WO2023213405A1

WO2023213405A1 - Radio access network (ran) cell sleep management

Info

Publication number: WO2023213405A1
Application number: PCT/EP2022/062216
Authority: WO
Inventors: Paddy Farrell; Shubham Jain
Original assignee: Telefonaktiebolaget Lm Ericsson (Publ)
Priority date: 2022-05-05
Filing date: 2022-05-05
Publication date: 2023-11-09

Abstract

A method for managing a capacity cell of a wireless communications network. The method includes obtaining a first set of quality-of-service, QoS, values for the capacity cell. The method also includes, after obtaining the first set of QoS values for the capacity cell, configuring the capacity cell to implement a first sleep threshold configuration (STC), wherein the first STC comprises one or more sleep thresholds. The method also includes, after configuring the capacity cell to implement the first STC: i) obtaining a second set of QoS values for the capacity cell, ii) calculating a first reward value using the first and second set of QoS values, and iii) obtaining a first state vector for the capacity cell. The method also includes, based on the first reward value, selecting a second STC from a set of STCs associated with a state identified by the first state vector. The method also include configuring the capacity cell to implement the second STC if the selected second STC is different than the first STC.

Description

RADIO ACCESS NETWORK (RAN) CELL SLEEP MANAGEMENT

TECHNICAL FIELD

[001] Disclosed are embodiments related to the management of sleep parameters for cells of a radio access network (RAN).

BACKGROUND

[002] A typical radio access network (RAN) comprises many base stations, each serving one or more cells to provide network coverage and meet the ever-increasing mobile traffic demand in the 5G era. A RAN is typically designed to accommodate peak traffic. Accordingly, having a cells in the network active during “off-peak” hours would likely waste much energy. Deactivating some idle base stations that do not service any traffic demand is an effective approach to reduce cellular network energy usage. Cell sleep optimization is one of the main factors to enhance energy and cost-efficiency in cellular networks. Accordingly, a cell sleep feature is typically deployed in densely populated networks to automatically activate and deactivate cells based on the traffic load.

[003] Cell Sleep Mode Feature

[004] The main benefit of the Cell Sleep Mode feature is automatic energy saving. In densely populated networks, so called “capacity” cells are deployed. A capacity cell is a cell at least partly within the coverage area of a coverage cell. Without this feature, cells are constantly turned on. With the Cell Sleep Mode feature, the capacity cells can detect low traffic conditions by themselves and turn themselves off to save energy, with confirmed support from coverage cells. That is, a capacity cell automatically enters sleep state when its traffic decreases below certain thresholds during an allowed time interval. A coverage cell monitors traffic conditions to determine whether or not to re-activate a sleeping capacity cell. This is a Self-Organizing Network (SON) capability that is adaptive to the network traffic conditions.

[005] When a capacity cell is turned off, the traffic load existing in the cell is off-loaded to one or more coverage cells. Coverage cells can be configured manually or automatically. The Cell Sleep Mode feature provides a capability, Coverage Cell Discovery, which can automatically detect neighbor cells that are capable to serve as coverage cells. Enabling the Cell Sleep Mode feature is not required on coverage cells. Once capacity cells are turned off for energy saving reasons, interference in the network is reduced. Therefore, the throughput of user equipments (UEs) can be improved. Capacity cells are dynamically and automatically turned on and off based on current traffic load. This can reduce power consumption and reduce inter-cell interference. Therefore the Cell Sleep Mode feature can reduce operating expenses (OPEX). The target deployment scenario for this Cell Sleep Mode feature is described in 3GPP Technical Report (TR) 36.927 V12.0.0, Potential solutions for energy saving for E-UTRAN.

SUMMARY

[006] Certain challenges presently exist. For instance, current approaches for configuring cell sleep parameters area based on only a few traffic parameters and this can lead to performance degradation. For example, the current sleep functions do not consider any Quality of Service (QoS) parameters (such as, for example, throughput, packet loss, inter-cell interference, latency, etc.) before or after cells sleep is active. There is a high expert level workload involved in configuring the current cell sleep features. While there are parameters to control Cell Sleep, it is not configured per cell due to the workload and complexity involved. With the advancement of technologies such as 2G, 3G, 4G, and the complex architecture of 5G, optimization of cell sleep to save energy becomes more complex and affects QoS. Hence, there is a need for an approach that can optimize cell sleep and retain the QoS.

[007] Accordingly, in one aspect there is provided a method for managing a capacity cell of a wireless communications network. The method includes obtaining a first set of quality- of-service, QoS, values for the capacity cell. The method also includes, after obtaining the first set of QoS values for the capacity cell, configuring the capacity cell to implement a first sleep threshold configuration (STC), wherein the first STC comprises one or more sleep thresholds. The method also includes, after configuring the capacity cell to implement the first STC: i) obtaining a second set of QoS values for the capacity cell, ii) calculating a first reward value using the first and second set of QoS values, and iii) obtaining a first state vector for the capacity cell. The method also includes, based on the first reward value, selecting a second STC from a set of STCs associated with a state identified by the first state vector. The method also include configuring the capacity cell to implement the second STC if the selected second STC is different than the first STC.

[008] In another aspect there is provided a computer program comprising instructions which when executed by processing circuitry of a network node causes the network node to perform any of the methods disclosed herein. In one embodiment, there is provided a carrier containing the computer program wherein the carrier is one of an electronic signal, an optical signal, a radio signal, and a computer readable storage medium. In another aspect there is provided a network node that is configured to perform the methods disclosed herein. The network node may include memory and processing circuitry coupled to the memory.

[009] An advantage of the embodiments disclosed herein is that they provide, a the cell level, automatic configuration of cell sleep parameters that are unlikely to adversely affect QoS. Thus, compared to conventional cell sleep methods, the embodiments provide a better QoS while still providing the advantage of reduced OPEX.

BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings, which are incorporated herein and form part of the specification, illustrate various embodiments.

[0011] FIG. 1 illustrates a system according to an embodiment.

[0012] FIG. 2 is a message flow diagram according to an embodiment.

[0013] FIG. 3 is a flowchart illustrating a process according to an embodiment.

[0014] FIG. 4 is a flowchart illustrating a process according to an embodiment.

[0015] FIG. 5 illustrates an example of a Q-table.

[0016] FIG. 6 is a block diagram of a network node according to an embodiment.

DETAIEED DESCRIPTION

[0017] FIG. 1 illustrates a system 100 according to an embodiment. System 100 includes a first base station 104 (e.g., a macro base station) that serves a first cell 114 and a second base station 102 (e.g., a femto base station) that serves a second cell 112. In this example, cell 112 is within cell 114. Hence, cell 114 is a coverage cell and cell 112 is a capacity cell. FIG. 1 shows a user equipment (UE) 101 located in cell 112, which means UE 101 is also located in cell 114. Because UE 101 is located in cells 112 and 114, UE 101 could be served by either base station 104 or base station 102. In one example, cell 112 can enter a sleep state when base station 104 has the resources to serve all of the UEs within cell 114. But if base station 104 cannot server all the UEs within cell 114, then base station 112 can put cell 112 in an active state so that base station 112 can server at least some of the UEs within cell 114 (i.e., the UEs within cell 112), thereby reducing the load on base station 112.

[0018] As noted above, the currently implemented cell sleep feature may adversely affect QoS and is not optimized. Accordingly, this disclosure provides a cell sleep management function (CSMF) 106 that implements an intelligent method to optimize the sleep mode parameters of a cell (e.g., sleep thresholds) such that the QoS parameters (e.g., throughput, packet loss, handover (HO) success rate) are not adversely affected. The objective is to sleep for as long as possible without affecting QoS within a tolerance that is accepted and tunable. First, the coverage and capacity cells are identified. Then the Cell Sleep threshold parameters are optimized per cell using a state-action-reward policy definition based on a calculated coverage area QoS. Quieter cells with less utilization will be put to sleep first and based on rewards the sleep threshold would be configured in a coverage area.

[0019] FIG. 2 is a message flow diagram illustrating a message flow according to some embodiments. The message flow begins with CSMF 106 obtaining from base station 102 state information for cell 112 and obtaining form base station 104 state information for cell 114.

[0020] The state information for cell 112 may include: i) information indicating the number of UE’s currently active in cell 112 (e.g., the total number of RRC connection in cell 112); ii) a resource usage ratio (e.g., Physical Resource Block (PRB) utilization) for cell 112; and iii) QoS information. Likewise, the state information for cell 114 may include: i) information indicating the available capacity of cell 114 (e.g., the maximum number of RRC connections cell 114 can handle minus the current total number of RRC connections in cell 114; ii) a resource usage ratio (e.g., Physical Resource Block utilization) for cell 114; and iii) QoS information.

[0021] The QoS information may include: i) a latency value including latency of the cell; ii) throughput value indicating an average throughput; iii) a packet loss ratio; iv) radio bearer drop rate; and v) interference level value indicating a level of interference experienced by the cell.

[0022] After collecting the state information for cell 112 and cell 114, CSMF 106 uses the state information to select a sleep threshold configuration (STC), which comprises one or more sleep threshold values that are used by base station 102 to detect whether or not a sleep opportunity exists. After selecting the STC, CSMF 106 provides the selected STC to base station 102.

[0023] Base station 102 then uses the STC to determine whether or not a sleep opportunity exists. For example, in one embodiment, STC consists of two threshold values: a PRB utilization threshold value (PrbSleep) and an RRC connection threshold value (denoted RrcConSleep). For example, in one embodiment a sleep opportunity may exist if: 1) the downlink (DL) PRB usage percentage (utilization) is less than or equal to PrbSleep; 2) the number of RRC connectdion in the cell is less than or equal to RrcConnSleep; and 3) both conditions remain true for at least a predetermined period of time.

[0024] Assuming that base station 102 detects that a sleep opportunity exists, base station 102 transmits to base station 104 a cell sleep request and then receives from base station 104 a cell sleep response. Assuming the cell sleep response grants the cell sleep request, base station 102 puts cell 112 in a sleep state to save energy (e.g., power provided to the components that serve cell 112 is reduced).

[0025] CSMF 106, meanwhile, periodically obtain up-to-date state information from base station 102 and base station 104 for cells 112 and 114, respectively. Using the up-to-date state information, CSMF 106 again selects an STC for cell 112 and provides the selected STC to base station 102, which then uses the STC to determine whether another sleep opportunity exists. This process may continue indefinitely.

[0026] FIG. 3 is a flowchart illustrating a process 300 performed by CSMF 106 according to some embodiments. Process 300 may begin in step s302 . step s302 comprises obtaining a first set of QoS values for a capacity cell (e.g., cell 112). Step s304 comprises, after obtaining the first set of QoS values for the capacity cell, configuring the capacity cell to implement a first STC (e.g., by providing the STC to base station 102), wherein the first STC comprises one or more sleep thresholds. Step s306 comprises, after configuring the capacity cell to implement the first STC: i) obtaining a second set of QoS values for the capacity cell, ii) calculating a first reward value using the first and second set of QoS values, and iii) obtaining a first state vector for the capacity cell. Step s3O8 comprises, based on the first reward value, selecting a second STC from a set of STCs associated with a state identified by the first state vector. Step s310 comprises configuring the capacity cell to implement the second STC if the selected second STC is different than the first STC (e.g., by providing the second STC to base station 102).

[0027] In some embodiments process 300 also includes, after selecting the second STC from the set of STCs associated with the state identified by the first state vector: obtaining a third set of QoS values for the capacity cell; calculating a second reward value using the second and third set of QoS values; obtaining a second state vector for the capacity cell; based on the second reward value, selecting a third STC from a set of STCs associated with a state identified by the second state vector; and if the selected third STC is different than the second STC, configure the capacity cell to implement the third STC.

[0028] In some embodiments, the first set of QoS values comprises one or more of: a latency value indicating an average latency associated with the capacity cell, a throughput value indicating an average throughput for the capacity cell, a packet loss value indicating an average packet loss for the capacity cell, bearer drop rate value indicating a bearer drop rate for the capacity cell, or an interference value indicating a level of interference experienced by the capacity cell.

[0029] In some embodiments, each STC included in the set of STCs associated with the state identified by the first state vector has a Q-value, and selecting the second STC from the set of STCs comprises selecting, from the set of STCs, the STC having the largest Q value.

[0030] In some embodiments, a first energy value indicating a current traffic load for the capacity cell (e.g., physical resource block, PRB, utilization) for the capacity cell; and a second energy value indicating a resource availability (e.g. PRB availability) for a coverage cell that covers the capacity cell. In some embodiments, the first state vector further comprises: a first traffic value indicating a total number of radio resource control, RRC, connections in the capacity cell; and a second traffic value indicating an available connection capacity for the coverage cell. [0031] In some embodiments process 300 also includes, prior to selecting the second STC from the set of STCs associated with a state identified by the first state vector, determining whether the first state vector is in a look-up table. In some embodiments process 300 also includes, as a result of determining that the first state vector is not in the look-up table, using a Directed Acyclic Graph, DAG, to add to the look-up table a record comprising the first state vector and the set of STCs.

[0032] In some embodiments, each said STC comprises a traffic load sleep threshold (e.g., a PRB sleep threshold), wherein the capacity cell has an opportunity to enter a cell sleep state if a set of sleep conditions are satisfied, wherein a first one the of the sleep conditions is that the current traffic load of the capacity cell must be less than or equal to the traffic load sleep threshold. In some embodiments, each said STC further comprises an RRC connection sleep threshold and a second one the of the sleep conditions is that the current number of RRC connection in the capacity cell is less than or equal to the RRC connection sleep threshold.

[0033] FIG. 4 is a flowchart illustrating a process 400 performed by CSMF 106 according to some embodiments.

[0034] Process 400 may begin in step s401. Step s401 comprises initializing a look-up table (a.k.a., the “Q-table”). FIG. 5 illustrates an example Q-table 500. As shown in FIG. 5, Q- table 500 (or “table 500” for short) contains a list of state vectors (statel, state2, ...). Each state vector uniquely identifies a different state. For each state vector, table 500 associates the state vector with a set of one or more STCs. Additionally, each [state vector, STC] tuple has a Q- value. For instance the Q-value for the tuple [statel, STC2] is 0.65.

[0035] Step s402 comprises collecting state information for cell 112 (capacity cell) and cell 114 (coverage cell).

[0036] Step s404 comprises creating a state vector (S) based on the collected state information. The state vector may contain the following: [an energy value for the coverage cell (e.g., coverage cell PRB availability), an energy value for the capacity cell (e.g., capacity cell PRB usage percentage), a vector of QoS values (e.g., a latency value, a throughput value, a packet loss value, a radio bearer drop rate value, and an interference level value), a traffic value for the coverage cell (e.g., coverage cell RRC connection availability), and a traffic value for the capacity cell (e.g., number of RRC connection in the capacity cell). [0037] Step s406 comprises determining whether Q-table 500 contains a state vector matching the state vector created in step s404. If it does, then process 400 proceeds to step s408, otherwise process 400 proceeds to step s416.

[0038] Step s408 comprises using a policy (a.k.a., “rule” or “model”) to select one of the STCs associated with the state vector. For example, each STC associated with the state vector has a Q-value, and selecting one of the STCs associated with the state vector comprises selecting the STC associated with the state vector that has the largest Q value.

[0039] Step s410 comprises providing the STC to the capacity cell, which then uses the STC determine whether a sleep opportunity exists.

[0040] Step s412 comprises collecting up-to-date state information for the capacity cell and coverage cell and creating a state vector (S’) using the up-to-date state information.

[0041] Step s414 comprises updating the Q-value for the tuple [S, STC selected in step s408] (denoted Q(S,STC)). For example:

Q(S,STC) = Q(S,STC) + a (R + y(Q(S’,STC’) - Q(S,STC), where a and y are predetermined coefficients, R is a reward value determined based on the up-to-date state information (e.g., based on the up-to-date QoS values). For example, if the STC for cell 112 was increased in first iteration and QoS parameters do not degrade by 10%, then reward would be 1, else if the QoS parameters degrade by 10% or more, then the reward would be -1, and STC’ is the STC associated with S’ that has the highest Q-value of all the STCs associated with S’.

[0042] Step s416 comprises using a Bayesian network to add a new record to Q-table 500. A Directed Acyclic Graph (DAG) is used to represent the Bayesian Network and like any other statistical graph, a DAG contains a set of nodes and links, where the links denote the relationship between the nodes. A DAG models the uncertainty of an event occurring based on the Conditional Probability Distribution (CDP) of each random variable. In this case, the random variables are: capacity cell PRB usage, number of RRC connections in the capacity cell, coverage cell PRB availability, coverage cell RRC connection availability, and STC for each cell. A posterior probability which is a measure of the likelihood of an event occurring provided that another event has already occurred (through assumption, supposition, statement, or evidence) is calculated. This probability generates a measure of possible STC that is optimal for the cell.

[0043] FIG. 6 is a block diagram of network node 600, according to some embodiments, that can be used to implement CSMF 106. As shown in FIG. 6, network node 600 may comprise: processing circuitry (PC) 602, which may include one or more processors (P) 655 (e.g., one or more general purpose microprocessors and/or one or more other processors, such as an application specific integrated circuit (ASIC), field-programmable gate arrays (FPGAs), and the like), which processors may be co-located in a single housing or in a single data center or may be geographically distributed (i.e., network node 600 may be a distributed computing apparatus); at least one network interface 648 (e.g., a physical interface or air interface) comprising a transmitter (Tx) 645 and a receiver (Rx) 647 for enabling network node 600 to transmit data to and receive data from other nodes connected to a network 110 (e.g., an Internet Protocol (IP) network) to which network interface 648 is connected (physically or wirelessly) (e.g., network interface 648 may be coupled to an antenna arrangement comprising one or more antennas for enabling network node 600 to wirelessly transmit/receive data); and a storage unit (a.k.a., “data storage system”) 608, which may include one or more non-volatile storage devices and/or one or more volatile storage devices. In embodiments where PC 602 includes a programmable processor, a computer readable storage medium (CRSM) 642 may be provided. CRSM 642 may store a computer program (CP) 643 comprising computer readable instructions (CRI) 644.

CRSM 642 may be a non-transitory computer readable medium, such as, magnetic media (e.g., a hard disk), optical media, memory devices (e.g., random access memory, flash memory), and the like. In some embodiments, the CRI 644 of computer program 643 is configured such that when executed by PC 602, the CRI causes network node 600 to perform steps described herein (e.g., steps described herein with reference to the flow charts). In other embodiments, network node 600 may be configured to perform steps described herein without the need for code. That is, for example, PC 602 may consist merely of one or more ASICs. Hence, the features of the embodiments described herein may be implemented in hardware and/or software.

[0044] While various embodiments are described herein, it should be understood that they have been presented by way of example only, and not limitation. Thus, the breadth and scope of this disclosure should not be limited by any of the above-described exemplary embodiments. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the disclosure unless otherwise indicated herein or otherwise clearly contradicted by context.

[0045] Additionally, while the processes described above and illustrated in the drawings are shown as a sequence of steps, this was done solely for the sake of illustration. Accordingly, it is contemplated that some steps may be added, some steps may be omitted, the order of the steps may be re-arranged, and some steps may be performed in parallel.

Claims

1. A method (300) for managing a capacity cell (112) of a wireless communications network (100), the method comprising: obtaining (s302) a first set of quality-of-service, QoS, values for the capacity cell; after obtaining the first set of QoS values for the capacity cell, configuring (s304) the capacity cell to implement a first sleep threshold configuration, STC, wherein the first STC comprises one or more sleep thresholds; after configuring the capacity cell to implement the first STC: i) obtaining a second set of QoS values for the capacity cell, ii) calculating a first reward value using the first and second set of QoS values, and iii) obtaining a first state vector for the capacity cell (s306); based on the first reward value, selecting (s3O8) a second STC from a set of STCs associated with a state identified by the first state vector; and if the selected second STC is different than the first STC, configuring (s310) the capacity cell to implement the second STC.

2. The method of claim 1, further comprising, after selecting the second STC from the set of STCs associated with the state identified by the first state vector: obtaining a third set of QoS values for the capacity cell; calculating a second reward value using the second and third set of QoS values; obtaining a second state vector for the capacity cell; based on the second reward value, selecting a third STC from a set of STCs associated with a state identified by the second state vector; and if the selected third STC is different than the second STC, configure the capacity cell to implement the third STC.

3. The method of claim 1 or 2, wherein the first set of QoS values comprises one or more of: a latency value indicating an average latency associated with the capacity cell, a throughput value indicating an average throughput for the capacity cell, a packet loss value indicating an average packet loss for the capacity cell, bearer drop rate value indicating a bearer drop rate for the capacity cell, or an interference value indicating a level of interference experienced by the capacity cell.

4. The method of any one of claims 1-3, wherein each STC included in the set of STCs associated with the state identified by the first state vector has a Q-value, and selecting the second STC from the set of STCs comprises selecting, from the set of STCs, the STC having the largest Q value.

5. The method of any one of claims 1-4, wherein the first state vector comprises: a first energy value indicating a current traffic load for the capacity cell for the capacity cell; a second energy value indicating a resource availability for a coverage cell that covers the capacity cell.

6. The method of claim 5, wherein the first state vector further comprises: a first traffic value indicating a total number of radio resource control, RRC, connections in the capacity cell; and a second traffic value indicating an available connection capacity for the coverage cell.

7. The method of any one of claims 1-6, further comprising, prior to selecting the second STC from the set of STCs associated with a state identified by the first state vector, determining whether the first state vector is in a look-up table.

8. The method of claim 7, further comprising: as a result of determining that the first state vector is not in the look-up table, using a Directed Acyclic Graph, DAG, to add to the look-up table a record comprising the first state vector and the set of STCs.

9. The method of any one of claims 1-8, wherein each said STC comprises a traffic load sleep threshold, wherein the capacity cell has an opportunity to enter a cell sleep state if a set of sleep conditions are satisfied, wherein a first one the of the sleep conditions is that the current traffic load of the capacity cell must be less than or equal to the traffic load sleep threshold.

10. The method of claim 9, wherein each said STC further comprises an RRC connection sleep threshold and a second one the of the sleep conditions is that the current number of RRC connection in the capacity cell is less than or equal to the RRC connection sleep threshold.

11. A network node (600) for managing a capacity cell (112) of a wireless communications network (100), the network node being configured to: obtain (s302) a first set of quality-of-service, QoS, values for the capacity cell; after obtaining the first set of QoS values for the capacity cell, configure (s304) the capacity cell to implement a first sleep threshold configuration, STC, wherein the first STC comprises one or more sleep thresholds; after configuring the capacity cell to implement the first STC: i) obtain a second set of QoS values for the capacity cell, ii) calculate a first reward value using the first and second set of QoS values, and iii) obtain a first state vector for the capacity cell (s306); based on the first reward value, select (s3O8) a second STC from a set of STCs associated with a state identified by the first state vector; and if the selected second STC is different than the first STC, configure (s310) the capacity cell to implement the second STC.

12. The network node of claim 11, wherein the network node is further configured to, after selecting the second STC from the set of STCs associated with the state identified by the first state vector: obtain a third set of QoS values for the capacity cell; calculate a second reward value using the second and third set of QoS values; obtain a second state vector for the capacity cell; based on the second reward value, select a third STC from a set of STCs associated with a state identified by the second state vector; and if the selected third STC is different than the second STC, configure the capacity cell to implement the third STC.

13. The network node of claim 11 or 12, wherein the first set of QoS values comprises one or more of: a latency value indicating an average latency associated with the capacity cell, a throughput value indicating an average throughput for the capacity cell, a packet loss value indicating an average packet loss for the capacity cell, bearer drop rate value indicating a bearer drop rate for the capacity cell, or an interference value indicating a level of interference experienced by the capacity cell.

14. The network node of any one of claims 11-13, wherein each STC included in the set of STCs associated with the state identified by the first state vector has a Q-value, and selecting the second STC from the set of STCs comprises selecting, from the set of STCs, the STC having the largest Q value.

15. The network node of any one of claims 11-14, wherein the first state vector comprises: a first energy value indicating a current traffic load for the capacity cell for the capacity cell; a second energy value indicating a resource availability for a coverage cell that covers the capacity cell.

16. The network node of claim 15, wherein the first state vector further comprises: a first traffic value indicating a total number of radio resource control, RRC, connections in the capacity cell; and a second traffic value indicating an available connection capacity for the coverage cell.

17. The network node of any one of claims 11-16, wherein the network node is further configured to, prior to selecting the second STC from the set of STCs associated with a state identified by the first state vector, determine whether the first state vector is in a look-up table.

18. The network node of claim 17, wherein the network node is further configured to: as a result of determining that the first state vector is not in the look-up table, use a

Directed Acyclic Graph, DAG, to add to the look-up table a record comprising the first state vector and the set of STCs.

19. The network node of any one of claims 11-18, wherein each said STC comprises a traffic load sleep threshold, wherein the capacity cell has an opportunity to enter a cell sleep state if a set of sleep conditions are satisfied, wherein a first one the of the sleep conditions is that the current traffic load of the capacity cell must be less than or equal to the traffic load sleep threshold.

20. The network node of claim 19, wherein each said STC further comprises an RRC connection sleep threshold and a second one the of the sleep conditions is that the current number of RRC connection in the capacity cell is less than or equal to the RRC connection sleep threshold.

21. A computer program (643) comprising instructions (644) which when executed by processing circuitry (602) of a network node (600) causes the network to perform the method of any one of claims 1-10.

22. A carrier containing the computer program of claim Al, wherein the carrier is one of an electronic signal, an optical signal, a radio signal, and a computer readable storage medium (642).