U.S. patent application number 12/772702 was filed with the patent office on 2010-08-26 for metering in packet-based telephony networks.
This patent application is currently assigned to Nortel Networks Limited. Invention is credited to William Bishop, Kevin Boyle, II, David P. Ress, Costin Strugaru.
Application Number | 20100215038 12/772702 |
Document ID | / |
Family ID | 34523248 |
Filed Date | 2010-08-26 |
United States Patent
Application |
20100215038 |
Kind Code |
A1 |
Ress; David P. ; et
al. |
August 26, 2010 |
METERING IN PACKET-BASED TELEPHONY NETWORKS
Abstract
One embodiment of the present invention facilitates efficient
metering in a packet network environment by providing a single
metering message, which contains sufficient information to provide
the complete call tariff model for a particular call. The media
gateway receiving the message can analyze the information provided
in the message to determine how to provide metering pulses for all
phases of the call, as well as any one-time charges, such as setup
and add-on charges. Another embodiment of the invention provides a
way for handling fractional pulse counts in an efficient manner.
Yet another embodiment facilitates the handling of situations where
charge intervals do not divide evenly into the phase duration of
the phase associated with the call. In still another embodiment,
the amount of information necessary to deliver the parameters of
the call tariff model is minimized to reduce the overhead necessary
for facilitating the metering process.
Inventors: |
Ress; David P.; (Cary,
NC) ; Boyle, II; Kevin; (Apex, NC) ; Bishop;
William; (Raleigh, NC) ; Strugaru; Costin;
(Immenstaad, DE) |
Correspondence
Address: |
WITHROW & TERRANOVA, P.L.L.C.
100 REGENCY FOREST DRIVE, SUITE 160
CARY
NC
27518
US
|
Assignee: |
Nortel Networks Limited
St. Laurent
CA
|
Family ID: |
34523248 |
Appl. No.: |
12/772702 |
Filed: |
May 3, 2010 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10742324 |
Dec 19, 2003 |
7715537 |
|
|
12772702 |
|
|
|
|
Current U.S.
Class: |
370/352 |
Current CPC
Class: |
H04M 15/56 20130101;
H04M 15/43 20130101; H04M 15/41 20130101; H04M 2215/202 20130101;
H04M 15/00 20130101; H04M 15/8264 20130101; H04M 15/81 20130101;
H04M 2215/0152 20130101; H04M 2215/0164 20130101; H04M 15/80
20130101; H04M 2215/7866 20130101; H04M 2215/0112 20130101 |
Class at
Publication: |
370/352 |
International
Class: |
H04L 12/66 20060101
H04L012/66 |
Claims
1. A method for providing metering from a gateway in a packet
network comprising: a) generating a message comprising a complete
call tariff model for controlling all metering in association with
a call; and b) sending the message over a packet network to a media
gateway supporting the call.
21. A system for providing metering from a gateway in a packet
network comprising: a) a packet interface to facilitate
communication with a media gateway over a packet network; and b) a
control system associated with the packet interface and adapted to:
i) generate a message comprising a complete call tariff model for
controlling all metering in association with a call; and ii) send
the message over the packet network to the media gateway supporting
the call.
Description
[0001] This application is a divisional application of U.S. patent
application Ser. No. 10,742,324, filed Dec. 19, 2003, now issued as
U.S. Pat. No. ______, the disclosure of which is incorporated
herein by reference in its entirety.
FIELD OF THE INVENTION
[0002] The present invention relates to communications, and in
particular to providing metering for communication services in a
packet-based telephony network.
BACKGROUND OF THE INVENTION
[0003] Classic Time Division Multiplex (TDM)-based telephony
switches allow service providers to provision tariff, or billing,
information. These tariffs apply to any number of applications,
such as local calls, long distance calls, operator-assisted calls,
service calls, Integrated Services Digital Network (ISDN) calls,
and other types of calls. Through a capability referred to as
hardware metering, tariff information may be sent to a user's
telephone or a device associated therewith, which processes the
information to provide charge information to the user or help
account for the call. For example, the information may be used to
alert a pay phone that additional money is necessary to establish
or continue the call. The tariff information is generally provided
to the telephone or to an associated device as a series of pulses,
which correspond to the tariff being incurred. The pulses are used
to calculate the tariff and control how to process the call or
control information provided to the user.
[0004] Generally, the call tariff models divide a call into a
number of phases. Each phase represents a different tariff rate.
The duration of a phase may be explicitly defined or may be
infinite. An infinitely long phase duration means the tariff
defined for the phase applies for the remainder of the call. The
tariff itself is defined by a charge rate, expressed in an amount
of currency per unit of time, for each phase and optionally a
charge interval. The charge interval allows the provider to specify
how frequently charges are assessed during the phase. Call tariff
models may also include one-time charges, such as a call setup
charge, or a mid-call add-on charge, which may be applied to charge
for events, such as invocation of a supplementary feature during
the call. In addition to the tariff rate information, the service
provider must be able to specify the conversion rate for mapping
currency charges into pulse counts. The result of this conversion
is referred to as a tariffed pulse rate (TPR).
[0005] Generally, the TPR is specified in units of pulses per
second, and each phase has a specific TPR. In a traditional s the
likelihood of inaccuracy in the metering process.
[0006] Current metering processPublic Switched Telephone Network
(PSTN), these pulses are provided over a dedicated circuit-switched
connection. As the PSTN is replaced by packet-based network
architectures, this process of providing pulses in a metering
process faces new challenges. Given the inherent nature of
packet-based networks, errors in metering accuracy are exacerbated
due to latency caused by routing equipment in the packet network,
as well as the potentially greater distances over which metering
information must be conveyed. Further, the potential for packet
loss and the subsequent need for packet retransmission increases in
packet networks rely heavily on the use of different messages to
handle different aspects of the call, such as the setup and add-on
events, as well as each phase of the call. The messaging associated
with the metering process impacts available bandwidth and
processing resources. Further, the complexity of the metering
process often requires an effective fractional pulse rate over a
given phase. At this point, there is no effective mechanism for
handling fractional pulse rates in a packet-based environment.
Accordingly, there is a need for an efficient metering process in a
packet environment that is capable of addressing one or more of the
limitations set forth above.
SUMMARY OF THE INVENTION
[0007] The present invention facilitates efficient metering in a
packet network environment by providing a single metering message,
which contains sufficient information to provide the complete call
tariff model for a particular call. The media gateway receiving the
message can analyze the information provided in the message to
determine how to provide metering pulses for all phases of the
call, as well as any one-time charges, such as setup and add-on
charges. The setup charges are associated with initiating the call
and the add-on charges are other charges related to activating
features during the call. One embodiment of the invention provides
an efficient way for handling fractional pulse counts in an
efficient manner. Another embodiment facilitates the handling of
situations where charge intervals do not divide evenly into the
phase durations of the phases associated with the call. In yet
another embodiment, the amount of information necessary to deliver
the parameters of the call tariff model is minimized to reduce the
overhead necessary for facilitating the metering process.
[0008] Those skilled in the art will appreciate the scope of the
present invention and realize additional aspects thereof after
reading the following detailed description of the preferred
embodiments in association with the accompanying drawing
figures.
BRIEF DESCRIPTION OF THE DRAWINGS FIGURES
[0009] The accompanying drawing figures incorporated in and forming
a part of this specification illustrate several aspects of the
invention, and together with the description serve to explain the
principles of the invention.
[0010] FIG. 1 is a block representation of a communication
environment according to one embodiment of the present
invention.
[0011] FIG. 2 illustrates an exemplary pulse configuration
according to one embodiment of the present invention.
[0012] FIG. 3 illustrates multiple phases for a metering process
according to one embodiment of the present invention.
[0013] FIG. 4 illustrates a metering process wherein charge
intervals do not divide evenly into a particular phase.
[0014] FIG. 5 illustrates the use of pulse windows to address
situations wherein the charge intervals do not divide evenly into a
phase according to one embodiment of the present invention.
[0015] FIG. 6 illustrates the distribution of pulses for a partial
charge interval when the charge intervals do not divide evenly into
a phase.
[0016] FIG. 7 illustrates the distribution of pulses in a remaining
portion of a charge interval according to an alternative embodiment
of the present invention.
[0017] FIG. 8 is a block representation of a media gateway
controller according to one embodiment of the present
invention.
[0018] FIG. 9 is a block representation of a media gateway
according to one embodiment of the present invention.
[0019] FIG. 10 is a block representation of a metering device
according to one embodiment of the present invention.
[0020] FIG. 11 is a block representation of a telephony device
having a metering function according to one embodiment of the
present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0021] The embodiments set forth below represent the necessary
information to enable those skilled in the art to practice the
invention and illustrate the best mode of practicing the invention.
Upon reading the following description in light of the accompanying
drawing figures, those skilled in the art will understand the
concepts of the invention and will recognize applications of these
concepts not particularly addressed herein. It should be understood
that these concepts and applications fall within the scope of the
disclosure and the accompanying claims.
[0022] The present invention provides an effective and efficient
technique for facilitating a metering process for a telephony
endpoint, which is supported by a packet network via a media
gateway. Metering is the process of applying signals, preferably in
the form of pulses, over a telephone line to a telephony endpoint.
The signals are received by the telephony endpoint or equipment
associated therewith to provide information bearing on charges
associated with the call. The information may be used to advise a
user of the charges for the call, or to instruct the telephony
endpoint or the device associated therewith to collect fees from
the user or payment associated with a call. The charges associated
with a call are generally referred to as a tariff, and the present
invention allows a complete description of the tariff through the
various phases of a call to be provided to the media gateway in a
single message, wherein the media gateway can take the information
in the message and provide all of the necessary pulses for the
entire call. The pulses are provided as necessary for each phase of
the call, as well as for any setup or additional charges incurred
at the beginning of or during the call. Prior to delving into the
details of the present invention, an overview of an exemplary
communication environment 10, which incorporates the concepts of
the present invention, is illustrated in FIG. 1.
[0023] The communication environment 10 is centered around a packet
network 12 and a Public Switched Telephone Network (PSTN) 14, which
support telephony communications between any number of telephony
endpoints 16(A), 16(B), such as telephones or computing devices
enabled to support telephony communications. When calls to or from
the telephony endpoints 16(A), 16(B) require metering, metering may
be facilitated by a separate metering device 18, which is
associated with a standard telephony endpoint 16(A), or a more
sophisticated telephony endpoint 16(B) may be configured to include
a metering function 20. In either case, the metering device 18 or
metering function 20 will operate to receive pulses from entities
in the PSTN 14 or from a supporting media gateway (MG) 22, which
facilitates communications over the packet network 12. The media
gateway 22 supports telephony circuits, which connect directly to a
telephony endpoint 16(B) or to a metering device 18, which supports
a traditional telephony endpoint 16(A).
[0024] In traditional metering systems, the PSTN 14 will send
metering pulses to the telephony endpoint 16(B) or metering device
18 over a dedicated, circuit-switched connection. In a packet
environment, metering pulses are provided by the media gateway 22
to either the metering device 18 or the telephony endpoint 16(B).
The present invention is directed to controlling the media gateway
22 to efficiently provide the metering pulses to the metering
device 18 or the telephony endpoint 16(B).
[0025] Continuing with FIG. 1, communications between the PSTN 14
and the packet network 12 are facilitated by a trunking media
gateway 24, which will convert circuit-switched communications to
packet-based communications, and vice versa, between trunks going
to the PSTN 14 and the packet network 12. Control of the media
gateways 22 and the trunking media gateway 24 is provided by a
media gateway controller (MGC) 26, which will also play a role in
the metering process. The media gateway controller 26 is
essentially a session or call control entity, and may interact with
other signaling entities on a signaling network 28, such as a
Signaling Systems 7 (SS7) network, directly or indirectly via a
signaling gateway (SG) 30 providing the necessary translations for
call or session control signaling between the signaling network 28
and the media gateway controller 26.
[0026] As noted, to facilitate interworking of PSTN and packet
communications, there are two primary elements: the media gateways
22, 24 and the media gateway controller 26. The media gateways 22,
24 provide the actual interface between the packet network 12 and
the telephony endpoints 16(A), 16(B), directly or indirectly via
the metering device 18, or provide translation in the trunks
between the packet network 12 and the circuit-switched telephony
switches in the PSTN 14, respectively. The lines connecting to the
telephony endpoints 16(B) or the metering devices 18 may be
traditional telephony lines and use traditional pulse code
modulated or analog signaling to carry voice information
therebetween, as well as to facilitate the provision of pulses for
metering from the media gateways 22 to the metering devices 18 or
the telephony endpoints 16(B). In turn, the media gateways 22 will
facilitate corresponding packet sessions to other media gateways to
enable a bi-directional communication session.
[0027] In operation, the media gateway controller 26, upon
coordinating with the media gateway 22 to establish a call
involving a telephony endpoint 16(A) or 16(B), will send a single
metering message, which will provide sufficient information to the
media gateway 22 supporting the particular telephony endpoint 16(A)
or 16(B) to allow the media gateway 22 to provide the necessary
metering throughout the call. Metering pulses are generally
described with the following four characteristics: [0028] pulse
frequency, which is based on a peak-to-peak value of a sinusoidal
signal and may typically have values between 12 KHz and 16 KHz;
[0029] pulse duration, which defines the period over which the
sinusoidal signal is applied to represent a pulse; [0030] pulse
repetition interval (PRI), which defines an interval between the
leading edge of a pulse to the leading edge of the next pulse; and
[0031] pulse level, which defines the relative amplitude of the
sinusoidal signal used to represent the pulse, wherein typical
values are between 2 and 2.6 Volts root mean square (RMS). These
characteristics are illustrated in FIG. 2, which depicts two pulses
having the same pulse duration and occurring over a pulse
repetition interval at a given pulse frequency. Each pulse is
uniform in pulse level, pulse duration, and pulse frequency in the
illustrated embodiment.
[0032] Referring now to FIG. 3, a call may include multiple phases,
which are associated with different tariff rates. The duration of a
phase may be explicitly defined as a finite period, or it may be
infinite. An infinitely long phase duration means the tariff for
the phase will be constant for the remainder of the call. The
tariff is generally defined by a charge rate, which may be based on
currency per unit of time for a given phase or perhaps for a charge
interval. The charge interval defines the frequency with which
charges are assessed in accordance with the tariff rate. For
example, the tariff rate may be ten cents per minute, wherein the
charge interval is every six seconds. In this example, a charge of
one cent is incurred every six seconds for a call. In addition,
tariffs for a call may include one-time charges, such as a call
setup charge for initiating a call, or an add-on charge, which may
be applied midway through a call for invoking a supplementary
feature. An example tariff rate data table is provided in Table
1.
TABLE-US-00001 TABLE 1 Example Tariff Rate Data Table Charge Rate
Phase Duration Charge Interval ($/time unit) (time unit) (time
unit) Phase 1 $1.00/min 1 min 6 sec Phase 2 $0.50/min 5 min 6 sec
Phase 3 $0.25/min 10 min 6 sec Phase 4 $0.10/min .infin. 6 sec
[0033] Since the tariff rates are implemented using pulses, there
must be a conversion from a tariff rate to a pulse rate, wherein
each pulse is associated with the tariff, such that the rate at
which pulses are provided during the metering process corresponds
to the tariff rates. The resultant pulse rate, which is based on
the tariff rate, is referred to as a tariffed pulse rate (TPR). The
TPR is generally specified in units of pulses per second, wherein
each phase has a unique TPR.
[0034] With reference to FIG. 3, a metering model is illustrated.
The depicted call includes two phases, Phase 1 and Phase 2, which
have unique TPRs. Each phase will have a phase duration defining
the length of the phase, if the phase is not infinite, which would
indicate a given TPR will be applied for the remainder of the call.
Select metering processes will define charge intervals, which again
define the frequency with which charges are incurred. Based on the
TPR, each charge interval will define a period in which a select
number of pulses will occur, wherein the charge intervals and the
pulses therein repeat throughout the phase. The number of pulses
occurring during any charge interval is referred to as a pulse
count per charge interval (PC-CI). When a new phase begins, the
charge intervals, as well as the PC-CI, may change to reflect the
TPR. In the illustrated example, each charge interval in Phase 1
has a PC-CI of three, wherein each charge interval in Phase 2 has a
PC-CI of one.
[0035] Because the tariff rates may be complex and vary from one
service provider to another for different types of services, the
conversion of the tariff rates to pulse counts can produce
non-integer values, such as 2.333 pulses per charge interval or
phase. Since a fraction of a pulse cannot be signaled to the
metering device 18 or the telephony endpoint 16(B), there is
inherent inaccuracy in the metering process. Due to the sensitivity
of billing matters to both the service provider and the subscriber
or customer to be billed, one of the key challenges for a metering
process is to be able to provide metering that is as accurate as
possible. In addition to dealing with fractional pulse counts,
another obstacle in the way of accurate metering is the messaging
overhead, packet loss, and latency associated with transporting
packets carrying metering instructions over the packet network 12
between the various media gateways 22, 24 as well as between the
media gateway 22 and the media gateway controller 26. To address
these concerns, one embodiment of the present invention provides a
single message to the media gateway 22 from the media gateway
controller 26 to provide all of the information necessary for the
metering associated with a given call. Thus, the message will
include, if necessary, parameters defining how to provide metering
for each possible phase associated with a call, as well as setup
and add-on charges. In essence, the entire (or complete) call
tariff model is expressed in a single message, which is sent to the
media gateway 22 in association with the call. An exemplary message
including an entire call tariff model follows:
TABLE-US-00002 Message Header TariffRate_SignalList={ {Setup
Charges}, {Phase 1 Tariff Data, Phase 1 Duration} {Phase 2 Tariff
Data, Phase 2 Duration} . . {Phase N Tariff Data, Phase N Duration}
}
After the message header, the TariffRate_SignalList defines the
call tariff model for the call. Initially, parameters for setup
charges are defined, followed by parameters necessary for metering
phases 1-N of the call. Notably, each phase will include a phase
duration, in addition to any tariff data, which essentially defines
the rate at which pulses are provided throughout the phase
duration, in general or in defined charge intervals.
[0036] Setup and add-on charges may be handled in a multitude of
ways within the call tariff model. In one embodiment, one-time
charges for call setup may be applied using a burst of pulses
provided at the beginning of the first phase, wherein pulses are
provided at a maximum rate until the one-time charge is reached and
any pulses required during the first phase during which the pulse
burst occurs are provided. In essence, the pulse burst will recover
the one-time charge and then catch up with the normal TPR for the
phase. In the following call tariff model, the
TariffRate_SignalList defines the setup charge parameters as
requiring five pulses using a metering pulse burst (MPB) at a 3
millisecond pulse rate interval (PRI):
TABLE-US-00003 TariffRate_SignalList={ {PRI=3, MPB=5} {Phase 1
Tariff Data, Phase 1 Duration} {Phase 2 Tariff Data, Phase 2
Duration} . . {Phase N Tariff Data, Phase N Duration} }
[0037] For add-on charges, which occur during the middle of the
call, the call tariff model may be further modified to define the
parameters for providing pulses for an add-on charge. In the
following example, eight pulses are provided using a metering pulse
burst, where the pulses are provided at three millisecond pulse
rate intervals. The original call tariff descriptor may be included
in the new message as a keep alive mechanism, which informs the
media gateway 22 that it should continue to process that signal
list rather than aborting and starting over:
TABLE-US-00004 {PRI=3, MPB=8} %% Add-on Charge
TariffRate_SignalList={ %% Keep alive original call tariff
descriptor {PRI=3, MPB=5} {Phase 1 Tariff Data, Phase 1 Duration}
{Phase 2 Tariff Data, Phase 2 Duration} . . {Phase N Tariff Data,
Phase N Duration} }
Notably, the above call tariff model still defines the call setup
charges as previously defined, wherein five pulses are provided at
three millisecond intervals. When the add-on charges occur, the
media gateway 22 will provide the burst of pulses for the add-on
charge as well as any additional pulses necessary for the TPR
normally associated with the given phase.
[0038] In an alternative to intermingling the normal pulses
associated with the TPR and any one-time charges such as setup and
add-on charges, virtual phases may be defined to accommodate the
pulses associated with one-time charges. These virtual phases are
generically referred to as pulse windows, and may correspond
directly to a regular phase, or may be a portion of a regular
phase. The following call tariff model, as embodied in a
TariffRate_SignalList, follows wherein the first regular phase is
divided into two pulse windows. The first pulse window defines the
parameters necessary to provide pulses associated with setup
charges and some of the pulses necessary for Phase 1, and the
second pulse window defines the regular pulses based on the TPR for
the first regular phase that are necessary for the balance of Phase
1. Notably, the first pulse window has a corresponding pulse window
duration, and the second pulse window is the balance of the
duration for the first regular phase. The remaining pulse windows
correspond directly to the remaining regular phases of the call.
The TarrifRate_SignalList is as follows:
TABLE-US-00005 TariffRate_SignalList={ {Pulse Window 1 Data (Setup
Charges + Some of Phase 1), Pulse Window 1 Duration} {Pulse Window
2 Data (Balance of Phase 1 Pulse Rate), Balance of Phase 1
Duration} {Pulse Window 3 Data (Phase 2 Tariff Data, Phase 2
Duration)} . . {Pulse Window N+1 Data (Phase N Tariff Data, Phase N
Duration)} }
Add-on charges are handled in a similar fashion.
[0039] As noted, handling fractional tariff pulse rates may require
a fractional pulse count throughout a defined charge interval or
phase. In one embodiment of the present invention, the fractional
pulse count is converted into discrete integer components, which
describe how many actual pulses to provide during consecutive
charge intervals to approximate the desired fractional pulse count
over a period of time. In essence, the pulse count per charge
interval may vary from one charge interval to another to
approximate the fractional tariff pulse rate over a series of
charge intervals. For example, if a phase consisting of 10 charge
intervals has a PC-CI of 2.333 pulses per second (pps), then a
pattern of applying 3 pps and 2 pps across the 10 intervals can
yield a very accurate approximation of 2.333 pps.
[0040] As implied in the previous example, the integer values used
as pulse counts in successive charge intervals are simply the
truncated and rounded values of the PC-CI value. These are referred
to respectively as PC-CImin and PC-CImax. Considering once again
the example from above with a phase of 10 charge intervals and a
PC-CI of 2.333 pps, PC-CImin=2 and PC-CImax=3 can be applied across
the 10 charge intervals as follows:
[0041] CI1: 3 pps (PC-CImax)
[0042] CI2: 2 pps (PC-CImin)
[0043] CI3: 2 pps
[0044] CI4: 3 pps
[0045] CI5: 2 pps
[0046] CI6: 2 pps
[0047] CI7: 3 pps
[0048] CI8: 2 pps
[0049] CI9: 2 pps
[0050] CI10: 2 pps
This results in 23 applied pulses across the phase which is a very
accurate representation of 2.333 pps times 10 charge intervals.
This pattern of applying pulses can be represented by a pulse map
as follows:
{3 2 2 3 2 2 3 2 2 2}.
Since the number of charge intervals could in theory be much
greater than 10, a simplified shorthand expression is desired for
representing the pulse map. This expression is of the form:
{PC-CImax, PC-CImaxrep} {PC-CImin, PC-CIminrep},
where PC-CImax and PC-CImin are as described above and PC-CImaxrep
and PC-CIminrep represent the number of occurrences of PC-CImax and
PC-CImin respectively that appear in the pulse map. The pulse map
illustrated above is represented in this format as:
{3 3} {2 7}.
[0051] While this format concisely describes the number of
instances of PC-CImax and PC-CImin to be applied across the phase,
it does not describe the interleaving pattern required.
Consequently, a number of viable pattern permutations exist. Three
general patterns are discussed briefly.
[0052] The first pattern, max loading, places the PC-CImax values
at the front of the pattern. This pattern eliminates the
possibility of under-pulsing on calls which end during the middle
of the phase. Conversely however, it may result in over-pulsing,
particularly in cases where the fraction is less than or equal to
0.5. Due to this tendency to overpulse as opposed to underpulse,
this model works more to the benefit of the service provider than
the subscriber. Using the max loading pattern, the previous example
is expressed as follows:
{3, 3, 3, 2, 2, 2, 2, 2, 2, 2}.
The second pattern, min loading, places the PC-CImin values at the
front of the pattern. As the max loading pattern tends to
overpulse, the min loading pattern tends to underpulse, thereby
benefiting the subscriber at the expense of the service provider.
Using the min loading pattern, the previous example is expressed as
follows:
{2, 2, 2, 2, 2, 2, 2, 3, 3, 3}.
A third pattern, weighted interleaving, interleaves the PC-CImax
and PC-CImin values in such a way that the most accurate pulse
count is achieved regardless as to what point in the phase the call
ends. The algorithm for creating the most optimal pulse map pattern
is a function of the number of elements in the pulse map and the
fractional part of the PC-CI value. Applying such an algorithm to
the previous example results in the following weighted interleaving
pulse map:
{3 2 2 3 2 2 3 2 2 2}.
[0053] Additional examples applying the weighted interleaving
pattern and algorithm are as follows: [0054] PC-CI=2.666 pps over
10 charge intervals yields a pulse map expression of {3, 7} {2, 3}.
Applying weighted interleaving the expression expands into the
following pulse map: {3, 3, 2, 3, 3, 2, 3, 3, 2, 3}; [0055]
PC-CI=8.333 pps over 10 charge intervals yields a pulse map
expression of {9, 3} {8, 7} with weighted interleaving expansion as
{9, 8, 8, 9, 8, 8, 9, 8, 8, 8}; [0056] PC-CI=8.333 pps over 100
charge intervals yields a pulse map expression of {9, 33} {8, 67}
with weighted interleaving expansion as {9, 8, 8, 9, 8, 8, 9, 8, 8,
9, 8, 8, 9, 8, 8, . . . } [0057] PC-CI=23.333 pps over 7 charge
intervals yields a pulse map expression of {24, 2} {23, 5} with
weighted interleaving expansion as {23, 23, 24, 23, 23, 24,
23}.
[0058] It is possible to specify which of the three described pulse
map patterns should be followed for any given application as
follows:
{PC-CImax, PC-CImaxrep} {PC-CImin, PC-CIminrep} {Pulse Map
Pattern}.
For example:
{3 3} {2 7} {weighted interleaving}.
In practice, media gateways may implement the weighted interleaving
pattern, or other pattern, algorithm and thereby remove the need to
specify the pattern as part of the media control protocol signaling
describing the call tariff model. Those skilled in the art will
recognize various techniques for providing different PC-CI over a
given period to approximate a fractional pulse count.
[0059] In many instances, the charge interval does not evenly
divide into the phase duration of a phase, as illustrated in FIG.
4. In essence, at the end of the phase there is a partial charge
interval remaining in the phase. Depending on the desires of the
service provider, there are two primary approaches to providing
pulses during the partial charge interval. The first is to try to
maintain the TPR as best as possible during the partial charge
interval. The second is to provide pulses at the beginning of the
partial charge interval as if the charge interval had a normal
duration.
[0060] This second approach tends to en on the side of
over-pulsing. With the first method, the part of the phase for
which the series of complete charge intervals may occur is defined
as a first pulse window (Pulse Window 1), and the partial charge
interval is defined as a second pulse window (Pulse Window 2), as
illustrated in FIG. 5. For Pulse Window 1, the PC-CI is provided in
a normal fashion. For example, if a fractional PC-CI were
necessary, the pulse counts for the respective charge intervals
would be calculated as described above for Pulse Window 1. For
Pulse Window 2, the total number of pulses that should be provided
during the phase, which includes Pulse Window 1 and Pulse Window 2,
is determined. From this value, the number of pulses provided in
Pulse Window 1 is subtracted to determine the number of pulses to
provide in Pulse Window 2. If the pulse count for Pulse Window 2 is
a fraction, a decision is made as to whether to round up or down to
the nearest integer and then provide that number of pulses in Pulse
Window 2. This method provides the most accurate pulsing throughout
a given phase.
[0061] With reference to FIG. 6, assume that Pulse Window 1
includes seven charge intervals, wherein a fractional pulse count
of 2.333 was implemented using a charge interval expression of {3,
2} {2, 5} with a pulse map of {3, 2, 2, 3, 2, 2, 2}. Assuming that
the TPR for the phase would require 16.8 pulses, Pulse Window 2
should theoretically provide 0.8 pulses, since Pulse Window 1
provided the first 16 pulses. Rounding 0.8 pulses up to one pulse,
one pulse may be provided in Pulse Window 2. In this example, the
entire phase is over-pulsed by 0.2 pulses, which provides a very
accurate metering process.
[0062] In the second method, pulse windows are not used, and the
partial charge interval is simply assigned a pulse count
corresponding to one of the previous charge intervals. In one
embodiment, the pulse map for a given number of charge intervals
will simply repeat throughout a phase. As such, the partial charge
interval will simply include either the next pulse count in the
pulse map, or if the pulse map ends just prior to the partial
charge interval, the partial charge interval will be assigned the
pulse count associated with the first charge interval in the pulse
map. With reference to FIG. 7, if the pulse map is seven charge
intervals long, and the first charge interval had a pulse count of
three, the partial charge interval will have a pulse count of
three.
[0063] As described above, the call tariff model is provided in a
single message from the media gateway controller 26 to the media
gateway 22 using an appropriate media control protocol, such as
H.248 or the Session Initiation Protocol, among others. The
information in the message will define sufficient parameters for
controlling all phases of the call, and optionally, any setup or
add-on charges. The media gateway 22 may be provisioned to make
certain decisions to assist in the metering process. For example,
the media gateway 22 may determine the pulse map based on the
charge interval expression, and as such, the media gateway
controller 26 will not need to provide information on how to
arrange the various pulses throughout the charge intervals, but
will only need to provide information necessary to define the
number of charge intervals having a first pulse count and the
number of charge intervals having a second pulse count for any and
all phases. Given the significant overhead associated with
signaling between the media gateway controller 26 and the media
gateway 22, one embodiment of the present invention attempts to
minimize the amount of information necessary to define various
parameters in the call tariff model as follows:
[0064] PCx=PC-CImax
[0065] PCn=PC-CImin
[0066] REPx=PC-CImaxRep
[0067] REPn=PC-CIminRep
[0068] PD=Phase Duration
[0069] CI=Charge Interval
[0070] PRI=Pulse Rate Interval
[0071] MPB=Metering Pulse Burst
Based on these abbreviated parameter definitions, a phase of a call
tariff model may be defined as PC-CI=3.6 pulses, CI=30 seconds,
Phase Duration=1 hr, PRI max=5 pulses per second. Based on these
parameters, the message provided to the media gateway 22 may be
provided using the following syntax:
(PRI=5, PCX=4, REPx=6, PCn=3, REPn=4, CI=30 PD=1 hour).
[0072] From this information, the media gateway 22 can construct a
pulse map and apply pulses as defined. Again, the media gateway 22
can derive the pulse map or pattern in which the various pulse
counts are provided in the charge intervals. Notably, not all cases
will include fractional pulse counts. As such, many cases will
include a PC-CI having an integer value. In this case, PCn and REPn
will be set to zero, for example, PC-CI=3 pulses, CI=30 seconds,
Phase Duration=1 hour, PRI max=5 pulses, which can be sent to the
media gateway 22 using the following syntax:
(PRI=5, PCx=3, REPx=10, PCn=0, REPn=0, CI=30, PD=1 hour).
[0073] Assuming the last two examples are parameters for
consecutive phases in a call tariff model, the message sent to the
media gateway 22 may define the call tariff model using the
following syntax:
TABLE-US-00006 {SignalList{PRI=5, PCx=4, REPx=6, PCn=3, REPn=4,
CI=30, PD=1 hr} {PRI=5, PCx=3, REPx=10, PCn=0, REPn=0, REPn=0,
CI=30, PD=1hr}}.
[0074] To minimize the size of the message sent to the media
gateway 22, the above information may be reduced, using a sublist
construction, as follows:
TABLE-US-00007 {SignalList{PRI=[5, 5], PCx=[4, 3], REPx=[6, 10],
PCn=[3, 0], REPn=[4, 0], CI=[30, 30], PD=[1 hr, 1 hr]}}.
As reduced, each parameter is simply assigned consecutive values
for consecutive phases. The reduction process is infinitely
extensible for as many phases as are necessary to define the call
tariff model. It is further noted that this sublist constructor is
generally applicable to other signal lists which may be utilized
within a media control protocol.
[0075] With reference to FIG. 8, a block representation of a media
gateway controller 26 is illustrated. The media gateway controller
26 may include a control system 32 having memory 34 with sufficient
software 36 to provide a call tariff model in a single message as
described above. The control system 32 will include a packet
interface 38 for facilitating communications with the media
gateways 22 and 24, and perhaps with the signaling gateway 30.
[0076] With reference to FIG. 9, a media gateway 22 is illustrated
as including a control system 40 having memory 42 with sufficient
software 44 to provide the metering process described above. The
control system 40 will be associated with a packet interface 46 for
facilitating communications over the packet network 12, as well as
a telephony interface 48, which will be coupled directly or
indirectly to a metering device 18, a telephony endpoint 16, or a
combination thereof
[0077] A metering device 18 according to one embodiment of the
present invention is illustrated in FIG. 10. The metering device 18
may include a control system 50 having memory 52 with sufficient
software 54 to facilitate the metering process. The metering
process may include receiving pulses from the media gateway 22, and
processing these pulses as necessary to control accounting or to
deliver information to the user, directly or via the telephony
endpoint 16(A). The control system 50 is associated with one or
more communication interfaces 56 to facilitate communications with
the media gateway 22, the telephony endpoint 16(A), or a
combination thereof The software 54 may operate to provide
functionality similar to that provided by the metering function 20,
which is resident in the telephony endpoint 16(B).
[0078] With reference to FIG. 11, a telephony endpoint 16(B) is
illustrated according to one embodiment of the present invention.
The telephony endpoint 16(B) will include a control system 58
having memory 60 with sufficient software 62 to operate as
described above. Preferably, the software 62 will provide the
metering function 20 to support the metering process and control
accounting or information delivery to the user. The control system
58 will be associated with a line interface 64, which will connect
to the media gateway 22. If the metering function 20 is not
provided in the telephony terminal 16(B), the control system 58 may
be associated with a metering device interface 66, which may
communicate with a metering device 18. In either embodiment, the
control system 58 may also cooperate with a microphone 68, speaker
70, keypad 72, and display 74 to provide a user interface.
[0079] Those skilled in the art will recognize improvements and
modifications to the preferred embodiments of the present
invention. For example, the media gateway controller 26 and media
gateway 22 may represent various types of call control entities,
such as call agents and servers, terminal adaptors, and the like,
taking the form of server and client or master and slave. The
invention is applicable with technologies incorporating H.248,
Megaco, Session Initiation Protocol, and other analogous protocols.
All such improvements and modifications are considered within the
scope of the concepts disclosed herein and the claims that
follow.
* * * * *