Title of Invention

SWITCHING SYSTEM AND SWITCHING METHOD BASED ON THE SWITCHING OF SERVICE GRADE

Abstract A switching system based on the switching of service class comprises multiple RPR apparatus composing one RPR, each RPR apparatus comprising a RPR frame processing unit for processing the incoming RPR frame from the RPR; a service classification processing unit for classifying the service class of the incoming Ethernet service from the local port; an enter ring switching unit for selecting the output ring with respect to the shortest route principle; and a scheduling unit for scheduling the incoming frame from the RPR which should bypass this RPR station and the frame which is in A/B/C queue at this outring of this RPR station, and outputs them through the RPR. The present invention realizes the bridging of the Ethernet service and the RPR, and can combine various service classification techniques to classify data frames and perform different processes to the services of different classes.
Full Text SWITCHING SYSTEM AND SWITCHING METHOD
BASED ON THE SWITCHING OF SERVICE CLASS
TECHNICAL FIELD
The present invention relates to data communication technique, specifically to the
Resilient Packet Ring (RPR) technique of the metropolitan area network, in particular to a
system and method of adapting the Ethernet service to three types of service classes in the
RPR technique and scheduling the service based on different service classes.
TECHNICAL BACKGROUND
The Resilient Packet Ring (RPR) technique is a newly-rising technique intended to build
a data transferring network, which has multiplexing bandwidth, fair algorithm of each site so
as to ensure the occupancy rate of each site, and the ability to ring protection and QoS
(Quality of Service), mainly aiming at metropolitan hard-core ring network and metropolitan
access ring network. IEEE designedly established an IEEE802.17 standard group in Dec. 2000
to draft the RPR MAC (Media Access Control) layer standard that is able to build RPR on
the basis of the RPR technique.
One of the most principal characters of the RPR technique is to classify the transferring
services, which are divided into three types of A, B, C, wherein the type A is the instant
service; the type B is further divided into two types: B-CIR (Committed Information Rate) is
the committed rate class service, B-EIR (Excess Information Rate) is the excess of committed
rate class service; the type C is the "do-the-best" service. The above three types of services are
distinguished by Service Class defined in the frame structure.
Nowadays, a variety of services are borne on the Ethernet. If the metropolitan area
network is built with the RPR technique, the problem that needs to be solved exigently is how
to bridge the Ethernet service to the RPR, and to adapt the Ethernet service to the three types
(A/B/C) of service classes of the RPR technique, so as to ensure the user service to be
transmitted according to the priority level. Currently, the corresponding technical documents
have not been found yet.
SUMMARY OF THE INVENTION
The technical problem intended to be solved by the present invention is to provide a
switching system and switching method based on the switching of service class, comprising a

switching and bridging system based on service class between the Ethernet system and the
Resilient Packet Ring network and a scheduling system based on service class, and comprising
a service classification method for adapting the Ethernet service to the Resilient Packet Ring
network and a preferable method while various classification methods exists simultaneously,
to realize that the Ethernet service is bridged to the Resilient Packet Ring ring.
The technical scheme is as followed:
A switching system based on the switching of service class comprises multiple RPR
apparatus, said RPR apparatus comprising two ports at the RPR side and a set of ports at the
user side, said RPR apparatus composing a RPR through the ports at the RPR side and
transmitting and receiving the Ethernet data frames through the ports at the user side; it is
characterized in that said RPR apparatus further comprises a RPR frame processing unit, a
service classification processing unit, an enter ring switching unit, and a scheduling unit;
said RPR frame processing unit is used for processing the incoming RPR frame from a
port at the RPR side, and scheduling the RPR frame to the corresponding port at the user
side or transmitting the RPR frame to the primary transmission queue or to the secondary
transmission queue according to whether the RPR frame should be dropped from the ring;
said service classification processing unit is used for processing the service class
classification for the incoming Ethernet service from a port at the user side and transmitting
it to the corresponding service queue;
said enter ring switching unit is used for selecting the output ring of the data frame and
transmitting it to A/B/C service queue of the output ring, queuing and waiting for being
scheduled;
said scheduling unit is used for unifiedly scheduling the data frames in the primary
transmission queue and the secondary transmission queue and the data frames in the A/B/C
service queue of the output ring, and outputs them from a port at the RPR side.
In said switching system based on the switching of service class, it further comprises a set
of service classification mode configuration registers, in one-to-one correspondence with the
ports at the user side of the RPR apparatus; said service classification mode configuration
register is used for storing the service classification modes corresponding to the ports at the
user side, and providing the classification mode of an incoming data frame from the
corresponding port at the user side to said service classification processing unit.


A switching method based on the switching of service class, comprising:
step 1: determine whether the RPR frame should be dropped from the ring according to
the destination address of the incoming RPR frame from the RPR; if so, the RPR frame is
scheduled to the corresponding port at the user side; if not, the RPR frame is transmitted to
the primary transmission queue or to the secondary transmission queue according to the
service class of the RPR frame; turn to the step 4;
step 2: process the service class classification for the incoming Ethernet data frame from
a port at the user side;
step 3: select the output ring of the data frame, and transmitting the data frame to A/B/C
service queue of the output ring;
step 4: unifiedly schedule the data frames in the primary transmission queue and the
secondary transmission queue and the data frames in the A/B/C service queue of the output
ring, and outputs them to the corresponding RPR.
The switching system based on the switching of service class and the switching method
thereof described by the present invention realizes the bridging of the Ethernet service and
the RPR through the service classification and the unified scheduling of all services, provides
an efficient switching system and switching method with respect to the RPR techniques. The
present invention can combine various service classification techniques, process different
disposes to the services with different classes, thereby acquiring various strategies for
providing different services and prices.
BRIEF DESCRIPTION OF THE ACCOMPANYING DRAWINGS
Fig.l is a structural schematic drawing of the switching system of the present invention;
Fig.2 is a schematic flowchart of the switching method of the present invention.
DETAILED DESCRIPTION OF THE INVENTION
The technical scheme of the present invention will be further described in detail in
combination with the embodiments and the drawings.
As illustrated in the Fig. 1, the switching system of the present invention comprises a
RPR composed of multiple RPR apparatus. The RPR apparatus comprise a RPR frame
processing unit, a service classification processing unit, an enter ring switching unit, a
scheduling unit, two ports at the RPR side, and a set of ports at the user side. All incoming
frames from and outputted to the ports at the RPR side are RPR frames, each of which has a
i

service class classification identification (ID), indicating the service class of the RPR frame is
A, B, or C; while the ports at the user side are generally Ethernet ports, and all incoming
frames from and outputted to such ports are Ethernet frames, in which the frame structure
has no service class classification ID of A, B, C. The multiple RPR apparatus compose a RPR
through the ports at the RPR side. The two principal work modes of the RPR apparatus are
switching mode and routing mode. The following embodiments are aiming at the service
classification under the switching mode, but are also applicable for the RPR apparatus
adopting the routing mode.
The incoming RPR frame from a port at the RPR enters the RPR frame processing unit
for processing. The RPR frame processing unit determines whether the RPR frame should be
dropped from the ring according to the destination MAC address of the RPR frame. If so, the
RPR frame is scheduled to the corresponding port at the user side according to the extended
destination address in the RPR frame structure and the correspondence relationship of the
destination MAC address and the port at the user side. If the RPR frame should not be
dropped from the ring the RPR frame is put into the primary transmission queue (PTQ) or to
the secondary transmission queue (STQ) according to its service class classification ID.
The service class (A/B/C) classification, however, is processed for the incoming Ethernet
data frame from a port at the user side of the RPR apparatus by the service classification
processing unit. The data frame after service class classification is outputted to the enter ring
switching unit, which selects the output ring of the data frame and outputs the data frame to
the A/B/C service queue of the corresponding output ring, waiting for being scheduled. The
services in the primary transmission queue (PTQ) and the secondary transmission queue
(STQ) and the services in the three types (A, B, C) of class service queue of the output ring are
unifiedly scheduled by the scheduling unit, and outputted through the ports at the RPR side.
The service classification processing unit processes the classification for the Ethernet
data frame in two manners. One is to directly process the service classification for the
Ethernet data frame. As the Ethernet structure has no RPR service class classification ID, the
classification is processed according to the information of the header of the Ethernet frame,
and the corresponding service class ID is put into the control field corresponding to the
Ethernet data frame. Then the Ethernet data frame is converted into the RPR frame prior to
the scheduling. Another manner is to convert the Ethernet data frame into the RPR frame
prior to the sen ice class classification. At this time, the sen ice class ID of the RPR frame is
endued with the default senice class, such as the senice of type C; then the senice class


classification is processed to re-assign the service class ID of the RPR frame.
The switching system may further comprise a set of service classification mode
configuration registers, each corresponding to each port at the user side, used for storing the
service class classification modes corresponding to each port at the user side. When an
incoming Ethernet frame from a port at the user side enters the service classification
processing unit, the service classification processing unit queries the service classification
mode configuration register corresponding to the port, acquires the service classification
mode of the incoming data frame from the port. Then the service classification processing unit
processes the service classification again according to the corresponding service classification
mode.
A schematic flowchart of the switching method based on the switching of service class of
the present invention is presented in the Fig. 2, in which initially determining the incoming
data frame is from a port at the RPR or from a port at the user side. If the incoming data
frame is from a port at the RPR, then the RPR frame processing is carried forward. If the
incoming data frame is from a port at the user side, then the service classification processing
and the enter ring switching processing is carried forward; finally the scheduling processing is
carried forward for the above-described data frame. Each of the above steps is described in
detail in the following.
The step of processing the incoming RPR frame from a port at the RPR comprises:
determine whether the RPR frame should be dropped from the ring according to the
destination MAC address of the RPR frame; as to be dropped from this ring, the RPR frame
is scheduled to the corresponding port at the user side; as to bypass this RPR station, the RPR
frame is transmitted to the primary transmission queue (PTQ) or to the secondary
transmission queue (STQ) according to the service class classification ID taken with the RPR
frame; if the service is belonging to type A, it is transmitted to the primary transmission
queue (PTQ); as to the service of type B or type C, it is transmitted to the secondary
transmission queue (STQ).
Processing the service class classification for the incoming Ethernet data frame from a
port at the user side refers to classifying the Ethernet data frame corresponding to the A/B/C
service class of the RPR.
The service classification may be processed for the Ethernet data frame in two manners.


One is to directly process the classification for the Ethernet frame. Then the Ethernet frame is
converted into the RPR frame prior to the scheduling operation. Another manner is to
convert the Ethernet data frame into the RPR frame prior to the service class classification.
Namely, a new RPR frame is created and the contents of the Ethernet frame are filled into the
corresponding section of the RPR frame. At this time, the service class ID taken with the RPR
frame is endued with the default service class ID value, such as the service of type C; then the
service class classification is processed to re-assign the value.
The service classification methods for Ethernet data frames may be the following: port
based service classification method; virtual local area network identification (VLAN ID)
based service classification method; source/destination MAC address group based service
classification method; and the classification method according to the mapping of the
IEEE802.3 MAC service class (for short, COS) bit to the A, B, C services of IEEE802.17MAC
Each of the above service classification methods is illustrated respectively in the
following.
The port based service classification method means that the service classification is
processed with respect to the number of the port at the user side of the incoming data frame,
in particular the method comprising: setting a default service type configuration item for each
port at the user side, as illustrated in Table 1. The service grade of the incoming data frame
from the port at the user side is configured according to the default service type configuration
item of that port. If the default service type of the port is 00, the data frame of the port is
classified as the service of type C; if the default service type of the port is 01, the data frame of
the port is classified as the service of type B; if the default service type of the port is 10, the
data frame of the port is classified as the service of type Al; if the default service type of the
port is 11, the data frame of the port is classified as the service of type A0. In the standard, the
service of type A is divided into sub-types Al and A0, wherein the priority level of the service
of type Al is higher.


The VLAN ID based service classification method means that the classification is
processed with respect to the VLAN ID taken with the incoming data frame from the port at
the user side, in particular the method comprising: setting a mapping table of the VLAN ID
and the service grade, as illustrated in Table 2. The incoming data frame from the port at the
user side is configured to corresponding service class according to the mapping of the VLAN
ID taken with it and the service class. If a 802.1Q tag is taken with the data frame and the
VLAN ID thereof is valid, the service classification is processed by using the VLAN ID; if the
VLAN ID is invalid, Table 3 is looked through to acquire the default VLAN ID of the port at
the user side, and setting the default VLAN ID to that of the data frame, then the service
classification is processed. If the VLAN ID is 00, the data frame of the port is classified as the
service of type C; if the VLAN ID is 01, the data frame of the port is classified as the service of
type B; if the VLAN ID is 10, the data frame of the port is classified as the service of type Al;
if the VLAN ID is 11, the data frame of the port is classified as the service of type A0.

The source/destination MAC address group based service classification method means
that the service grade classification is processed for the data frame with respect to the source
MAC address and the destination MAC address taken with the data frame, in particular the
method comprising: setting a mapping table of the source/destination MAC address group
and the service type, as illustrated in Table 4, wherein the source/destination MAC address
group may be the MAC address group of type MAC-MAC (namely both the source MAC
address and the destination MAC address are specific), or the MAC address group of type
MAC-ANY (namely the source MAC address is specific but the destination MAC address is
arbitrary), or the MAC address group of type ANY-MAC (namely the destination MAC

address is specific but the source MAC address is arbitrary). The service classification is
processed to the incoming data frame from the port at the user side according to the mapping
table illustrated by Table 4. Firstly, the destination MAC address and the source MAC
address is extracted from the incoming data frame from the port at the user side. Then they
are processed with the MAC stream matching. At this point, the type MAC-MAC is processed
with the MAC stream matching, namely matching the extracted destination MAC address
and source MAC address with the DesMAC + SrcMAC in Table 4. If the matching is
successful, then the configuration value of the ServiceClass item corresponding to the entry is
set as the service class of the data frame; if the matching is unsuccessful, then the type MAC-
ANY is processed with the MAC stream matching, namely whether the matching source MAC
address equals to SrcMAC, and the destination MAC address equals to DesMac, equaling to
all F; if the matching is successful, then the configuration value of the ServiceClass item
corresponding to the entry is set as the service class of the data frame;
if the matching is unsuccessful, then the type ANY-MAC is processed with the MAC
stream matching, namely whether the matching source MAC address equals to SrcMAC,
equaling to all F, and the destination MAC address equals to DesMac; if the matching is
successful, then the configuration value of the ServiceClass item corresponding to the entry is
set as the service class of the data frame; otherwise, the service class of the data frame is
determined as type C.

The classification method according to the mapping of the IEEE802.3 MAC service class
(COS) bit to the A, B, C services of IEEE802.17 MAC comprises: determining whether the
802.1Q tag is taken with the data frame; if the 802.1Q tag is taken with the data frame, the

mapping table of the A/B/C services of 802.1P-802.17 is looked through with respect to the
COS bit in the tag to acquire the particular service classification value as the service class of
the data frame; if the 802.1Q tag is not taken with the data frame, the value of the COS bit is
determined as 0, then the mapping table is looked through to acquire the service class of the
data frame. The correspondence relationship of the 802.3 COS bit and the A/B/C services of
802.17 is recorded in the mapping table of the A/B/C services of 802.1P-802.17.
There may be various methods of service classification to Ethernet frames, which can
exist in the same system. There is provided a preferable port based method with respect to the
present invention. It may, therefore, control the service classification method at the port at the
user side. In particular, the method comprises: at each port, a port based service classification
mode configuration register is set, as illustrated in Table 5. When the incoming data frame is
from the port at the user side, the service classification mode configuration register of the port
at the user side is initially queried, then the service classification is processed by the
corresponding method according to the service classification mode of the particular
configuration.

The enter ring switching processing is made to the data frame after the service
classification processing, comprising: selecting the output ring of the data frames according to
the shortest path principle to the destination site, and respectively transmitting the data
frames to the service queues of type A/B/C of the corresponding output ring, queuing and
waiting for being scheduled.
The step of scheduling comprises: unifiedly scheduling the incoming data frames which
should not be dropped from the RPR, namely the data frames in the primary transmission
queue and the secondary transmission queue, and the data frames in the incoming service
queue of type A/B/C from the port at the user side, and outputting them to the corresponding
RPR.
A better embodiment of the present invention is given below.

When a data frame enters from some port at the user side, the service classification is
initially processed.
1. Service classification processing
1) Checking from which port the data frame entered. It is assumed as port i.
2) Searching the service classification mode configuration register of the port i for the
service classification mode value of the port i, as the SCMode value in Table 5. If the value is
00, it indicates that the service class is distinguished according to 802.1P, then turn to step 3);
if the value is 01, it indicates that the service class is distinguished according to the port, then
turn to step 4); if the value is 10, it indicates that the service class is distinguished according to
the Vlan ID, then turn to step 5); if the value is 11, it indicates that the service class is
distinguished according to the MAC address group, then turn to step 6).
3) If the 802.1Q tag is taken with the data frame, the mapping table of the A/B/C services
of 802.1P-802.17 is looked through with respect to the value of the COS bit in the tag to
acquire the particular service classification value as the service class of the data frame; if the
802.1Q tag is not taken with the data frame, the value of the COS bit is determined as 0, then
the mapping table is looked through to acquire the service class of the data frame; then turn
to step 7).

4) Searching the port default service type configuration of the port i, as illustrated in
Table 1. The acquired particular service type value is set to the service class of the service;
then turn to step 7).
5) If a 802.1Q tag is taken with the data frame, then it is determined whether the VLAN
ID in the tag is valid. If it is valid, the VLAN ID based service type configuration is searched
according to the VLAN ID value, to acquire the corresponding service type value as the
service class of the data frame; if it is invalid, the default VLAN ID value of the port is set to
the VLAN ID of the data frame, then performing the search again to acquire the service class
of the data frame; then turn to step 7).
6) Searching the service type configuration according to the source MAC address and the
destination MAC address of the data frame. If the matching item is found, the service type
value corresponding to the item is set to the service class of the data frame; otherwise the
default service type value, type C, is used. Then turn to step 7).
7) Performing the enter ring switching processing.
2. Enter ring switching processing
1) Searching the enter ring selection table according to the destination MAC address

taken with the data frame. The enter ring selection table provides the correspondence
relationship of the destination MAC address and the RPR selection which is obtained
according to the shortest path principle . The output RPR j is acquired through looking
through the table.
2) Transmitting the service class acquired after the service classification processing to the
A/B/C service queue of the RPR j.
3) Converting the Ethernet data frame into the RPR frame, waiting for being scheduled.
3. RPR frame processing
The incoming RPR frame from the RPR will be processed.
1) Checking from which RPR the data frame entered. It is assumed as ring k.
2) Determining whether the destination MAC address of the RPR frame is belonging to
the local site. If so, turn to step 3); otherwise turn to step 5).
3) Extracting the destination MAC address of the 802.3 frame which is encapsulated in
the RPR frame, searching the drop ring MAC address table, which contains the
correspondence relationship of the MAC address and the local port, to acquire the number of
the local site m needed to output the data frame.
4) Stripping the 802.3 frame from the RPR frame and outputting it through the local
port m. The process is ended.
5) Putting the RPR frame into the primary transmission queue (PTQ) and the secondary
transmission queue (STQ) according to the service class of the RPR frame, wherein all
services of type A are put into PTQ, while other services are put to STQ. Then the scheduling
processing is performed.
4. Scheduling processing
Scheduling the services in the 5 queues of PTQ/STQ/A/B/C in turn, and outputting them
from the corresponding ring. Such scheduling follows the definition of IEEE802.17.
The system and method of the present invention is an efficient switching system with
respect to RPR techniques, and the method may combine various service classification
techniques, to perform service classification for end-user and perform different processing
with respect to the services of different classes, thereby acquiring various strategies for
providing different services and prices to users of different classes.
Although the service class classification method of the present invention has been
illustrated and described in the particular embodiments thereof, it is apparent that the
present invention is not restricted by such classification method. The contents of a variety of

modifications, replacements, variations, substitutions, and equivalents, which should fall into
the protected range of the present invention, will be considered by those skilled in the art
without departing from the spirit and scope of the scheme of the present invention as
determined by the appended claims.

We Claim :
1. A switching system based on the switching of service class, comprising
multiple Resilient Packet Ring (RPR) apparatus, said RPR apparatus having two
ports at the RPR side and a set of ports at the user side, said RPR apparatus
composing a RPR through the ports at the RPR side and transmitting and receiving
the Ethernet data frames through the ports at the user side; the switching system
being characterized in that said RPR apparatus comprises a RPR frame processing
unit, a service classification processing unit, an enter ring switching unit, and a
scheduling unit; wherein
said RPR frame processing unit is used for processing an incoming RPR frame
from a port at the RPR side, and scheduling the RPR frame to the corresponding port
at the user side or transmitting the RPR frame to the primary transmission queue or
to the secondary transmission queue, depending on whether the RPR frame should be
dropped from the ring;
said service classification processing unit is used for classifying the service class
of an incoming Ethernet data frame from a port at the user side and transmitting the
data frame to the corresponding service queue;
said enter ring switching unit is used for selecting the output ring of the data
frame and transmitting the data frame to the service queue corresponding to the class
of the output ring, queuing and waiting for being scheduled;
said scheduling unit is used for unifiedly scheduling the data frames in the
primary transmission queue and the secondary transmission queue and the data
frames in the A/B/C service queue of the output ring, and outputting them through the
ports at the RPR side.
2. The switching system as claimed in claim 1, wherein a set of service
classification mode configuration registers are provided in one-to-one correspondence

with the ports at the user side of the RPR apparatus; said service classification mode
configuration register being used for storing the service classification modes
corresponding to the ports at the user side, and providing the classification mode of an
incoming data frame from the corresponding port at the user side to said service
classification processing unit.
3. The switching system as claimed in claim 1 or 2, wherein said service
classification processing unit is adapted for converting an incoming Ethernet frame
into a RPR frame and then processing its service class classification.
4. The switching system as claimed in claim 3, wherein said service
classification processing unit is adapted to process the service class classification
with respect to the number of the port at the user side of the incoming data frame,
and, in particular by setting a default service type configuration item for each port at
the user side; the service class of the incoming data frame being configured from the
port at the user side according to the default service type configuration item of that
port.
5. The switching system as claimed in claim 3, wherein said service
classification processing unit is adapted to process the service class classification
with respect to a virtual local area network identification taken with the incoming
data frame from the port at the user side, in particular by setting a mapping table of
the virtual local area network identification and the service class; such that in the
event of a tag802.1Q tag being taken with the incoming data frame from the port at
the user side, and its virtual local area network identification being valid, the virtual
local area network identification is caused to be used, by looking through the said
mapping table, to acquire the service class of the data frame; and in the event of the
virtual local area network identification being invalid, the default virtual local area
network identification of the port is caused to be set to that of the data frame, by

looking through the mapping table, to acquire the service class of the data frame.
6. The switching system as claimed in claim 3, wherein said service
classification processing unit is a source/destination MAC address group based service
classification unit, whereby the service class classification is caused to be processed
for the data frame with respect to the source MAC address and the destination MAC
address taken with the data frame, in particular, for; setting a mapping table of the
source/destination MAC address group and the service type; extracting the source
MAC address and the destination MAC address from the incoming data frame from
the port at the user side; performing the MAC stream matching to the
source/destination MAC address in the mapping table; so that in the event of the
matching being successful, the corresponding service class item is used as the service
class of the data frame; and in the event of the matching being unsuccessful, the
default service class value is used as the service class of the data frame.
7. The switching system as claimed in claim 6, wherein said source/destination
MAC address group is in any of the following catagories; both the source MAC
address and the destination MAC address are specific; or the source MAC address is
specific but the destination MAC address is arbitrary; or the destination MAC address
is specific but the source MAC address is arbitrary.
8. The switching system as claimed in claim 3, wherein said service
classification processing unit is adapted to classify according to the mapping of
IEEE802.3 MAC service class bit with the A, B, C services of IEEE802.17 MAC, in
particular, by determining whether the tag802.1Q tag is taken with the data frame,
so that in the event of the tag802.1Q tag being taken with the data frame, the
mapping table of the A/B/C services of 802.1P-802.17 is caused to be looked through
according to the service class bit in the tag to acquire the particular service
classification value as the service class of the data frame; and in the event of the

tag802.1Q tag being not taken with the data frame, the value of the service class bit is
caused to be determined as 0; and the above-mentioned mapping table is caused to be
looked through to acquire the service class of the data frame.
9. A switching method based on the switching of service, said method
comprising the steps of
(a) determining whether the Resilient Packet Ring (RPR) frame should be
dropped from the ring according to the destination address of the RPR frame output
from the RPR; distributing the RPR frame to the corresponding port at the user side,
in the event of that being determined to be affirmative; while transmitting the RPR
frame to the primary transmission queue or to the secondary transmission queue
according to the service class of the RPR frame, in the event of that being determined
to be negative;
(b) classifying the service class of the Ethernet data frame output from a port at
the user side;
(c) selecting the output ring of the data frame, and transmitting the data frame
to A/B/C service queue of the output ring;
(d) unifiedly scheduling the data frames in the primary transmission queue and
the secondary transmission queue and the data frames in the A/B/C service queues of
the output ring, and outputting them to the corresponding RPR.

10. The switching method as claimed in claim 9, wherein, in said step (a), in the
event of the RPR frame being dropped from the ring at the local site, destination MAC
address of 802.3 frame, which is encapsulated in the RPR frame, is extracted; the
drop ring MAC address table is looked through to acquire the number of the port at
the user side, as needed to output the data frame; and the 802.3 frame is stripped
from the RPR frame and outputted through the said port at the user side.
11. The switching method as claimed in claim 9, wherein, in said step (a), in the

event of the RPR frame being determined to be the service of type A, it is transmitted
to the primary transmission queue; and in the event of the RPR frame being
determined to be the service of type B or type C, it is transmitted to the secondary
transmission queue.
12. The switching method as claimed in claim 9, wherein said step (b),
comprises; converting an incoming Ethernet frame into a RPR frame, that is, by
creating a new RPR frame, prior to the service class classification, filling the contents
of the Ethernet frame into the corresponding section of the RPR frame, and enduing
the service class identification of the RPR frame with the default service class
identification value.
13. The switching method as claimed in claim 12, wherein the service class
classification method of said step (b) is the port based service classification method,
that is, the service classification is processed with respect to the number of the port at
the user side of the incoming data frame, said method, in particular, involving :
setting a default service type configuration item for each port at the user side; and
configuring the service class of the incoming data frame from the port at the user side
according to the default service type configuration item of that port.
14. The switching method as claimed in claim 12, wherein the service class
classification method of said step (b) is a virtual local area network identification
based service classification method, whereby the classification is processed with
respect to the virtual local area network identification taken with the incoming data
frame from the port at the user side, said method, in particular, involving : setting a
mapping table of the virtual local area network identification and the service class; if a
tag802.1Q tag is taken with the incoming data frame from the port at the user side
and its virtual local area network identification is valid, the said mapping table is
looked through by using the virtual local area network identification to acquire the

service class of the data frame; if the virtual local area network identification is
invalid, the default virtual local area network identification of the port is set to that of
the data frame, and the said mapping table is looked through to acquire the service
class of the data frame.
15. The switching method as claimed in claim 12, wherein the service class
classification method of said step (b) is a source/destination MAC address group based
service classification method, whereby the service class classification is processed for
the data frame with respect to the source MAC address and the destination MAC
address taken with the data frame, said method, in particular, involving : setting a
mapping table of the source/destination MAC address group and the service type;
extracting the source MAC address and the destination MAC address from the
incoming data frame from the port at the user side; performing the MAC stream
matching to the source/destination MAC address in the mapping table,' so that in the
event of the matching being successful, the corresponding service class item is the
service class of the data frame; and in the event of the matching being unsuccessful,
the default service class value is used as the service class of the data frame.
16. The switching method as claimed in claim 15, wherein the source/destination
MAC address group is in any of the following categories : both the source MAC
address and the destination MAC address are specific MAC address groups; or the
source MAC address is specific but the destination MAC address is an arbitrary MAC
address group; or the destination MAC address is specific but the source MAC address
is an arbitrary MAC address group.
17. The switching method as claimed in claim 12 , wherein the service class
classification method of said step (b) is a classification method according to the
mapping of IEEE802.3 service class bit with the A, B, C services of IEEE802.17, in
particular, by determining whether the tag802.lQ tag is taken with the data frame;

so that in the event of the tag802.1Q tag being taken with the data frame, the
mapping table of the A/B/C services of 802.1P-802.17 is looked through with respect to
the value of the service class bit in the tag to acquire the particular service
classification value as the service class of the data frame; and in the event of the
tag802.1Q tag being not taken with the data frame, the value of the service class bit is
determined as 0; and the above-described mapping table is looked through to acquire
the service class of the data frame.
18. The switching method as claimed in any one of claim 12-17, wherein said
step (b) comprises: setting a port based service classification mode configuration
register at each port; querying the service classification mode configuration register of
a port at the user side when the data frame passes the port at the user side, and then
processing the service classification according to the particularly configured service
classification mode.
19. The switching method as claimed in claim 9, wherein said step (c) comprises:
looking through the enter ring selection table according to the destination MAC
address taken with the data frame, wherein the enter ring selection table provides the
correspondence relationship of the destination MAC address and the RPR selection
according to the shortest route principle; acquiring the output RPR of the data frame;
and transmitting the data frame to the corresponding service queue of the output RPR
with respect to the service class obtained by the service classification processing.
20. The switching method as claimed in claim 9, wherein the scheduling
principle of said step (d) is that, by means of the order of primary transmission
queue/secondary transmission queue/ service queue of type A/service queue of type
B/service queue of type C, the data frames in the above-described service queues are
sequentially scheduled and outputted from the corresponding output RPR.

Documents:

00238-kolnp-2006-abstract.pdf

00238-kolnp-2006-claims.pdf

00238-kolnp-2006-description complete.pdf

00238-kolnp-2006-drawings.pdf

00238-kolnp-2006-form 1.pdf

00238-kolnp-2006-form 3.pdf

00238-kolnp-2006-form 5.pdf

00238-kolnp-2006-international publication.pdf

238-kolnp-2006-assignment.pdf

238-KOLNP-2006-CORRESPONDENCE 1.1.pdf

238-kolnp-2006-correspondence-1.2.pdf

238-KOLNP-2006-CORRESPONDENCE.pdf

238-kolnp-2006-examination report.pdf

238-kolnp-2006-form 13-1.1.pdf

238-KOLNP-2006-FORM 13.pdf

238-kolnp-2006-form 18.pdf

238-kolnp-2006-form 3.pdf

238-kolnp-2006-form 5.pdf

238-KOLNP-2006-FORM-27.pdf

238-kolnp-2006-gpa.pdf

238-kolnp-2006-granted-abstract.pdf

238-kolnp-2006-granted-claims.pdf

238-kolnp-2006-granted-description (complete).pdf

238-kolnp-2006-granted-drawings.pdf

238-kolnp-2006-granted-form 1.pdf

238-kolnp-2006-granted-form 2.pdf

238-kolnp-2006-granted-specification.pdf

238-KOLNP-2006-OTHER PATENT DOCUMENT.pdf

238-KOLNP-2006-OTHERS 1.1.pdf

238-kolnp-2006-others-1.2.pdf

238-kolnp-2006-reply to examination report.pdf

abstract-00238-kolnp-2006.jpg


Patent Number 247510
Indian Patent Application Number 238/KOLNP/2006
PG Journal Number 16/2011
Publication Date 22-Apr-2011
Grant Date 13-Apr-2011
Date of Filing 31-Jan-2006
Name of Patentee ZTE CORPORATION
Applicant Address ZTE PLAZA, KEJI ROAD SOUTH, HI-TECH INDUSTRIAL PARK, NANSHAN DISTRICT, SHENZHEN CITY, GUANGDONG PROVINCE
Inventors:
# Inventor's Name Inventor's Address
1 YU JIE ZTE PLAZA, KEJI ROAD SOUTH, HI-TECH INDUSTRIAL PARK, NANSHAN DISTRICT, SHENZHEN CITY, GUANGDONG PROVINCE 518057
2 MA KAI ZTE PLAZA, KEJI ROAD SOUTH, HI-TECH INDUSTRIAL PARK, NANSHAN DISTRICT, SHENZHEN CITY, GUANGDONG PROVINCE 518057
PCT International Classification Number H04L 12/28
PCT International Application Number PCT/CN2004/000464
PCT International Filing date 2004-05-10
PCT Conventions:
# PCT Application Number Date of Convention Priority Country
1 03140106.6 2003-08-06 China