DESCRIPTION
Title of Invention
SLEEPING CORE NETWORK NODES FOR ENERGY SAVING IN 3G NETWORKS
Technical Field
[0001]
The present invention relates to energy saving for Core Network (CN) in 3rd Generation
(3G) networks.
Background Art
[0002]
Mobile coverage across the globe is growing at a rapid pace and now covers the
majority of the Earth's population. With more and more people accessing the network, the
power consumption created by the mobile operators may become a heavy burden to the
environment and needs consideration. Energy saving equipments and mechanisms are needed
to reduce the energy consumption to prevent further pollution to the environment.
[0003]
A mechanism for the efficient use of mobile CNs is provided in this invention. Since
the load upon mobile core networks moves according to human activities, it can easily be
speculated that the CN will have a lower load during night time. By making unused nodes
"sleep" during those hours, the total power consumption of the CN can be decreased.
Citation List
Non Patent Literature
[0004]
NPL 1: 3GPP TS 23.251, "Network Sharing; Architecture and functional description
(Release 10)", VI0.0.0, 2010-12, pages 7-8, clause 4.1
NPL 2: 3GPP TS 23.236, "Intra-domain connection of Radio Access Network (RAN) nodes
to multiple Core Network (CN) nodes (Release 10)", VIO.2.0, 2010-12, page 12, clause 4.5a.l
NPL 3: GISFI, GE-20100020, "Greening the mobile core network", 2010-12
Summary of Invention
Technical Problem
[0005]
Current standards do not provide means to shutdown nodes with active calls. The
lu-flex mechanism may provide means to shutdown nodes with no calls, but if there are long
time connections (e.g. video steaming) on the node, it may take time for that node to shutdown.
[0006]
This invention gives methods to solve this problem by using handover mechanisms to
transfer active calls to different nodes, enabling a shorter time to sleep.
[0007]
Note that as related technologies, Network Sharing is depicted in Non Patent Literature
(NPL) 1. Intra-domain connection of Radio Access Network nodes to multiple Core Network
nodes is depicted in NPL 2. Sleeping EPC is depicted in NPL 3.
Solution to Problem
[0008]
The mobile network consumes energy even in time periods when no users or fewer
users are using the network (e.g. night time). In order to cut unneeded power consumption, this
invention proposes that CN nodes "sleep (shuts down or runs in a low power state)". Here, we
define a sleeping CN node for the Packet Switched (PS) network, i.e. the sleeping SGSN
(Serving GPRS (General Packet Radio Service) Support Node).
[0009]
There are included novelty means for SGSNs sharing their resource information by
using operation and maintenance (O&M) or GPRS tunneling protocol's C-plane (GTP-C)
messages, for triggering Radio Network Controller (RNC)/Base Station Controller (BSC)
relocation by Power Down Notification messages, and for RNC/BSCs being able to identify
resources that "relocation required'V'relocation request" were sent/received belong to the same
call.
Advantageous Effects of Invention
[0010]
According to the present invention, the following effects can be achieved.
[0011]
1. Operators will be able to shutdown or lower the power of unused network nodes.
Since this will achieve lower power consumption, operators can realize an eco-friendly system.
[0012]
2. Operators will be able to scale the size of their CN dynamically according to the
amount of user's accessing network.
Brief Description of Drawings
[0013]
[Fig. 1]
Fig. 1 is a graph chart showing GHG emission of the Mobile communications sector in
2020.
[Fig- 2]
Fig. 2 is a block diagram showing Basic Configuration of a 3GPP Access Network.
[Fig. 3]
Fig. 3 is a block diagram showing Basic Configuration of LTE network.
[Fig. 4]
Fig. 4 is a block diagram showing Gateway Core Network configuration for Network
Sharing.
[Fig. 5]
Fig. 5 is a block diagram showing Sleeping core network node operation.
[Fig. 6]
Fig. 6 is a sequence diagram showing a Detach/Attach method which is applicable for
IDLE mode conditions.
[Fig. 7]
Fig. 7 is a sequence diagram showing a Re-routing method which is applicable for
CONNECTED mode conditions.
[Fig. 8]
Fig. 8 is a block diagram showing a configuration example of a core network node
according to an exemplary embodiment of the present invention.
Description of Embodiments
[0014]
Hereafter, an exemplary embodiment of a CN node according to the present invention,
and a mobile communication system to which this CN node is applied will be described with
reference to the drawings. In the 3GPP based network where Iu-flex is available, a group of
RAN (Radio Access Network) nodes inside a common pool-area can be connected to a group of
CN nodes. Every RAN node inside the same pool-area can be controlled by any CN node
looking over that pool-area. The CN nodes will not be fully operational at certain times of the
day, and the total amount of unused resources in the pool-area may exceed the capacity of a CN
node. In these cases a specific CN node may decide to pass its load to another CN node (an
active CN node) and then shutdown or turn down the power consumption (e.g. entering a
standby mode or a hibernation mode). Here, we focus on the PS network.
[0015]
[Sleeping SGSN]
Before falling asleep, a "sleepy" SGSN or the operator will check the amount of open
resources of other active SGSN nodes inside the same PS-pool-area (Step SI0 1 shown in each of
Figs. 6 and 7). This can be accomplished by using O&M messages or by defining messages
over the GTP-C protocol. In more detail, the SGSNs communicate with each other by use of
the O&M or GTP-C messages, thereby sharing information on open resources (hereinafter,
sometimes referred to as "resource information") between the SGSNs. Having done this, the
sleepy SGSN or the operator would know if there is an adequate amount of resources for the
sleepy SGSN to transfer its load and sleep (Step SI02 shown in each of Figs. 6 and 7).
[0016]
Once the sleepy SGSN decides to sleep, it would declare that it is going to sleep by
sending a power down notification to the connected RNC/BSC and SGSNs in the same pool
(Step SI03 shown in each of Figs. 6 and 7). This prevents the RNC/BSC from selecting the
sleepy SGSN for new connections and handovers.
[0017]
Next, the sleepy SGSN would transfer its load to other SGSNs in the same PS-pool-area.
There are two methods available: one would be to requesting the User Equipment (UE) to
detach/attach to the network (the RNC would not re-select the sleepy SGSN from the power
down notification) or re-routing the connection to a new SGSN.
[0018]

Fig. 6 describes the detach/attach method. This method is most applicable for IDLE
mode conditions. It can be also used for CONNECTED mode conditions but may cause
interruptions to active data transfers. Here, after the SGSN sends the power down notification,
it would send detach requests for each of its connections to the UE with "detach type: reattach
required". The detach process will be conducted between the related nodes (Step S104). After
that the UE will trigger an attach procedure. Here, the RNC/BSC will know that the sleeping
SGSN is no longer in use, will select an active SGSN for the attach request (Step SI05).
[0019]

Fig. 7 describes the re-routing method. This method is most applicable for
CONNECTED mode conditions for it partly uses handover methods to forward UE contexts
from sleepy SGSNs to active SGSNs. There will be no interactions between the UE and
RNC/BSC for there will be no change in the selected RNC/BSC. The power down notification
triggers the RNC/BSC to initiate a relocation procedure by sending a relocation message to the
sleepy SGSN (Step S201). The sleepy SGSN will select an active SGSN to switchover thereto
and sends a forward relocation request to the active SGSN (Step S202). The active SGSN will
send a relocation request to the RNC/BSC to establish a new Radio Access Bearer (RAB). The
RNC/BSC will identify that the relocation request is for one of its current resources and will
prepare for redirecting the communication (not all resources required for handovers will be
allocated) (Step S203). After the RAB is established, it will respond to the forward relocation
request and the sleepy SGSN will trigger the relocation command to the RNC/BSC (Step S204).
The RNC/BSC will send a relocation detect message to the active SGSN and also send a
relocation complete message later (Step S205). The active SGSN will send forward relocation
complete notification to the sleepy SGSN and will update the Packet Data Protocol (PDP)
context to redirect U-plane data through the active SGSN (Step S206). Finally, the sleepy
SGSN will be able to cut the Iu connection with the RNC/BSC (Step S207).
[0020]
After all connections has been removed from the sleepy SGSN, it can shutdown or turn
to a low power mode, making it a Sleeping SGSN (Step S208). When the Sleeping SGSN by a
given timer or manually by an operator, it would send power up notification messages by GTP or
O&M messages to show it is online again.
[0021]
Based on the above description, the following document will be submitted to Global
ICT Standardization Forum for India (GISFI).
[0022]
1. Abstract
Mobile communications is a fast growing sector not only in India but the whole world.
It can help bring down the Green House Gas (GHG) emission of other sectors, but will increase
the amount of emission of the mobile communication sector itself along with its growth. In
order to assure sustainable growth of this sector, means to alleviate GHG emissions must be
proposed. This document is an update of a previous proposal GE-20100020[1], and also
focuses on the core network part of mobile communications. We propose this document to be
accepted for the deliverable 2 of Green Energy activity, "Study on potential enhancements of ICT
(Information and Communications Technology)".
[0023]
2. Introduction
Utilizing ICT can be considered as an effective method to decrease the GHG emissions
of other sectors (as discussed in GE-20 10001 1 [2]). However, it is difficult for ICT to reduce
GHG emissions of the ICT sector itself. Therefore, further methods to decrease GHGs must be
considered for a "green" ICT. This document will focus on the core network of the mobile
communication part of ICT.
[0024]
According to SMART2020 [3], it is said that mobile communications will emit 201
Mega-tons of GHG into the earth's atmosphere by 2020 (see Fig. 1). Mobile communication is
consisted of mobile terminals, Radio Access Networks (RAN), and core networks. Even
though the majority of these emissions are from the RAN, the core network should take
measures to reduce its energy consumption and GHG emission.
[0025]
In this document, we first discuss how the current core network is structured for 3G and
LTE. Second, we discuss what elements are missing from the current networks. Finally, we
propose a high level solution that will efficiently reduce the amount of GHG emission.
[0026]
3. Current Analysis
Before we discuss how we can reduce the amount of GHG emission for the core
network, it is important to know the current architecture of the network. In this section, we will
take conventional 3rd Generation networks and Long Term Evolution (LTE) networks as
examples, and describe measures taken by the 3rd Generation Partnership Project (3GPP) to
reduce GHG emission.
[0027]
3.1. Current Status of the Core Networks
Fig. 2 describes the 3G network architecture [4] (LTE is left out of this picture). The
network architecture is consisted of the RAN and the core network. The RAN consists of
Radio Network Controllers (RNC) and NodeBs for 3 Both circuit switched (CS) and packet
switched (PS) calls go through the RAN and are directed to different nodes in the core network
(i.e. CS is served by the Mobile Switching Centre (MSC), PS is served by the Serving GPRS
Support Node (SGSN)). After CS or PS data enter the core network, they are processed and
sent to the appropriate destinations.
[0028]
Fig. 3 describes the LTE network [5]. Unlike conventional 3G networks, the LTE
network only handles PS services. The RAN consists of eNodeBs and their controls are
handled by Mobility Management Entities (MME). The Serving Gateway (S-GW) and P-GW
(Packet Data Network Gateway) handle the user plane traffic and the MMEs handle the control
packets. In the case where circuit switched services are needed, it is possible to fallback to the
3G network if coverage is available.
[0029]
3.2. Current Measures for Green Core Networks
One method being standardized in 3GPP that results in a Green Core Network would be
network sharing (TS 23.251 [6]). Network sharing enables different operators to share the same
RAN and nodes at the edge of the core network. This reduces over lapping equipment covering
the same area. It results in cutting deployment and operation costs for operators as it gives
green effect in having less equipment and power consumption. However, when there already is
an existing network, operators will end up throwing away their equipment. Thus this solution
may only be attractive to operators when applying it for deployment in new areas.
[0030]
Fig. 4 shows the Gateway Core Network (GCNW) configuration of a conventional
3GPP network. The SGSNs/MSCs/MMEs are shared between operators at the edge of their
core networks, and RNCs/eNodeBs are shared for RAN.
[0031]
4. Gap Analysis
Additional methods for saving power in the core network must also be considered.
First, we will discuss the problems that cannot be solved with the current solution. Then we
will elaborate on the potential solutions to cope with the problem.
[0032]
4.1. Areas needing additional saving
The current network sharing solution given by the 3GPP is very beneficial as a method
to cut equipment and operation costs. However, there are issues that still need to be handled,
creating more areas where power reduction is possible. There also would be demands from
standardization aspects to cope with the problem.
[0033]
One significant issue would be the power used in systems that are not being used
constantly throughout the day. For example, in commercial areas, the system is in its busiest
state during the day time, and the network load will start decreasing after business hours.
Residential areas will probably have a different peak hour, and most areas would have a
significantly lower load during the smaller hours of the day.
[0034]
Current systems do not change their processing capacity according to their load. They
are built based on the busy hour call attempts (BHCA) and are operated so that it can process its
maximum capacity during any time of the day. This leads to unneeded power consumption
making its countermeasure a requirement in achieving a green core network.
[0035]
4.2. High level Proposal
In order to solve the requirement discussed in the previous section, a new solution is
needed. The requirement in the last chapter was to cut power consumption when the system
has smaller loads compared with its maximum capacity. Some potential solutions for cutting
power consumption are: lowering CPU performance, putting the system in suspended/standby
mode, and powering off the system. These apply for parts of the network not being used (in
this case, the SGSN for 3G, and the MME and S-GW for LTE). However, these solutions may
create communication problems when the target node is accessed from other nodes in the
network. In order to power down nodes in the network, the negotiation between nodes would
be crucial.
[0036]
Fig. 5 presents a high level architecture of a system that enables core nodes to power
down (or to enter a lower powered state that does not handle actual transactions). First, when a
certain core node (SGSN, MME or S-GW) decides to "sleep" for various reasons (e.g. small
number of signalling traffic in midnight, lack of users, etc), it will send out "sleeping
declarations" to connected nodes. If the core node that receives the message is possible to
handle all the traffic covered by the sleeping node, the sleeping node handovers all of its
transactions to the receiving node. Thus, the users' traffic remains connected to the network via
a different node. Since now the sleeping node has no transactions left, it can be turned off or
into a low powered state. The sleeping node can be "waked up" by a given timer or a wake up
message.
[0037]
5. Conclusions
Throughout this document, we have discussed how the core network may be using
power inefficiently. As the mobile industry is a fast growing sector in India, there must be
measures to avoid increasing GHG emissions. We proposed a high level architecture that will
help eliminate unneeded energy consumption of the core network. We propose that this
document is accepted for deliverable 2 of Green Energy activity, "Study on potential
enhancements of ICT".
[0038]
6. References
[1] GISFI, GE1-20100020, Greening the mobile core network, Dec, 2010.
[2] GISFI, GE1-20100011, Making things Green with ICT, Sept, 2010.
[3] The Climate Group: Global e-Sustainability Initiative report, "Smart 2020 Enabling Low
Carbon Economy in the Information Age", 2008, .
[4] 3GPP, TS 23.002 v9.1.0, Network Architecture, Sept, 2009.
[5] 3GPP, TS 23.401 v9.6.0, General Packet Radio Service (GPRS) enhancements for Evolved
Universal Terrestrial Radio Access Network (E-UTRAN) access, Sept, 2010.
[6] 3GPP, TS 23.251 v 9.2.0, Network sharing; Architecture and functional description, Mar,
2010.
[0039]
Next, a configuration example of the CN node according to this exemplary embodiment
(i.e., the Sleeping SGSN shown in Figs. 6 and 7) will be described with reference to Fig. 8.
[0040]
As shown in Fig. 8, an SGSN 10 includes a sharing unit 11, a deciding unit 12, a
notifying unit 13, and requesting units 14 and 15.
[0041]
The sharing unit 11 shares the resource information between other SGSNs (i.e., active
SGSNs), for example, by using the above-mentioned O&M or GTP-C messages. The deciding
unit 12 decides whether or not to make the SGSN 10 sleep based on the resource information as
described above. The notifying unit 13 transmits the power down notification to the RNC/BSC
and the active SGSNs, when the SGSN 10 sleeps. The requesting unit 14 performs processing
according to the above-mentioned detach/attach method. Specifically, the requesting unit 14
transmits the detach request message with the detach type which indicates "reattach required" to
a UE attached to the SGSN 10 through the RNC/BSC. The requesting unit 15 performs
processing according to the above-mentioned re-routing method. Specifically, the requesting
unit 15 receives from the RNC/BSC the relocation message (Relocation Required message
shown at Step S201 in Fig. 7) as a response to the power down notification. Then, the
requesting unit 15 transmits the Forward Relocation Request message to one of the active
SGSNs.
[0042]
These units 11 to 15 can be configured by, for example, interfaces which communicate
with other SGSNs and the RNC/BSC, and a controller which controls these interfaces to execute
the processes shown in Figs. 6 and 7 or processes equivalent thereto.
[0043]
Note that the present invention is not limited to the above-mentioned exemplary
embodiment, and it is obvious that various modifications can be made by those of ordinary skill
in the art based on the recitation of the claims.
[0044]
This application is based upon and claims the benefit of priority from Japanese patent
application No. 2011-038779, filed on February 24, 2011, the disclosure of which is incorporated
herein in its entirety by reference.
[0045]
The whole or part of the exemplary embodiment disclosed above can be described as,
but not limited to, the following supplementary notes.
[0046]
(Supplementary note 1)
Methods to decide which core node should sleep
There are several ways to decide the sleeping node. An operator can decide the
sleeping node by checking SGSN resources via an O&M terminal through pre-defined O&M
messages. Otherwise, SGSN can check resources for themselves by checking resources
through defining GTP-C or O&M messages.
[0047]
(Supplementary note 2)
Power down notification and power up notification messages
Power down/up notification messages can be sent by defining messages on the RANAP
(Radio Access Network Application Part), GTP-C, and O&M messages. Power up messages
can be triggered by a given timer, disasters, overload messages from other SGSNs/RNCs, or
manually by an operator.
[0048]
(Supplementary note 3)
Power down preparation
In order for the sleepy node to power down efficiently, it will block new connections,
and only accept disconnect or handover related messages. Once all the user connections are
transferred or disconnected, the sleeping node will power down.
[0049]
(Supplementary note 4)
Re-use of Serving RNS (Radio Network Subsystem) relocation procedures
This procedure is different from serving RNS relocation for the source RNC and the
target RNC are the same RNC. Therefore, the RNC will be able to omit unneeded inter RNC
messages and prevent itself from securing unneeded resources. The RNC will be able to
specify the relocation messages belong to the same call by comparing the relocation required and
relocation request messages that are sent/received at the RNC.
Reference Signs List
[0050]
10 SGSN
11 SHARING UNIT
12 DECIDING UNIT
13 NOTIFYING UNIT
14, 15 REQUESTING UNIT
CLAIMS
[Claim 1]
A mobile communication system comprising a plurality of nodes that form a core
network,
wherein at least one of the plurality of nodes decides to sleep, based on resource
information shared between the plurality of nodes.
[Claim 2]
A node that forms a core network in a mobile communication system, the node
comprising:
first means for sharing resource information between one or more different nodes that
form the core network; and
second means for deciding whether or not to make the node itself sleep, based on the
resource information.
[Claim 3]
The node according to Claim 2, wherein the second means decides to make the node
itself sleep when the resource information indicates that any one of the different nodes has an
adequate mount of resources for transferring a load on the node itself.
[Claim 4]
The node according to Claim 2 or 3, further comprising:
third means for notifying the different nodes that the node itself sleeps, when the second
means decides to make the node itself sleep.
[Claim 5]
The node according to Clam 4, wherein the third means further notifies a node forming
a RAN (Radio Access Network) that the node itself sleeps.
[Claim 6]
The node according to Claim 5, further comprising:
fourth means for requesting a UE (User Equipment) that is attached to the node itself
through the RAN to re-attach to any one of the different nodes, prior to making the node itself
sleep.
[Claim 7]
The node according to Claim 6, wherein the fourth means performs the request by
transmitting a detach request message with a detach type indicating re-attach required.
[Claim 8]
The node according to any one of Claims 5 to 7, further comprising:
fifth means for requesting, upon receiving a response to the notification from the node
forming the RAN, any one of the different nodes to establish connection with the node forming
the RAN as a substitute for the node itself.
[Claim 9]
The node according to Claim 8,
wherein the response comprises a Relocation Required message used in handover
procedure,
wherein the fifth means performs the request by transmitting a Forward Relocation
Request message used in the handover procedure.
[Claim 10]
Amethod for energy saving in a mobile communication system, the method comprising:
deciding to cause at least one of a plurality of nodes that form a core network in the
mobile communication system to sleep, based on resource information shared between the
plurality of nodes.
[Claim 11]
Amethod of controlling a node that forms a core network in a mobile communication
system, the method comprising:
sharing resource information between one or more different nodes that form the core
network; and
deciding whether or not to make the node itself sleep, based on the resource
information.
[Claim 12]
The method according to Claim 11, wherein it is decided to make the node itself sleep
when the resource information indicates that any one of the different nodes has an adequate
mount of resources for transferring a load on the node itself.
[Claim 13]
The method according to Claim 11 or 12, further comprising:
notifying the different nodes that the node itself sleeps, when it is decided to make the
node itself sleep.
[Claim 14]
The method according to Clam 13, further comprising:
notifying a node forming a RAN (Radio Access Network) that the node itself sleeps.
[Claim 15]
The method according to Claim 14, further comprising:
requesting a UE (User Equipment) that is attached to the node itself through the RAN to
re-attach to any one of the different nodes, prior to making the node itself sleep.
[Claim 16]
The method according to Claim 15, wherein the request is performed by transmitting a
detach request message with a detach type indicating re-attach required.
[Claim 17]
The method according to any one of Claims 14 to 16, further comprising:
requesting, upon receiving a response to the notification from the node forming the
RAN, any one of the different nodes to establish connection with the node forming the RAN as a
substitute for the node itself.
[Claim 18]
The method according to Claim 17,
wherein a Relocation Required message used in handover procedure is received as the
response,
wherein the request is performed by transmitting a Forward Relocation Request message used in
the handover procedure.