, "Towards dynamic energy-efficient operation of cellular network infrastructure", IEEE Commun. Mag., Vol. 49, no. 6, pp. 56-61, June 2011.

21. P. Rost and G. Fettweis, "On the transmission-computation-energy tradeoff in wireless and fixed networks ", in Proc. 2010 IEEE GreenComm.

22. W. Yang, L. Li, Y. Wang, and W. Sun, "Energy-efficient transmission schemes in cooperative cellular systems", in Proc. 2010 IEEE Green-Comm.

23. Z. Niu, Y. Wu, J. Gong, and Z. Yang, "Cell zooming for cost-efficient green cellular networks", IEEE Commun. Mag., Vol. 48, no. 11, pp. 74-79, Nov. 2010.

24. K. Son, S. Chong, and G. de Veciana, "Dynamic association for load balancing and interference avoidance in multi-cell networks", IEEE Trans. Wireless Commun., Vol. 8, no. 7, pp. 3566-3576, July 2009.

25. K. Son, H. Kim, Y. Yi, and B. Krishnamachari, "Base station operation and user association mechanisms for energy-delay tradeoffs in green cellular networks", IEEE J. Sel. Areas Commun., Vol. 29, no. 8, pp. 1525-1536, Sept. 2011.

26. R. M. Karp, "Reducibility among combinatorial problems ", Complexity Comput. Comput., Vol. 40, no. 4, pp. 85-103, 1972.

27. "IEEE draft amendment standard for local and metropolitan area networks-part 16: air interface for broadband wireless access systems-advanced air interface", IEEE, Piscataway, NJ, Sept. 2010, 802.16m.

28. L. Sun, H. Tian, and P. Zhang, "Decision-making models for group vertical handover in vehicular communications", Telecommun. Syst., Dec. 2010.

29. Q. Lu and M. Ma, "Group mobility support in mobile WiMAX networks ", J. Netw. Comput. Appl., Vol. 34, no. 4, pp. 1272-1282, July 2011.

30. H. Cheon, B. Park, and D. Hong, "Adaptive multicarrier system with reduced feedback information in wideband radio channels ", in Proc. 1999 IEEE VTC.

31. G. Bianchi, L. Fratta, and M. Oliveri, "Performance evaluation and enhancement of the CSMA / CA MAC protocol for 802.11 wireless LANs", in Proc. 1996 IEEE PIMRC, vol. 2, pp. 392-396.

32. G. P. Pollini, "Trends in handover design ", IEEE Commun. Mag., Vol. 34, no. 3, pp. 82-90, Mar. 1996.

33. K. Son, S. Lee, Y. Yi, and S. Chong, "REFIM: a practical interference management in heterogeneous wireless access networks", IEEE J. Sel. Areas Commun., Vol. 29, no. 6, pp. 1260-1272, June 2011.

34. "IEEE 802.16m evaluation methodology document (EMD) ", IEEE, Piscataway, NJ, Tech. Rep. IEEE 802.16m-08 / 004r5, Sept. 2009.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a method and apparatus for BS switching that is known to require high computational complexity In order to solve the conventional energy minimization problem, it is an object of the present invention to provide a dynamic base station switching method for a green cellular network for reducing energy consumption in a wireless cellular network.

Further, the present invention introduces a new concept of network-impact as described later, and is constructed so that it can be operated in a distributed manner with a low computational complexity, thereby realizing a switching on / off based energy saving based on a switching-on / off-based energy save (SWES) algorithm, considering the additional load increase of neighboring BSs, And to provide a dynamic base station switching method for a green cellular network configured to significantly reduce total energy consumption.

In order to achieve the above object, according to the present invention, based on a switching-on / off based energy save (SWES) algorithm, an additional load of neighboring base stations A dynamic base station switching method for a green cellular network configured to be able to reduce the total energy consumption in a wireless cellular network by switching off a base station whose influence on the network is minimized considering an increase in the number of base stations, A switching off processing step of sequentially determining the influence of a plurality of base stations on the network on the network and sequentially switching off the base stations having the least effect on the network; And a switching on processing step of switching on a base station switched off from the active base station again when a system load of an active base station that has not been switched off reaches a predetermined threshold value, A dynamic base station switching method for a green cellular network is provided.

Here, the switching-off processing step may include periodically switching the information on the received signal strength and the base station ID from a plurality of user equipment (UE) connected to the base station (BS b) And receives information on the system load density from a plurality of neighboring base stations BS b 'adjacent to the BS b and shares the preamble with the neighbor BSs, processing; A switching off decision step of determining an influence of each base station on the network based on the information received in the switching off pre-processing step to determine whether the base station can be turned off; And a switching-off post-processing step in which the respective base stations are turned off in order of decreasing influence on the network according to a result of the determination in the switching-off determination step.

In addition, the switching-off determining step calculates a network impact F _b that indicates the influence of each base station on the network, using the following equation,

은 기지국의 스위치 오프로 인해 인접하는 기지국에 발생하는 추가적인 시스템 부하 증가임)
(Where N _b is the set of base stations neighboring the base station BS b, ρ _n is the system load,

Is an additional system load increase that occurs at neighboring base stations due to the switch-off of the base station)

When the calculated network impact is smaller than a predetermined system load threshold, processing is performed to determine that it is possible to be off.

In addition, in the switching off post-processing step, when each base station first transmits a request to switching-off (RTSO) and receives CTSO (clear to switching-off) from all neighboring base stations, And a process of transmitting a CLSO (conformation of switching-off) to neighboring base stations and then performing a process of being turned off is performed.

Further, in the switching off post-processing step, before the respective base stations are turned off, the corresponding base station and adjacent base stations exchange information about a switching-off status including the RTSO and the CTSO, And the neighboring base stations are aware of what conditions and states are off.

Further, the switching off post-processing step may be performed such that a user terminal, which has been served by a base station that is switched off after switching off each base station, moves to an adjacent base station that provides the strongest signal strength Is performed.

In addition, when the CLSO is received from one of the base stations (BS b) adjacent to an arbitrary base station (BS b ') on the network, the base station (BS b' A process of determining that the base station BS b is to be switched off and recording its own system load including the handover traffic from the neighboring base station BS b after the neighboring base station BS b is turned off A switching on preprocessing step performed; If it is determined that the system load of the base station BS b 'is increased and it is necessary to switch on the adjacent base station BS b switched off, A switching-on determining step of performing a process of transmitting a signal; And when the neighboring base station BS b receives the switching on request and is switched on, a process of the user terminals UE located in the service coverage area reselects their service base station based on the best signal strength And performing a switching-on post-processing step to be performed.

Furthermore, the switching-on preprocessing step is configured to be performed simultaneously with the switching-off post-processing step of the switching-off algorithm.

In addition, the switching on determining step may be performed when the system load of the base station BS b 'reaches the recorded system load when the adjacent base station BS b is switched off in the preprocessing step, The base station BS B is configured to perform a process of waking up the neighboring BS b by sending RTSON (request to switching-on).

In addition, in the case where a plurality of system loads are recorded for a plurality of switched-off adjacent base stations, the switching-on determining step may be such that a base station that is finally switched off based on the last recorded system load is first switched on .

Further, according to the present invention, a dynamic base station switching method for the green cellular network described above is used to switch off a base station whose influence on the network is minimized considering an additional load increase of adjacent base stations And is configured to reduce the total energy consumption in the wireless cellular network. ≪ RTI ID = 0.0 > [0002] < / RTI >

As described above, according to the present invention, there is provided a dynamic base station switching method for a green cellular network for reducing energy consumption in a wireless cellular network, thereby achieving high computational complexity and large signaling overhead. Can solve the conventional energy minimization problem associated with BS switching known to be required.

In addition, according to the present invention, a new concept of network-impact can be introduced, so that it can be operated in a distributed manner with a low computational complexity, thereby realizing a switching on / off based switching-on a method of switching a BS in which the influence on a network is minimized in consideration of an additional load increase of neighboring BSs is proposed in order to provide a dynamic base station switching method for a green cellular network constituted using an off- Thereby significantly reducing the total energy consumption in the wireless cellular network.

1 is a diagram illustrating an actual traffic profile from a cellular wireless network.
Fig. 2 is a view showing the formula symbols used in the embodiment of the present invention.
3 is a diagram schematically illustrating the effect of a system load due to switching off of a base station.
4 is a diagram schematically showing a configuration of a switching algorithm according to the present invention.
5 is a table summarizing the SWES algorithm according to the present invention.
6 is a diagram illustrating a result of comparing energy savings of the SWES algorithm according to the present invention using a composite traffic profile.
Fig. 7 is a diagram showing the energy saving rates of weekdays and weekends.
8 is a diagram showing the effect of the hysteresis margin.
9 is a diagram showing a change in the average system load through the switching-off process for SWES _{(1, 1)} .
10 is a diagram showing changes in the BS topology characteristic through the switching-off process for SWES _{(1, 1)} .
11 is a diagram showing a change in the BS topology characteristic through the switching-off process for the SWES _{(1, 1)} .
12 is a flowchart schematically showing the overall configuration of a dynamic base station switching method for a green cellular network according to the present invention.
FIG. 13 is a flowchart schematically illustrating a specific configuration of a switching-off processing step of a dynamic base station switching method for a green cellular network according to the present invention shown in FIG.
FIG. 14 is a flowchart schematically illustrating a specific configuration of a switching-on processing step of a dynamic base station switching method for a green cellular network according to the present invention shown in FIG.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a dynamic BS switching method for a green cellular network according to the present invention will be described with reference to the accompanying drawings.

Hereinafter, it is to be noted that the following description is only an embodiment for carrying out the present invention, and the present invention is not limited to the contents of the embodiments described below.

In the following description of the embodiments of the present invention, parts that are the same as or similar to those of the prior art, or which can be easily understood and practiced by a person skilled in the art, It is important to bear in mind that we omit.

In the following description of the embodiments of the present invention, the same or similar components are denoted by the same reference numerals for brevity, and a detailed description thereof is omitted.

That is, in order to solve the conventional energy minimization problem related to BS switching, which is known to require a high computational complexity and a large signaling overhead as described later, And more particularly, to a dynamic base station switching method for a green cellular network for reducing energy consumption in a cellular network.

Further, the present invention introduces a new concept of network-impact as described later, and is constructed so that it can be operated in a distributed manner with a low computational complexity, thereby realizing a switching on / off based energy saving based on a switching-on / off-based energy save (SWES) algorithm, considering the additional load increase of neighboring BSs, To a dynamic base station switching method for a green cellular network configured to significantly reduce total energy consumption.

Next, with reference to the drawings, the detailed contents of the dynamic base station switching method for the green cellular network according to the present invention configured as described above will be described.

Referring first to Figure 1, Figure 1 is a diagram illustrating an actual traffic profile from a cellular wireless network.

More specifically, in FIG. 1, the traffic profile shown in FIG. 1 represents the normalized actual traffic load for one week recorded by any cellular operator, (Averaged) with a resolution of 1 second and a time-scale of 30 minutes.

That is, as shown in FIG. 1, it can be seen that the traffic profile at night is much lower than that during the week, and it can be seen that there is a difference in the traffic between the weekday and the weekend / holiday, It also coincides with what is presented in the literature (see reference 14).

Here, the operator has to set up the base station to support peak time traffic, but this configuration is inevitably left in an under-utilized state at other times of the day, especially at night and on weekends there is a problem.

However, the base station consumes most of its maximum power consumption even when there is little or no operation (see ref. 15), and thus the power of the entire cellular network by switching the base station dynamically according to the usage rate of the base station The consumption can be greatly reduced.

That is, the traffic profile changes according to time and space, and there is a specific period and location with a low usage rate of the base station. Accordingly, the present inventors have found that when a certain base station is switched on / What are the important parameters to consider in determining the switching decision? ", We proposed a method for managing the base station in an energy efficient manner.

In other words, the general problem of saving energy by switching base stations appears in a complex form, and a central controller is needed to get the optimal solution to this complex problem.

On the other hand, the present inventors have introduced the concept of 'network-impact' which defines how much influence a particular base station has on a network when a base station is switched off.

More specifically, the network effect is configured with deterministic parameters for each base station, such as internal (own) and external (neighbor base station) system load, and system parameters such as traffic profile and threshold The present inventors have found that it is possible to solve an energy saving problem such as a base station selection problem with linear computational complexity and to solve a distriduted switching on / off based energy saving -on / off based energy saving (SWES) algorithm.

In order to implement the SWES algorithm as described above, the present inventors have proposed three heuristic algorithms that can operate without any partial feedback or feedback, The experiment shows that the energy consumption is reduced by 10% in real traffic environment.

Next, the detailed contents of the dynamic base station switching method for the green cellular network according to the present invention will be described.

That is, in recent years, energy-efficient designs for cellular wireless networks have received particular attention (see References 1, 7, 8, 11, 12, 16-23), among which dynamic base station operation (dynamic BS opreation), and a simple analytical model has been proposed (see References 1, 8, and 16).

However, these conventional methods have focused only on ideal networks, such as hexagonal or Manhattan model networks, and have only found that energy consumption is reduced by dynamic base station switching.

The concept of a base station operation considering a network shared among networks, such as macro / macro, macro / micro, macro / femto-cell, (See references 11, 12, and 17), but they only showed how much energy was saved compared to existing methods.

Here, the basic concept of the dynamic base station operation can be referred to, for example, Reference 7 and Reference 20, and some algorithms are shown in References 21 to 23. However, There is a limit to the point that I did not do it.

Accordingly, the present inventors have proposed a practical distributed on-line algorithm for dynamic base station operation and proposed important parameters that can exhibit remarkable energy saving effect in consideration of implementation difficulty such as information feedback.

More specifically, in the embodiment of the present invention described below, first, the network model of the present invention assumes a wireless cellular network B composed of a plurality of base stations located in a two-dimensional area A, The present invention is not limited to the case where the present invention is applied to a mobile Internet, that is, a downlink communication from a base station to a user equipment (hereinafter also referred to as 'UE'). However, It should be noted that it is applicable to other types of networks.

In addition, in the embodiment of the present invention described below, the traffic model is a packet-based traffic model used for analysis and simulation. The traffic model includes a traffic arrival rate is modeled by an independent Poisson distribution with an average arrival rate λ (x, t) and the average requested file size has an average of 1 / μ (x, t) It is an exponentially distributed random variable.

Here, referring to FIG. 2, FIG. 2 is a diagram showing the mathematical symbols used in the embodiment of the present invention.

In addition, the spatial traffic variability can be obtained by setting different reach rates or file sizes for different users, and the traffic load of the UE is defined as follows: < EMI ID = 1.0 >

[Equation 1]

From each UE, the traffic load can be interpreted as a Quality of Service (QoS) request, since it is the amount of traffic that the user has to receive in order to be satisfied.

Further, the BS Selection Rule for selecting a base station to which a UE located at x EA is associated and provides bset signal strength is shown in Equation (2) below.

&Quot; (2) "

는 시간 t에서 활성 기지국(activ BS)의 집합이며, g(b, x)는 기지국 b로부터 x에 위치한 UE까지의 경로손실(path loss) 및 슬로우 페이딩(slow fading)(예를 들면, 로그노멀 셰도윙(log-normal shadowing))과 같은 다른 요인을 포함하는 평균 채널이득(average channel gain)이고, P_b는 기지국 b의 전송 전력(transmission power)이다.
here,

Is a set of active BSs at time t and g (b, x) represents a path loss and slow fading from the base station b to a UE located at x (for example, (Log-normal shadowing), and P _b is the transmission power of the base station b.

In the channel model, the physical capacity is assumed by the Shannon capacity, and the srevice rate of the UE at time t from BS b at position x is given by [ Equation 3].

&Quot; (3) "

Where BW is the system bandwidth and SINR _b (x, t) is the received signal-to-interference and noise ratio (SINR) from BS b at time t, (4). &Quot; (4) "

&Quot; (4) "

Here,? ² is noise power.

In addition, the system load must allocate a certain amount of resources, such as time or frequency, for example, according to the user's traffic and service rate, in order to ensure QoS of the UE From each system, the system load of BS b at time t is defined as a fraction of resources to provide the total traffic load within the coverage, as follows:

&Quot; (5) "

Here, A _b is the coverage of BS b (i.e., the set of UEs served by BS b), and the mathematical symbols used in the present invention are summarized and summarized in FIG.

Subsequently, the BS switching algorithm according to the embodiment of the present invention for reducing the total energy consumption of the cellular network for time T may be expressed as Equation (6) below.

&Quot; (6) "

Here, E _BS is a BS power _{consumption, a b (t) ∈ {} 0, 1} is determined by the BS switching method as the active indicator (activity indicator) of the BS b at time t ∈ [0, T], a Is the vector of the activity indicator of all BSs during time T.

In general, an energy saving problem considering BS switching can be expressed as Equation (7) below.

&Quot; (7) "

That is, according to the present invention, the system load threshold ρ ^th () for the system load for a balanced tradeoff between system stability / reliability and energy efficiency, such as the constraint shown in Equation (7) ≤ 1).

For example, at low threshold values, the BS operates in a typical manner with an average low system load (i.e., large free space), resulting in a user feeling less delay and also bursty it can be expected that the BS will be robust against traffic arrivals and call dropping will be reduced while a high threshold value close to 1 (i.e., a loose threshold) , A slight reduction in performance can result in more energy savings.

That is, at any given time interval t, the energy minimization problem of Equation (7) becomes to determine the set of active BSs with respect to system load constraints.

Here, the above problem can be reduced to the NP-complete vertex cover problem (see Reference 26), but there are two difficulties in obtaining the optimal solution to this problem Lt; / RTI >

온/오프 조합 사이의 최적 활성 BS 집합을 찾기 위해 높은 계산 복잡도가 요구되고, 또한, 둘째는, 모든 BS로부터 실제로 정보를 요청하는 중앙 처리장치가 필요하다는 점이다.
First, theoretically,

High computational complexity is required to find the optimal active BS set between on / off combinations, and second, a central processing unit is required to actually request information from all the BSs.

The inventors of the present invention overcome the above two problems and propose a distributed algorithm that can be practically implemented at present. In the following description, it should be noted that the time slot index t is excluded in order to simplify the explanation do.

Generally, the BS is placed based on the peak traffic volume and is always turned on regardless of the traffic load, so switching off some BSs with low utilization while not peak time Energy can be saved greatly.

Here, assuming that one BS is turned off, not only the UE to which the originally connected UE has to move to another BS, but also the service rate s _b (x) due to the distance between the UE and the new BS, The system load of the adjacent BS is increased.

On the other hand, when BS is turned off, inter-cell interference decreases, which is a positive aspect of the system load. In particular, some UEs exhibit a higher service rate s _b (x) in a newly connected BS than the original BS It is possible.

That is, referring to FIG. 3, FIG. 3 is a diagram schematically illustrating the influence of a system load due to switching off of a base station.

More specifically, FIG. 3 shows that a traffic load is delivered to neighboring BSs when the BS is switched off, and as shown in FIG. 3, when BS1 is off, The load is reduced, but in most cases the overall system load increases, such as BS2, BS3, BS4 are switched off.

Here, when referring to BS adjacent (other than BS b) to a set of BS adjacent to a BS b N _b Definition, provides the highest signal strength to the UE at the position of x ∈ Α _b with n ∈ N _b, specific The possibility that the BS is turned off or not is defined as the following Equation (8).

&Quot; (8) "

Also, BS n can be interpreted as a BS in which the traffic load transits after BS b is turned off, and thus BS b is off only when the adjacent BS satisfies the constraint as shown in the following equation (9) .

&Quot; (9) "

은 BS b가 오프 되었을 때 BS b로부터 인접하는 BS n으로 핸드오버(handed over) 되는 UE의 커버리지를 나타낸다.
here,

Represents the coverage of the UE handed over from BS b to the adjacent BS n when BS b is off.

은 BS b로부터 BSn으로의 외부 시스템 부하(external system load)로 정의된다.
In Equation (9), the original system load? _N is defined as the internal system load of BS n, and the system load increase due to the switch-off of the adjacent BS

Is defined as an external system load from BS b to BSn.

That is, in the example shown in FIG. 3, since the system load of the BS 1 exceeds the threshold by the load of the external system, BS 2 and BS 3 should not be switched off. Considering the system load of the adjacent BS after switching off, Only BS1 and BS4 are switchable.

In addition, it is desirable to select BS1 (for subsequent additional traffic) since turning off BS1 has a larger spare space than BS4.

을 고려한 "네트워크 영향(network impact)"의 개념을 도입하였다.
Moreover, since quantizing the system load of the network (more specifically, the neighboring BSs) is affected by the switching off process, the present inventors have found that the original system load ρ _{n < / RTI &gt}; plus additional load _{< RTI ID = 0.0 >}

The concept of "network impact" is introduced.

More specifically, the network influence for the determination of the switching off BS b is defined as: " (10) "

&Quot; (10) "

및 전체 시스템 부하의 증가량

등과 같이, 여러 가지 다른 방법으로 모델링 될 수 있음에 유념해야 한다.
Here, in the conventional manner, it is assumed that BS n ∈ N _b is the largest among the neighbor BSs, and then selects the worst BS having the smallest free space for the subsequent traffic request, Average system load of

And the overall system load increase

And the like, may be modeled in several different ways.

That is, the SWES algorithm according to the present invention for turning off the BS having the minimum network influence can be expressed by Equation (11) below.

&Quot; (11) "

집합(set)으로 매우 낮은 계산(선형) 복잡도를 가지나, 구현을 위해 중앙 제어장치가 요구된다.
Here, this process is repeated until there is no active BS having an adjacent BS satisfying the constraint shown in Equation (9)

A set has very low computational (linear) complexity, but a central control unit is required for implementation.

The system load is also a simple yet powerful metric capturing network impacts that are dependent on the neighboring environment, such as traffic load and the number of adjacent BSs, distance and load, A switching on / off algorithm based on a similar type of system load with slightly different formulas is presented in various technical literatures (see References 1, 8 and 25), but the present invention has developed a disibuted algorithm This is in contrast to the prior art, which only presents a centralized algorithm.

Next, the concrete contents of the distributed BS switching algorithm according to the present invention and its protocol-level implementation will be described.

Referring to the switching off algorithm, the decision criterion defined in Equation (10) above depends only on the information about the BS and neighboring BSs, so that the decision of switching off is a problem in each BS Can be limited.

The distributed switching-off algorithm proposed in the present invention is simple in that the system information such as signal strength and system load is periodically shared between the BS and the UE and whether or not each BS is to be turned off is simple, No device is required.

More specifically, referring to FIG. 4, FIG. 4 is a diagram schematically showing a configuration of a switching algorithm according to the present invention.

As shown in FIG. 4, the switching-off algorithm according to the present invention is roughly divided into a pre-processing step, a decision step, and a post-processing step.

First, the pre-processing step is generally performed in a cellular network, such as 3GPP (-LTE) and IEEE 802.16 (e / m) Feedback on the received signal strength is periodically fed back.

If BS b is off, the user in the coverage is moved to the lane's BS (i.e., the best BS after turning off the currently serving BS b), and until the movement is completed, With the BS ID.

This information is also used to calculate how much load increase will occur to adjacent BSs, and as described below, feedback can be further reduced at the cost of some performance loss in energy savings.

In addition, the system load is shared between adjacent BSs periodically (every few minutes) and / or when sudden system load fluctuations occur.

Next, in the determination step, each BS first calculates a network influence as shown in Equation (10) based on the information received from the user and the neighboring BS, and determines whether or not it can be turned off.

That is, the switching off decision is as follows.

[Determination of Switching Off]

F _b <? ^th , a switch-off request is sent to the adjacent BS, N _b .

Here, in a distributed operation, two or more BSs in which adjacent zones overlap can be switched off at the same time, which causes an overload to an adjacent base station.

To prevent this problem, each BS is configured to be switched off only when it first sends a request to switching-off (RTSO) and receives CTSO (clear to switching-off) from all BSs adjacent thereto, Prior to off, the BS sends a conformation, i.e., CLSO (conformation of switching-off), to its neighboring BSs.

Next, the post-processing step is turned off only when BS b receives CTSO from all neighboring BSs as described above, so that the UE serviced by the switching off BS transmits the neighbor BS Which is similar to a conventional hand-over, except that groups of UEs move at the same time.

Here, the primary effect of the group handover is to be able to predict / prepare the handover first, for example, the latest state-of-the-art group handover techniques, as shown in references 28 and 29, Can be used in conjunction with the switching-off algorithm according to the present invention to support it.

Also, for example, a control signal for switching off decisions such as RTSO / CTSO and CLSO can be used to pre-trigger group handover.

Next, the switching-on algorithm according to the present invention will be described.

One way to implement the switching-on algorithm is to reverse the switching-off algorithm, i.e., the basic idea of the switching-on algorithm is to have the BS switched on when the system load becomes equal to the value at which the BS is switched off .

However, since the OFF-state BS does not have information on the current system load, it can not make a switching-on decision by itself. Thus, the switching-on algorithm according to the present invention is configured to make a switching ON decision based on the neighbor BS.

That is, before the BS is turned off, the BS and its neighboring BSs exchange information about a switching-off status, such as, for example, RTSO and CTSO, And knows whether the corresponding BS is off in the state.

The switching-on algorithm also has three steps of pre-processing, determination and post-processing, like the above-mentioned switching-off algorithm.

First, the preprocessing step works in conjunction with the post-processing step of the switching-off algorithm described above. When the BS (i.e., BS b ') receives the CLSO from one of the neighboring BSs (i.e., BS b) 'Knows that BS b is to be switched off, and after BS b is off, BS b' records its system load including handover traffic from BS b.

Then, the determining step is such that when the system load of BS b 'reaches the recorded system load when the adjacent BS b is switched off, BS b' sends a request to switching-on (RTSON) to wake up BS b ).

Here, when a plurality of system loads are recorded for a plurality of switched-off neighboring BSs, the last recorded system load is taken into consideration, so that the last switched-off BS is first switched on.

라 하면, 스위칭 온 결정은 다음과 같다.
That is, when BS b is switched off, the system load recorded in BS b '

, The switching on decision is as follows.

[Determination of Switching On]

이면, 스위치 오프된 인접하는 BS b에 스위칭 온 요청을 보낸다.

, It sends a switching-on request to the adjacent BS b that is switched off.

은 작은 상수값(small constant value)이다.
here,

Is a small constant value.

Subsequently, the post-processing step wakes up when the BS receives the RTSON, so that the UEs located in the service coverage area reselect (i.e., hand over) their service BSs based on the best signal strength.

Here, if the traffic pattern changes at the same rate over the entire space, the switching-on algorithm according to the present invention is simply a reverse operation of the switching-off process.

Next, an actual implementation of the algorithm configured as described above will be described.

In order to determine BS switching, some feedback is required from the UEs and neighboring BSs, and feedback information may reduce system performance and increase the difficulty of actual implementation, so that it is required to reduce feedback information efficiently.

First, the load of the handover system will be described. Generally, the UE transmits feedback information regarding the received signal strength from the BS functioning for adaptive modulation (see Reference 27), but the handover system load Additional feedback is required, such as the lane signal strength and its associated BS ID.

This additional information increases the burden on the system, for example, more than 6 bits per channel are required to feed back the overall channel condition for adaptive modulation, and one way to reduce feedback is to load the handover system load (12).

&Quot; (12) &quot;

Where k is a compensation factor that depends on the placement of the BS and neighboring BSs.

Assuming that the network has a homogeneous deployment like an ideal hexagonal cellular network, the coefficients are estimated to be one.

Based on the approximation shown in [Equation 12], the network influence for determining the switching off BS can be modified as in Equation 13 below.

&Quot; (13) &quot;

Next, describing the system load of adjacent BSs, the amount of feedback for the system load between the BSs compared to the load of the handover system can be exchanged through a high-speed wired backhaul, which is not a major problem Do not.

However, since it is possible to increase the system burden to implement a message actually exchanged, the present inventors can simply estimate the system load of an adjacent BS, as shown in Equation (14) below, We have proposed a method to reduce the head.

&Quot; (14) &quot;

This approximation is valid when the traffic load is uniformly distributed and is not valid for non-homogeneous traffic loads, but the user's traffic pattern may change continuously in the spatial domain rather than abruptly The error is small.

Therefore, from the above description, the network influence for determining the switching off BS can be expressed by the following equation (15).

&Quot; (15) &quot;

Further, by combining [Equation 12] and [Equation 14], the network effect can be calculated without information feedback as in Equation 16 below.

&Quot; (16) &quot;

Here, it should be noted that the simplest algorithm, SWES _{(0, 0)} , exactly matches the contents of the prior art presented by the present inventors (see Reference 1).

Referring to FIG. 5, FIG. 5 is a table summarizing the SWES algorithm according to the present invention.

That is, as shown in Figure 5 and above, the required information depends on what network influence is used to determine the switching off BS, and in fact, for example, a wired BS-BS connection And a wireless BS-UE connection, as well as system performance such as feedback burden and loss in data rate, for example.

In addition, in the actual implementation, the above-mentioned algorithm, in the actual implementation, performs the message exchange such as RTSO, CTSO and CLSO with the switching off BS and its adjacent BSs, It is possible to prevent a plurality of BSs having BSs from being switched off at the same time. However, there is a problem in that the SWES algorithm can operate inefficiently due to a synchronous operation for allowing such messages to be exchanged at the same time have.

For example, BS A and B send RTSO to the same neighboring BS C at the same time, BS C sends CTSO to BS B, and BS B sends RTSO to another neighboring BS (i.e., BS D) , Then both BS A and BS B can not be switched off.

In order to alleviate this problem, the inventors of the present invention have assumed a network operating asynchronously.

That is, in order to implement the above-described algorithm as a synchronous operation, for example, an RTSO waiting with RTSO transmitting a RTSO at a random waiting time (BS A and BS B) Similar to traditional methods for mitigating collisions at the protocol design stage (see reference 31), such as multi-step CTSO where BS D responds CTSO to BS A if CLSO does not arrive within random latency Additional processing is required to prevent conflicts of message exchanges.

Further, another problem arising from system load is that it is prone to fluctuate, and due to the high change in system load, the BS may be able to perform switching (or switching) operations similar to the ping-pong effect Off and on may be repeated inefficiently.

The present inventors, to solve this problem, was introduced into the △ _h hysteresis margin (hysteresis margin) for an actual implementation, i.e., the formula below the system load threshold for BS switching off and on, respectively 17] can be expressed as follows.

&Quot; (17) &quot;

Here, the determination process with the hysteresis margin can reduce ineffective switching off and on, while the amount of energy savings is reduced by the low system load threshold, and therefore, inefficient switching and energy A trade-off between savings should be considered.

Next, numerical analysis of the SWES algorithm according to the present invention will be described.

That is, in order to simulate the SWES algorithm according to the present invention as described above, the present inventors considered an actual 3G network structure composed of 18 BSs in an area of 5 × 5 km ² (refer to Reference 33) A wrap-around technique was applied to avoid the edge effect (see reference 34).

In addition, it is assumed that the traffic load is spatially homogeneous and varies by scaling the traffic arrival rate. When the system load of any BS reaches ρ ^th as the traffic arrival rate increases, Was treated as relative system load = 1.

Further, in the simulation, the threshold value ρ ^th for the SWES algorithm according to the present invention is set to 0.6 in consideration of the system reliability, and the traffic load of the peak traffic time is applied to the SWES algorithm The system load was generalized to be 1.

Also, using the general values of transmission power and operational energy for BS per unit time, namely P _i = 20 W and E _BS = 865 W (see ref. 13 ).

In addition, other parameters, including the channel propagation model and BS characteristics, may be used to determine the propagation characteristics of the urban macro model (e.g., the modified COST 234 Hath path loss model) urban macro model according to the IEEE 802.16m evaluation methodology document (EMD) (see reference 34).

More specifically, referring to FIG. 6, FIG. 6 is a diagram illustrating a result of comparing energy savings of the SWES algorithm according to the present invention while changing a relative system load from 0 to 1 using a composite traffic profile.

6, the lower the relative system load is low can be seen that the expected high energy savings, that is, compared to the energy consumption of the exhaustive search optimization _{(Optimal exhaustive search), SWES (} 1 having a full feedback information _{, 1)} consumes 8% more energy than the optimization algorithm for all system loads.

This means that a performance similar to that of a conventional central processing optimization algorithm that investigates all 2 ¹⁸ on / off BS combinations with a simple distribution algorithm with linear complexity can be obtained.

In addition, the algorithm having a portion feedback grows and has performance difference of the one operating well when the smaller the relative system load, in particular, the more the relative system load is close to the first optimization, this trend _{SWES (0, 1), SWES} (1, 0 ₎ , And the performance difference between SWES _{(0, 0)} is about 4-5%.

In addition, a complementary coefficient k is required for SWES _{(1, 0)} and SWES _{(0, 0)} to capture the effect of signal strength degradation when a traffic load is transmitted from a switched-off BS to an adjacent BS. , The simulations were performed assuming that the complement coefficient is 1.

Next, a simulation result of the energy saving effect of the SWES algorithm according to the present invention will be described in consideration of the actual traffic profile shown in FIG.

7 and 8, FIG. 7 is a diagram showing the energy saving rate of weekday and weekend, and FIG. 8 is a diagram showing the effect of the history margin.

More specifically, FIG. 7 shows the energy savings over a day under the actual traffic profile when the hysteresis margin is zero, and shows a large energy savings of about 55% and 80% during weekdays and weekends, respectively, as shown in FIG. It can be seen that the performance difference between the SWES algorithm according to the present invention and the conventional total irradiation optimization is less than 10%.

This is because the traffic load for the main time zone during the day is low, that is, based on the traffic load profile shown in FIG. 1, the time zone during which the traffic per day is less than 10% of the peak is about 30% during weekdays and about 43% (See Reference 20), and as shown in Fig. 6, the performance difference for such a period is relatively small.

8, the influence of the hysteresis margin on the SWES _{(1, 1)} appears, and when the hysteresis margin _h increases from 0.01 to 0.25, It is possible to prevent frequent switching on / off, but this causes a slight loss in energy saving.

Therefore, in system design, it is important to select an appropriate hysteresis margin that leads to a good trade-off between energy savings and system stability.

As described above, the energy savings of the SWES algorithm are closely related to the system load depending on the environment such as the internal system load, the external system load, the number of adjacent BSs, and the distance between BSs.

If that is, with reference to FIGS. 9 to 11, in Fig. 9 is a diagram showing a change in the average system load through the switching-off process on SWES _{(1, 1),} 10 and 11 are SWES _{(1, 1)} Fig. 8 is a diagram showing a change in the BS topology characteristic through the switching-off process.

11 shows the average of the number of BSs adjacent to the switching off BS and all the BSs, and in Figures 9 to 11, a very low system &lt; RTI ID = 0.0 &gt; Starting with a load (= 0.04), the BS was turned off one by one according to the SWES algorithm.

First, referring to FIG. 9, FIG. 9 shows three average system loads through the switch-off process, i.e., the internal (Int.) And external (Ext.) System loads of the switching off- Represents the average system load of neighboring BSs, and the average (system load) represents the average system load of all BSs in the network.

As shown in Figure 9, the average system load of the switching off BS and neighboring BSs is lower than the average system load of all BSs, which is less influential on the network than switching off the other BSs, The BS having the external system load is switched off first.

Also, in the initial switching process, the average system load of the switching off BS is slightly lower than that of adjacent BSs, but the opposite result is obtained at the end of the process. This is because the internal factor of the switching off BS in the low system load region With more network effects, external factors by neighboring BSs are significant in high system load regions.

10 and 11 show the average distance from the BS to the neighboring BSs and the average of the neighboring BSs, respectively. This is because the operation of the switching-off algorithm and the BS topology Relationship.

10 and 11, the average distance / number is shown for the switching off BS and all BSs, and the average distance from the BS to the adjacent BS increases monotonically as the density of the active BS decreases .

In particular, the average distance from the BS to the neighboring BSs has a value lower than that for all BSs, which means that the switching off BS in the higher BS density region has less impact on the network than the BS in the lower BS density region. .

However, on the other hand, the average of the number of adjacent BSs remains approximately the same during the switching off process (e.g., the variance is less than one), which is why the coverage of each BS increases even if the BS density decreases.

Also, since the system load of the BS increases due to interference from neighboring BSs as the number of neighboring BSs increases, the average of the number of switching offs BSs is slightly lower than that for all BSs, i.e., This implies that the distance between BSs is a more important factor than the number of neighboring BSs in designing network effects.

As described above, the SWES algorithm according to the present invention can be implemented as described above, and the dynamic BS switching method for the green cellular network according to the present invention can be implemented using the SWES algorithm.

12 and 14 are flowcharts schematically showing the overall configuration of a dynamic base station switching method for a green cellular network according to the present invention. Figs. 13 and 14 are flowcharts FIG. 3 is a flowchart schematically showing specific configurations of a switching-off processing step and a switching-on processing step of a dynamic base station switching method for a green cellular network according to the invention;

That is, as shown in FIG. 12, a dynamic base station switching method for a green cellular network according to the present invention determines the influence of a plurality of base stations on a network on each network, Off state of the active base station (active BS) is switched on again when the system load of the active base station (active BS) that has not been switched off reaches a predetermined threshold value And a switching-on processing step S12.

13, each of the base stations BS b on the network receives a signal strength from a plurality of user equipment (UE) connected to the base station BS b, Receives information on the system load density from a plurality of neighbor base stations (BS b ') adjacent to the base station (BS b) and receives information on the system load density from the neighbor base stations A switching-off pre-processing step (S21) for calculating an influence of each base station on the network based on the information received in the switching-off preprocessing step, Off decision step S22 and the determination result in the switching off decision step, And a switching-off post-processing step S23 in which the switch-off processing is turned off.

14, when the CLSO is received from one of the base stations (BS b) adjacent to an arbitrary base station (BS b ') on the network, b ') determines that the neighboring base station (BS b) is to be switched off, and after the neighboring base station (BS b) is turned off, its own system load including the handover traffic from the neighboring base station A switching on preprocessing step S31 for performing a process of recording a BS b 'on a base station BS b', and if it is determined that it is necessary to switch on an adjacent base station BS b switched off due to an increased system load of the BS b ' A switching on decision step S32 in which a process of sending a switching on request to an adjacent base station BS b switched off is performed and a neighboring base station BS b receives a switching on request and is switched on, There is user equipment (UE) located in the station can be configured to include a switch-on after the treatment step (S33) to be re-selection process is performed to their serving base station based on the best signal strength.

In addition, more detailed contents of each of the above steps can be configured with reference to the contents of the SWES algorithm according to the present invention with reference to FIGS. 1 to 11 and [Equation 1] to [Equation 17].

Therefore, the dynamic BS switching method for the green cellular network according to the present invention can be implemented as described above.

Also, by implementing a dynamic base station switching method for a green cellular network according to the present invention as described above, according to the present invention, dynamic base station switching for a green cellular network for reducing energy consumption in a wireless cellular network A conventional energy minimization problem related to BS switching, which is known to require a high computational complexity and a large signaling overhead, can be solved.

In addition, according to the present invention, a new concept of network-impact can be introduced, so that it can be operated in a distributed manner with a low computational complexity, thereby realizing a switching on / off based switching-on a method of switching a BS in which the influence on a network is minimized in consideration of an additional load increase of neighboring BSs is proposed in order to provide a dynamic base station switching method for a green cellular network constituted using an off- Thereby significantly reducing the total energy consumption in the wireless cellular network.

Moreover, according to the present invention, a plurality of SWSEs using an approximation of network influence as a decision metric to further reduce signaling and implementation overhead due to air and backhaul, And shows how the above algorithm is implemented at the protocol level.

In addition, as a result of the simulation for evaluating the energy saving amount, the SWES algorithm according to the present invention is evaluated as having a potential energy saving effect of 50 to 80% or more based on the actual traffic profile in the urban area of the metropolitan area, .

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. It will be understood by those skilled in the art that various changes, modifications, combinations, and substitutions can be made in the present invention based on design requirements and various other factors.

Claims (11)

Based on a switching-on / off based energy save (SWES) algorithm, a base station that minimizes the influence on a network in consideration of an additional load increase of neighboring base stations (BSs) A method for dynamic base station switching for a green cellular network configured to reduce total energy consumption in a wireless cellular network by switching-off,
A switching off processing step of sequentially determining the influence of a plurality of base stations on the network on the network and sequentially switching off the base stations having the least effect on the network; And
And a switching on processing step of switching on a base station switched off from the active base station again when a system load of an active base station that has not been switched off reaches a predetermined threshold,
Wherein the switching-
Each base station BS b on the network periodically receives information on a received signal strength and a base station ID from a plurality of user equipment (UE) connected to the base station BS b, A switching off preprocessing step of receiving information on a system load density from a plurality of neighboring base stations BS b 'adjacent to the base station BS b' and sharing the system load density with neighboring base stations;
A switching off decision step of determining an influence of each base station on the network based on the information received in the switching off pre-processing step to determine whether the base station can be turned off; And
And a switching-off post-processing step in which the respective base stations are turned off in order of decreasing influence on the network in accordance with a result of the determination in the switching off decision step,
Wherein the switching-off determination step comprises:
(F _b ) indicating the influence of each base station on the network itself, using the following equation: &lt; EMI ID =

(Where N _b is the set of base stations neighboring the base station BS b, ρ _n is the system load,

Is an additional system load increase that occurs at neighboring base stations due to the switch-off of the base station)

And if the calculated network influence is less than a predetermined system load threshold, processing is performed to determine that it is off-possible.
delete delete The method according to claim 1,
Wherein the switching off post-
When each base station first transmits a request to switching-off (RTSO) and receives CTSO (clear to switching-off) from all adjacent base stations, the corresponding base station transmits a CLSO (conformation of switching -off) is sent and then the process is turned off.
5. The method of claim 4,
Wherein the switching off post-
Before the respective base stations are turned off, the base station and neighboring base stations exchange information on the switching-off status including the RTSO and the CTSO, Wherein the base stations are configured to be aware of the base stations.
6. The method of claim 5,
Wherein the switching off post-
Characterized in that a process is carried out such that a user terminal, which has been served by a base station switched off after switching off each base station, is moved to an adjacent base station providing the strongest signal strength except for the switched off base station A dynamic base station switching method for a green cellular network.
The method according to claim 6,
The switching-
When the CLSO is received from one of the base stations BS b 'adjacent to an arbitrary base station BS b' on the network, the base station BS b 'switches off the neighboring base station BS b A switching on preprocessing step of performing a process of recording its own system load including handover traffic from the neighboring base station (BS b) after the neighboring base station (BS b) is turned off;
If it is determined that the system load of the base station BS b 'is increased and it is necessary to switch on the adjacent base station BS b switched off, A switching-on determining step of performing a process of transmitting a signal; And
When the neighboring base station BS b receives the switching on request and is switched on, a process of the user terminals UE located in the service coverage area reselects their service base stations based on the best signal strength is performed And a processing step after the switching-on of the base station.
8. The method of claim 7,
The switching-on pre-
Wherein the step of performing the switching-off post-processing of the switching-off algorithm is performed simultaneously with the switching-off post-processing step of the switching-off algorithm.
9. The method of claim 8,
The switching-
When the system load of the base station BS b 'reaches the recorded system load when the adjacent base station BS b is switched off in the preprocessing step, the base station BS b' on of the neighboring base station (BS b) to send a signal to the base station (BS b) to wake up the neighboring base station (BS b).
10. The method of claim 9,
The switching-
Characterized in that when a plurality of system loads are recorded for a plurality of switched off adjacent base stations, the last switched off base station based on the last recorded system load is configured to be first switched on The base station switching method.
delete

Images