WO2009127734A1

WO2009127734A1 - Network apparatus and controlling method therefore

Info

Publication number: WO2009127734A1
Application number: PCT/EP2009/054652
Authority: WO
Inventors: Wen Jing Liang; Junbiao Zhang; Jin Fei Yu
Original assignee: Thomson Licensing
Priority date: 2008-04-18
Filing date: 2009-04-20
Publication date: 2009-10-22

Abstract

The present invention provides a network apparatus in a multi-rate shared medium network and controlling method therefore. The apparatus accesses the shared medium using a contention based shared medium access protocol, wherein said apparatus is able to use a plurality of physical layer rates over the shared access medium. One of the plurality of physical layer rates is set as the rated physical layer rate of the apparatus corresponding to a rated application layer bandwidth. The method comprises the steps of determining the real physical layer rate of the apparatus; and upon detection of the real physical layer rate being lower than the rated physical layer rate, adjusting the real application layer bandwidth of the apparatus to be lower than the rated application layer bandwidth. According to the invention, the shared medium is controlled and managed as a communication resource to prevent one or some faulty CPEs from negatively influencing the application layer bandwidth of normal ones so that the normal CPEs and the network on the whole can work properly under such a condition.

Description

NETWORK APPARATUS AND CONTROLLING METHOD THEREFORE

FIELD OF THE INVENTION The present invention relates generally to the network technology, and more particularly, to a network apparatus in a multi-rate shared medium network and controlling method therefore.

BACKGROUND OF THE INVENTION In a multi-rate shared medium network, a plurality of client premise equipments (CPEs) access to a communication medium with a physical layer (modulation) rate among a plurality of optional rates by a CSMA/CA (Carrier Sense Multiple Access with Collision Avoidance) protocol.

The physical layer rate of a CPE in the network, during its real operation, may drop to a lower level due to some reasons, such as faults in communication lines and hardware. In such a case, since the opportunities of access to the shared medium will remain unchanged under the principle of equity, the fault CPE will increase its bus time occupation ratio. This will lead to a decrease of the bus time occupation ratio and in turn the application layer bandwidth of other CPEs. As a result, the operation and QoS of the whole network are impacted.

SUMMARY OF THE INVENTION

An object of the invention is to provide a network apparatus in a multi- rate shared medium network and controlling method therefore. According to the invention, the shared medium is controlled and managed as a communication resource to prevent one or some faulty apparatus from negatively influencing the application layer bandwidth of normal ones so that the normal apparatus and the network on the whole can work properly under such a condition.

According to one aspect of the invention, a method for controlling an apparatus in a multi-rate shared medium network is provided. The apparatus accesses the shared medium using a contention based shared medium access protocol, wherein said apparatus is able to use a plurality of physical layer rates over the shared access medium. One of the plurality of physical layer rates is set as the rated physical layer rate of the apparatus corresponding to a rated application layer bandwidth. The method comprises the steps of determining the real physical layer rate of the apparatus; and upon detection of the real physical layer rate being lower than the rated physical layer rate, adjusting the real application layer bandwidth of the apparatus to be lower than the rated application layer bandwidth. According to another aspect of the invention, an apparatus in a multi-rate shared medium network is provided, which accesses the shared medium using a contention based shared medium access protocol, wherein said apparatus is able to use a plurality of physical layer rates over the shared access medium. One of the plurality of physical layer rates is set as the rated physical layer rate of the apparatus corresponding to a rated application layer bandwidth. The apparatus comprises a detecting unit for detecting the real physical layer rate of the apparatus; and a control unit for receiving a detecting result from the detecting unit, and upon detection the real physical layer rate being lower than the rated physical layer rate, adjusting the real application layer bandwidth of the apparatus to be lower than the rated application layer bandwidth.

BRIEF DESCRIPTION OF DRAWINGS

These and other aspects, features and advantages of the present invention will become apparent from the following description in connection with the accompanying drawings in which:

Figure 1 is a diagram showing a system infrastructure for access to the internet through existing cable TV cable network;

Figure 2 is a diagram showing the application layer bandwidth of CPEs in a multi-rate bus network; Figure 3 is a block diagram showing the structure of a CPE 300 in a multi-rate bus network according to an embodiment of the present invention; and

Figure 4 is a diagram showing the workflow of a method for controlling a CPE in a multi-rate bus network according to an embodiment of the present invention.

DETAILED DESCRIPTION OF PREFERRED EMBODIMENTS

In the following description, various aspects of an embodiment of the present invention will be described. For purposes of explanation, specific configurations and details are set forth in order to provide a thorough understanding. However, it will also be apparent to one skilled in the art that the present invention may be practiced without the specific details present herein. In recent years, the number of broadband access network users increased dramatically. With the significant price drop of optical fibre and optical transmission equipments, cable TV operators were capable of providing the broadband access with the EPON (Ethernet Passive Optical Network) + EoC (Ethernet over Coaxial Cable) infrastructure to efficiently use the bandwidth resources of coaxial cable networks. Various applications of shared medium technology on the coaxial cable network have been developed and implemented, such as those based on WiFi (Wireless Fidelity), HPNA (Home Phoneline Networking Alliance), MoCA (Multimedia over Coax Alliance), PLC (Power Line Communication) and Ethernet. Figure 1 is an exemplary diagram showing a system infrastructure for access to the internet through an existing cable TV cable network. Here the internet is used in a broad sense and refers to the wide area network, which could be the Internet or the operator's network with walled garden applications. At the server end of the system as shown by sign 100, during the downlink transmission (from the server end 100 to the client end 100'), a headend apparatus 10 is provided between the internet and the cable TV network. Said headend apparatus 10 comprises multiple access points 20 (AP1 ...APn). The access points 20 are used to transform the Ethernet network signal received via a switch 12 into RF signal. The RF signals from the multiple access points 20 are combined together with cable TV signal by a splitter 30. Here the splitter 30 represents a set of power splitters and band splitters. The splitter 30 is connected to a cable 50. The access points 20 in the embodiment provide a data switching function over the DataLink Layer of the OSI (Open System Interconnect) Reference Model. As shown in Figure 1 , each client 40, for example client 2, at the remote client end 100' of the cable TV network, is provided with a splitter 60 for separating RF signal from the analog video signal of cable TV, and transmitting relevant signals to modem 70 and the TV receiver 90 at client 2 respectively. Here, the splitter 60 can be implemented using power splitters and/or band pass filters. Finally, the data signal is demodulated by the modem 70 and sent to a PC 80 at client 2.

A basic requirement for this coaxial cable network is that the operator should ensure the application layer bandwidth subscribed by the subscribers within a predetermined error range.

However, there is a contradiction in the same contention area of an EOC system between the limited application layer bandwidth on the bus (the shared medium) and a random or even unlimited bandwidth requirement. Therefore, the bandwidth needs to be managed and properly utilized. In the prior art, the above object is typically achieved by determining a maximum subscriber number of a network according to the bus bandwidth (equal to application layer bandwidth on the bus) and the subscribed bandwidth, the average online ratio at busy time and the average bandwidth occupation ratio at busy time of each subscriber.

For example, suppose the bus bandwidth of a cable TV data communication network is 3OM bit/s and each of its subscribers has a subscribed bandwidth of 2M bit/s, with an average online ratio at busy time of 50% and an average bandwidth occupation ratio at busy time of 25%, then the maximum subscriber number of the network is derived as follows:

30/ ( 2^*50%^*25% ) =120. In addition, the network can also reserve some bandwidth according to real conditions to cope with the influence of network peak flow. For the example described above, if 25% of bandwidth is reserved for the network, the available subscriber number of the network is derived as follows: 120^* ( 1 -25% ) =90.

For a baseband transmission having a constant physical layer rate, the above network planning scheme is applicable since each subscriber can be allocated a subscribed bandwidth by the configuration of network administrator. However, this scheme will have problems when it is used in a multi-rate shared medium access network.

This problem mainly arises from the fact that a CPE in a multi-rate bus access network can operate in one of a plurality of optional physical layer rates. If related conditions, such as of communication lines or hardware, of a CPE deteriorates, the CPE will be not suitable for operation in a rated physical layer rate (in most cases, the maximum rate among all the options) and have to reduce its rate to a lower level. However, since all the CPEs have the same opportunity to access to the bus under the principle of equity in the CSMA/CA protocol, the above described CPE will have a larger bus time occupation ratio. When the physical layer rate of this fault CPE drops to some extent, it will occupy the bus time of the other normal CPEs. Therefore, the application layer bandwidth of the normal CPEs will not be available and the quality of service ('QoS') will be impacted.

Another case is that the physical layer rate of a CPE will be reduced with increasing CPE packet loss rate. Similarly as above, this CPE will have a larger bus time occupation ratio, which will in turn impact the operation of the normal CPEs and the bus of the network. A non-normal operation of the bus will cause the packet loss rate to further increase, and consequently the physical layer rate of the above mentioned fault CPE will be further reduced.

Figure 2 is a diagram showing the application layer bandwidth of CPEs in a multi-rate bus network. Figure 2a shows a normal case in which two CPEs use a rated physical layer rate of 54Mbps to access to the bus. As shown in Figure 1 a, suppose that the two CPEs both send out four frames under the principle of equity, they occupy the same bus time and each frame carries same number of data, then two CPEs gets same application layer bandwidth. As shown in Figure 2b, the upper CPE operates in a physical layer rate of 18Mbps, as a result its frame length extends to three times of the initial one but carries the same number of data. In this case, two CPEs will send out the same number of frames, two frames in the example, but the upper one occupies more time than the lower one, and they both get a worse application layer bandwidth. In this case bandwidth of lower CPE, which is normal, is affected by the faulty upper one. A specific example in a cable TV access network will be described to further explain the above problem. In the network, an access point can associate with 25 CPEs, each of which is capable of operating with one of multiple physical layer rates such as

54Mbps_N 48Mbps_N 36Mbps_N 24Mbps_N 18Mbps_N 12 Mbps_^ 9 Mbps and 6 Mbps. In normal operation, each CPE will use the 54Mbps rate, called the rated physical layer rate, to access to the network. In this case, suppose that the application layer throughput of the network is 25Mbit/s and each CPE will have 1 Mbit/s uplink application layer bandwidth. A packet loss rate under 0.1 % will meet the operational requirement, in which case we consider the network works normally. But if the packet loss rate increases, for example to 5%, a drop of the physical layer rate will be triggered.

If at a certain time the channel condition of one CPE deteriorates (for example when a joint between two cables becomes loose) that in turn increase the packet loss rate to reach the 5% threshold, the physical layer rate of this CPE will be reduced from 54Mbps to 6 Mbps. However, since the CPE still has an uplink application layer bandwidth of 1 Mbit/s, it will occupy 1/3 of the bus time in view of the facts that practically a physical layer rate of 6 Mbps can support an application layer bandwidth of about 3Mbit/s. At the same time, the other 24 normal CPEs can only have about 2/3 of their subscribed bus time. Thus it can be seen that one faulty CPE impacts the operation of all the other normal CPEs. It will be helpful for understanding this problem to describe a simple access network only having one access point and two CPEs (CPE1 and CPE2). In a normal state of this network, both CPEs operate with the rated physical layer rate of 54Mbps. Suppose that each of the CPEs is allocated an application layer bandwidth of 15Mbit/s, then they respectively have a bus time occupation ratio of 1/2. At a certain time, the condition of the communication line between the CPE1 and the access point deteriorates to result in a reduction of its physical layer rate to 1 Mbps while the physical layer rate of CPE2 is not changed. According to the CSMA/CA protocol, the opportunity of CPE1 to access to the bus is the same as that of the CPE2. That is, CPE1 and CPE2 will transmit the same number of frames. Suppose the frame length of CPE1 and CPE2 is initially L. Since the physical layer rate of CPE1 drops from 54Mbps to 1 Mbps, its frame length becomes 54L. However, the frame length of CPE2 remains unchanged. It can be appreciated by a person skilled in the art that in this case the bus time occupation ratio of CPE1 is 54L/(L+54L) = 54/55 while that of CPE2 changes into 1/55. It is obvious that almost all the bus time of CPE2 was occupied by the fault CPE1 , which induces problems with CPE2.

In view of the above problem, an embodiment of the present invention proposes a solution to manage and control the bus in a multi-rate bus network to prevent at least one faulty CPE from occupying bus time of other CPEs so that the normal operation of the bus and the network can be ensured.

According to an embodiment of the present invention, the application layer bandwidth of a CPE in the network is adjusted according to its physical layer rate in order that the CPE will operate with lower application layer bandwidth in case of lower physical layer rate and operate with an extremely low application layer bandwidth or just keep management channels in case of minimum physical layer rate. Therefore, the faulty CPE will not impact the operation of other normal CPEs and the bus.

Refer again to Figure 2. Figure 2c shows a bandwidth reallocation when one embodiment of the invention works. The upper CPE whose physical layer rate drops from 54Mbps to 18Mbps reduces its application layer bandwidth, by which the bus time occupation ratio of this CPE is reduced. As shown in Figure 2c, according to this example, the lower CPE can still have the bus time occupation ratio of 1/2. As a result, lower CPE remains its subscribed application layer bandwidth.

Preferably, the application layer bandwidth of a faulty CPE is controlled to reduce in proportion to the reduction of the physical layer rate in order to prevent the bus time of other normal CPEs from being occupied.

Figure 3 is a block diagram showing the structure of a CPE 300 in a multi-rate bus network according to an embodiment of the present invention. The CPE 300 accesses to the network with one of a plurality of optional physical layer rates according to the CSMA/CA protocol. One of the plurality of physical layer rates is set as the rated physical layer rate of the CPE 300, under which condition the CPE 300 has a rated application layer bandwidth. As shown in Figure 3, the CPE 300 comprises a detecting unit 301 for detecting the real physical layer rate of the CPE 300. The CPE 300 also comprises a control unit 302 for receiving the detecting result from the detecting unit 301. If the detected result shows that the real physical layer rate is lower than the rated physical layer rate of the CPE 300, the control unit 302 will adjust the real application layer bandwidth of the CPE 300 to be lower than the rated application layer bandwidth.

Figure 4 is a flow chart showing the procedure 400 carried out by the CPE 300 according to an embodiment of the present invention. As shown in Figure 4, in the step 401 , the real physical layer rate of the CPE 300 is detected. Then the procedure goes to the step 403 in which the detected real physical layer rate is compared with the rated physical layer rate of the CPE 300. If the real physical layer rate is larger than or equal to the rated physical layer rate, the procedure goes to step 401. Otherwise (the real physical layer rate is lower than the rated physical layer rate), the procedure goes to the step 405 in which the real application layer bandwidth of the CPE is adjusted to be lower than its rated application layer bandwidth. For the step 405, an adjustment factor can be introduced to adjust the application layer bandwidth of the CPE 400. The adjustment factor can be defined to be the ratio between the real physical layer rate and the rated physical layer rate of the CPE 300. Therefore, the real application layer bandwidth of the CPE 300 = A x the rated application layer bandwidth.

For the step 405, if the CPE 300 is in the state of transmission, the control unit 302 will reduce the application layer bandwidth upon detection of the real physical layer rate being lower than the rated one. If the CPE 300 is in the state of reception, the control unit 302 will send a message to the transmitter (access point in the case), asking for the transmitter to reduce the application layer bandwidth upon detection of the real physical layer rate being lower than the rated one.

Next, the principle of the present invention will be explained with reference to the simple bus network described above, which only has one access point and two CPEs. The network comprising of more access points and CPEs can operate under the same principle.

In this network, the initial physical layer rates of the CPE1 and CPE2 with the access point are both 54Mbps, which corresponds to a total application layer bandwidth of about 25Mbit/s. If both the CPE1 and CPE2 are allocated an uplink bandwidth of 10Mbit/s, the total bus occupation ratio is about 80%, with CPE1 and CPE2 having about 40% respectively.

During the operation of the network, the condition of the bus changes for some reasons. That is, the channel between the access point and CPE2 remains normal while the channel between the access point and the CPE1 deteriorates to result in a reduction of the physical layer rate into 18Mbps.

In this case, the application layer bandwidth of CPE1 should be reduced to prevent the CPE1 from occupying the bus time of CPE2 according to an embodiment of the present invention. If the application layer bandwidth adjustment factor A is introduced, this adjustment procedure can be represented by the equation of the real application layer bandwidth = A x rated application layer bandwidth, wherein A is the application layer adjustment factor and 0<A<1. As described above, the application layer bandwidth adjustment factor A can be set as the ratio between the real physical layer rate and the rated physical layer rate of the CPE1. In this specific example, since the rated physical layer rate of the CP E1 is 54Mbps and its real physical layer rate is 18Mbps, A=18/54.

Then, the real application layer bandwidth of CPE1 is reduced to be 10 x (18/54) M bit/s. In this case, the real bus time occupation ratio of CPE should be 40% x (10 x (18/54)/10) x (54/18) = 40%, which remains unchanged.

Therefore, the application layer bandwidth of the normal CPE2 suspends and is not impacted by the faulty CPE1.

Please note that if the minimum physical layer rate preset for the network is 18Mbps, then the preset retransmission times of the network should also be considered for the application layer bandwidth adjustment of CPE1 to keep the CPE1 to operate in an extremely low rate. That is, when the sender does not receive the ACK from receiver, it will resend at maximum for the preset number of times. If the frame is not received correctly after N trials, then the frame is dropped.

Generally, the real retransmission times of the network do not change over different physical layer rates. In this case, only the physical layer rate needs to be considered in view of the application layer bandwidth adjustment. However, when the CPE1 operates in a minimum physical layer rate of the network, the preset retransmission times of the network should also be considered as well as the physical layer rate. The reason for this lies in the fact that in this case the real retransmission times of CPE1 can not be reduced by the reduction of the physical layer rate so that the real retransmission times of CPE1 is larger than that of CPE2. In this extreme case, retransmission times of each packet of CPE1 are likely to be the preset retransmission times of the network which therefore should be considered in the application layer bandwidth adjustment of CPE1. That is, if CPE operates with a minimum physical layer rate (18Mbps in this embodiment), the application layer bandwidth adjustment factor A should be set as the ratio between the real physical layer rate and the rated physical layer rate of the CPE1 divided by the preset retransmission times of the network. According to this example, if the preset retransmission times of the network is set as R=3, then A = ( 18/54 ) /3.

Similarly, the real application layer bandwidth of CPE1 is reduced to be 10 x (18/54)/3 M bit/s. In this case, as can be appreciated by a person skilled in the art, the application layer bandwidth of CPE1 will not be impacted by the fault CPE1.

In addition, for an extreme condition the application layer bandwidth adjustment factor A can be set as 0, in which case only management channel was maintained for CPE1 and the application layer bandwidth will be recovered after the fault removal.

The principle of the present invention was described with reference to various embodiments. As described above, an embodiment of the present invention proposes an apparatus in a multi-rate shared medium network with and controlling method therefore which adjust the real application layer bandwidth upon reduction of the physical layer rate of the apparatus to prevent bus time of other normal apparatus in the network from being occupied.

While the embodiments are presented in the context of the coaxial cable access network of EPON + EOC based on WiFi, those skilled in the art will recognize that the principles of the invention are not limited to WiFi system but also applicable to other shared medium network systems such as MoCA, HPNA and PLC. The principles of the invention are not limited to coaxial cable transmission but also applicable to other transmission mode such as wireless, power line, twisted-pair cable and Hub. It is to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims.

Claims

1. A method (400) for controlling an apparatus in a multi-rate shared medium network, wherein the apparatus accesses the shared medium using a contention based shared medium access protocol, wherein said apparatus is able to use a plurality of physical layer rates over the shared access medium, one of the plurality of physical layer rates being set as the rated physical layer rate of the apparatus corresponding to a rated application layer bandwidth, characterized in that the method (400) comprises the steps of determining (401 ) the real physical layer rate of the apparatus; and upon detection of the real physical layer rate being lower than the rated physical layer rate (403), adjusting (405) the real application layer bandwidth of the apparatus to be lower than the rated application layer bandwidth.

2. The method (400) according to claim 1 , wherein the highest of the plurality of optional physical layer rates is set as the rated physical layer rate.

3. The method (400) according to claim 1 or 2, wherein the contention based shared medium access protocol is CSMA/CA.

4. The method (400) according to any one of claims 1 to 3, wherein the real application layer bandwidth is adjusted as a function of the rated application layer bandwidth and a first correction coefficient, wherein the first correction coefficient represents the ratio between the real physical layer rate and the rated physical layer rate of the apparatus.

5. The method (400) according to any one of claims 1 to 3, wherein upon detection of the real physical layer rate being equal to the minimum one of the plurality of optional physical layer rates, the real application layer bandwidth is adjusted as a function of the rated application layer bandwidth and a first correction coefficient, wherein the first correction coefficient represents the value of the ratio between the real physical layer rate and the rated physical layer rate of the apparatus divided by the retransmission times of the network.

6. The method (400) according to any one of claims 1 to 3, wherein upon detection the real physical layer rate being equal to the minimum one of the plurality of optional physical layer rates, the real application layer bandwidth is adjusted to be 0 to only maintain a management channel between the apparatus and the network.

7. An apparatus (300) in a multi-rate shared medium network, which accesses the shared medium using a contention based shared medium access protocol, wherein said apparatus is able to use a plurality of physical layer rates over the shared access medium, one of the plurality of physical layer rates being set as the rated physical layer rate of the apparatus corresponding to a rated application layer bandwidth, characterized in that the apparatus (300) comprises a detecting unit (301 ) for detecting the real physical layer rate of the apparatus; and a control unit (302) for receiving a detecting result from the detecting unit (301 ), and upon detection the real physical layer rate being lower than the rated physical layer rate, adjusting the real application layer bandwidth of the apparatus to be lower than the rated application layer bandwidth.

8. The apparatus (300) according to claim 7, wherein the highest of the plurality of optional physical layer rates is set as the rated physical layer rate

9. The apparatus (300) according to claim 7 or 8, wherein the contention based shared medium access protocol is CSMA/CA.

10. The apparatus (300) according to any one of claims 7 to 9, wherein the control unit (302) adjusts the real application layer bandwidth as a function of the rated application layer bandwidth and a first correction coefficient, wherein the first correction coefficient represents the ratio between the real physical layer rate and the rated physical layer rate of the apparatus.

11. The apparatus (300) according to any one of claims 7 to 9, wherein upon detection by the detecting unit of the real physical layer rate being equal to the minimum one of the plurality of optional physical layer rates, the control unit (302) adjusts the real application layer bandwidth as a function of the rated application layer bandwidth and a first correction coefficient, wherein the first correction coefficient represents the value of the ratio between the real physical layer rate and the rated physical layer rate of the apparatus divided by the retransmission times of the network.

12. The apparatus according to any one of claims 7 to 9, wherein upon detection by the detecting unit of the real physical layer rate being equal to the minimum one of the plurality of optional physical layer rates, the control unit (302) adjust the real application layer bandwidth to be 0 to only maintain a management channel between the apparatus and the network.

13. The apparatus (300) according to any one of claims any one of claims 7 to 12, wherein the apparatus is client premise equipment.