Cross Reference to Related Applications
[001] The present application is related to co-pending United States Patent
Application entitled "ADMISSION CONTROL FOR SHARED LTE NETWORK"
(Chu et al.), filed concurrently herewith, the entire content of which is
incorporated herein by reference.
Field of the invention
[002] The invention is directed to packet switching communication networks,
and particularly to call admission control and preemption for multiple bit-rate
applications.
Background of the Invention
[003] Call Signaling
[004] In many packet switching applications, a user may require some minimal
quality of service (QoS) from the network so that it can work properly. In general,
an application will request the network to provide the required QoS by encoding
the QoS parameters in the call signaling messages.
[005] For example, consider Long Term Evolution (LTE) from 3rd Generation
Partnership Project (GPP), which is the next generation technology and
architecture for 3GPP.
[006] In LTE, user equipment (UE) will send and receive user packets through
its designated PDN-GW (Public data network-gateway). The PDN-GW will
forward the data packet from the UE to its intended destination. It also accepts
packets on the behalf of the UE, and then forwards the arriving packets to the
UE. (Note that there are other network elements such as the eNodeB and the SGW
between an UE and the PDN-GW)
[007] The logical connection between the UE and the PDN-GW is referred to as
the EPS (Evolved Packet System) "bearer" (or bearer for short). Associated with
each bearer is a QoS (Quality of Service) profile which governs how packets of
this bearer should be treated by the network. As a UE may have multiple
concurrent sessions, each with a different QoS needs, multiple bearers can be
set up between an UE and the PDN-GW, each supporting a different QoS.
Multiple sessions of the same QoS class can be mapped onto the same bearer.
[008] When a UE wants to set up an EPS bearer to the PDN-GW, it will send a
Bearer Resource Modification request message to the PDN-GW. The message
includes the following information (with other information):
• Service data flow templates which are used by the gateways to detect
packets that belongs to a data flow, upstream and downstream.
• QCI - QoS class identifier which defines the level of service required (such
as the packet loss rate and delay)
• Upstream (UL) and downstream (DL) maximum bit rate (MBR)
• Upstream and downstream guaranteed bit rate (GBR)
• Allocation and Retention Priority (ARP) parameter, which specifies the
priority of the call for call admission and, when in congestion, whether it
can preempt a lower priority call or be preempted by higher priority call.
The QCI, MBR, GBR, ARP are collectively referred to as the QoS parameters of
the bearer.
[009] There are other call signal protocols that provides QoS request. A well
known example is the Session Initiation Protocol (SIP) from the Internet
Engineering Task Force (IETF).
Call Admission and Call Preemption
[0010] When a user requests a specific QoS level for a call from the network, the
network may not have enough resources to support the call and maintain the
QoS level as requested. In these instances, the network would not admit the call
and the call is blocked. However, the incoming call may be of high priority and
network operator may want to admit the incoming high priority call at the expense
of the lower priority established calls. In these instances, preemption would take
place. Lower priority call(s) would be disconnected to free up enough resources
so the higher priority call would be admitted. The selection of which calls to be
preempted depends on the policy of network operator, as well as the capabilities
of the equipment.
Multiple Bit-Rate Applications
[001 1] Many applications can operate at multiple bit-rate levels. Consider
streaming video as an example. The nominal speed for a particular video
encoder may be 10 Mbps. An application may be willing to transmit at ¼ the
nominal speed by reducing the screen size (1/2 the height and 1/2 the width).
Therefore, the speed now is 2.5 Mps. By accepting B&W and forgoing color, a
reduction of another 50% could be achieved. Reducing the frame rate of a
streaming video application can be another method of reducing the bandwidth
required, as would using a different compression algorithm or controlling
parameters of a compression algorithm.
[0012] In view of the foregoing it would be desirable to provide an improved
system and method for call admission and preemption to take advantage of
flexibility of multiple bit-rate applications.
Summary of the Invention
[0013] For multiple bit-rate applications, it may not be efficient to preempt an
entire call to free up resources to order to allow a higher priority call. Depending
on the resources required by the incoming call, it may be sufficient for these calls
to operate at lower data rates to free up enough resources. Therefore,
embodiments of the present invention provide a method that allows these calls to
drop back speed during preemption to free up bandwidth instead of being
disconnected completely.
[0014] Embodiments of the present invention are directed to a method of
providing call admission control of multiple bit-rate applications in a digital
communication system. The method comprising steps of: identifying allowable bit
rates levels and associated priority levels of an incoming multiple bit-rate call;
determining if the incoming multiple bit-rate call can be admitted, at a first,
maximum requested bit-rate level and associated minimum priority level; and if
the incoming multiple bit-rate call can not be admitted, selecting a lower
allowable bit-rate level and associated priority level for the incoming call, and
repeating the determining step until the incoming multiple bit-rate call can be
admitted; admitting the multiple bit-rate call at the last-selected lower allowable
bit-rate level and associated priority level.
[0015] In some embodiments, the lower of the allowable bit-rate levels are
associated with a corresponding higher priority level.
[0016] In some embodiments, the admitting step further comprises sending a
signaling message to a User Equipment (UE) initiating the incoming call,
indicating the last-selected lower allowable bit-rate.
[0017] In some embodiments, the step of identifying allowable bit-rate levels and
associated priority levels of an incoming multiple bit-rate call further comprises
identifying individual call components which comprise the allowable bit-rate
levels, wherein each individual call component has an associated priority level.
[0018] In some embodiments, the determining step further comprises a step of
preempting components of existing multiple bit-rate calls having a lower priority
than the priority level of the incoming call.
[0019] In some embodiments, the step of preempting components comprises
making the call components inactive.
[0020] In some embodiments, if a call is dropped by the communication system,
the method further comprises steps of: Determining available bandwidth of the
communication system; compiling a candidate list of current inactive call
components of active multiple bit-rate calls along with their associated bit-rate
and priority level; sorting the candidate list by priority level, higher priority level
first; selecting from the candidate list, in order, inactive call components whose
bit-rate can be accommodated by the available bandwidth of the communication
system; activating the selected inactive call components in order to upgrade the
service of the active calls associated with the selected inactive call components.
[0021] In some embodiments, the digital communications system comprises a
packet data network.
[0022] In some embodiments, the digital communications system comprises a
Long Term Evolution (LTE) packet data network.
[0023] In some embodiments, the process occurs at an eNodeB of an LTE
network.
Brief Description of the drawings
[0024] Some embodiments of apparatus and/or methods in accordance with
embodiments of the present invention are now described, by way of example
only, and with reference to the accompanying drawings in which:
Figure 1 illustrates a flow chart for an embodiment of a call admission and
preemption process for a multiple bit-rate call; and
Figure 2 illustrates a flow chart for an embodiment of a process to
increase bit-rate for a multiple bit-rate call.
In the figures like features are denoted by like reference characters.
Detailed Description
[0025] There are many applications that can operate at different bandwidth
levels, which are referred to as multiple bit-rate applications, and associated calls
as multiple bit-rate calls. These applications and calls are sometimes referred to
as multi-rate; or variable bit-rate applications and calls, and for the purposes of
this document these term scan be considered interchangeable. When a new calls
arrives at a network, and there are insufficient resources to support the call,
preemption could be used to free up network resources to admit the incoming
call. However, in many circumstances, it would be inefficient, by default, to
disconnect a multiple bit-rate call completely to support a higher priority call. In
many situations, it may sufficient to request a multiple bit-rate call to operate at
lower level. By operating at a lower rate, it may be possible that sufficient
resources are released to support the incoming call.
[0026] It is impractical, when a new call arrives and preemption is needed, for
the eNodeB to request from the UEs whether the application is a multiple bit-rate
application. Therefore, embodiments of this invention specify that the following
information, encoded as new information elements (IE), be added to the bearer
set up messages (e.g., IP-CAN Session Modification, Bearer Set-up Request,
and Request Bearer Resource Modification in LTE):
• An indication that the application would support multiple QoS levels.
• The different QoS levels that this application is willing to support.
• For each QoS level, the priority level of the QoS level.
[0027] Consider the stream video described previously as an example. The UE,
UE A, may encode the QoS levels as follows:
• 10 Mps, priority 3 , drop back OK
• 5 Mbps, priority 2 , drop back OK
• 1.25 Mps, priority, no drop-back, non-preemptable
[0028] In LTE, the bearer set up signaling message contains the following
parameters:
• Service data flow templates which are used by the gateways to detect
packets that belongs to a data flow, upstream and downstream. These
templates are referred to as the TFT (traffic flow template).
• QCI - QoS class identifier
• Upstream (UL) and downstream (DL) maximum bit rate (MBR)
• Upstream and downstream guaranteed bit rate (GBR)
• The allocation and retention priority (ARP) parameter,
[0029] It is possible that the MBR and GBR (uplink or downlink) for a bearer to
have different values. In many LTE implementations, the eNodeB would map the
pair of values to a single value which is often referred to as the effective
bandwidth (uplink or downlink respectively) of the call. To simplify description, in
the subsequent paragraphs, the term bandwidth is used to mean the effective
bandwidth of the call (or the value pair). Usually, effective bandwidth is
expressed in bits per second, the same as regular bandwidth.
[0030] In many applications, the bandwidth required the uplink and downlink
directions are different. Also, the physical bandwidth of uplink and the downlink is
different. The downlink has higher capacity than the uplink. Therefore, the logic
for admission control and preemption are handled separately for the uplink and
the downlink. However, the processing logic is the same for both directions. In
the subsequent description, the invention is described for a single direction. In
actual implementation, two identical processes will be implemented in the
eNodeB to managing call admission and preemption for uplink and downlink
respectively.
[0031] The following notation will be used to describe a multiple bit-rate call with
N levels:
• (P1 , B 1 ; P2, B2; PN, BN), where Pi is the priority level of level i , with
P 1 being the highest priority;
• Bi is the bandwidth required for the ith level, with B 1 < B2; < BN.
A normal call can be considered logically as a multiple bit-rate call with one level.
Call Admission for Multiple bit-rate Call
[0032] Figure 1 is an illustration of an embodiment of a call admission process
for a multiple bit-rate call. At step 100, a multiple bit-rate call arrives with N levels
(P1 , B 1 ; P2, B2; PN, BN). At step 110 , the process sets k = N. At step 120,
the process determines whether a call with priority Pk and with bandwidth Bk can
be admitted. If so, the process proceeds to step 130, where the incoming call is
admitted at level k . A signaling message is sent to the UE indicating the call is
admitted at level k . Note that it is possible that, by admitting this call, other calls
(regular calls as well as other multiple bit-rate calls) may be preempted.
Preemption for multiple bit-rate calls may mean that the call operates at a lower
bit rate. After step 130, the process ends.
[0033] If at step 120, the process determines that the call, at the current selected
level, can not be admitted, the process proceeds to step 140, where the process
sets k = k - 1. Step 150 checks whether k = 0 . If k > 0 , the process goes back to
step 120 to process the next level of the incoming multiple bit-rate call. If at step
150, the process determines that k=0, this means that the algorithm has
processed all the levels of the incoming multiple bit-rate call without success, and
the process proceeds to step 160, where the call is not admitted. A signaling
message is sent back to the UE to indicate the call is not admitted.
[0034] Once a multiple bit-rate call with N levels is admitted, it is considered to
have N components:
Component 1 (C1 ) has priority P 1 and bandwidth D 1 = B 1,
Component 2 (C2) has priority 1 and bandwidth D2 = B2 - B 1,
Component N (CN) has priority 1 and bandwidth DN = BN - Bn-1 .
If the call is admitted at level k , then components C 1 to Ck are active, with a total
bandwidth of Bk which is the total of the incremental bandwidths of each active
component (Bk = Dk + Dk-1 + ... + D 1) . Components Ck+1 to CN will be inactive.
Preemption of Multiple bit-rate Call
[0035] When a new call arrives, preemption may be necessary. In the
preemption process, each component of a multiple bit-rate call is treated as a
separate call. Thus, if a multiple bit-rate call has three active components C 1, C,
and C3, they are treated as three separate calls. Since C 1 has higher priority
than C2 and C3, C 1 can not be preempted unless C2 and C3, having lower
priority, are also preempted. At the end of the preemption process, some or all of
the active components of a multiple bit-rate call may be preempted. If some of
the components are preempted to level k , the preempted components will be Ck,
Ck+1 , Cn. In this case, the eNodeB sends a mid-call signaling message to
the UE indicating that multiple bit-rate call should operate at level (k-1 ) , i.e., at
priority Pk-1 and bandwidth Ck-1 . Components Ck, Ck+1 , Cn will become
inactive. If all the active components of a multiple bit-rate call are preempted, the
call will be dropped.
Example
[0036] The following example illustrates an embodiment of a call admission
process of multiple bit-rate calls. Multiple bit-rate call 1 has 4 levels with the
following characteristics: (P1 , 2Mbps; P2, 4Mbps; P3, 6 Mbps; P4, 8 Mbps).
Assume the capacity of the link is 10 Mbps. Assume that the system has no other
calls when call 1 arrives. Call 1 is then admitted at level 4 because at level 4 , the
call requires 8 Mbps and the system has 10 Mbps. After admission, call 1 has 4
components (C1 , C2, C3, C4), each having 2 Mbps of bandwidth.
[0037] Assume that a second multiple bit-rate call, call 2 , arrives, with identical
characteristics as call 1. The admission control process at eNodeB would be as
follows:
• First, the call admission process evaluates call 2 at level 4 , the lowest
level of call 2 . At level 4 , call 2 requires 8 Mbps, but the system currently
only has 2 Mbps not in use. Therefore, the call can not be admitted
without preemption. However, Call 2 being of priority P4 can not preempt
any of the components of call 1. Therefore, call 2 can not be admitted at
level 4 .
• Then, the eNodeB would determine whether call 2 can be admitted at
level 3 . The required bandwidth is 6 Mbps. The available bandwidth is 2
Mbps. Therefore, the call can not be admitted without preemption. Call 2 ,
at level 3 , can preempt component 4 of call 1 which has a lower priority
than call 2 at level 3 . However, this would only free up 2 Mbps, resulting in
4 Mbps of available bandwidth, which would not sufficient. Therefore, call
2 can not be admitted at level 3 .
• Next, eNodeB will determine whether call 2 can be admitted at level 2 . At
level 2 , call 2 requires 4 Mbps. Pre-emption would be necessary to admit
this call. Call 2 at level 2 can preempt component 3 and component 4 of
call 1, thus freeing up 4 Mbps of bandwidth (resulting in 6 Mbps of
available bandwidth). However, since only 4 Mbps is needed, the system
will only preempt component 4 of call 1. Therefore, call 2 can be admitted
at level 2 by preempting components 4 of call 1.
[0038] The eNodeB will send signaling messages to the UEs of call 2 indicating
that the call is admitted at level 2 (components 1 and 2 active, components 3 and
4 inactive). It also will send mid-call signaling messages to UEs of call 1,
indicating that the UEs should now operate at level 3 (components 1, 2 and 3
active, components 4 inactive). Note than the system will remember that both
calls have 4 components at 2 Mbps each. Thus, in this example, after call 2 is
admitted with the required preemption, call 1 will have the first 3 components
active wile the last component is inactive. Call 2 will have the first two
components active while the last two components are inactive.
Automatic Up-speed
[0039] As calls depart the network, network resources will be released. Under
these circumstances, the system may want to upgrade the bandwidth of a
multiple bit-rate call which is not operating at maximum requested bandwidth (i.e.
some components of the call is inactive).
[0040] When a call departs the system, the eNodeB can automatically check
whether some of the multiple bit-rate calls could be upgraded. An embodiment of
such as procedure is illustrated by the flowchart in Figure 2 . At step 200, the
process initializes the following parameters as follows:
• V = current available bandwidth of the system
• The selected list which represents current inactive multiple bit-rate call
components that could be upgraded (i.e. make active). This list is
initialized as an empty list.
• The candidate list: This is the list of currently inactive components of
multiple bit-rate calls arrange in the following order:
o Components with higher priority are placed before components with
lower priority.
o Among components with the same priority, the order is arranged
according to the time of call arrival; the ones that arrive first are
placed before those which arrive latter.
[0041] At step 2 10 , the process checks if the candidate list is empty. If the
candidate list is not empty, the process continues to step 220 where it
determines the bandwidth requirements (F) of the first member of the candidate
list. At step 230 the process checks if F is less than or equal to V, the current
available bandwidth. If so, this means that the first member can be upgraded (i.e.
make active), and the process continues to step 240 where the first member is
added to the selected list and removed from the candidate list. At step 250, V is
updated to V- F, and process goes back to step 2 10 to evaluate the next
member.
[0042] If at step 230, it is determined that F > V, this means that the first member
can not be upgraded, in which case the process proceeds to step 260 where the
process removes this member and all other components in the candidate list that
belongs to the same call, from the candidate list. The process then goes back to
step 2 10 to evaluate the next member.
[0043] If at step 2 10 it is determined that the candidate list is empty, the process
continues to step 270, where the process checks whether the selected list is
empty. If the selected list is empty, no upgrade is possible and the algorithm
terminates.
[0044] If at step 270 it is determined that the selected list is not empty, the
process continues to step 280, where the process identifies components in the
selected list that belong to the same call, and merges them together per call. At
step 290, the process identifies the multiple bit-rate calls that can be upgraded
and the appropriate level of upgrade. At step 300, the process then executes the
upgrade by:
• Sending signaling messages to the appropriate UE indicating to them that
the bearer is upgrade to allow more bandwidth.
• The status of the components that are upgraded is changed to active.
• Update other pertinent parameters such current available bandwidth of the
system.
The algorithm then terminates.
[0045] The process of Figure 2 can be executed at any time, although right after
a call departure is a natural and useful point of time as this is when network
resources are released. However, for busy systems, calls may depart frequently.
To avoid frequent execution of the above algorithm, in some embodiments, the
eNodeB only executes the above algorithm after a predetermined time interval
from the last execution of the process. An example of a predetermined time
interval is 30 seconds.
[0046] In another embodiment, the eNodeB may maintain a parameter which
represents the minimum bandwidth of all the inactive components. The above
algorithm will only be executed if the current available bandwidth exceeds this
minimum. This would reduce the amount of processing needed to support
automatic upgrade.
Variations
[0047] In this document, LTE is used as the context to describe embodiments of
the invention. However, it should be apparent to persons skilled in the art that
the invention is applicable to all digital communications systems that support call
admission and preemption based on network resource availability.
[0048] The scope of the present invention encompasses many other
embodiments. For example, the call components with the same priority in the
candidate list may be arranged differently (e.g., component with the least
bandwidth is placed first, instead of components with an earlier of time of arrival).
It is appreciated that these variations although not explicitly described or shown
herein, embody the principles of the invention, and are included within its spirit
and scope.
[0049] A person of skill in the art would readily recognize that steps of various
above-described methods can be performed by programmed computers. Herein,
some embodiments are also intended to cover program storage devices, e.g.,
digital data storage media, which are machine or computer-readable and encode
machine-executable or computer-executable programs of instructions, wherein
said instructions perform some or all of the steps of said above-described
methods. The program storage devices may be, e.g., digital memories, magnetic
storage media such as a magnetic disks and magnetic tapes, hard drives, or
optically readable digital data storage media. The embodiments are also
intended to cover computers programmed to perform said steps of the abovedescribed
methods.
[0050] The description and drawings merely illustrate the principles of the
invention. It will thus be appreciated that those skilled in the art will be able to
devise various arrangements that, although not explicitly described or shown
herein, embody the principles of the invention and are included within its spirit
and scope. Furthermore, all examples recited herein are principally intended
expressly to be only for pedagogical purposes to aid the reader in understanding
the principles of the invention and the concepts contributed by the inventor(s) to
furthering the art, and are to be construed as being without limitation to such
specifically recited examples and conditions. Moreover, all statements herein
reciting principles, aspects, and embodiments of the invention, as well as specific
examples thereof, are intended to encompass equivalents thereof.
[0051] The functions of the various elements shown in the figures, including any
functional blocks labeled as "processors", may be provided through the use of
dedicated hardware as well as hardware capable of executing software in
association with appropriate software. When provided by a processor, the
functions may be provided by a single dedicated processor, by a single shared
processor, or by a plurality of individual processors, some of which may be
shared. Moreover, explicit use of the term "processor" or "controller" should not
be construed to refer exclusively to hardware capable of executing software, and
may implicitly include, without limitation, digital signal processor (DSP) hardware,
network processor, application specific integrated circuit (ASIC), field
programmable gate array (FPGA), read only memory (ROM) for storing software,
random access memory (RAM), and non volatile storage. Other hardware,
conventional and/or custom, may also be included. Similarly, any switches shown
in the figures are conceptual only. Their function may be carried out through the
operation of program logic, through dedicated logic, through the interaction of
program control and dedicated logic, or even manually, the particular technique
being selectable by the implementer as more specifically understood from the
context.
[0052] It should be appreciated by those skilled in the art that any block
diagrams herein represent conceptual views of illustrative circuitry embodying the
principles of the invention. Similarly, it will be appreciated that any flow charts,
flow diagrams, state transition diagrams, pseudo code, and the like represent
various processes which may be substantially represented in computer readable
medium and so executed by a computer or processor, whether or not such
computer or processor is explicitly shown.
[0053] Numerous modifications, variations and adaptations may be made to the
embodiment of the invention described above without departing from the scope
of the invention, which is defined in the claims.
CLAIMS:
1. A method of providing call admission control of multiple bit-rate
applications in a digital communication system, the method comprising steps
of:
identifying allowable bit-rate levels and associated priority levels of an
incoming multiple bit-rate call;
determining if said incoming multiple bit-rate call can be admitted at a
first, maximum requested bit-rate level and an associated minimum priority
level, and, if said incoming multiple bit-rate call cannot be admitted;
selecting a lower allowable bit-rate level and associated priority level
for said incoming call;
repeating said determining step until said incoming multiple bit-rate
call can be admitted; and
admitting said multiple bit-rate call at the last-selected lower
allowable bit-rate level and associated priority level.
2. The method of claim 1, wherein lower allowable bit-rate levels are
associated with a respectively higher priority levels.
3. The method of claim 1, further comprising:
sending a signaling message to a User Equipment (UE);
initiating said incoming call; and
indicating a last-selected lower allowable bit-rate.
4. The method of claim 2, further comprising:
identifying individual call components which comprise said allowable
bit-rate levels, wherein each individual call component has an associated
priority level.
5. The method of claim 4, further comprising:
preempting components of existing multiple bit-rate calls having a
lower priority than the priority level of said incoming call.
6. The method of claim 5, further comprising:
making said call components inactive.
7. The method of claim 6, wherein if a call is dropped by said communication
system, the method further comprises:
determining available bandwidth of said communication system;
compiling a candidate list of current inactive call components of active
multiple bit-rate calls along with their associated bit-rate and priority level;
sorting said candidate list by priority level, higher priority level first;
selecting from said candidate list, in order, inactive call components
whose bit-rate can be accommodated by said available bandwidth of said
communication system; and
activating the selected inactive call components in order to upgrade the
service of the active calls associated with said selected inactive call
components.
8. The method of claim 1, wherein said digital communications system
comprises a packet data network.
9. The method of claim 8, wherein said digital communications system
comprises a Long Term Evolution (LTE) packet data network.
10. The method of claim 9, wherein said process occurs at an eNodeB of an
LTE network.