WO2014180350A1 - 基站非连续发送节能控制方法、小站、宏站及异构网络 - Google Patents

基站非连续发送节能控制方法、小站、宏站及异构网络 Download PDF

Info

Publication number
WO2014180350A1
WO2014180350A1 PCT/CN2014/077175 CN2014077175W WO2014180350A1 WO 2014180350 A1 WO2014180350 A1 WO 2014180350A1 CN 2014077175 W CN2014077175 W CN 2014077175W WO 2014180350 A1 WO2014180350 A1 WO 2014180350A1
Authority
WO
WIPO (PCT)
Prior art keywords
station
total cost
control
small
macro
Prior art date
Application number
PCT/CN2014/077175
Other languages
English (en)
French (fr)
Inventor
陈雁
崔颖
刘坚能
徐修强
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2014180350A1 publication Critical patent/WO2014180350A1/zh

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W52/00Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02Power saving arrangements
    • H04W52/0203Power saving arrangements in the radio access network or backbone network of wireless communication networks
    • H04W52/0206Power saving arrangements in the radio access network or backbone network of wireless communication networks in access points, e.g. base stations
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W52/00Power management, e.g. TPC [Transmission Power Control], power saving or power classes
    • H04W52/02Power saving arrangements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W76/00Connection management
    • H04W76/20Manipulation of established connections
    • H04W76/28Discontinuous transmission [DTX]; Discontinuous reception [DRX]
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00Reducing energy consumption in communication networks
    • Y02D30/70Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • Base station discontinuous transmission energy-saving control method, small station, macro station and heterogeneous network
  • the present invention belongs to the field of heterogeneous network technologies, and in particular, to a base station discontinuous transmission control method, a macro station, a small station, and a heterogeneous network. Background technique
  • heterogeneous networks In order to support the rising volume of business and improve the throughput of the network, heterogeneous networks (English: Heterogeneous Networks) have become a recognized and effective solution.
  • a typical heterogeneous network consists of a macro station and multiple small stations.
  • the small stations can be divided into micro cells, pico cells, and femto cells according to the coverage.
  • the basic coverage is provided, and different small stations provide different ranges and different levels of capacity enhancement within the coverage of the macro station.
  • the opening of the small station can effectively improve the energy efficiency of the whole network, because short-distance coverage greatly reduces the energy consumption of the transmission; however, when the traffic is low, the capacity of the macro station is sufficient to serve the user, and There is no need for a small station to increase capacity, and the static energy consumption introduced by the small station at this time will become an important part of the entire network energy consumption. Therefore, in an ideal situation, at low traffic, the station should be put into sleep mode to save network power consumption.
  • BS-DTX Base Station-Discontinuous Transmission
  • the information of the load changes, that is, a traffic-adaptive BS-DTX energy-saving control method, and the discontinuous transmission control of the base station is adaptive to the traffic volume.
  • Load related parameters such as the number of calls per unit time (voice), the degree of user activity, the rate of arrival of traffic, etc.
  • the small station can decide to sleep at its own discretion, and transfer the remaining service to the macro station providing coverage, or other small station service that can provide coverage; After the load rises to a certain value, the station is woken up and continues to provide service.
  • the performance of the system depends not only on the size of the traffic (the arrival of service requirements), but also on the channel conditions (in the case of service). Therefore, BS-DTX is only based on the traffic volume. Control, the resulting control strategy usually does not achieve good energy savings. Summary of the invention
  • an object of the present invention is to provide a base station discontinuous transmission energy-saving control method, a macro station, a small station, and a heterogeneous network, which aims to solve the traffic-based adaptive of the existing heterogeneous network.
  • the BS-DTX energy-saving control method because the obtained control strategy is only related to the size of the traffic, can not achieve a good energy-saving effect.
  • the base station discontinuous transmission energy-saving control method includes the following steps: at a start time of each control cycle, the small station acquires a total cost required to service its associated user, and feeds the total cost to the macro station. And causing the macro station to make a preferred control policy according to the total cost, where the total cost includes the average delay information of the associated user of the small station; and receiving the control instruction corresponding to the preferred control policy sent by the macro station, corresponding to turning on or off the small station.
  • control period is divided into multiple scheduling time slots, and the control instruction corresponding to the preferred control policy sent by the receiving macro station is corresponding to the opening or closing of the small station, Including: At the beginning of each scheduling slot, the opened stations assign their respective subband resources to their associated users.
  • the small station at the beginning of each control cycle, obtains the total cost required to service its associated user, and The total cost is fed back to the macro station, and specifically includes: at the beginning of each control period, the small station acquires an optimal bandwidth allocation of its associated user; and obtains a total generation under the optimal bandwidth allocation condition. Price, and feedback the total cost to the macro station.
  • each small station includes multiple sub-bands, and the small time is started at the beginning of each scheduling time slot.
  • the station allocates its own sub-band resources to its associated users, including: At the beginning of each scheduling time slot, the opened small station is on each sub-band, and selects the equivalent rate of the instantaneous throughput weighted queue length in its associated users. The largest user; re-enters the next control period when the control period ends or when the associated user set changes in the heterogeneous network, otherwise enters the next scheduled time slot.
  • the base station discontinuous transmission energy-saving control method includes: the macro station receives a total cost from each small station, and obtains a total network total cost under different control policies according to the total cost, and the whole network is When the total cost is minimum, the corresponding control strategy is used as a preferred control strategy; the control command corresponding to the preferred control policy is sent to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • the receiving a total cost from each small station, and acquiring a total cost of the entire network under different control policies according to the total cost When the cost is the minimum, the corresponding control strategy is used as the preferred control strategy, and the method includes: the macro station receives the total cost from each small station; traverses all possible control policies, and obtains an optimal bandwidth allocation of the user served by the macro station; Total cost and optimal bandwidth allocation, obtain the total cost of the whole network under various control strategies; select the corresponding control strategy when the total cost of the whole network is the smallest as the preferred control strategy.
  • the receiving a total cost from each small station, and acquiring a total cost of the entire network under different control policies according to the total cost When the cost is the smallest, the corresponding control strategy is used as the preferred control strategy, which specifically includes: the macro station receives the total cost from each small station; obtains the total cost of the entire network when all the small stations are turned on, and the total cost in the small station that is turned off sequentially is the highest. The small station, until the total cost of the whole network is no longer reduced or all the small stations are closed, the control strategy of the small station at this time is taken as the preferred control strategy.
  • the control corresponding to the preferred control policy After the instruction is sent to the small station to enable the small station to be turned on or off according to the control instruction, the method further includes: at each start time of the scheduling time slot, the macro station allocates the sub-band resource to its associated user and closes the association of the small station. user.
  • the macro station includes multiple sub-bands, and the macro station allocates sub-band resources at the beginning of each scheduling time slot.
  • the macro station allocates sub-band resources at the beginning of each scheduling time slot.
  • the following includes: At the beginning of each scheduling time slot, the macro station is on each sub-band, and the associated users of the associated users and the closed stations are selected, the instantaneous throughput The user with the highest equivalent rate after the weighted queue length; when the control period ends or the associated user set changes in the heterogeneous network, the next control period is re-entered, otherwise the next scheduled time slot is entered.
  • the small station includes: a total cost obtaining unit, configured to acquire a total cost required to serve an associated user at a start time of each control period, and feed back the total cost to the macro station, so that the macro station Determining, according to the total cost, a preferred control policy, where the total cost includes an average delay information of the associated user of the small station, and an instruction receiving and executing unit, configured to receive a control instruction corresponding to the preferred control policy sent by the macro station, corresponding to opening or closing Small station.
  • a total cost obtaining unit configured to acquire a total cost required to serve an associated user at a start time of each control period, and feed back the total cost to the macro station, so that the macro station Determining, according to the total cost, a preferred control policy, where the total cost includes an average delay information of the associated user of the small station
  • an instruction receiving and executing unit configured to receive a control instruction corresponding to the preferred control policy sent by the macro station, corresponding to opening or closing Small station.
  • the small station further includes: a first sub-band allocation unit, configured to allocate respective sub-band resources to its associated users at a start time of each scheduling time slot .
  • the total cost obtaining unit includes: a bandwidth allocation acquiring module, configured to start at a start time of each control period, Obtaining an optimal bandwidth allocation of its associated user; a total cost acquisition module, configured to obtain a total cost under the optimal bandwidth allocation condition, and feed back the total cost to the macro station.
  • the first sub-band allocation unit includes: a first sub-band allocation module, configured in each scheduling At the beginning of the time slot, on each sub-band, select the user with the highest equivalent rate after the instantaneous throughput weighted queue length among the associated users of the small station; the judgment execution module is used at the end of the control period or different When the associated user set changes in the fabric, the next control period is re-entered, otherwise it enters the next scheduling slot.
  • the macro station includes: a policy generating unit, configured to receive a total cost from each small station, and obtain a total cost of the entire network under different control policies according to the total cost, where the total network is total When the cost is minimum, the corresponding control strategy is used as the preferred control strategy; the command sending unit sends the control command corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • a policy generating unit configured to receive a total cost from each small station, and obtain a total cost of the entire network under different control policies according to the total cost, where the total network is total
  • the corresponding control strategy is used as the preferred control strategy
  • the command sending unit sends the control command corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • the policy generating unit includes: a total cost receiving module, configured to receive a total cost from each small station; and a bandwidth allocation module, configured to traverse all possible control policies And obtaining an optimal bandwidth allocation of the user served by the macro station; a total network cost obtaining module, configured to obtain a total network cost under various control strategies according to the total cost and the optimal bandwidth allocation; The module is used to select a corresponding control strategy when the total cost of the entire network is minimum as a preferred control strategy.
  • the policy generating unit includes: a total cost receiving module, configured to receive a total cost from each small station; and a sub-optimal selecting module, configured to acquire all small stations At the time of the total cost of the whole network, the station with the highest total cost among the opened small stations is closed in turn, until the total cost of the whole network is no longer reduced or all the small stations are all closed, and the control strategy of the small station at this time is taken as the preferred control strategy.
  • the macro station further includes: a second sub-band allocation unit, configured to start at each scheduling time slot, Assign subband resources to their associated users and associated users who close the station.
  • the second sub-band allocation unit includes: a second sub-band allocation module, configured to start at each scheduling time slot On each sub-band, select the user with the largest equivalent rate after the instantaneous throughput weighted queue length among the associated users of the macro station associated user and the closed small station; judge the execution module at the end of the control period or in the heterogeneous network When the associated user set changes, re-enter the next control cycle, otherwise Enter the next scheduling slot.
  • the heterogeneous network comprises: the small station and the macro station, and the macro station is connected to each small station by wire or wirelessly.
  • the invention has the following beneficial effects:
  • the present invention provides a BS-DTX energy-saving control scheme with traffic and channel conditions adaptive.
  • the small station needs to obtain the total cost required to serve its associated users, and the total cost includes the small station.
  • the average delay information of the associated users the macro station makes a preferred control policy according to the total cost of the feedback of each small station, so in the technical solution of the present invention, not only the traffic volume (ie, the average throughput) but also the channel state is considered.
  • the information (the average delay information reflects the channel state information), which makes a good compromise between the network delay and the total power consumption, and further reduces the system energy consumption under the condition of meeting the delay requirement;
  • the total cost of obtaining in the present invention is realized by each small station. This distributed design reduces the workload of the macro station, is easy to implement, and greatly reduces the network control overhead and improves the network efficiency.
  • Figure 1 is a network topology diagram of a heterogeneous network
  • FIG. 2 is a flow chart of a method for controlling a discontinuous transmission energy saving of a base station according to a first embodiment of the present invention
  • FIG. 3 is a flow chart of a method for controlling a discontinuous transmission energy saving of a base station according to a second embodiment of the present invention
  • FIG. 4 is a flow chart of a method for controlling a discontinuous transmission energy saving of a base station according to a third embodiment of the present invention.
  • FIG. 5 is a flow chart of a method for controlling a discontinuous transmission energy saving of a base station according to a fourth embodiment of the present invention.
  • FIG. 6 is a flow chart of a method for controlling a discontinuous transmission energy saving of a base station according to a fifth embodiment of the present invention.
  • FIG. 8 is a structural block diagram of a small station according to a seventh embodiment of the present invention.
  • FIG. 9 is a structural block diagram of a small station according to an eighth embodiment of the present invention.
  • Figure 10 is a block diagram showing the structure of a macro station according to a ninth embodiment of the present invention.
  • Figure 11 is a block diagram showing the structure of a macro station according to a tenth embodiment of the present invention.
  • Figure 12 is a block diagram showing the structure of a macro station according to an eleventh embodiment of the present invention.
  • FIG. 13 is a system diagram of a heterogeneous network according to a twelfth embodiment of the present invention.
  • Figure 14 is a simulation curve relationship diagram of transmission power-average delay. detailed description
  • the embodiment of the invention provides a BS-DTX energy-saving control scheme based on traffic volume and channel condition adaptation, takes into account the traffic volume and channel state, obtains a preferred control strategy, and reduces system power consumption as much as possible under the condition of delay requirement.
  • each analysis model is established as follows:
  • the network topology model of the heterogeneous network As shown in Figure 1, assume that the heterogeneous network includes one macro station and B small stations, and the coverage of all the small stations is within the coverage of the macro station. Provides basic coverage and capacity for heterogeneous networks, while small stations provide capacity improvements in some areas.
  • Let 0 denote the label of the macro station, and let 1,... denote the labels of the B stations in the macro station respectively.
  • Let ⁇ 1, ⁇ ⁇ ⁇ , ⁇ denote a collection of all stations.
  • Let all mobile user sets be represented, let 7 ⁇ denote the set of mobile users in the bth station, ⁇ " ⁇ .
  • Each user (English: mobile station, abbreviation: MS) chooses to logically receive the strongest signal
  • a macro station or a small station is associated.
  • the MS associated with the small station may be served by the macro station or the small station. If the control policy determines that the small station is turned on, these associated users are served by the small station, otherwise the macro station services
  • the embodiment of the present invention only considers covering the macro station. Small stations with active users in the range perform switch optimization control. For those stations with no active users, they are automatically closed by default.
  • the power consumption model of the base station can be expressed as:
  • ⁇ , and ⁇ are the total power consumption, fixed power consumption (relative static power consumption), and transmit power of the b-th base station, respectively. Indicates the power amplification factor due to the power amplifier, feeder (not present), cooling, or power conversion efficiency.
  • the total time axis is logically divided into a number of discrete scheduling moments of length ⁇ , labeled t.
  • Pb Q indicates that the bth station is closed, and the macro station is always on to ensure coverage of the area and access by new users.
  • the average throughput between the b-th station and the k-th user ( e ) can be estimated by:
  • the average throughput between a macro station and its user k can be estimated as:
  • the average delay of the user k served by the bth base station can be defined as
  • the function of P consists of the following two parts: 1) the average total power of the macro station; 2) the average total power of all open stations.
  • 0 is a trade-off between power and time delay, specified by system requirements.
  • the time delay is reduced, and when the focus of the system design is to reduce the power, Increase.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • Fig. 2 is a flowchart showing the flow of the discontinuous transmission power saving control method of the base station according to the first embodiment of the present invention. For the convenience of description, only the parts related to the embodiment of the present invention are shown.
  • the base station discontinuous transmission energy-saving control method is described by the small station side, and includes:
  • Step S201 At the beginning of each control cycle, the small station acquires the total cost required to service its associated user, and feeds back the total cost to the macro station, so that the macro station makes a preferred control policy according to the total cost.
  • the total cost includes the average delay information of the associated users of the station.
  • Each small station serves the associated users in its scope.
  • the total cost of the associated user of the small station service is obtained, that is, the total service cost, and the total cost includes the average delay information of the associated user of the small station.
  • the average delay information reflects the channel state information to a certain extent, and each small station sends the total cost obtained to the macro station, and the macro station obtains the total cost of the whole network under different control policies according to all the total costs, and the When the total cost of the whole network is the smallest, the corresponding control strategy is used as the preferred control strategy, and the opening and closing of the small station is controlled according to the preferred control strategy.
  • Step S202 Receive a control instruction corresponding to a preferred control policy sent by the macro station, corresponding to turning on or off the small station.
  • the small station After the macro station makes a preferred control policy, the small station receives the control instruction corresponding to the preferred control policy, and performs corresponding opening or closing of the small station.
  • the method in this embodiment is a time-aware traffic and channel condition adaptive BS-DTX energy-saving control scheme.
  • each small station considers the traffic volume and the average delay, and calculates the total cost required by the small station.
  • the macro station obtains the total cost of the whole network under different control strategies according to the total cost, and takes the corresponding control strategy as the preferred control strategy when the total cost of the whole network is minimized.
  • this embodiment considers the channel condition (ie, the user The average delay) enables the macro station to obtain a preferred control strategy, which can further reduce the system energy consumption under the condition of meeting the delay requirement.
  • the method further includes:
  • Step S203 At the beginning of each scheduling time slot, the opened small stations allocate respective sub-band resources to their associated users.
  • the macro station generates a preferred control strategy for each control cycle, and opens or closes the small station according to the preferred control strategy.
  • This is a long-cycle BS-DTX energy-saving control strategy, considering the average of long periods. Traffic and average channel quality, which is a long-term scale control scheme, but in the actual network, there are many delay-sensitive services with burst arrival characteristics. Therefore, in addition to the traditional physical layer throughput, consider The random burst arrival of the service and the latency performance of the MAC (Media Access Control) layer also become important. Therefore, the preferred step further considers the random burstiness of the service, divides the control period into a number of scheduling slots, and the scheduling slot label is represented by t.
  • the small station performs subband allocation, and the subband resources are allocated to Its associated users.
  • the preferred embodiment further needs to perform sub-band resource allocation of a short period (ie, each scheduling slot), and the dual-scale BS-DTX control and user
  • the selection method can further reduce network power consumption while ensuring network delay.
  • Fig. 3 is a flowchart showing the flow of the discontinuous transmission power saving control method of the base station according to the second embodiment of the present invention. For the convenience of description, only the parts related to the embodiment of the present invention are shown.
  • the base station discontinuous transmission energy saving control method includes:
  • Step S301 At the beginning of each control cycle, the small station acquires an optimal bandwidth allocation of its associated user.
  • Each station obtains the optimal bandwidth allocation of its associated users, and the formula is as follows: Which satisfies the limitation in equation (5) and can be expressed as Vfc and W
  • the base station in the network shares the common spectrum bandwidth, indicating the average throughput between the bth station and the kth user ( k G , U is the average arrival rate of the service, so 4(t) indicates that user k is at the serving base station side t
  • k G the average throughput between the bth station and the kth user
  • U the average arrival rate of the service
  • Step S302 Acquire a total cost under the optimal bandwidth allocation condition, and feed back the total cost to the macro station.
  • each small station obtains the total cost calculated under the optimal bandwidth allocation condition, according to the network average total delay and power consumption.
  • the model can know the average total delay and average power consumption of the small station under the optimal bandwidth allocation condition. Specifically, the total cost under the optimal bandwidth allocation:
  • the power is transmitted for the small station, indicating the power amplification factor, indicating the average delay of the user k of the service of the small station b under the optimal bandwidth allocation;
  • is a trade-off parameter between power and delay, The specific value is determined by the delay requirement.
  • Step S303 Receive a control instruction corresponding to a preferred control policy sent by the macro station, corresponding to turning on or off the small station.
  • Step S304 At each start time of each scheduling time slot, the opened small station is on each sub-band, and selects the user whose associated rate is the largest in the associated user after the instantaneous throughput weighted queue length;
  • Step S305 when the control period ends or when the associated user set changes in the heterogeneous network, Newly enters the next control cycle, otherwise enters the next scheduled time slot.
  • Each small station includes multiple sub-bands, and the opened small station is on each sub-band, and the associated users are selected, and the instantaneous throughput is weighted. The user with the highest equivalent rate after the queue length.
  • the instantaneous throughput between the macro station and its user k on the sub-band at time t can be expressed as
  • the expression of the allocation indicator indicates that the mth subband of the tth scheduling slot is allocated to the user with the highest equivalent rate after the instantaneous throughput weighted queue length.
  • the method can further reduce network power consumption while ensuring network delay.
  • Embodiment 3 is a diagrammatic representation of Embodiment 3
  • Fig. 4 is a flowchart showing a flow of a discontinuous transmission power saving control method for a base station according to a third embodiment of the present invention. For the convenience of description, only parts related to the embodiment of the present invention are shown.
  • the base station discontinuous transmission energy-saving control method provided by this embodiment is described by the macro station side, and specifically includes:
  • Step S401 The macro station receives the total cost from each small station, and obtains the total cost of the whole network under different control policies according to the total cost, and uses a corresponding control strategy as the preferred control strategy when the total cost of the entire network is minimized. ;
  • Step S402 Send a control instruction corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • the two steps in this embodiment correspond to the steps S201 and S202 of the embodiment, and are described by the macro station side.
  • the macro station After each small station obtains the total cost Ub and feeds back to the macro station, the macro station obtains the total cost of the whole network under various control strategies according to the total cost, and then obtains a corresponding control strategy when the total cost of the entire network is minimum.
  • the control strategy is preferred, and the opening and closing of the small station is controlled according to the preferred control strategy corresponding to issuing a control command to the small station.
  • the present embodiment Compared with the existing traffic adaptive BS-DTX energy saving control scheme, the present embodiment considers the channel condition (ie, the average user delay), so that the macro station obtains the preferred control strategy, and under the condition that the delay requirement is met, Further reduce system energy consumption.
  • the channel condition ie, the average user delay
  • the overhead is very small, only the total cost of feedback from each station, and the control commands issued by the macro station to control the opening and closing of the station. This distributed working method can improve network efficiency.
  • the method further includes:
  • Step S403 At the beginning of each scheduling time slot, the macro station allocates the sub-band resource to its associated user and the associated user of the closed station.
  • the macro station generates a preferred control strategy for each control cycle, and opens or closes the small station according to the preferred control strategy.
  • This is a long-cycle BS-DTX energy-saving control strategy, considering the average of long periods. Traffic and average channel quality, which is a long-term scale control scheme, but in the actual network, there are many delay-sensitive services with burst arrival characteristics. Therefore, in addition to the traditional physical layer throughput, consider The random burst arrival of the service and the latency performance of the MAC (Media Access Control) layer also become important. Therefore, the preferred step further considers the random burstiness of the service, divides the control period into a number of scheduling slots, and the scheduling slot label is represented by t.
  • the preferred control strategy of the macro station in the preferred embodiment is to obtain a long period (ie, control period). After that, the sub-band resource allocation of the short period (ie, each scheduling slot) needs to be further performed.
  • This dual-scale BS-DTX control and user selection method can further reduce the network power consumption on the basis of ensuring the network delay.
  • FIG. 5 is a flowchart of a method for controlling a discontinuous transmission power saving of a base station according to a fourth embodiment of the present invention. For the convenience of description, only parts related to the embodiment of the present invention are shown.
  • the base station discontinuous transmission energy-saving control method provided by this embodiment is described by the macro station side, and specifically includes:
  • Step S501 The macro station receives a total cost from each small station
  • Step S502 Traverse all possible control policies, and obtain an optimal bandwidth allocation of the user served by the macro station.
  • W indicates that the base station in the network shares the common spectrum bandwidth.
  • is the average arrival rate of the service.
  • Pb is an indicator of the opening or closing of the b-th station.
  • Step S503 Acquire a total cost of the entire network under various control policies according to the total cost and the optimal bandwidth allocation;
  • Step S504 Select a control strategy corresponding to a total cost of the entire network as a preferred control strategy.
  • the above steps S501-S504 are a specific preferred implementation manner of step S401 in the third embodiment.
  • the whole network under various control strategies P is calculated.
  • the total cost, the total cost of the whole network also includes three parts: 1) the cost of the macro station serving its associated users; 2) the cost of the associated user opened by the small station under the p policy; 3) the macro station under the p policy
  • the service is associated with the small station but is served by the macro station to serve the user. Therefore, the total cost of the entire network can be expressed as: ) + £ P b U b + (lp b ) ⁇ D kfl (wl fi )
  • the macro station then takes the control strategy based on the minimum total cost of the whole network as the preferred control strategy, namely:
  • Step S505 Send a control instruction corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction;
  • Step S506 At each start time of each scheduling time slot, the macro station selects the user whose associated user and the closed small station are associated with each other, and the user with the highest equivalent rate after the instantaneous throughput weighted queue length;
  • Step S507 When the control period ends or the associated user set changes in the heterogeneous network, the next control period is re-entered, otherwise the next scheduling time slot is entered.
  • steps S506 and S507 are a specific preferred embodiment of the step S403 in the third embodiment, and a seed band allocation method is implemented.
  • the macro station also includes multiple sub-bands, and the macro station is in the scheduling time slot t for the kth user.
  • ⁇ 0 represents the queue status information in the service queue when the user k starts at the t-th scheduling time slot of the serving base station, that is, the remaining bit number QSI (bit), and the QSI indicates the queue.
  • the expression of the length, allocation indicator indicates the user with the highest equivalent rate after assigning the mth subband of the tth scheduling slot to the instantaneous throughput weighted queue length.
  • the control strategy is selected as the preferred control strategy when the total cost of the whole network is the smallest among all the possible control policies. Therefore, the preferred control strategy selected in this embodiment is the optimal control strategy, but the total cost of the entire network is calculated.
  • the complexity is exponentially related to the number of small stations B. When the number of small stations is not large, such as two, it is better to use this scheme.
  • this embodiment also implements a specific short-period subband allocation algorithm. Considering the random burstiness of the service, after obtaining the long-term preferred control strategy, the random burst of the service is further considered. Subdividing the control period into a number of scheduling slots for short-cycle (ie, each scheduling slot) sub-band resource allocation.
  • FIG. 6 is a flowchart of a method for controlling a discontinuous transmission energy saving of a base station according to a fifth embodiment of the present invention. For the convenience of description, only parts related to the embodiment of the present invention are shown.
  • the base station discontinuous transmission energy-saving control method provided by this embodiment is described by the macro station side, and specifically includes:
  • Step S601 The macro station receives a total cost from each small station
  • Step S602 Obtain the total cost of the whole network when all the small stations are opened, and sequentially close the small station with the highest total cost among the opened small stations, until the total cost of the whole network is no longer reduced or all the small stations are all closed, and then the d,
  • the station's control strategy serves as the preferred control strategy.
  • steps S601 and S602 are another specific preferred embodiment of step S401 in the third embodiment.
  • Step S603 Send a control instruction corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction;
  • Step S604 At each start time of each scheduling time slot, the macro station selects the user whose associated user and the closed small station are associated, and the user with the highest equivalent rate after the instantaneous throughput weighted queue length;
  • Step S605 When the control period ends or the associated user set changes in the heterogeneous network, the next control period is re-entered, otherwise the next scheduling time slot is entered.
  • the above steps S603-S605 are the same as S505-S507 in the fourth embodiment, and are not described herein again.
  • the difference from the fourth embodiment is that the fourth embodiment selects the best among all possible control strategies.
  • the control strategy has an exponential relationship with the number of small stations B.
  • the preferred control strategy selected in this scheme is a suboptimal control strategy.
  • the method of this embodiment is more suitable. .
  • Example 6
  • FIG. 7 is a flowchart showing an interaction process of a base station discontinuous transmission energy saving control method according to a sixth embodiment of the present invention. For the convenience of description, only parts related to the embodiment of the present invention are shown.
  • the method for controlling the discontinuous transmission energy-saving of the base station is described from the perspective of the entire heterogeneous network system, and specifically includes:
  • Step S701 At the beginning of each control cycle, the small station acquires an optimal bandwidth allocation of its associated user;
  • Step S702 The small station acquires a total cost under the optimal bandwidth allocation condition, and feeds back the total cost to the macro station.
  • Step S703 The macro station receives the total cost from each small station, and obtains the total network cost under different control policies according to the total cost, and uses a corresponding control strategy as the preferred control strategy when the total network total cost is minimum. ;
  • Step S704 The macro station sends the control instruction corresponding to the preferred control policy to the small station.
  • Step S705 The small station receives the control instruction corresponding to the preferred control policy sent by the macro station, and correspondingly turns on or off the small station.
  • the method further includes:
  • Step S706 At the start time of the current scheduling time slot, the opened small station is on each sub-band, and among the associated users of the small station, the user with the highest equivalent rate after the instantaneous throughput weighted queue length; and the sub-band of the macro station The resource is allocated to the associated user of the macro station and the associated user of the closed station, and the user with the highest equivalent rate after the instantaneous throughput weighted queue length;
  • Step S707 when the control period ends or when the associated user set changes in the heterogeneous network, Newly enters the next control cycle, otherwise enters the next scheduled time slot.
  • steps S701-S705 are a long-period BS-DTX control method for delay sensing, which can obtain a preferred control strategy in a control period, and steps S906 and S907 further consider the random burstiness of the service, and divide the control period into A number of scheduling slots are used to perform sub-band resource allocation in short periods (ie, each scheduling slot).
  • This dual-scale BS-DTX control and user selection method can further reduce network power consumption while ensuring network delay.
  • Example 7 Example 7:
  • the present application further describes a base station discontinuous transmission energy-saving control method provided by this embodiment by using a specific application scenario, which is a dual-time-scale BS-DTX control method, including a long-period BS-DTX algorithm and a short-period sub-control.
  • a specific application scenario which is a dual-time-scale BS-DTX control method, including a long-period BS-DTX algorithm and a short-period sub-control.
  • the first preferred control strategy under long-term period is to make the system reduce the system power consumption as much as possible while ensuring the network delay.
  • Zhang Zhonghong station performs the method described in Embodiment 4
  • the small station performs the method described in Embodiment 1
  • the heterogeneous network includes a macro station (BS0) and two small stations (BS1, BS2) in the macro.
  • the average throughput rate between the macro station BS0 and its service user MS0 is ⁇
  • the average throughput rate between the macro station and MS1 is ⁇
  • the average throughput rate between the small station BS1 and MS1 is ⁇
  • the average data of MS0 and MS1 are two users.
  • the arrival rate is . , ⁇ .
  • the macro station BSO calculates the total generation of the whole network according to the formula (16)
  • This embodiment further illustrates a base station discontinuous transmission energy-saving control method provided by this embodiment by using another specific application scenario, which is also a dual-time-scale BS-DTX control method, including a long-period BS-DTX algorithm and a short cycle.
  • another specific application scenario which is also a dual-time-scale BS-DTX control method, including a long-period BS-DTX algorithm and a short cycle.
  • the average throughput rate between its service user MS0. the average throughput rate between the macro station and MS1, MS2. with.
  • the method steps of this embodiment are specifically as follows:
  • the macro station BSO obtains the optimal long-period BS-DTX control strategy according to equation (17).
  • the foregoing embodiment 9 and the embodiment 10 are examples of the base station discontinuous transmission energy-saving control method in two typical application scenarios. If the number of small stations is large (for example, greater than 2), the macro station may follow the fifth embodiment. The method selects a sub-optimal control strategy, and then performs user selection under the sub-optimal control strategy, and the specific implementation process is not repeated here.
  • Example 9
  • Fig. 8 shows the structure of a small station provided by a ninth embodiment of the present invention, and only parts related to the embodiment of the present invention are shown for convenience of explanation.
  • the total cost obtaining unit 801 is configured to acquire a total cost required to service its associated user at the start time of each control cycle, and feed back the total cost to the macro station, so that the macro station makes a preferred control according to the total cost Strategy, the total cost includes average delay information of the associated user of the small station;
  • the instruction receiving execution unit 802 is configured to receive a control instruction corresponding to the preferred control policy sent by the macro station, corresponding to turning on or off the small station.
  • the small station further includes:
  • the first subband allocation unit 803 is configured to allocate respective subband resources to its associated users at the beginning of each scheduling slot.
  • the functional units provided in this embodiment respectively implement the three steps in the first embodiment.
  • This embodiment considers the channel conditions (ie, the average user delay), so that the macro station obtains the preferred control policy, and only the existing services are considered. Compared with the BS-DTX energy-saving control scheme, a better optimal control strategy can be obtained to achieve energy saving.
  • the control period is further divided into a plurality of scheduling slots, and the sub-band allocation is performed in each slot to allocate sub-band resources to its associated users.
  • the scaled BS-DTX control and user selection method can further reduce network power consumption while ensuring network delay.
  • Fig. 9 shows the structure of a small station according to a tenth embodiment of the present invention, and only parts related to the embodiment of the present invention are shown for convenience of explanation.
  • the total cost obtaining unit 91 is configured to acquire a total cost required to service its associated user at the start time of each control cycle, and feed back the total cost to the macro station, so that the macro station makes a preferred control according to the total cost Strategy, the total cost includes average delay information of the associated user of the small station;
  • the instruction receiving and executing unit 92 is configured to receive a control instruction corresponding to the preferred control policy sent by the macro station, corresponding to turning on or off the small station.
  • the first subband allocation unit 93 is configured to allocate respective subband resources to its associated users at the beginning of each scheduling slot.
  • the total cost obtaining unit 91 includes:
  • a bandwidth allocation obtaining module 911 configured to acquire an optimal bandwidth allocation of an associated user at a start time of each control period
  • the total cost acquisition module 912 is configured to obtain a total cost under the optimal bandwidth allocation condition, and feed back the total cost to the macro station.
  • the first sub-band allocation unit 93 includes:
  • a first sub-band allocation module 931 at each start time of each scheduling time slot, on each sub-band, Selecting the user with the highest rate of the instantaneous throughput weighted queue length, and determining the execution module 932, for re-entering the next control period when the control period ends or the associated user set changes in the heterogeneous network , otherwise enter the next scheduled time slot.
  • the functional modules provided in this embodiment respectively implement the steps in the second embodiment.
  • the embodiment further discloses the structure of the total cost obtaining unit 91 and the first sub-band allocation unit 93 on the basis of the ninth embodiment.
  • the total cost of the total cost acquisition module 912 under the optimal bandwidth allocation condition, the system power consumption is the smallest, so the power consumption can be further reduced under the premise of ensuring the delay.
  • Example 11
  • Fig. 10 shows the structure of a macro station according to an eleventh embodiment of the present invention, and only parts related to the embodiment of the present invention are shown for convenience of explanation.
  • the policy generating unit 101 is configured to receive a total cost from each small station, and obtain a total network cost under different control policies according to the total cost, and use a corresponding control strategy as a minimum when the total network total cost is minimum. Control Strategy;
  • the instruction sending unit 102 sends a control instruction corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • the macro station further includes:
  • the second sub-band allocation unit 103 is configured to allocate the sub-band resources to its associated users and the associated users of the closed stations at the beginning of each scheduling slot.
  • the functional unit provided in this embodiment implements two steps in the third embodiment, and each small station obtains the total cost and feeds back to the macro station, and the policy generating unit 121 obtains different according to the total cost received.
  • the total cost of the whole network under the control policy is used as a preferred control strategy when the total network cost is minimized, and the command sending unit 122 sends the control command corresponding to the preferred control policy to the small station to be small.
  • the station is turned on or off according to the control instruction.
  • Embodiments take into account channel conditions (ie, user average delay), such that the macro station obtains a preferred control strategy, Under the condition of meeting the delay requirement, the system energy consumption can be further reduced.
  • the data transmission overhead between the macro station and the small station is small, only the total cost of feedback from each station, and the control command issued by the macro station to control the opening and closing of the station.
  • This distributed working method can improve network efficiency.
  • the control period is divided into a plurality of scheduling slots, and the sub-band allocation is performed in each slot, and the sub-band resources are allocated to the associated users.
  • a dual-scale BS-DTX control and user selection method can further reduce network power consumption while ensuring network delay.
  • Fig. 11 shows the structure of a macro station according to a twelfth embodiment of the present invention, and only parts related to the embodiment of the present invention are shown for convenience of explanation.
  • the policy generating unit 111 is configured to receive a total cost from each small station, and obtain a total network cost under different control policies according to the total cost, and use a corresponding control strategy as a minimum when the total network total cost is minimum. Control Strategy;
  • the command sending unit 112 sends a control command corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • the second sub-band allocation unit 113 is configured to allocate the sub-band resources to its associated users and the associated users of the closed stations at the beginning of each scheduling slot.
  • the policy generating unit 111 includes:
  • the total cost receiving module 1111 is configured to receive a total cost from each small station
  • a bandwidth allocation module 1112 configured to traverse all possible control policies and obtain an optimal bandwidth allocation of a user served by the macro station;
  • the total network total cost obtaining module 1113 is configured to obtain a total network cost under various control policies according to the total cost and the optimal bandwidth allocation;
  • the optimal selection module 1114 is configured to select a corresponding control strategy when the total cost of the entire network is minimum Choose a control strategy.
  • the second sub-band allocation unit 113 includes:
  • the second sub-band allocation module 1131 is configured to select an equivalent rate of the instantaneous throughput weighted queue length among the associated users of the macro station associated user and the closed small station at each subband at the start time of each scheduling time slot. Largest user
  • the judgment execution module 1132 re-enters the next control period when the control period ends or when the associated user set changes in the heterogeneous network, otherwise enters the next scheduled time slot.
  • each of the functional units and the functional modules provided in this embodiment implements the respective steps in the fourth embodiment.
  • the embodiment further discloses the policy generating unit 111 and the second sub-band based on the thirteenth embodiment.
  • the allocation unit 113 is a specific preferred structure.
  • the control strategy of selecting the total cost of the whole network is the best control strategy among all the possible control strategies. Therefore, the preferred control strategy selected in this embodiment is the most Excellent control strategy, but because the complexity of calculating the total cost of the whole network is exponentially related to the number of small stations B, when the number of small stations is not large, for example, two, the scheme using this embodiment is better.
  • Example 13 Example 13:
  • Fig. 12 is a view showing the configuration of a macro station according to a thirteenth embodiment of the present invention, and only parts related to the embodiment of the present invention are shown for convenience of explanation.
  • the policy generating unit 121 is configured to receive a total cost from each small station, and obtain a total network cost under different control policies according to the total cost, and use a corresponding control strategy as a minimum when the total network total cost is minimum. Control Strategy;
  • the command sending unit 122 sends a control command corresponding to the preferred control policy to the small station, so that the small station is correspondingly turned on or off according to the control instruction.
  • the second sub-band allocation unit 123 is configured to allocate the sub-band resource to its associated user and the associated user of the closed station at the start time of each scheduling time slot.
  • the policy generating unit 121 includes:
  • the total cost receiving module 1211 is configured to receive a total cost from each small station
  • the sub-optimal selection module 1212 is configured to obtain the total cost of the entire network when all the small stations are turned on, and sequentially close the small stations with the highest total cost among the opened small stations, until the total cost of the whole network is no longer reduced or all the small stations are all closed.
  • the control strategy of the small station at this time is taken as the preferred control strategy.
  • the second sub-band allocation unit 123 includes:
  • the second sub-band allocation module 1231 is configured to select an associated user of the associated user and the closed small station on each sub-band at the start time of each scheduling time slot, and the instantaneous throughput weighted queue length is the maximum equivalent rate.
  • the judgment execution module 1232 re-enters the next control period when the control period ends or when the associated user set changes in the heterogeneous network, otherwise enters the next scheduled time slot.
  • each of the functional units and the functional modules provided in this embodiment implements the respective steps in the fifth embodiment.
  • the present embodiment further discloses another specific preference of the policy generating unit 141 on the basis of the thirteenth embodiment. Structure, in this embodiment, after all the small stations are turned on, the small stations with the largest total cost among the remaining small stations are sequentially closed, so the complexity of the present embodiment has a linear relationship with the number of small stations B. Therefore, the solution is selected.
  • the preferred control strategy is a suboptimal control strategy. When the number of small stations B is large (such as 4, 8, 10, etc.), it is more suitable to select the macro station of the structure of the embodiment.
  • Embodiment 14 is a suboptimal control strategy. When the number of small stations B is large (such as 4, 8, 10, etc.), it is more suitable to select the macro station of the structure of the embodiment.
  • This embodiment provides a heterogeneous network, as shown in FIG. 13, including a plurality of small stations as described above, and a macro station as described in the above embodiment, and the macro station and the small station pass Wireless or wired connection.
  • the small station includes a first processor, a first memory, and a first communication interface connected to the data bus, where the first memory is configured to store the required execution steps of the small stations in the above embodiments.
  • the program is specifically controlled by a first processor, the macro station including a second processor, a second memory, and a second communication interface connected to the data bus, the second memory
  • the user stores the program corresponding to the execution step of the macro station in the above embodiments, and is specifically controlled by the second memory, and the small station and the macro station are connected through the first communication interface and the second communication interface to implement signal transmission, including The total cost, the control command corresponding to the preferred control strategy, and the like.
  • the heterogeneous network provided in this embodiment may use only the long-period BS-DTX control method, or the long-cycle BS-DTX control method and the short-cycle user selection method combined with the dual-scale energy-saving control method, and
  • the scale control method is superior to the long period BS-DTX control method.
  • less data is transmitted between the small station and the macro station, including the total cost of the small station feeding back to the macro station and the control instruction sent by the macro station to the small station, and the distributed structure has less processing overhead and improved. Network transmission efficiency.
  • the base station discontinuous transmission energy-saving control method is a BS-DTX energy-saving control scheme based on traffic volume and channel condition adaptation, in particular, delay-aware dual-scale BS-DTX control and user selection method. It can reduce the power consumption of the network very well.
  • the large-scale path loss models of macro stations and small stations are 128.1+37.61oglO(R) and 140.7+ 36.71oglO(R) (R units km), respectively.
  • the arrival of user data obeys the uniform distribution under the set ⁇ 2.5, 3, 3.5, 4, 4.5 ⁇ packet/second, and the packet size is 0.5 MByte.
  • curve 1 is a relationship between the small station shutdown control strategy based on the number of users and the transmission power-average delay corresponding to the slot user scheduling strategy based on the maximization of the subband channel gain, and curve 2 is based on the number of users.
  • the small station shutdown control strategy and the transmission power-average delay curve corresponding to the proportional fair time slot user scheduling strategy curve 3 is the small station shutdown control strategy based on the user arrival rate and the subband based channel gain
  • the transmission power-average delay curve corresponding to the maximized time slot user scheduling policy curve 4 is based on the user arrival rate
  • the small station shutdown control strategy and the transmit power-average delay curve corresponding to the proportional fair time slot user scheduling strategy curve 5 is the dual-scale BS-DTX control and the user-selected method corresponding to the transmit power-average
  • the relationship between delays as can be seen from the figure, when the same total average delay requirement (such as 40ms) is used, the proposed method saves the network average total transmit power by about 6dB (4 times) compared with the traditional method.

Abstract

本发明适用于异构网络,提供一种基站非连续发送节能控制方法、宏站、小站及异构网络,在小站侧所述方法包括:在每个控制周期开始时刻,小站获取服务其关联用户所需的总代价,并将所述总代价反馈至宏站;接收宏站发送的优选控制策略对应的控制指令,对应开启或关闭小站。在宏站侧所述方法包括:接收来自于各个小站的总代价,并根据所述总代价获取在不同控制策略下的整网总代价,将所述整网总代价最小时对应的控制策略作为优选控制策略;将所述优选控制策略对应的控制指令发送至小站。本发明考虑到了信道条件对网络功耗的影像,使得宏站得到优选控制策略,在满足时延要求的条件下,可以进一步降低系统能耗。

Description

基站非连续发送节能控制方法、 小站、 宏站及异构网络
技术领域
本发明属于异构网络技术领域, 尤其涉及一种基站非连续发送控制方 法、 宏站、 小站及异构网络。 背景技术
蜂窝网络能耗上升和二氧化碳排放量的上升已成为移动业务量指数上 升后的必然趋势, 也成为网络运营商保持利润降低成本所亟待解决的问题。 国际标准组织也将此问题列入到未来通信标准中需要解决的重要问题之一, 并引入了能效作为能量有效性的一种度量, 其中能效是指每焦耳能量可传输 的比特数。
为了支撑指数上升的业务量, 提升网络的吞吐量, 异构网络(英文: Heterogeneous Networks )已成为一种公认的有效解决方案。一个典型的异构 网络包括一个宏站和多个小站组成, 小站根据覆盖范围大小可以分为小站 ( micro cell ) 、 微小站 (pico cell ) 以及超小站 ( femto cell ) , 宏站提供了 基础覆盖, 而不同的小站在宏站的覆盖范围之内提供了不同范围和不同程度 的容量提升。 在高业务量时, 小站的开启能有效提升整网的能量效率, 因为 短距离覆盖极大降低了传输的能量消耗; 然而当业务量较低时, 宏站的容量 已足以服务用户, 并不需要小站来提升容量, 而此时由小站开启引入的静态 能耗会成为整网能耗的重要组成部分。 因此,在理想情况下,在低业务量时, 应该让小站进入休眠模式来节省网络能耗。
为此,目前有一些关于基站非连续发送(英文: Base Station-Discontinuous Transmission, 缩写: B S-DTX )节能控制的研究, 但是目前的研究方案中, BS-DTX的控制都只是随着业务量负载的信息变化而变化, 也就是一种业务 量自适应的 BS-DTX节能控制方法,基站的非连续发送控制自适应于业务量 负荷相关的参数, 比如单位时间内电话数(语音) , 用户活动程度, 业务量 到达速率等。当小站平均业务量低于某个预设的值,小站可以自行决定休眠, 并将剩余业务转移到提供覆盖的宏站, 或由其他可提供覆盖的小站服务; 当 此网络的业务量负载上升到某个值后, 此小站被唤醒, 继续提供服务。 但是 实际情况下, 系统的性能, 不光取决于业务量的大小 (服务需求的到达) , 同时还取决于信道条件好坏(被服务的情况) , 因此, 仅根据业务量大小进 行的 BS-DTX控制, 得到的控制策略通常不能达到很好的节能效果。 发明内容
鉴于上述问题,本发明的目的在于提供一种基站非连续发送节能控制方 法、 宏站、 小站及异构网络, 旨在解决现有异构网络的基于业务量自适应的
BS-DTX节能控制方法, 由于得到的控制策略仅与业务量大小有关, 无法达 到很好的节能效果。
第一方面, 所述基站非连续发送节能控制方法包括下述步骤: 在每个控 制周期开始时刻, 小站获取服务其关联用户所需的总代价, 并将所述总代价 反馈至宏站, 使得宏站根据所述总代价做出优选控制策略, 所述总代价包括 小站关联用户的平均时延信息; 接收宏站发送的优选控制策略对应的控制指 令, 对应开启或关闭小站。
在第一方面的第一种可能的实现方式中,所述控制周期划分为多个调度 时隙, 所述接收宏站发送的优选控制策略对应的控制指令, 对应开启或关闭 小站之后, 还包括: 在每个调度时隙开始时刻, 开启的小站将各自的子带资 源分配给其关联用户。
结合第一方面或第一方面的第一种可能的实现方式,在第二种可能的实 现方式中, 在每个控制周期开始时刻, 小站获取服务其关联用户所需的总代 价, 并将所述总代价反馈至宏站, 具体包括: 在每个控制周期开始时刻, 小 站获取其关联用户的最优带宽分配; 获取在所述最优带宽分配条件下的总代 价, 并将所述总代价反馈至宏站。
结合第一方面的第一种或第二种可能的实现方式,在第三种可能的实现 方式中, 每个小站包含多个子带, 所述在每个调度时隙开始时刻, 开启的小 站将各自的子带资源分配给其关联用户, 具体包括: 在每个调度时隙开始时 刻, 开启的小站在每个子带上, 选择其关联用户中瞬时吞吐量加权队列长度 后等效速率最大的用户; 在控制周期结束时或者异构网络中关联用户集改变 时, 重新进入下一控制周期, 否则进入下一调度时隙。
第二方面, 所述基站非连续发送节能控制方法包括: 宏站接收来自于各 个小站的总代价, 并根据所述总代价获取在不同控制策略下的整网总代价, 将所述整网总代价最小时对应的控制策略作为优选控制策略; 将所述优选控 制策略对应的控制指令发送至小站, 以使小站根据所述控制指令对应开启或 关闭。
在第二方面的第一种可能的实现方式中,所述接收来自于各个小站的总 代价, 并根据所述总代价获取在不同控制策略下的整网总代价, 将所述整网 总代价最小时对应的控制策略作为优选控制策略, 具体包括: 宏站接收来自 于各个小站的总代价; 遍历所有可能的控制策略, 并获取宏站所服务用户的 最优带宽分配; 根据所述总代价和最优带宽分配, 获取在各种控制策略下的 整网总代价; 选择整网总代价最小时对应的控制策略作为优选控制策略。
在第二方面的第二种可能的实现方式中, 所述接收来自于各个小站的总 代价, 并根据所述总代价获取在不同控制策略下的整网总代价, 将所述整网 总代价最小时对应的控制策略作为优选控制策略, 具体包括: 宏站接收来自 于各个小站的总代价; 获取开启所有小站时的整网总代价, 顺次关闭开启的 小站中总代价最高的小站, 直至整网总代价不再降低或者所有小站全部关 闭, 将此时小站的控制策略作为优选控制策略。
结合第二方面或第二方面的第一种可能的实现方式或第二种可能的实 现方式, 在第三种可能的实现方式中, 所述将所述优选控制策略对应的控制 指令发送至小站,以使小站根据所述控制指令对应开启或关闭之后,还包括: 在每个调度时隙开始时刻,宏站将子带资源分配给其关联用户以及关闭小站 的关联用户。
结合第二方面的第三种可能的实现方式, 在第四种可能的实现方式中, 所述宏站包含多个子带, 所述在每个调度时隙开始时刻, 宏站将子带资源分 配给其关联用户以及关闭的小站的关联用户, 具体包括: 在每个调度时隙开 始时刻,宏站在每个子带上,选择其关联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户; 在控制周期结束时或者异 构网络中关联用户集改变时, 重新进入下一控制周期, 否则进入下一调度时 隙。
第三方面, 所述小站包括: 总代价获取单元, 用于在每个控制周期开始 时刻, 获取服务其关联用户所需的总代价, 并将所述总代价反馈至宏站, 使 得宏站根据所述总代价做出优选控制策略, 所述总代价包括小站关联用户的 平均时延信息; 指令接收执行单元, 用于接收宏站发送的优选控制策略对应 的控制指令, 对应开启或关闭小站。
在第三方面的第一种可能的实现方式中, 所述小站还包括: 第一子带分 配单元, 用于在每个调度时隙开始时刻, 将各自的子带资源分配给其关联用 户。
结合第三方面或第一方面的第一种可能的实现方式,在第二种可能的实 现方式中, 所述总代价获取单元包括: 带宽分配获取模块, 用于在每个控制 周期开始时刻, 获取其关联用户的最优带宽分配; 总代价获取模块, 用于获 取在所述最优带宽分配条件下的总代价, 并将所述总代价反馈至宏站。
结合第三方面的第一种或第二种可能的实现方式,在第三种可能的实现 方式中, 所述第一子带分配单元包括: 第一子带分配模块, 用于在每个调度 时隙开始时刻, 在每个子带上, 选择小站关联用户中, 瞬时吞吐量加权队列 长度后等效速率最大的用户; 判断执行模块, 用于在控制周期结束时或者异 构网络中关联用户集改变时, 重新进入下一控制周期, 否则进入下一调度时 隙。
第四方面, 所述宏站包括: 策略生成单元, 用于接收来自于各个小站的 总代价, 并根据所述总代价获取在不同控制策略下的整网总代价, 将所述整 网总代价最小时对应的控制策略作为优选控制策略; 指令发送单元, 将所述 优选控制策略对应的控制指令发送至小站, 以使小站根据所述控制指令对应 开启或关闭。
在第四方面的第一种可能的实现方式中, 所述策略生成单元包括: 总代 价接收模块, 用于接收来自于各个小站的总代价; 带宽分配模块, 用于遍历 所有可能的控制策略, 并获取宏站所服务用户的最优带宽分配; 整网总代价 获取模块, 用于根据所述总代价和最优带宽分配, 获取在各种控制策略下的 整网总代价; 最优选择模块, 用于选择整网总代价最小时对应的控制策略作 为优选控制策略。
在第四方面的第二种可能的实现方式中, 所述策略生成单元包括: 总代 价接收模块, 用于接收来自于各个小站的总代价; 次优选择模块, 用于获取 开启所有小站时的整网总代价, 顺次关闭开启的小站中总代价最高的小站, 直至整网总代价不再降低或者所有小站全部关闭,将此时小站的控制策略作 为优选控制策略。
结合第四方面或第二方面的第一种可能的实现方式或第二种可能的实 现方式, 所述宏站还包括: 第二子带分配单元, 用于在每个调度时隙开始时 刻, 将子带资源分配给其关联用户以及关闭小站的关联用户。
结合第四方面的第三种可能的实现方式, 在第四种可能的实现方式中, 所述第二子带分配单元包括: 第二子带分配模块, 用于在每个调度时隙开始 时刻, 在每个子带上, 选择宏站关联用户以及关闭的小站的关联用户中, 瞬 时吞吐量加权队列长度后等效速率最大的用户; 判断执行模块, 在控制周期 结束时或者异构网络中关联用户集改变时, 重新进入下一控制周期, 否则进 入下一调度时隙。
第五方面, 所述异构网络包括: 上述小站和宏站, 所述宏站与各个小站 间有线或无线连接。
本发明的有益效果是: 本发明提供了一种业务量与信道条件自适应的 BS-DTX节能控制方案, 首先小站需要获取服务其关联用户所需的总代价, 所述总代价包括小站关联用户的平均时延信息,宏站根据各个小站反馈的总 代价来做出优选控制策略, 因此在本发明技术方案中, 不仅考虑到了业务量 (即平均吞吐量), 还考虑到了信道状态信息(所述平均时延信息反应了信 道状态信息) , 使得网络时延与总功耗之间有了较好折中, 在满足时延要求 的条件下进一步降低系统能耗; 另一方面, 本发明中获取总代价是由各个小 站实现, 这种分布式设计降低了宏站的工作量, 易于实现, 而且还大大降低 了网络控制开销, 提高了网络效率。 附图说明
图 1是异构网络的一种网络拓朴图;
图 2 是本发明第一实施例提供的基站非连续发送节能控制方法的流程 图;
图 3 是本发明第二实施例提供的基站非连续发送节能控制方法的流程 图;
图 4 是本发明第三实施例提供的基站非连续发送节能控制方法的流程 图;
图 5 是本发明第四实施例提供的基站非连续发送节能控制方法的流程 图;
图 6 是本发明第五实施例提供的基站非连续发送节能控制方法的流程 图;
图 7是本发明第六实施例提供的基站非连续发送节能控制方法的交互流 程图;
图 8是本发明第七实施例提供的小站的结构方框图;
图 9是本发明第八实施例提供的小站的结构方框图;
图 10是本发明第九实施例提供的宏站的结构方框图;
图 11是本发明第十实施例提供的宏站的结构方框图;
图 12是本发明第十一实施例提供的宏站的结构方框图;
图 13是本发明第十二实施例提供的异构网络的系统图;
图 14是发送功率-平均时延的一种仿真曲线关系图。 具体实施方式
为了使本发明的目的、 技术方案及优点更加清楚明白, 以下结合附图及 实施例, 对本发明进行进一步详细说明。 应当理解, 此处所描述的具体实施 例仅仅用以解释本发明, 并不用于限定本发明。
本发明实施例提供了一种基于业务量与信道条件自适应的 BS-DTX 节 能控制方案, 考虑到了业务量和信道状态, 得到优选控制策略, 在满足时延 要求条件下尽可能降低系统功耗, 为了说明本发明所述的技术方案, 下面建 立各个分析模型:
首先定义异构网络的网络拓朴模型, 如图 1所示, 假设异构网络中包括 一个宏站和 B个小站, 且所有小站的覆盖范围均在宏站的覆盖范围内,宏站 为异构网络提供基础覆盖和容量, 而小站则在部分地区提供容量提升。 令 0 表示宏站的标号,令 = 1,... 分别表示宏站内 B个小站的标号。令 = {1,· · ·, } 表示所有小站的集合。 令 表示所有移动用户集合, 令 7 ^表示在第 b个小站 内的移动用户集合, Κ = " Λ。每个用户(英文: mobile station, 缩写: MS ) 选择逻辑上与其接收到信号最强的一个宏站或一个小站相关联。 与小站相关 联的 MS可能被宏站或者该小站服务, 如果控制策略决定此小站开启, 则这 些关联用户由小站服务, 否则由宏站服务。 本发明实施例只考虑对宏站覆盖 范围内有活动用户 ( active users ) 的小站进行开关优化控制, 对那些无活动 用户的小站, 则默认其已自动关闭。
对于系统功耗模型, 基站的功耗模型可表达成:
( 1 )
其中 4 、 ^ 和 ^分别是第 b个基站的总功耗、 固定功耗(相对 静态的功耗) 以及发送功率。 表示由功率放大器、 馈线 (小站没有)、 制 冷或电源转换效率等带来的功率放大因子。
对于系统物理层模型, 假设异构网络中所有的基站共享公共的 WHz的 频谱, 总频带被分为 M独立等分, 每个独立子带的带宽 HZ。 令 H4 ¾ m(t)和
M
分别表示第 b个基站和第 k个用户之间、 在第 m个子带、 在 t时刻的小 尺度信道衰落和大尺度路径损耗。 此外, 时间尺度上, 总的时间轴被逻辑地 划分为时间长度为 ^的多个离散调度时刻, 标号为 t。 在一个控制周期, 即一 长周期 T内, 令 Pb {0,1}来指示第 b个基站 = l,...,iO BS-DTX控制动作, 其中 A = 1表示第 b个小站开启, Pb Q则表示第 b个小站关闭, 并且宏 站始终开启来保证该区域的覆盖和新用户的接入。 当第 b个小站开启时, 该 小站覆盖范围内的用户集 ^ ^由此小站服务, 否则由宏站服务。 在一个长周期 内, 考虑大尺度路径损耗来估计平均吞吐量 (在时间和频带上的平均). 第 b 个小站与第 k个用户 ( e ) 间的平均吞吐量可由下式估计:
Figure imgf000010_0001
表示噪声功率。 类似的, 宏站与其用户 k间的平均吞吐量可估计为:
p T
1 + : k e !C ( 3 )
Ρ、' 对于网络平均总时延和功率消耗模型, 令 ww 表示第 b个小站分配给 用户 k的平均总带宽, 这个带宽分配满足下述的规则:
∑¾o+∑A∑¾0=W (4)
∑wkJ>=W, beB (5 ) 宏站与小站之间复用所有的带宽, 式(4)表示宏站服务的用户所使用 的带宽相加等于总带宽 W, 式( 5 )表示每个开启的小站也使用全部的带宽。
另外, 被第 b个基站服务的用户 k的平均时延可定义为
(6) wk,k, k
因此,可以把整个异构网络的时延代价表示为长周期 BS-DTX控制策略 p = {pb:bsB} 以及长周期带宽分配 w {W l: :e ,beS,:eX:,b = 0}的函数。 因 此, 网络平均总时延(定义为网络中每个活动用户 k的平均时延总和)可以 表示为如下三部分的和: 1 )与宏站相关联且被宏站服务用户的平均时延总 和; 2 )与小站相关联但由于小站关闭而被宏站服务的用户的平均时延总和; 3 )与小站相关联并被小站服务的用户的平均时延的总和。 网络平均总时延:
£)(p,W)=∑Z 。(¾。) +∑ pb∑ DkJ,(wk,)+(l-pb)∑ Dkfi(wkp) ( 7 ) 类似的,整网的平均总功率消耗也可以表示为长周期 BS-DTX控制策略
P的函数, 包括如下两部分: 1 )宏站的平均总功率; 2)所有开启的小站的 平均总功率。
P(P) = P +∑PbPtOtal, (8) 本发明实施例就是为了降低式(7 ) 中的网络平均总时延和式(8) 中的 平均总功率消耗。 具体的, 结合式(4)和式(5), 通过帕累托最优(Pareto optimal ) 的折衷解来解决:
min P,w (P,w) + ^P(p)
其中 0是表示在功率和时延之间折衷参数, 具体由系统需求给定。 当 系统设计的重点在降低时延时, 减小, 而当系统设计的重点在降低功率时, 增大。 极限情况当 = 0, 则表示系统设计以时延优化为唯一目标; 反之, 当 趋向无穷,则表示系统设计以最小化功率开销为目标。 因此本发明实施 例可以在满足时延条件下, 尽可能的降低系统能耗。
下面通过具体实施例来进行说明。
实施例一:
图 2示出了本发明第一实施例提供的基站非连续发送节能控制方法的流 程, 为了便于说明仅示出了与本发明实施例相关的部分。
所述基站非连续发送节能控制方法以小站侧进行描述, 包括:
步骤 S201、 在每个控制周期开始时刻, 小站获取服务其关联用户所需 的总代价, 并将所述总代价反馈至宏站, 使得宏站根据所述总代价做出优选 控制策略, 所述总代价包括小站关联用户的平均时延信息。
每个小站都服务其范围内的关联用户, 本步骤首先要获取小站服务其关 联用户的总代价, 即服务总开销, 所述总代价包括小站关联用户的平均时延 信息, 所述平均时延信息在一定程度上反应了信道状态信息, 各个小站再将 得到的总代价发送至宏站,宏站根据所有的总代价在不同的控制策略下获取 整网总代价, 将所述整网总代价最小时对应的控制策略作为优选控制策略, 根据优选控制策略控制小站的开启和关闭。
步骤 S202、 接收宏站发送的优选控制策略对应的控制指令, 对应开启 或关闭小站。
在宏站做出优选控制策略后, 小站接收到所述优选控制策略对应的控制 指令, 执行相应开启或关闭小站工作。
本实施例方法是一种时延感知的业务量与信道条件自适应 BS-DTX 节 能控制方案, 在一控制周期下, 各个小站考虑到业务量和平均时延, 计算小 站所需总代价, 宏站再根据总代价得到在不同的控制策略下的整网总代价, 并且将整网总代价最小时对应的控制策略作为优选控制策略。 与现有的业务 量自适应 BS-DTX节能控制方案相比, 本实施例考虑到了信道条件 (即用户 平均时延) , 使得宏站得到优选控制策略, 在满足时延要求的条件下, 可以 进一步降低系统能耗。
优选的, 在所述步骤 S202之后还包括:
步骤 S203、 在每个调度时隙开始时刻, 开启的小站将各自的子带资源 分配给其关联用户。
宏站在每个控制周期都会生成一个优选控制策略, 并根据所述优选控制 策略对应开启或关闭小站, 这是一种长周期下的 BS-DTX节能控制策略, 考 虑了长周期下的平均业务量和平均信道质量, 这是一种长时间尺度的控制方 案,但实际网络中,有很多时延敏感的业务, 且具有突发到达的特性, 因此, 除了传统的物理层吞吐量, 考虑业务的随机突发到达以及 MAC (媒体访问 控制)层的时延性能, 也变得十分重要。 因此本优选步骤进一步考虑到业务 的随机突发性, 将控制周期分成若干调度时隙, 调度时隙标号用 t表示, 在 每个时隙中小站都进行子带分配, 将子带资源分配给其关联的用户。 本优选 实施方式在得到长周期(即控制周期)的优选控制策略后, 还需进一步进行 短周期(即每个调度时隙 )的子带资源分配, 这种双尺度的 BS-DTX控制及 用户选择方法可以在保证网络时延的基础上进一步降低网络功耗。 实施例二:
图 3示出了本发明第二实施例提供的基站非连续发送节能控制方法的流 程, 为了便于说明仅示出了与本发明实施例相关的部分。
所述基站非连续发送节能控制方法包括:
步骤 S301、 在每个控制周期开始时刻, 小站获取其关联用户的最优带 宽分配。
各个小站获取其关联用户的最优带宽分配, 公式如下:
Figure imgf000013_0001
其中 满足式(5 ) 中限制, 且可表示为 Vfc , W表示
Figure imgf000014_0001
网络中基站共享公共频谱带宽, 表示第 b个小站与第 k个用户 ( k G 间的平均吞吐量, U为业务平均到达速率, 令 4(t)表示用户 k在其服务基 站端第 t个调度时隙开始时, 新到达的业务量(单位: bit ) , 则 λ¾ = ¾^
τ (bit/second)。
步骤 S302、 获取在所述最优带宽分配条件下的总代价, 并将所述总代 价反馈至宏站。
上述步骤 S301-S302是实施例一中步骤 S201的一种具体优选步骤, 本 实施例中, 各个小站获取在最优带宽分配条件下计算总代价, 根据上述的网 络平均总时延和功率消耗模型, 可以得知在最优带宽分配条件下小站的平均 总时延和平均功率消耗, 具体的, 最优带宽分配下的总代价:
Ub = ( ) + 7 bPaJ,
( 10 ) 其中, 为小站发送功率, 表示功率放大因子, 表示在 最优带宽分配下, 小站 b所述服务的用户 k的平均时延, ; κ为功率和时延之 间折衷参数, 其具体值按照时延需求所定。 各个小站在获取到总代价后 Ub 后, 将 Ub反馈至宏站, 然后宏站根据所述总代价, 在不同的控制策略下获 取整网总代价, 将所述整网总代价最小时对应的控制策略作为优选控制策 略, 所述优选控制策略可以控制小站的开启和关闭。
步骤 S303、 接收宏站发送的优选控制策略对应的控制指令, 对应开启 或关闭小站。
步骤 S304、 在每个调度时隙开始时刻, 开启的小站在每个子带上, 选 择其关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户;
步骤 S305、 在控制周期结束时或者异构网络中关联用户集改变时, 重 新进入下一控制周期, 否则进入下一调度时隙。
上述步骤 S304-S305是实施例一中步骤 S203的一种具体优选实施方式, 每个小站包含有多个子带, 开启的小站在每个子带上, 选择其关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户。
令 , (0 {0,1}表示第 b个基站对第 k个用户在调度时隙 t对第 m个子 带第 m个子带的分配(即用户选择)指示因子, = 1表示在调度时隙 t 将第 m个子带分配到第 k个用户, 即在第 m个子带上选择第 k个用户, 否 则 ;^) = 0。 本实施例是基于瞬时吞吐量进行的, 第 b个小站与用户 k在第 t个调度时隙, 在第 m个子带上的瞬时吞吐量(单位: bit/sec/Hz )可表示为
Figure imgf000015_0001
类似的, 宏站与其用户 k之间在 t时刻 m子带上的瞬时吞吐量可以表示为
Figure imgf000015_0002
第 b个基站对第 k个用户在调度时隙 t对第 m个子带的分配(即用户选 择)指示因子:
, fc = argmaxfce j¾(tD )
[0, otherwise
( 13 )
其中, (0表示用户 k在其服务基站端第 t个调度时隙开始时刻时, 月良 务队列中的队列状态信息, 即剩余比特数 QSI ( bit ), QSI表示了队列长度, 队列在下一时隙的更新可 示为:
Figure imgf000015_0003
分配指示因子的表达式表示了将第 t个调度时隙的第 m个子带分配给了 瞬时吞吐量加权队列长度后等效速率最大的用户。 在一个控制周期 T的当前调度时隙将子带资源分配完毕后,进入下一调 度时隙子带资源分配, 直至所述每个控制周期 T的结束时刻, 或者异构网络 中关联用户集改变时, 重新进入下一长周期进行所述基于业务量与信道条件 的自适应 BS-DTX节能控制, 得到优选控制策略, 然后进行各个调度时隙的 子带资源分配, 即用户选择。 本实施例在得到长周期(即控制周期)的优选 控制策略后, 还需进一步进行短周期 (即每个调度时隙) 的子带资源分配, 这种双尺度的 BS-DTX控制及用户选择方法可以在保证网络时延的基础上 进一步降低网络功耗。
实施例三:
图 4示出了本发明第三实施例提供的基站非连续发送节能控制方法的流 程, 为了便于说明仅示出了与本发明实施例相关的部分。
本实施例提供的基站非连续发送节能控制方法以宏站侧进行描述, 具体 包括:
步骤 S401、 宏站接收来自于各个小站的总代价, 并根据所述总代价获 取在不同控制策略下的整网总代价,将所述整网总代价最小时对应的控制策 略作为优选控制策略;
步骤 S402、 将所述优选控制策略对应的控制指令发送至小站, 以由小 站根据所述控制指令对应开启或关闭。
本实施例中的两个步骤与实施例一种步骤 S201、 S202对应, 以宏站侧 进行描述。 各个小站在获取到总代价 Ub并反馈给宏站后, 宏站根据所述总 代价获取在各种不同控制策略下的整网总代价, 然后获取整网总代价最小时 对应的控制策略作为优选控制策略, 并根据所述优选控制策略对应向小站发 出控制指令来控制小站的开启和关闭。
与现有的业务量自适应 BS-DTX节能控制方案相比,本实施例考虑到了 信道条件(即用户平均时延) , 使得宏站得到优选控制策略, 在满足时延要 求的条件下, 可以进一步降低系统能耗。 另外, 宏站和小站之间的数据传输 开销很少, 只有各个小站反馈的总代价, 以及宏站发出的用于控制小站开启 和关闭的控制指令。 这种分布式工作方式可以提高了网络效率。
进一步作为优选的实施方式, 所述步骤 402之后还包括:
步骤 S403、 在每个调度时隙开始时刻, 宏站将子带资源分配给其关联 用户以及关闭小站的关联用户。
宏站在每个控制周期都会生成一个优选控制策略, 并根据所述优选控制 策略对应开启或关闭小站, 这是一种长周期下的 BS-DTX节能控制策略, 考 虑了长周期下的平均业务量和平均信道质量, 这是一种长时间尺度的控制方 案,但实际网络中,有很多时延敏感的业务, 且具有突发到达的特性, 因此, 除了传统的物理层吞吐量, 考虑业务的随机突发到达以及 MAC (媒体访问 控制)层的时延性能, 也变得十分重要。 因此本优选步骤进一步考虑到业务 的随机突发性, 将控制周期分成若干调度时隙, 调度时隙标号用 t表示, 本 优选实施方式宏站在得到长周期(即控制周期)的优选控制策略后, 还需进 一步进行短周期(即每个调度时隙)的子带资源分配,这种双尺度的 BS-DTX 控制及用户选择方法可以在保证网络时延的基础上进一步降低网络功耗。 实施例四:
图 5示出了本发明第四实施例提供的基站非连续发送节能控制方法的流 程, 为了便于说明仅示出了与本发明实施例相关的部分。
本实施例提供的基站非连续发送节能控制方法以宏站侧进行描述, 具体 包括:
步骤 S501、 宏站接收来自于各个小站的总代价;
步骤 S502、 遍历所有可能的控制策略, 并获取宏站所服务用户的最优 带宽分配。
一个异构网络里有 B个小站,那么就有 2s种小站控制策略, 本步骤中要 便利所有的控制策略, 并获取在每种策略下的最优带宽分配, 表示为:
Figure imgf000018_0001
(14)
其中^满足对一个给定控制策略, 满足(4) 式限制, 这里:
Figure imgf000018_0002
(15)
其中, W表示网络中基站共享公共频谱带宽, ,。表示宏站与第 k个用 户间的平均吞吐量, ^为业务平均到达速率。 Pb为第 b个小站开启或关闭 的指示因子。
步骤 S503、根据所述总代价和最优带宽分配,获取在各种控制策略下的 整网总代价;
步骤 S504、 选择整网总代价最小时对应的控制策略作为优选控制策略。 上述步骤 S501-S504为实施例三中步骤 S401的一种具体优选实施方式, 根据每个小站的反馈 Ub以及宏站获取的最优带宽分配, 计算得到在各种控 制策略 P下的整网总代价, 所述整网总代价同样包括三部分: 1 )宏站服务 其关联用户的代价; 2)在 p策略下开启的小站服务其关联用户的代价; 3) 在 p策略下宏站服务关联在小站但被宏站服务用户的代价。 因此所述整网总 代价可以表示为:
Figure imgf000018_0003
) + £ PbUb + (l-pb)∑Dkfl(wlfi )
(16)
宏站再根据整网总代价最小时的控制策略作为优选控制策略, 即:
p* = argminpf/(p,w*(p)) 步骤 S505、 将所述优选控制策略对应的控制指令发送至小站, 以由小 站根据所述控制指令对应开启或关闭;
步骤 S506、 在每个调度时隙开始时刻, 宏站在每个子带上, 选择其关 联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度后等效速率 最大的用户;
步骤 S507、 在控制周期结束时或者异构网络中关联用户集改变时, 重 新进入下一控制周期, 否则进入下一调度时隙。
上述步骤 S506、 S507为实施例三中步骤 S403的一种具体优选实施方式, 实现了一种子带分配的方法, 宏站同样包含有多个子带, 宏站对第 k个用户 在调度时隙 t对第 m个子带的分配(即用户选择)指示因子:
il, : = argmax^ y !¾(t ,m (t)
¾,0,m( = 6
0, otherwise ( 18 ) 其中, β<0表示用户 k在其服务基站端第 t个调度时隙开始时刻时, 服 务队列中的队列状态信息, 即剩余比特数 QSI ( bit ), QSI表示了队列长度, 分配指示因子的表达式表示了将第 t个调度时隙的第 m个子带分配给了瞬时 吞吐量加权队列长度后等效速率最大的用户。
本实施例是在所有可能的控制策略中选出整网总代价最小时的控制策 略作为优选控制策略, 因此本实施例选出的优选控制策略为最优控制策略, 但是由于计算整网总代价的复杂度与小站个数 B呈指数关系,当小站的个数 不大时, 比如 2个, 使用本方案较好。 另一方面, 本实施例还实现了一种具 体的短周期的子带分配算法, 虑到业务的随机突发性, 在得到长周期的优选 控制策略后, 进一步考虑到业务的随机突发性, 将控制周期分成若干调度时 隙,进行短周期(即每个调度时隙 )的子带资源分配,这种双尺度的 BS-DTX 控制及用户选择方法可以在保证网络时延的基础上进一步降低网络功耗。 实施例五: 图 6示出了本发明第五实施例提供的基站非连续发送节能控制方法的流 程, 为了便于说明仅示出了与本发明实施例相关的部分。
本实施例提供的基站非连续发送节能控制方法以宏站侧进行描述, 具体 包括:
步骤 S601、 宏站接收来自于各个小站的总代价;
步骤 S602、 获取开启所有小站时的整网总代价, 顺次关闭开启的小站 中总代价最高的小站, 直至整网总代价不再降低或者所有小站全部关闭, 将 此时 d、站的控制策略作为优选控制策略。
上述步骤 S601、 S602是实施例三中步骤 S401的另一中具体优选实施方 式, 具体实现时, 宏站将每个小站的反馈 {C b e B}进行降序排列, 并令 u0 = if ), 其中对所有的小站 = ι,即 υο是所有小站均开启下的代价函数 值, 设置计数器 b=l。
在进行第 b次循环计算,宏站计算此时 p策略下的总代价 fT(p), 此时的 ρ策略中, 对于 b'≤b, b'≤b , 否则; ¾Ί = 1。 如果此时的策略 ρ下获得的总代 价小于初始上一轮的代价, 且总的轮数小于 则令 b=b+l,并将此时的策略 赋给最终策略, = P 并进入下一轮循环计算, 直至整网总代价不再降低, 或者轮数达到 B时终止计算。
步骤 S603、 将所述优选控制策略对应的控制指令发送至小站, 以使小 站根据所述控制指令对应开启或关闭;
步骤 S604、 在每个调度时隙开始时刻, 宏站在每个子带上, 选择其关 联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度后等效速率 最大的用户;
步骤 S605、 在控制周期结束时或者异构网络中关联用户集改变时, 重 新进入下一控制周期, 否则进入下一调度时隙。
上述步骤 S603-S605与实施例四中 S505-S507相同, 此处不再赘述。 与实施例四不同之处在于, 实施例四在所有可以的控制策略中选择优选 控制策略,复杂度与小站个数 B呈指数关系,而本实施例将所有小站开启后, 顺次关闭剩余小站中总代价最大的小站, 因此本实施方案的复杂度与小站个 数 B呈线性关系, 因此, 本方案所选出的优选控制策略为次优控制策略, 在 小站个数 B较大时 (比如 4、 8、 10等) , 选择本实施例方法更合适。 实施例六:
图 7示出了本发明第六实施例提供的基站非连续发送节能控制方法的交 互流程, 为了便于说明仅示出了与本发明实施例相关的部分。
本实施例从整个异构网络系统的角度来描述基站非连续发送节能控制 方法, 具体包括:
步骤 S701、 在每个控制周期开始时刻, 小站获取其关联用户的最优带 宽分配;
步骤 S702、 小站获取在所述最优带宽分配条件下的总代价, 并将所述 总代价反馈至宏站;
步骤 S703、 宏站接收来自于各个小站的总代价, 并根据所述总代价获 取在不同控制策略下的整网总代价,将所述整网总代价最小时对应的控制策 略作为优选控制策略;
步骤 S704、 宏站将所述优选控制策略对应的控制指令发送至小站; 步骤 S705、 小站接收宏站发送的优选控制策略对应的控制指令, 对应 开启或关闭小站。
优选的, 所述步骤 S705之后还包括:
步骤 S706、 在当前调度时隙开始时刻, 开启的小站在每个子带上, 选 择小站的关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户; 以 及将宏站的子带资源分配给宏站的关联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户;
步骤 S707、 在控制周期结束时或者异构网络中关联用户集改变时, 重 新进入下一控制周期, 否则进入下一调度时隙。
上述步骤 S701-S705是一种时延感知的长周期 BS-DTX控制方法,可以 得到在一控制周期内的优选控制策略,步骤 S906和 S907进一步考虑到业务 的随机突发性, 将控制周期分成若干调度时隙, 进行短周期(即每个调度时 隙 )的子带资源分配, 这种双尺度的 BS-DTX控制及用户选择方法可以在保 证网络时延的基础上进一步降低网络功耗。 实施例七:
本实施通过一具体应用场景来进一步说明本实施例提供的基站非连续 发送节能控制方法, 这是一种双时间尺度的 BS-DTX控制方法, 包括长周期 的 BS-DTX算法和短周期的子带分配算法,首先得到长周期下的优选控制策 使得系统在保证网络时延的前提下, 尽可能降低系统功耗。 本实施例张中宏 站执行实施例四所述方法, 小站执行实施例一所述方法, 4叚设异构网络中包 括一个宏站(BS0 )和两个小站(BS1、BS2 ),在宏站中有一个活动用户( MS0 ), 小站 1 中有一个活动用户 ( MS1 ) , 在小站 2 中没有用户, 即 Β = {1},ΚΌ = {0},Κι = {1},}^ = φ , 因此小站 BS2将直接被关闭, 我们只需考虑对 小站 BS1的 BS-DTX控制策略。
宏站 BS0与其服务用户 MS0之间的平均吞吐率为^, 宏站与 MS1之 间的平均吞吐率为^, 小站 BS1 与 MS1 之间的平均吞吐率为 ^, MS0和 MS1两用户平均数据到达速率为 。,^。 本实施方法步骤具体如下:
A ) 首先在一个长周期的第一个调度时隙, 即令 t=l。
B ) 然后执行长周期的 BS-DTX算法得到最优控制策略, 具体如下: Bl、 小站 BS1根据式(9 )计算得分配给其关联用户 MS1的最优带宽 ^ = ^。基于此, 小站 BS1根据式 ( 10 ) 计算此时服务 MS1的总代价
Figure imgf000022_0001
) + alPtx l , 并将此值反馈给宏站; B2、 对于 A = i, 宏站 BSO根据式(14)获得对用户 MSO和 MSI的最 优带宽分配; 。(1) = 和; 。(1) = 0, 基于此, 宏站 BS0根据式( 16 )计算 整网总代价 f/(l, = D00(wlil)) + Ub, 此处 Ub = Ul。
对于 A = 0, 宏站 BSO根据式(14)获得对用户 MSO和 MSI的最优带 宽分配, 分别为 + , 并且满足
Figure imgf000023_0001
v0= + , 基于此, 宏站 BS0根据式 ( 16 ) 计算整网总代
0,0 0
(0, w(o)) = ( 0》 + Α,0(«Ο));
Β3、 宏站 BS0根据式( 17 )得到最优的长周期 BS-DTX控制策略 P = argmin^ ;
C)最后执行短周期的子带分配算法进行用户选择,具体的, 当 A* = l, 小 站 BS1 处于开启状态, 小站 BS1 根据式(13 )选择用户, 对所有的 m, ,m(0 = l, 宏站 BS0 根据式(18 )选择用户,对于所有的 m, ¾。,m(0 = l; 如 果 A* = o, 小站 BSl处于关闭状态,小站 BSl不服务用户 MSI,宏站 BS0 根 据式(is)为不同子带 m选择用户: 。,。 =1, 。,™=0
Figure imgf000023_0002
D)如果 t=T 或者任何一个关联用户集改变,回到步骤 A ,否则令 t=t+l 回步骤 B。 这样在一个控制周期(即长周期)的结束时隙或者当关联用户集 改变时, 进入下一控制周期进行最优策略选择和用户选择, 否则进入下一调 度时隙进行最优策略选择和用户选择。 实施例八:
本实施通过另一具体应用场景来进一步说明本实施例提供的基站非连 续发送节能控制方法, 这是也是一种双时间尺度的 BS-DTX控制方法, 包括 长周期的 BS-DTX算法和短周期的子带分配算法,本实施例张中宏站执行实 施例四所述方法, 小站执行实施例一所述方法, 假设异构网络中包括一个宏 站(BS0)和两个小站(BS1、 BS2) , 在宏站中有一个活动用户 (MS0) , 小 站 1 和 2 中 各有一个活动用 户 ( MS1 和 MS2 ) , 即 B = {1,2},/ = {0},ς = {1},/ς = {2} , 宏站 BS0与其服务用户 MS0之间的平均吞 吐率为 。, 宏站与 MS1、 MS2之间的平均吞吐率为 。 和 。 小站 BS1 与 MS1 之间的平均吞吐率为 ,i, 小站 BS2 与 MS2 之间的平均吞吐率为 MS, MSI和 MS2三个用户平均数据到达速率为 , ^。 本实施例 方法步骤具体如下:
A) 首先在一个长周期的第一个调度时隙, 即令 t=l。
B) 然后执行长周期的 BS-DTX算法得到最优控制策略, 具体如下:
Bl、 小站 BS1根据式(9)计算得分配给其关联用户 MS1的最优带宽 w = W, 基于此, 小站 BS1根据式 ( 10 ) 计算此时服务 MS1的总代价
Figure imgf000024_0001
) + «Λ, 并将此值反馈给宏站; 小站 BS2根据式 (9)计算 得分配给其关联用户 MS2的最优带宽^^^, 基于此, 小站 BS1根据式 ( 10)计算此时服务 MS2的总代价 f/2 = 22 2 ) + α2Ρ 1, 并将此值反馈给 宏站;
Β2、 遍历控制策略p = Cp1Jp2) = (0,0),(0,l),(l,0),(l,l) (分别表示所述两 个小站的开关状态), 宏站 BS0根据式( 14)获得对用户 MS0、 MSI以及 MS2的最优带宽分配, 分别为 。 ),0),0), 基于此, 宏站 BS0根 据 式 ( 16 ) 计 算 整 网 总 代 价
U (p, w*0(p)) = D00 (w ip)) + plUl + (l- pl )Dl0 (Μ¾ρ)) + p2U2 + (l-p2 )D20 (¾p))
B3、 宏站 BSO根据式( 17 )得到最优的长周期 BS-DTX控制策略
Figure imgf000024_0002
C)最后执行短周期的子带分配算法进行用户选择, 在已获得的长周期 的 BS-DTX最优策略 P*下, 宏站 BS0、 小站 BS1和 BS2 (如果开启的话) 分别执行每时隙的子带分配算法: 具体的, 如果 ^ = 1 , 小站 BS 1 处于开 启状态, 小站 BS1 根据式(13 )选择用户, 对于所有的 m, sl l m(t) = h 宏站
BS0 根据式(18 )选择用户, 对于所有的 m, ™( = 1 , 如果 ^ = 1, 小站 BS2 处于开启状态, 小站 BS2 也根据式(13 )为每个子带 m选择用户, 宏 站 BS0 根据式( 18 )选择用户。
D )如果 t=T 或者任何一个关联用户集改变,回到步骤 A ,否则令 t=t+l 回步骤 B。 这样在一个控制周期(即长周期)的结束时隙或者当关联用户集 改变时, 进入下一控制周期进行最优策略选择和用户选择, 否则进入下一调 度时隙进行最优策略选择和用户选择。
上述实施例九和实施例十是列举的是在两种典型的应用场景下的基站 非连续发送节能控制方法, 若小站个数较多 (比如大于 2时) , 宏站可以按 照实施例五所述方法选择出次优控制策略, 然后再在所述次优控制策略下进 行用户选择, 具体的实现过程此处不再赘述。 实施例九:
图 8示出了本发明第九实施例提供的小站的结构, 为了便于说明仅示出 了与本发明实施例相关的部分。
本实施例提供的小站包括:
总代价获取单元 801, 用于在每个控制周期开始时刻, 获取服务其关联 用户所需的总代价, 并将所述总代价反馈至宏站, 使得宏站根据所述总代价 做出优选控制策略, 所述总代价包括小站关联用户的平均时延信息;
指令接收执行单元 802, 用于接收宏站发送的优选控制策略对应的控制 指令, 对应开启或关闭小站。
优选的, 所述小站还包括:
第一子带分配单元 803, 用于在每个调度时隙开始时刻, 将各自的子带 资源分配给其关联用户。 本实施例提供的各个功能单元对应实现了实施例一中的三个步骤, 本实 施例考虑到了信道条件 (即用户平均时延) , 使得宏站得到优选控制策略, 与现有只考虑到业务量的 BS-DTX节能控制方案相比,能得到更好的优选控 制策略, 达到节能的目的。 优选方案中, 进一步考虑到业务的随机突发性, 将控制周期分成若干调度时隙, 在每个时隙中小站都进行子带分配, 将子带 资源分配给其关联的用户,这种双尺度的 BS-DTX控制及用户选择方法可以 在保证网络时延的基础上进一步降低网络功耗。 实施例十:
图 9示出了本发明第十实施例提供的小站的结构, 为了便于说明仅示出 了与本发明实施例相关的部分。
本实施例提供的小站包括:
总代价获取单元 91,用于在每个控制周期开始时刻,获取服务其关联用 户所需的总代价, 并将所述总代价反馈至宏站, 使得宏站根据所述总代价做 出优选控制策略, 所述总代价包括小站关联用户的平均时延信息;
指令接收执行单元 92,用于接收宏站发送的优选控制策略对应的控制指 令, 对应开启或关闭小站。
第一子带分配单元 93,用于在每个调度时隙开始时刻,将各自的子带资 源分配给其关联用户。
其中, 所述总代价获取单元 91包括:
带宽分配获取模块 911, 用于在每个控制周期开始时刻, 获取其关联用 户的最优带宽分配;
总代价获取模块 912, 用于获取在所述最优带宽分配条件下的总代价, 并将所述总代价反馈至宏站。
其中, 所述第一子带分配单元 93包括:
第一子带分配模块 931,用于在每个调度时隙开始时刻,在每个子带上, 选择小站关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户; 判断执行模块 932, 用于在控制周期结束时或者异构网络中关联用户集 改变时, 重新进入下一控制周期, 否则进入下一调度时隙。
本实施例提供的功能模块对应实现了实施例二中的各个步骤, 另一方 面, 本实施例在实施例九的基础上进一步公开了总代价获取单元 91和第一 子带分配单元 93的结构, 总代价获取模块 912在最优带宽分配条件下的总 代价, 系统功耗最小, 因此可以进一步在保证时延的前提下降低功耗。 实施例十一:
图 10示出了本发明第十一实施例提供的宏站的结构, 为了便于说明仅 示出了与本发明实施例相关的部分。
本实施例提供的宏站包括:
策略生成单元 101, 用于接收来自于各个小站的总代价, 并根据所述总 代价获取在不同控制策略下的整网总代价,将所述整网总代价最小时对应的 控制策略作为优选控制策略;
指令发送单元 102, 将所述优选控制策略对应的控制指令发送至小站, 以使小站根据所述控制指令对应开启或关闭。
优选的, 所述宏站还包括:
第二子带分配单元 103, 用于在每个调度时隙开始时刻, 将子带资源分 配给其关联用户以及关闭小站的关联用户。
本实施例提供的功能单元对应实现了实施例三中的两个步骤, 各个小 站在获取到各自的总代价后并反馈给宏站, 策略生成单元 121根据接收到的 总代价, 获取在不同控制策略下的整网总代价, 将所述整网总代价最小时对 应的控制策略作为优选控制策略,指令发送单元 122再将所述优选控制策略 对应的控制指令发送至小站, 以由小站根据所述控制指令对应开启或关闭。 实施例考虑到了信道条件(即用户平均时延),使得宏站得到优选控制策略, 在满足时延要求的条件下, 可以进一步降低系统能耗。 另外, 宏站和小站之 间的数据传输开销很少, 只有各个小站反馈的总代价, 以及宏站发出的用于 控制小站开启和关闭的控制指令。 这种分布式工作方式可以提高了网络效 率。 进一步的, 优选方案中, 考虑到业务的随机突发性, 将控制周期分成若 干调度时隙, 在每个时隙中小站都进行子带分配, 将子带资源分配给其关联 的用户,这种双尺度的 BS-DTX控制及用户选择方法可以在保证网络时延的 基础上进一步降低网络功耗。 实施例十二:
图 11 示出了本发明第十二实施例提供的宏站的结构, 为了便于说明仅 示出了与本发明实施例相关的部分。
本实施例提供的宏站包括:
策略生成单元 111, 用于接收来自于各个小站的总代价, 并根据所述总 代价获取在不同控制策略下的整网总代价,将所述整网总代价最小时对应的 控制策略作为优选控制策略;
指令发送单元 112, 将所述优选控制策略对应的控制指令发送至小站, 以使小站根据所述控制指令对应开启或关闭。
第二子带分配单元 113, 用于在每个调度时隙开始时刻, 将子带资源分 配给其关联用户以及关闭小站的关联用户。
其中, 所述策略生成单元 111包括:
总代价接收模块 1111, 用于接收来自于各个小站的总代价;
带宽分配模块 1112,用于遍历所有可能的控制策略,并获取宏站所服务 用户的最优带宽分配;
整网总代价获取模块 1113,用于根据所述总代价和最优带宽分配,获取 在各种控制策略下的整网总代价;
最优选择模块 1114,用于选择整网总代价最小时对应的控制策略作为优 选控制策略。
其中, 所述第二子带分配单元 113包括:
第二子带分配模块 1131,用于在每个调度时隙开始时刻,在每个子带上, 选择宏站关联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度 后等效速率最大的用户;
判断执行模块 1132,在控制周期结束时或者异构网络中关联用户集改变 时, 重新进入下一控制周期, 否则进入下一调度时隙。
本实施例提供的各个功能单元和功能模块对应实现了实施例四中的各 个步骤, 另一方面, 本实施例在实施例十三的基础上, 进一步公开了策略生 成单元 111和第二子带分配单元 113—种具体优选结构, 本实施例是在所有 可能的控制策略中选出整网总代价最 d、时的控制策略作为优选控制策略, 因 此本实施例选出的优选控制策略为最优控制策略,但是由于计算整网总代价 的复杂度与小站个数 B呈指数关系, 当小站的个数不大时, 比如 2个, 使用 本实施例方案较好。 实施例十三:
图 12示出了本发明第十三实施例提供的宏站的结构, 为了便于说明仅 示出了与本发明实施例相关的部分。
本实施例提供的宏站包括:
策略生成单元 121, 用于接收来自于各个小站的总代价, 并根据所述总 代价获取在不同控制策略下的整网总代价,将所述整网总代价最小时对应的 控制策略作为优选控制策略;
指令发送单元 122, 将所述优选控制策略对应的控制指令发送至小站, 以使小站根据所述控制指令对应开启或关闭。
第二子带分配单元 123, 用于在每个调度时隙开始时刻, 将子带资源分 配给其关联用户以及关闭小站的关联用户。 其中, 所述策略生成单元 121包括:
总代价接收模块 1211, 用于接收来自于各个小站的总代价;
次优选择模块 1212,用于获取开启所有小站时的整网总代价,顺次关闭 开启的小站中总代价最高的小站, 直至整网总代价不再降低或者所有小站全 部关闭, 将此时小站的控制策略作为优选控制策略。
其中, 所述第二子带分配单元 123包括:
第二子带分配模块 1231,用于在每个调度时隙开始时刻,在每个子带上, 选择其关联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度后 等效速率最大的用户;
判断执行模块 1232,在控制周期结束时或者异构网络中关联用户集改变 时, 重新进入下一控制周期, 否则进入下一调度时隙。
本实施例提供的各个功能单元和功能模块对应实现了实施例五中的各 个步骤, 另一方面, 本实施例在实施例十三的基础上, 进一步公开了策略生 成单元 141另一种具体优选结构, 本实施例将所有小站开启后, 顺次关闭剩 余小站中总代价最大的小站,因此本实施方案的复杂度与小站个数 B呈线性 关系, 因此, 本方案所选出的优选控制策略为次优控制策略, 在小站个数 B 较大时(比如 4、 8、 10等) , 选择本实施例结构的宏站更合适。 实施例十四:
本实施例提供了一种异构网络, 如图 13所示, 包括若干个如上述所述 的小站, 以及一个如上述实施例所述的宏站, 所述宏站和小站之间通过无线 或有线连接。
在物理层面上, 所述小站包括连接到数据总线上的第一处理器、 第一存 储器和第一通讯接口, 所述第一存储器用于存储上述各个实施例中小站所需 执行步骤所对应的程序, 具体由第一处理器来控制执行, 所述宏站包括连接 到数据总线上的第二处理器、 第二存储器和第二通讯接口, 所述第二存储器 用户存储上述各个实施例中宏站所需执行步骤所对应的程序, 具体由第二存 储器来控制执行, 小站和宏站之间通过第一通讯接口和第二通讯接口连接, 实现信号传输, 包括总代价、 优选控制策略对应的控制指令等。
本实施例提供的异构网络可以只釆用长周期的 BS-DTX控制方法,也可 以釆用长周期的 BS-DTX控制方法和短周期的用户选择方法结合的双尺度 节能控制方法, 并且双尺度控制方法优于所述长周期的 BS-DTX控制方法。 另外, 本实施例中小站和宏站之间传输的数据较少, 包括小站向宏站反馈的 总代价以及宏站向小站发送的控制指令, 这种分布式结构处理开销较小, 提 高了网络传输效率。 综上, 本发明技术方案提供的基站非连续发送节能控制方法是基于业务 量与信道条件自适应的 BS-DTX节能控制方案,特别是时延感知的双尺度的 BS-DTX控制及用户选择方法能够很好的降低网络功耗, 参照如图 14所示 的发送功率-平均时延的仿真图,仿真条件为:假设宏站的覆盖半径为 250m, 总带宽总 W=10 MHz , 可分为 10 个独立的子带, 每个带宽 1M。 宏站覆盖 下共有 2个小站, 即 B=2。 宏站和小站的发送功率分别为 。 = 46 dBm 、 ^ = 37 dBm。宏站和小站的大尺度路径损耗模型分别为 128.1+37.61oglO(R) 和 140.7+ 36.71oglO(R)(R单位 km)。 网络中的总用户数 K=16。 用户在网络 中均匀分布, 每个用户的活动概率为 0.5 (即有一半的概率是活动的) 。 用 户数据的到达服从集合 {2.5, 3, 3.5, 4, 4.5} packet/second下的均匀分布, 包大 小为 0.5 MByte。 图示中, 曲线 1为基于用户数的小站关断控制策略以及基 于子带信道增益最大化的时隙用户调度策略所对应的发送功率 -平均时延的 关系曲线, 曲线 2为基于用户数的小站关断控制策略以及基于比例公平的时 隙用户调度策略所对应的发送功率-平均时延的关系曲线, 曲线 3 为基于用 户到达率的小站关断控制策略以及基于子带信道增益最大化的时隙用户调 度策略所对应的发送功率-平均时延的关系曲线, 曲线 4 为基于用户到达率 的小站关断控制策略以及基于比例公平的时隙用户调度策略所对应的送功 率-平均时延的关系曲线, 曲线 5为双尺度的 BS-DTX控制及用户选择方法 对应的发送功率-平均时延的关系曲线, 从图示中可看出, 当针对相同的总 平均时延要求(比如 40ms ) , 本方案所提方法比传统方法节省网络平均总 发送功率约 6dB ( 4倍) 。
本领域普通技术人员可以理解, 实现上述实施例方法中的全部或部分步 骤是可以通过程序来指令相关的硬件来完成, 所述的程序可以在存储于一计 算机可读取存储介质中, 所述的存储介质, 如 ROM/ AM、 磁盘、 光盘等。 以上所述, 仅为本发明较佳的具体实施方式, 但本发明的保护范围并不 局限于此, 任何熟悉本技术领域的技术人员在本发明揭露的技术范围内, 可 轻易想到的变化或替换, 都应涵盖在本发明的保护范围之内。 因此, 本发明 的保护范围应该以权利要求的保护范围为准。

Claims

权利要求
1、 一种基站非连续发送节能控制方法, 其特征在于, 所述方法包括: 在每个控制周期开始时刻, 小站获取服务其关联用户所需的总代价, 并 将所述总代价反馈至宏站, 使得宏站根据所述总代价做出优选控制策略, 所 述总代价包括小站关联用户的平均时延信息;
接收宏站发送的优选控制策略对应的控制指令, 对应开启或关闭小站。
2、 如权利要求 1所述方法, 其特征在于, 所述控制周期划分为多个调 度时隙, 所述接收宏站发送的优选控制策略对应的控制指令, 对应开启或关 闭小站之后, 还包括:
在每个调度时隙开始时刻,开启的小站将各自的子带资源分配给其关联 用户。
3、 如权利要求 1或 2所述方法, 其特征在于, 在每个控制周期开始时 刻, 小站获取服务其关联用户所需的总代价, 并将所述总代价反馈至宏站, 具体包括:
在每个控制周期开始时刻, 小站获取其关联用户的最优带宽分配; 获取在所述最优带宽分配条件下的总代价, 并将所述总代价反馈至宏 站。
4、 如权利要求 2或 3所述方法, 其特征在于, 每个小站包含多个子带, 所述在每个调度时隙开始时刻, 开启的小站将各自的子带资源分配给其关联 用户, 具体包括:
在每个调度时隙开始时刻, 开启的小站在每个子带上, 选择其关联用户 中瞬时吞吐量加权队列长度后等效速率最大的用户;
在控制周期结束时或者异构网络中关联用户集改变时,重新进入下一控 制周期, 否则进入下一调度时隙。
5、 一种基站非连续发送节能控制方法, 其特征在于, 所述方法包括: 宏站接收来自于各个小站的总代价, 并根据所述总代价获取在不同控制 策略下的整网总代价,将所述整网总代价最小时对应的控制策略作为优选控 制策略;
将所述优选控制策略对应的控制指令发送至小站, 以使小站根据所述控 制指令对应开启或关闭。
6、 如权利要求 5所述方法, 其特征在于, 所述接收来自于各个小站的 总代价, 并根据所述总代价获取在不同控制策略下的整网总代价, 将所述整 网总代价最小时对应的控制策略作为优选控制策略, 具体包括:
宏站接收来自于各个小站的总代价;
遍历所有可能的控制策略, 并获取宏站所服务用户的最优带宽分配; 根据所述总代价和最优带宽分配, 获取在各种控制策略下的整网总代 价;
选择整网总代价最小时对应的控制策略作为优选控制策略。
7、 如权利要求 5所述方法, 其特征在于, 所述接收来自于各个小站的 总代价, 并根据所述总代价获取在不同控制策略下的整网总代价, 将所述整 网总代价最小时对应的控制策略作为优选控制策略, 具体包括:
宏站接收来自于各个小站的总代价;
获取开启所有 d、站时的整网总代价,顺次关闭开启的小站中总代价最高 的小站, 直至整网总代价不再降低或者所有小站全部关闭, 将此时小站的控 制策略作为优选控制策略。
8、 如权利要求 5-7任一项所述方法, 其特征在于, 将所述优选控制策 略对应的控制指令发送至小站, 以使小站根据所述控制指令对应开启或关闭 之后, 还包括:
在每个调度时隙开始时刻,宏站将子带资源分配给其关联用户以及关闭 小站的关联用户。
9、 如权利要求 8所述方法, 其特征在于, 所述宏站包含多个子带, 所 述在每个调度时隙开始时刻,宏站将子带资源分配给其关联用户以及关闭的 小站的关联用户, 具体包括:
在每个调度时隙开始时刻, 宏站在每个子带上, 选择其关联用户以及关 闭的小站的关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户; 在控制周期结束时或者异构网络中关联用户集改变时,重新进入下一控 制周期, 否则进入下一调度时隙。
10、 一种小站, 其特征在于, 所述小站包括:
总代价获取单元, 用于在每个控制周期开始时刻, 获取服务其关联用户 所需的总代价, 并将所述总代价反馈至宏站, 使得宏站根据所述总代价做出 优选控制策略, 所述总代价包括小站关联用户的平均时延信息;
指令接收执行单元, 用于接收宏站发送的优选控制策略对应的控制指 令, 对应开启或关闭小站。
11、 如权利要求 10所述小站, 其特征在于, 所述小站还包括: 第一子带分配单元, 用于在每个调度时隙开始时刻, 将各自的子带资源 分配给其关联用户。
12、 如权利要求 10或 11所述小站, 其特征在于, 所述总代价获取单元 包括:
带宽分配获取模块, 用于在每个控制周期开始时刻, 获取其关联用户的 最优带宽分配;
总代价获取模块, 用于获取在所述最优带宽分配条件下的总代价, 并将 所述总代价反馈至宏站。
13、 如权要求 11或 12所述小站, 其特征在于, 所述第一子带分配单元 包括:
第一子带分配模块, 用于在每个调度时隙开始时刻, 在每个子带上, 选 择小站关联用户中, 瞬时吞吐量加权队列长度后等效速率最大的用户; 判断执行模块,用于在控制周期结束时或者异构网络中关联用户集改变 时, 重新进入下一控制周期, 否则进入下一调度时隙。
14、 一种宏站, 其特征在于, 所述宏站包括:
策略生成单元, 用于接收来自于各个小站的总代价, 并根据所述总代价 获取在不同控制策略下的整网总代价, 将所述整网总代价最小时对应的控制 策略作为优选控制策略;
指令发送单元, 将所述优选控制策略对应的控制指令发送至小站, 以使 小站根据所述控制指令对应开启或关闭。
15、 如权利要求 14所述宏站, 其特征在于, 所述策略生成单元包括: 总代价接收模块, 用于接收来自于各个小站的总代价;
带宽分配模块, 用于遍历所有可能的控制策略, 并获取宏站所服务用户 的最优带宽分配;
整网总代价获取模块, 用于根据所述总代价和最优带宽分配, 获取在各 种控制策略下的整网总代价;
最优选择模块,用于选择整网总代价最小时对应的控制策略作为优选控 制策略。
16、 如权利要求 14所述宏站, 其特征在于, 所述策略生成单元包括: 总代价接收模块, 用于接收来自于各个小站的总代价;
次优选择模块, 用于获取开启所有小站时的整网总代价, 顺次关闭开启 的小站中总代价最高的小站, 直至整网总代价不再降低或者所有小站全部关 闭, 将此时小站的控制策略作为优选控制策略。
17、如权利要求 14-16任一项所述宏站, 其特征在于, 所述宏站还包括: 第二子带分配单元, 用于在每个调度时隙开始时刻, 将子带资源分配给 其关联用户以及关闭小站的关联用户。
18、 如权利要求 17所述宏站, 其特征在于, 所述第二子带分配单元包 括:
第二子带分配模块, 用于在每个调度时隙开始时刻, 在每个子带上, 选 择宏站关联用户以及关闭的小站的关联用户中, 瞬时吞吐量加权队列长度后 等效速率最大的用户;
判断执行模块, 在控制周期结束时或者异构网络中关联用户集改变时, 重新进入下一控制周期, 否则进入下一调度时隙。
19、 一种异构网络, 其特征在于, 所述异构网络包括如权利要求 10-13 任一项所述的小站, 以及如权利要求 14-18任一项所述的宏站, 所述宏站与 各个小站间有线或无线连接。
PCT/CN2014/077175 2013-05-09 2014-05-09 基站非连续发送节能控制方法、小站、宏站及异构网络 WO2014180350A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310169157.2 2013-05-09
CN201310169157.2A CN104144478B (zh) 2013-05-09 2013-05-09 基站非连续发送节能控制方法、小站、宏站及异构网络

Publications (1)

Publication Number Publication Date
WO2014180350A1 true WO2014180350A1 (zh) 2014-11-13

Family

ID=51853525

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/077175 WO2014180350A1 (zh) 2013-05-09 2014-05-09 基站非连续发送节能控制方法、小站、宏站及异构网络

Country Status (2)

Country Link
CN (1) CN104144478B (zh)
WO (1) WO2014180350A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107005935B (zh) * 2014-12-22 2020-05-26 北京花旺在线商贸有限公司 网络能效优化方法和装置
CN108882269B (zh) * 2018-05-21 2021-07-30 东南大学 结合缓存技术的超密集网络小站开关方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102413554A (zh) * 2011-12-23 2012-04-11 浙江大学 一种基于异构蜂窝无线网络的节能方法及其实现结构
CN102625320A (zh) * 2012-04-13 2012-08-01 北京邮电大学 一种家庭基站网络节能的方法
CN102892125A (zh) * 2012-10-19 2013-01-23 北京邮电大学 异构网络节能通信的干扰协调方法

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2076069A1 (en) * 2007-12-27 2009-07-01 Thomson Telecom Belgium Method and system for performing service admission control
JP5418353B2 (ja) * 2010-03-25 2014-02-19 ソニー株式会社 通信制御方法、および中小規模基地局
CN102111816A (zh) * 2011-03-22 2011-06-29 北京邮电大学 节能的毫微微网络系统及节能方法
CN102883330B (zh) * 2011-07-13 2017-05-31 株式会社Ntt都科摩 一种异构网络中小区间干扰协调的方法以及异构网络
CN102711182B (zh) * 2012-06-05 2016-09-21 中国联合网络通信集团有限公司 业务均衡与节能控制方法、实现方法及装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102413554A (zh) * 2011-12-23 2012-04-11 浙江大学 一种基于异构蜂窝无线网络的节能方法及其实现结构
CN102625320A (zh) * 2012-04-13 2012-08-01 北京邮电大学 一种家庭基站网络节能的方法
CN102892125A (zh) * 2012-10-19 2013-01-23 北京邮电大学 异构网络节能通信的干扰协调方法

Also Published As

Publication number Publication date
CN104144478B (zh) 2018-11-13
CN104144478A (zh) 2014-11-12

Similar Documents

Publication Publication Date Title
JP5664882B2 (ja) 通信システムにおけるユーザスケジューリング及び送信電力制御方法、及び装置
Wen et al. Energy efficient power allocation schemes for device-to-device (D2D) communication
CN101815347A (zh) 无线网络中通过中控设备调度休眠/唤醒周期的方法
CN104584627A (zh) 一种网络资源控制方法、装置和系统
Gao et al. An online energy allocation strategy for energy harvesting cognitive radio systems
Panahi et al. Energy efficiency analysis in cache-enabled D2D-aided heterogeneous cellular networks
Toumi et al. An adaptive Q-learning approach to power control for D2D communications
WO2014180350A1 (zh) 基站非连续发送节能控制方法、小站、宏站及异构网络
CN103024748B (zh) 基于控制信道的单天线无线传感器网络动态频谱接入方法
Chen et al. Resource allocation based on dynamic hybrid overlay/underlay for heterogeneous services of cognitive radio networks
WO2015035927A1 (zh) 一种信道配置方法及系统、接入控制器
Xu et al. Offloading data traffic via cognitive small cells with wireless powered user equipments
Wang et al. QoS-aware energy-saving mechanism for hybrid optical-wireless broadband access networks
Wymeersch et al. Multiple access control in wireless networks
Bartoli et al. AI based network and radio resource management in 5G HetNets
Chen et al. Achieving energy saving with QoS guarantee for WLAN using SDN
WO2014000447A1 (zh) 宏微小区之间的频宽调整方法及基站
Cheng et al. A game approach for dynamic resource allocation in cognitive radio networks
Li et al. User scheduling and slicing resource allocation in industrial Internet of Things
Huang et al. Analysis and design of energy efficient traffic transmission scheme based on user convergence behavior in wireless system
Liu et al. On optimizing energy efficiency in multi-radio multi-channel wireless networks
Huang et al. QoE based SDN heterogeneous LTE and WLAN multi-radio networks for multi-user access
Zhang et al. AP load balance strategy in face of high user density
Liu et al. A multi-channel sensing order optimization algorithm based on markov prediction
Alvarez et al. Device-to-device resource scheduling by distributed interference coordination

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14795209

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14795209

Country of ref document: EP

Kind code of ref document: A1