TW411676B - Arrangement in a network repeater for monitoring link integrity and automatically down shifting link speed - Google Patents

Abstract

A network repeater having a plurality of repeater ports selectively establishes links with remote nodes at one of two data rates based on the capabilities of the remote network node and a determined link integrity. A network repeater establishes a link with a network mode using auto-negotiation techniques to establish a 100 Mb/s link. The network repeater then monitors the link for symbol errors, and determines an integrity of the link based on a detected number of symbol errors relative to a prescribed threshold. If the detected number of symbol errors reaches the prescribed threshold, indicating poor link integrity due to poor cable connection or condition, faulty network device, etc., the network repeater performs a down shifting operation by breaking the established 100Mb/s link, and restarting auto-negotiation to establish a 10Mb/s link. Hence, the network repeater may monitor high-speed links for link integrity, and selectively downshift a link encountering a substantial number of errors to a reduced data rate, without a necessity of a network manager or other remote management function.

Description

--4U67 £ V. Description of Invention (1) [Background of Invention] [Invention. Field]

The present invention relates to a network repeater. In particular, it is a matter of connection. It is a combination of two or more repeater units in accordance with IEEE 802.3, Yanshishan, ^ Valley repeaters are chosen for different rates. Monitor the integrity of the connection and adjust the operation rate. Based on the principle of Mitac's minimum error rate, it can also be applied to other network equipment, including (but not limited to): bridges, routers, network interfaces, and switching networks. Repeaters are generally used for connection between network nodes, such as network stations used as data terminal equipment (DTE) on the same network media. The repeater has the advantage of extending the actual distance between each network node by receiving data packets from one network medium, readjusting the actual signal, and in turn funding the packets to the second network medium. The repeater can also transmit a carrier sensing and contention (Co Uisi on) signal to effectively extend the contention domain of one medium to the second medium. — Repeaters connected between network nodes on different network media are usually configured to connect only to networks operating at the same transmission rate. One problem encountered on high-speed networks such as 100 Mb / s802. 3 networks is between one of the network repeater ports and a network repeater and a remote network node with a low level of integration , Identification of the connection. In particular, it is necessary to find the connection between the relay port and the network node. This node, for example, fails to reach the requirements due to poor electrical connections or conditions, and the wrong network interface device at the remote node. Performance specifications. There is a way to use network management capabilities to handle connection failures, for example

5. Description of the invention (2) For example, configure a remote management function (ie, network manager) for monitoring and controlling the operation of repeaters and network nodes. For example, a remote management technology can use management protocols to transfer management information between the repeater and the network manager. However, such management must rely on the network having management functions, and cannot be used for networks without management functions. In addition, the connection monitoring managed by the central network requires the management units to process the management agreement (the function of the intermediary), thus increasing the complexity and processing requirements of the management units (such as repeaters). [Summary of the Invention] * There is a need on the market for a connection configuration that uses repeaters for network nodes of different speeds. The repeaters and network nodes in this configuration can be reliably monitored without the need for a network manager. The integrity of the connection. There is also a need for a configuration in the network that has repeaters that can be used to monitor the connection status of high-speed data connections, and can automatically change based on the edge or unacceptable connection status measured on the selected high-speed data connection The selected high-rate data is connected to the low-rate data. The present invention can meet these and other needs. The network repeater of the present invention can monitor each repeater port-side connection that communicates with a remote network node at a corresponding data rate. The network repeater monitors each connection by counting the number of symbol errors detected, and determines the integrity of the connection based on the number of symbol errors measured relative to a specified threshold. Based on the threshold value measured by the specified measurement-a connection with poor integrity, the network repeater can choose to reconfigure the connection by reducing the data rate on the network media, this network media More reliable wiring with reduced number of symbol errors.

C: \ Program Files \ Patent \ 91524.ptd Page 7 ______ 5. Description of the Invention (3) According to a concept of the present invention, a method of a network repeater includes establishing a connection with a remote network node According to the specified data rate, the connection is monitored by the network media by counting the number of symbol errors detected, and the integrity of the connection is determined according to the number of symbol errors detected relative to the specified threshold. Monitoring the connection by counting the number of detected symbol errors allows the network repeater to use a relatively simple method without the need for external management functions such as the network manager. That is, the integrity of each network repeater port can be determined. Furthermore, monitoring the connection by counting the number of detected symbol errors allows the network repeater to determine the performance of each network repeater port with the least burden of processing that is encountered during the most uncomplicated general network monitoring process. Integrity. Another feature of this concept is the selective reduction of the data rate on the network refinement to a lower data rate in response to the number of sign errors exceeding a prescribed threshold. Therefore, based on the number of symbol errors detected by the network repeater exceeding the specified threshold, the network repeater can provide a more reliable connection as much as possible and reduce the data rate of identifying the connection. The number of sign errors is reduced relative to the connection. Another concept of the present invention is to provide a network repeater with a plurality of repeater ports configured to send and receive data packets between remote network nodes via respective network media. The network repeater includes a first repeater for sending and receiving data packets according to a first data rate between the first group of repeater ports. The network repeater also includes a second repeater core configured to send and receive data packets between the second group of repeater ports at a second data rate that is slower than the first data rate. There is an auto-negotiation unit configured to select one of the first and second data rates. This 1 ^ 1 1 ^ 1 C: \ Program Files \ Patent \ 91524.ptd page 8

V. Description of the invention (4) The data rate is used to establish the repeater port and the respective ^ «.I Μ ^ ^ ^ f through the remote network nodes to detect the phase cake connection. There is a symbol error detector that is configured for false counts. This repeater also includes a controller that selectively changes the number of sign errors that reach the critical threshold. Then, the repeater can detect the first data rate to the second data rate. The symbol error detector can detect the number of L symbol errors and the number of L symbol errors in at least one connection, enabling the management function of each connection. In addition, when there is a slower connection to a lower speed than the second controller, it is unnecessary to try to use ancient techniques to monitor the integrity of the connection or perform a more complicated suspension to try to overcome the bad connection. Overall traits Other advantages and novel features of the present invention: On the one hand, it is described in the following, and on the other hand, it can be reviewed by a person skilled in the art. 4 The following = explains and implements the present invention to learn the invention Zhenming. The advantages and advantages of the invention will be realized and attained by the means and combinations particularly pointed out in the appended claims. [Brief description of the drawings] Please refer to the attached drawings. The same reference numerals in the drawings represent corresponding parts, wherein: FIG. 1 is a block diagram of a network repeater according to an embodiment of the present invention; FIG. Is a flowchart illustrating a control transmission method in a network repeater according to an embodiment of the present invention. [Illustration of figure number] 10 repeaters 2a- 1 2d,

C: \ Program F les \ Patent \ 91524. Ptd page 9 ill £ 2 £ 5. Description of the invention (5) 14a-14 d Repeater port 18 Backplane 22 Heart (state machine) 26 Port exchange and control Interface 32 Management control logic 40 Processor 44 Physical layer transceiver 48 Counter 16a-16d Media 20 Backplane ... 24 Heart 30 Auto-negotiation unit 34 Connection controller 42 Physical layer transceiver 46 Symbol error detector 50 Table 52 Select Timer circuit 54 timer [detailed description of the preferred embodiment] FIG. 1 is a block diagram of a network repeater 10 according to an embodiment of the present invention, configured to transmit data packets to and from a remote network node 12 s_repeater 10 is a fully integrated multi-port repeater capable of operating at 10Mb / s and 100Mb / s. In detail, the repeater 10 includes four repeater ports 14, which can be used by respective remote network nodes 12 to send and receive data packets in accordance with the IEEE802.3 protocol. Each repeater port 14 and the corresponding network node 12 establish a connection at a specified data rate (for example, 10Mb / s or 100Mb / s) via a network medium 16, and this network medium 16 is, for example, type 3 Unshielded twisted pair (UTP) or Category 5 UTP cable. As described below, each repeater port 14 uses an auto-negotiation protocol and is automatically configured to a rate to the remote network node 12. As is well known in this technology, the 10Mb / s protocol specifically transmits Ethernet (IEEE 802.3) data packets at 1 OMb / s via two twisted pair UTP connections. This connection runs from node 12 to the repeater. The maximum cable distance of 10 is 100 meters ^ 100 Ethernet- (IEEE802.3) data packets required by the 100BASE-TX standard are transmitted at 1000Mb / s via two

C: \ Program Files \ Patent \ 91524.ptd Page 10 411676 V. Description of the invention (6) Category 5 twisted pair UTP connection, here the maximum cable distance from node 12 to repeater 10 is 100 metres. As explained below, the repeater 10 also includes a 10Mb / s backplane 18 and a 100Mb / s backplane 20, so that the repeater 10 can be connected to other similar repeaters, effectively forming a large number of Chun-end repeater. The repeater 10 also includes a 10 Mb / s repeater core 22 and a 100 MB / s repeater core 24. The repeater cores 22 and 24 are configured to send and receive data packets between the selected repeater ports at respective data rates. In particular, the 'repeater 10 includes a port switching and control interface 26 configured to select each network port to connect to one of the repeater cores 22 and 24 according to the connection rate of the corresponding repeater 璋 14. For example, if the repeater port 14a is configured to send and receive data packets via the media i6a at a 10Mb / s connection rate, the port exchange and control interface 26 connects the repeater port i4a to the repeater core 22. Similarly, 'if the repeater port i4b is configured to send and receive data packets via the media 16b at a connection rate of 100Mb / s', the port exchange and control interface 26 connects the repeater port 4b to 100MB / s Repeater heart 24. The port exchange and control interface 26 can be implemented as a plurality of multipliers. Depending on the connection rate determined for the corresponding relay pad 14, a port μ is selected to connect to the appropriate relay core 22 or 24. The repeater cores 22 and 24 are implemented as state machines and are configured to operate in accordance with IEEE802.3 section Θ and section 27, respectively. In detail, the 10Mb / s repeater state machine 22 is configured so that all repeater ports 4 operate in the 10Mb / s contention domain 'within the repeater or via coupling to the backplane 18 The 10-strong / 3 expansion bus' forms a single repeater operating in accordance with IEEE 8 02. 3, Section 9. If any

C: \ Program Files \ Patent \ 91524. Ptd Page 11 V. Description of the invention (7) The unit 14 connected to the state machine 22 in the 10Mb / s unit 14 detected the start of a valid packet 'repeater' Heart 2 2 will send and receive packets to all other ports connected to Heart 2 2 2 unless contention is detected. The repeater core 22 also supplies a packet to the 10Mb / s expansion bus that is lightly connected to the backplane 18 to facilitate the design of the multiple repeater 10. When the packet is retransmitted, the repeater core 22 ensures that the transmitted packet conforms to the IEEE802.3 signal amplitude, symmetry, and jitter specifications according to the internal clock of the repeater 10. In addition, the repeater core 22 will ensure that the preceding part is at least 56 bits before the start of the frame delimiter (SFD). The repeater core 2 2 also detects and responds to all contention conditions connected to the repeater core 2 2 port, as specified in Section 9 of IEEE 802.3, including detection of the contention status through the backplane 18 . Other identification functions are implemented by the repeater core 22 to ensure reliable transmission of content in the 10Mb / s contention domain, such as fragment extension and auto-segmentation / identification. The 10 OMb / s repeater core 24 is used as a state machine for configuration, so that all port ends operate in the 100 Mb / s contention domain in the repeater 10, or via the backplane 20 to form the first IEEE802.3u Single repeater as specified in Section 27. In detail, if any port 1 4 connected to the 100Mb / s heart 2 4 senses the start of a valid packet, the repeater heart 2 4 will send the received packet to all other ports Unless contention is detected. Repeated data is also supplied to the backplane 20 for sending to other t-relays connected to the backplane 20. As mentioned above, the 100Mb / s repeater core 24 ensures that the output packets on the transmit port comply with IEEE802.3u (Sections 24, 25, and 27) signal amplitude, symmetry, and jitter specifications.

Page 12 C: \ Program Files \ Patent \ 91524. Ptd 411676 5. Description of Invention (8). The signal sent is also retimed by the internal clock. Other identification functions are implemented by the repeater core 24 'to ensure reliable transmission of data in the 100Mb / s contention domain. The repeater 10 also includes an auto-negotiation unit 30, a management control logic 32, and a connection controller 34. The auto-negotiation unit 30 performs the auto-negotiation as defined in section 28 of iEEE802.2.3. The auto-negotiation unit 30 uses an auto-negotiation protocol to establish a connection between the repeater ports 14 corresponding to the node 12 according to the capacity of the corresponding node 12 and the selected data rate. For example, if the remote node 12a can transmit at 100Mb / s, the auto-negotiation unit 30 establishes a repeater port 14a and the remote network node 123 at a data rate of 100Mb / s (100BASE -TX) connection. However, if the remote node 2a cannot send and receive data packets at 100 Mb / s, the auto-negotiation unit 30 establishes a repeater 14a and the remote network node 12a at 10 Mb / s (10BASE-T) connection. In detail, the auto-negotiation unit 30 uses burst connection pulses called fast connection pulses (FLPs). The interval between this fast connection pulse is 55 microseconds and 100 to 14 microseconds. It can be ignored by the standard 10BASE-T receiver. The FLP burst contains information about the capabilities of the transmitting device (i.e. repeater 14). The remote network node 12 capable of sending and receiving al00Mb / s decodes the FLP burst to learn about the capabilities of the sending device 14. The repeater 10 also includes management control logic 32. The management control logic 32 is provided with an interface to the processor 40. This processor 40 is used as a network management medium. The management control logic 32 allows the network manager 40 to perform indirect storage of the control register in the network repeater 10 according to the configuration method of IEEE802.3u Section 22

C: \ Program Files \ Patent \ 91524. Ptd page 13 411676 5 'Explanation of invention (9) Take. The connection controller 34 'is configured to reduce the connection integrity based on the detection. ► The physical property is selected from 100Mb / s data rate to 10Mb / s data rate to change the repeater port 14 and the remote network node. The connection established between 12. In detail, the connection controller 34 is configured to detect and correct excess error rates on the Mb / s connection, for example due to cable connection or equipment errors. The integrity of each 100 Mb / s connection was determined by performing a high-speed connection integrity check on each 100 Mb / s connection. In detail, each repeater port 14 includes a 10Mb / s physical layer transceiver 42, a 100Mb / s physical layer transceiver 44, and a symbol error detector 46, a counter 48, and a timer 54. The symbol error detector α is configured to detect the symbol definition according to Table 1. When the repeater port 14 uses a 100Mb / s transceiver 44 to operate at a data rate of 100Mb / S, the corresponding connection symbol is incorrect. Number-ϋ Table 1: Symbol Definition Symbol (Hexadecimal) NRZ 4b / 5b ----- Explanation 0 0000 11110 ----- -—Data 0 --Data 1 ——ϋ £ 2__ Data 3 ~~ Material 4 ——ϋ5_ —— 1 0 0 0 1 01001 2 0010 10100 3 0011 " 1 — 10101 4 0100 01010 5 0101 01011 6 0110 01110

411676 V. Description of invention (ίο) Symbol (hexadecimal) NRZ 4 b / 5 b Description 7 0111 01111 Data 7 8 1000 10010 Data 8 9 1001 10011 Data 9 A 1010 10110 Data AB 1011 10111 Data BC 1100 11010 Data CD 1101 11011 Data DE 1110 11100 Data EF 1111 11101 Data F Idle undefined mu Idle symbol J 0101 11000 1 start delimiter 2 of the stream K 0101 1 0 0 0 1 start delimiter 2 of the stream 2 T undefined 01101 stream End of delimiter 2 1 R Undefined 00111 End of stream 2 of delimiter Η Undefined 00100 Error bit counter 4 8 can be configured to increase or decrease in response to each detected sign error from the PHY (Depending on the embodiment) Count value. As explained below,

C: \ Program Files \ Patent \ 91524. Ptd Page 15 411676 V. Description of the invention (11) The connection controller monitors the connection by counting the number of symbol errors detected on each port 1 and 4, and monitors the connection based on the detection. The number of sign errors relative to the specified threshold determines the integrity of the connection. If the number of detected symbol errors reaches the critical value of the relevant specified time interval, the connection controller 34 chooses to reduce the data rate corresponding to the corresponding network media 1 b to a reduced 1 Mbps data by disconnecting the connection. Rate, and the corresponding relay port 14 can only perform auto-negotiation by operating at 10 Mb / S. Therefore, the first connection is established at a data rate of 10Mb / s, reducing the probability of sign error β as shown in the first figure. 'The connection controller 34 includes Table 50 and the selector circuit 52. Table 50 stores a plurality of valid thresholds. The value 'selector circuit η selects one of the valid thresholds as a predetermined threshold corresponding to the unacceptable symbol error rate. The predetermined threshold is based on a selection signal from the management control logic, or in some other way. Choosing thresholds in an unmanaged system. Therefore, the repeater 10 can use a plurality of threshold values stored in Table 50t as the initial program planning, and the configuration register can also be programmed to select one of the registers, and choose to enable each register. Port-based drop operation. Figure 2 is a diagram showing the method of controlling transmission by monitoring the integrity of the connection and choosing to reduce the data transmission rate of repeater port 14. This repeater port U has the number of symbol errors detected by the connection. Specified threshold. Figure 2 illustrates the processing procedure for only one port. For any multi-bee repeater as shown in Fig. 丨, each has its own set of processing programs, and all processing will be performed simultaneously. The processing of a port starts at step 5 and 9 at this time.

C: \ Program Files \ Patent \ 91524. Ptd page 16 r--iu.676 ___ V. Description of the invention (12) It is decided that this port (for example, 14a) can be operated. In step 60, the error counter 48 is preset to the threshold value determined by the selection circuit 52. The timer 54 is set to generate a desired sampling interval. Then start to auto-negotiate according to IEEE802.3u. —Β At step 61, determine whether the connection at Mbps has been successfully established. Right is already established, then start processing steps 72 and 64 at the same time. If it is not established, the connection is set to lQMbps at step 62, and the port will try to establish a connection at that rate. If the establishment of the 10 Mb ps connection is completed immediately, the process of step 63 is performed '. In this step, any change state on the connection is continuously monitored. If no connection is detected, then step 63 is also processed, in which any symbol establishing a connection is monitored. At any state change (from 10MbpS2 good connection to no connection 'or from no connection to establishment of connection), the program execution will return to step 59. Steps 64, 72, 74 and 76 are different ways of trying to illustrate a process. The interval timer 54 counts down at a steady rate until zero. At the same time, the 'error monitor 46 continuously monitors the in-turn symbol stream, and decrements (decrements) the counter 48 once for each symbol error detected. Test the connection status according to the same rules 'the same these processes' as in step 6 3 as shown in step 7 4. When the timer 54 reaches 0 (step 76), the value of the test error counter 48 is as shown in step 78. If the value of the counter is greater than 0, the system (for example, a wired controller) encounters an acceptable error rate. Proceed to step, reset the counter and timer 'and start the monitoring cycle step by returning to steps 72 and 64.

Five Invention Explanatory Notes

The value of the right error counter drops to 0, which indicates that an unacceptable number of errors occurred during the sampling interval. The system proceeds to step 82. At this time, the temporary g bit referenced in step 59 is reset to notify the port end that it can only operate in the Mbps mode. The system then proceeds to step 62 'and attempts to operate in 10 Mbps mode. According to the disclosed embodiment, it is easy to monitor the 10 1) / 3 data connection to determine whether the connection integrity has not been performed at the required specifications, for example, due to poor cable or poor hardware circuit performance Wait. The disclosed configuration eliminates the need for a remote network manager 40 to continue to monitor the connection status to each network port. Moreover, depending on the detected symbol error rate reaching a predetermined threshold, the connection controller can effectively reduce the corresponding repeater by disconnecting the connection and establishing a new connection at a reduced data rate. Although the disclosed configuration illustrates a single connection controller 34 used for each repeater port, the connection controller 34 can be implemented alternately on each repeater port 14 and serves as a repeater port 14 It can independently monitor and control its own connection based on the detected symbol error rate. In addition, symbol error detectors, counters, and timers can be centrally configured with the link controller 34 to provide a more centralized architecture. Although the present invention has been described by way of examples, it should be understood that various changes or modifications without departing from the broad spirit and viewpoints defined in the scope of the patents attached to the present invention should all belong to the scope of the present invention.

Claims (1)

__ 4tl67e 6. Scope of patent application h A method for controlling transmission of a network relay, the method includes the following steps:. Establish a connection with a remote network node via a network medium at a specified data rate; detect by counting Monitor the connection; and determine the integrity of the connection based on the number of detected symbol errors relative to a predetermined threshold. 2. The method according to item 1 of the scope of patent application, wherein the monitoring step includes: reducing the counter value in response to detecting that it caused a symbol error. 3. The method according to item 2 of the scope of patent application, wherein the step of reducing comprises suspending the decrease of the counter if the counter is at a value of zero. 4. The method of claim 2 of the patent scope, wherein the judging step includes judging the integrity of a specified time interval equal to approximately one millisecond, and the time interval is controlled by a timer counting down to zero. 5. If the method in the first scope of the patent application includes the step of selecting a predetermined threshold value from a plurality of thresholds β 6. In the method in the first scope of the patent application, it also includes the provision of the required threshold through the serial interface Steps to the network repeater. 7. As for the method in the scope of applying for patent, the method still includes the step of generating a signal to indicate the overall problem of the connection in response to the number of sign errors of the lucky test reaching a prescribed threshold. δ. The method according to item 7 of the scope of patent application, which makes the signal correspond to at least one of the LED signal and the interrupt signal of the network management medium. 9_ If the method of applying for the scope of the first item of the patent scope, including the response to the detection C: \ Program Files \ Patent \ 91524. Ptd Page 19 4tl676 '-------- VI. Application number --- The number of errors exceeds the specified threshold and selectively reduces the data rate on the network media Steps to reduced data rate. 0 'The method according to item 9 of the patent application range, wherein the monitoring step includes a step of reducing the error count in response to detecting each of the symbol errors, and the method further includes resetting the counter value and responding to the timer reaching. The step of resetting the error counter to a critical value. 1 1. The method according to item 9 of the scope of patent application, wherein the step of selectively reducing comprises reducing the data rate according to the enable signal 'response to the number of detected symbol errors exceeding a predetermined threshold value ^ ~ 12. As in the first scope of patent application The method of the method, wherein the determining step includes: reducing a timer's timing value from a prescribed time interval; reducing a first counter that is initially set at a first threshold value in response to a detected sign error; and reacting when the timer reaches 0. Value, whether the first counter reaches a value of zero, and the integrity is determined according to the fact that the first counter that reaches a value of zero agrees with the timer that reaches a value of zero. 13, A network repeater 'has a plurality of repeater ports configured to send and receive data packets between remote network nodes through respective network media. The network repeater includes: A repeater core configured to send and receive data packets between the first group of repeater ports according to the first data rate; & The first data rate is slower. The second data rate of the data between the two groups is to send and receive repeater port packets. C: \ Program Files \ Patent \ 91524. Ptd Page 20 6. The patent application range auto-negotiation unit is configured to select one of the first and second data rates to establish a t-relay port and an individual remote network. Way nodes ^ via individual network media connections; a symbol error detector for detecting the number of symbol errors associated with a predetermined time interval operating on at least one of the connections; and a controller , Used to selectively change at least one connection from the first data rate to the second data rate according to the number of sign errors that reach a predetermined threshold. 1 4. The repeater according to item 13 of the patent application scope, wherein the symbol error detector includes a counter configured to decrease the count value in response to each detected symbol error. 15. If the selection interface of the management medium in item 14 of the scope of the patent application is used for the critical value, the controller outputs a notification signal to the network management in response to the arrival # The relay device also includes the connection to the network and the reception from the network. The sign error counter that manages the stipulated stipulated critical value.