CN101778464A - Communication power adjustment method and device - Google Patents

Abstract

The present invention provides a communication power adjustment method and device, which are used to detect the communication status between a network port and a link object, and adjust the data transmission power and/or data reception power of the network port according to the detected communication status, so as to achieve the purpose of taking into account the accuracy of data transmission and reception and the power saving of data transmission and reception.

Description

Communication power adjustment method and device

technical field

The present invention relates to a power regulation method and device, in particular to a power regulation method and device for communication.

Background technique

In a network environment, data transmission between a computer and a link object needs to be carried out in a wired or wireless manner. When using wired transmission, the energy-saving technology of the existing network link performs energy regulation and configuration on the network transmission equipment of the computer according to the length of the link cable, that is, the power consumption of the transmission equipment when the cable is short is lower than that when the cable is shorter. Long-term power consumption, so as to achieve the purpose of energy saving and environmental protection. However, since different link objects will affect the power consumption allocation of the network transmission equipment, if the conditions of the link objects are not considered, the error rate of data transmission will often be high.

Contents of the invention

Therefore, one of the objects of the present invention is to solve the problems of the prior art.

Another object of the present invention is to provide a communication power adjustment method and device that can save transmission power and ensure transmission quality.

An embodiment of the communication power regulating device of the present invention is in the form of an integrated circuit installed in a network switch, but it can also be installed on a network card or other communication devices. The power regulating device can be realized by hardware (such as a combination of logic circuits); it can also be realized by a combination of hardware and firmware (such as a combination of a computing unit and a storage unit loaded with firmware); it can also be realized by using hardware and software collocation (such as a computer system with communication functions) to achieve.

The foregoing switch comprises: a plurality of ports; a packet analyzer for receiving packets from the ports and analyzing packet information; an error detector receiving link reliability of the packet information from the packet analyzer; Energy-saving controller to adjust power.

The method executed by the detector includes: preset and store error rate critical values R _L and R _H , receive packet information from the port and calculate the error rate, and then perform one of the following judgments: determine whether the error rate is less than R _L 1. Judging whether the error rate is greater than R _H ; if the error rate is less than _RL , an adjustment command to lower the first-order power is generated; if the error rate is greater than R _H , an adjustment command to increase the first-order power is generated and sent to the energy-saving controller; if If the error rate is neither less than R _L nor greater than R _H , no adjustment instruction is generated. In addition, when the detector detects that the power is equal to a preset low power point CP _MIN , it notifies the energy-saving controller to fix the power at CP _MIN and no longer lowers it; when it detects that the power is equal to a preset high power point CP _MAX , the power will return to the preset CP _initial ; when it is detected that the error rate is oscillating between the two critical values _RL and R _H , and the number of oscillating times reaches the default value within the predetermined time, the power reduction will be turned off The energy-saving mechanism does not generate an adjustment command for reducing the first-order power to the energy-saving controller.

The energy saving controller adjusts the packet transmission or reception power of the port according to the adjustment instruction.

During the network connection stage, when the detector detects that the network is performing initialization operations such as DHCP or BOOTP, the energy-saving controller adjusts the power of the port to the highest value in advance; after the initialization operation is completed, it is adjusted back to the energy-saving mode. The preset CP _initial value. When the initialization is completed, if the link is successful within k seconds and the number of link failures within m seconds after the link is successful exceeds the default value n, an adjustment command to increase the first-order power will be generated.

The detector can calculate the error rate for the packets transmitted from the port and the packets received, and accordingly generate adjustment instructions to adjust the transmit power and receive power of the port, or only for the packets transmitted or received Groups, one of which calculates the error rate and adjusts the power, and the other only needs to be adjusted accordingly.

The method for calculating the error rate by the detector can define the error rate to be equal to the ratio of the number of packets Mcnt to the number of errors Ecnt. The value of Mcnt increases with the received packet, and the value of Ecnt increases when the received packet information shows a packet transmission error, and when the The detector counts the Mcnt value up to a preset Mmax value, divides Mcnt and Ecnt by a value, or resets Mcnt and Ecnt to zero. When the detector judges that the received packet has the characteristics of a request and a response, the number of packets is added and the packet information of the packet is recorded. The packet information includes the address and port number of the destination end and the source end, as well as the storage sequence number. The packet transmission direction can be seen from the packet information.

When the detector judges that the existing packet information has been recorded and receives another packet, the newly received packet is compared with the recorded packet information. For TCP/IP packets, the contents of the comparison include: destination Whether it matches the address and port sequence number of the source, and the relationship between the response sequence number or packet sequence number of the newly received packet and the saved sequence number of the recorded packet-when the transmission direction of the newly received packet is the same as the recorded packet, Then compare whether the packet sequence number of the newly received packet is equal to the saved sequence number of the recorded packet. If it is equal, it means that there is a packet retransmission phenomenon; when the transmission direction of the newly received packet is opposite to the recorded packet, compare the newly received packet Whether the response sequence number of the obtained packet is greater than the saved sequence number of the recorded packet, if not, it means that there is a phenomenon of packet retransmission. If there is a phenomenon of packet retransmission, the number of errors will be added, and the recorded packet information will be deleted.

Taking ICMP packets as an example, when the detector judges that the received packet has the characteristics of a request and a response, it then judges whether there is a response to the outgoing packet within a predetermined time. If there is no response within the time limit, the error count will be added.

For a packet with only a serial number for error detection, when judging the receiving capability of a port, the detector calculates the error rate by using the serial number to determine whether the packet is damaged, and taking the damage rate as the error rate.

Through the following embodiments of the present invention, it is possible to better understand how the present invention is implemented and the effects it achieves.

Description of drawings

Fig. 1 is the block diagram of the embodiment of power conditioning device of the present invention;

Fig. 2 is a flow chart of the linking phase of the embodiment of the power adjustment method of the present invention;

Fig. 3 is a flow chart of adjusting port power according to an embodiment of the power adjusting method of the present invention;

Fig. 4 is in Fig. 3, the flow chart about calculating transmission (Tx) error rate;

Fig. 5 is a schematic diagram of packet transmission between a switch and a link object B;

Fig. 6 is in Fig. 3, the relevant flow chart of calculating receiving (Rx) error rate; And

FIG. 7 is a schematic diagram of packet transmission between a switch and a link object B. FIG.

Explanation of main component symbols

1 Exchanger 3 Energy-saving controller

10a Port A Link object

10b Port B Link object

11 Packet analyzer S ₁₀ ~ S ₁₉ steps

12 Packet transmission engine S ₂₁ ~ S ₂₉ steps

100 Power adjustment device S ₂₅₁ , S ₂₆₁ steps

2 Detector S ₃₁ ～～S ₄₃ steps

Detailed ways

Please refer to FIG. 1 , which is a schematic diagram of an embodiment of a communication power regulating device 100 of the present invention, which is presented in the form of an integrated circuit installed in an Ethernet switch (network switch) 1 . This switch 1 is connected between a plurality of link objects. The switch 1 described in this embodiment includes a packet analyzer (packet parser) 11, the power regulator 100, a packet forwarding engine (packet forwarding engine) 12, and ports (ports) connected to link objects A and B respectively 10a, 10b. Switch 1 sends the packet of link object B to port 10b through port 10a, and then transmits it to link object A, and receives the packet of link object A by port 10b, transmits it to port 10a, and then transmits it to link object B.

In this embodiment, the power regulating device 100 mainly utilizes a protocol layer having packet error detection and control characteristics in the network International Standards Organization (ISO) model, such as a data link layer (data link layer) or a transport layer (transport layer). etc. to implement the data power adjustment method of the present invention, and further notify the physical layer (physical layer) to adjust the consumed power, such as adjusting the operating current and voltage of any circuit included in the physical layer. The power adjustment device 100 includes a link reliability error detector (reliable link error detection, hereinafter referred to as the detector) 2, and an energy-saving controller responsible for adjusting the transmission and reception power level (power level) of the physical layer in the network protocol 3. Please note that in this embodiment, the power regulating device 100 can be implemented by hardware (such as a combination of logic circuits), or by a combination of hardware and firmware (such as a combination of a computing unit and a storage unit loaded with firmware). To achieve this, those skilled in the art can select an appropriate method to implement the power adjustment device 100 according to the disclosure of this specification. In another embodiment of the present invention, the realization of the communication power adjusting device of the present invention is that a computer loads and executes a program through its operating system, and then controls the transmission power of a communication device coupled with the computer.

The packet analyzer 11 of the switch 1 analyzes each field in the header of each incoming packet, such as parameters such as a label (flag), destination/source address, protocol type, etc., and transmits these parameters to the detection of the power regulating device 100 Device 2. The packet transmission engine 12 is responsible for determining the packet path and normal packet transmission. Please note that the present invention focuses on the power conditioning device 100. As for the packet analyzer 11, the packet transmission engine 12 and other necessary circuits (not shown in FIG. 1 ) included in the switch 1, they belong to the scope of the prior art and will not be described here. It will be described in detail in this embodiment.

Detector 2 is used for long-term detection and the quality of transmission and reception between the link object, and mainly analyzes whether there is an error in each parameter analysis in the field introduced by packet analyzer 11, and judges whether the preset power adjustment condition is reached; To adjust the power, an adjustment command is generated and sent to the energy-saving controller 3 . The energy-saving controller 3 adjusts the packet transmission power and/or packet reception power of the ports 10a and/or 10b according to the adjustment instruction.

In this embodiment, the detector 2 judges whether there is an error (such as packet loss), whether the power adjustment condition is met, and the related technology of the power adjustment by the energy-saving controller 3 is described in detail below with the flow chart. The following is an example of the data transmission between the switch 1 and the link object B through the port 10a, and the adjustment of the receiving and transmitting power of the port 10a.

With reference to FIG. 2, when the energy-saving mode of the power regulating device 100 of the present invention is activated, in order to avoid link failure due to insufficient power, or multiple link success (link up) and link failure (link down) oscillations, the power regulating device 100 The power adjustment method of step S ₁₀ to step S ₁₉ is executed first, so as to adjust the power of the port 10a to make the link successful and stable.

Step S ₁₀ - when the detector 2 detects that the link objects are performing dynamic host configuration protocol (dynamic host configuration protocol, referred to as DHCP) or bootstrap protocol (bootstrap protocol, referred to as BOOTP) and other initialization operations (such as allowing the computer to obtain an address, network shielding, etc.) etc.), in order to avoid link failure due to insufficient power of port 10a or link instability due to power adjustment, this step pre-adjusts the power of port 10a to the maximum value, such as adjusting the operating current and/or voltage of port 10a to the maximum value; after the initialization operation is completed, adjust it back to a default value of CP _initial . The CP _initial value in this embodiment is determined based on the characteristics of the actual circuit and the physical layer information (such as the power range in which the physical layer can work normally), and it is based on the information disclosed in this manual by those with ordinary knowledge in the technical field. to implement. It should be noted that, if no initialization operation is required, the step S ₁₀ can also be skipped on the premise that the power of the port 10 a is preset as CP _initial .

Step S11—the detector 2 sets the link count number LinkCnt=0.

Step _S12 —the detector 2 detects the connection status of the port 10a, whether the connection is successful within k seconds (for example, 60 seconds)? If yes, go to step S ₁₃ , if not, it usually means that the cable is not connected (meaning if no initialization operation is required), go back to step S ₁₁ . In addition, in the case where the aforementioned initialization operation needs to be performed, if the detector 2 detects that the connection status of the port 10a is not successfully connected within k seconds, it may be that the preset power CP _initial is too small, so the preset power CP _initial Increase by one step (not shown in Figure 2), and then re-check whether the port 10a is successfully connected within k seconds, if so, enter step _S13 , if not, then increase the preset power CP _initial by one step, and so on until the preset power Set the power CP _initial to increase until CP _MAX .

Step _S13 - the detector 2 detects whether a link failure occurs within m seconds (for example, 30 seconds)? If yes, it means that the connection status may be unstable, and enter into step _S14 and the following process of increasing the number of connection counts; if not, it means that the connection status is stable, and proceed to steps _S18 and _S19 .

Step _S14 - Detector 2 increments the LinkCnt value by one.

Step S ₁₅ - the detector 2 judges whether the value of LinkCnt is greater than the default value n? If it is greater than the default value n, it means that the link status is indeed unstable, and go to step S ₁₆ ; if not, the link failure may be only a sporadic situation, which needs to be observed, and go to step S ₁₇ .

Step _S16 - the energy-saving controller 3 increases the power of the port 10a by one level, that is, CP=CP+1, and then returns to step _S11 to re-check whether the power is sufficient.

Step S ₁₇ - the energy-saving controller 3 maintains the current power CP, then returns to step S ₁₂ , and continuously observes whether the link succeeds or fails repeatedly.

Step _S18 - If the aforementioned step _S13 detects that there is no link failure within m seconds, it is determined that the link status is stable, and LinkCnt=0 is set.

Step S ₁₉ -Enter normal working state, establish reliable connection flow (reliable connection flow), or data flow (data flow).

Please refer to FIG. 1 and FIG. 3. When a reliable data flow is established, that is, it enters a normal working state. In addition to confirming whether the packet reception power (Rx power) of the port 10a of the switch 1 is sufficient to successfully receive the packet, the power adjustment device 100 also needs to Confirm whether the transmit power (Tx power) of the port 10a is sufficient for the link partner to receive packets smoothly, so as to adjust the power. The detector 2 in this embodiment continuously randomly samples and detects any established data stream, but as long as the calculation speed allows, other embodiments of the present invention can also detect all packets of the data stream. In the following, the monitoring of transmission (Tx) capability is taken as an example for illustration.

Step S ₂₁ - the energy saving controller 3 determines an initial power CP _initial . The initial power CP _initial mentioned in this step is the power CP value determined after calculation using the flow chart in FIG. 2 .

Step S ₂₂ - the detector 2 starts to detect the data stream and calculate the error rate at the same time. The calculation method of the error rate in the present invention is mainly to detect a reliable connection packet (reliable connection packet) with request (request) and response (response) characteristics, such as a transmission control/network communication protocol (hereinafter referred to as TCP/IP) packet , the grouping of address resolution communication protocol (hereinafter referred to as ARP), or the grouping of Internet group management communication protocol (hereinafter referred to as ICMP), after confirming the grouping information, promptly further judge whether the link object B has received the outgoing grouping of port 10a, If it is found that the packet is lost, the number of errors will be increased to increase the error rate. The method of calculating the error rate in detail will be described in detail later in conjunction with the flow chart in Figure 4.

Step S ₂₃ -When the error rate is known, the detector 2 judges whether the error rate is less than _RL ? In the present invention, _RL is a preset low threshold (threshold), such as 3/100,000. If the error rate is less than _RL , it means that the condition of transmitting packets is very stable, and step _S24 can be entered; if the error rate is not less than _RL , Then it is necessary to enter the right half of the process in Fig. 3 (steps S26-S29), and then determine whether the error rate is higher than the high critical value, that is, R _H , for example, 10/100,000.

Step _S24 - the energy-saving controller 3 first judges whether the power CP is equal to a preset CP _MIN ? If yes, go to step _S241 ; if not, go to step _S25 .

Step S ₂₄₁ - the energy-saving controller 3 fixes CP=CP _MIN , that is to say, when the transmission power has reached a preset low point, the transmission power is fixed and no downward adjustment is made, and the data is returned to step S ₂₂ to continue detecting flow. The default value of the power CP _MIN in this embodiment is determined by the minimum power supported by the physical layer, but the determination method is not limited to this embodiment.

Step _S25 - Since the error rate was judged to be less than _RL in the previous step _S23 , it represents that the packet transmission status is very stable, and in step _S24 it is judged that the power has not reached the minimum value, the detector 2 can generate an adjustment command and send it to the energy-saving The controller 3 enables the energy-saving controller 3 to adjust the transmission power of the port 10a to decrease by one level according to the adjustment instruction, CP=CP-1. The "level" of reducing or increasing the power in this embodiment refers to the preset level of the physical layer. Then return to step S22 to continue detecting the data flow.

When step _S23 determines that the error rate is not less than _RL , and the error rate may be equal to or greater than _RL , it is necessary to enter the right half of the process in Figure 3:

Step S ₂₆ - determine whether the error rate is greater than R _H ? In the present invention, R _H is a preset high critical value. If the error rate is higher than R _H , it means that the condition of transmitting packets is unstable, and it is necessary to enter step _S27 ; if the error rate is not higher than R _H , it means that the error rate is between Between _RL and R _H (including _RL and R _H ), then proceed to step S ₂₆₁ .

Step S ₂₆₁ - Since the error rate so far is acceptable, the current CP value is fixed, and then returns to step S ₂₂ to continue detecting data streams.

Step S ₂₇ - In the case that the error rate is higher than R _H , in this step the detector 2 judges whether the power CP is equal to a preset CP _MAX ? If yes, go to step _S28 ; if not, go to step _S29 .

Step S ₂₈ - Since the power CP has reached the high power point CP _MAX , the error rate is still greater than R _H , indicating that the error rate is not caused by insufficient transmission power, so the energy-saving controller 3 sets CP back to CP _initial , and returns to step S ₂₂ Continue to detect data flow. The default value of power CP _MAX in this embodiment is determined by testing the characteristics of the actual circuit or according to the specifications of the actual circuit or rules of thumb. The power required to transmit a packet will not be higher than CP _MAX in theory, but CP _MAX is determined in a different way. This example is limited.

Step S ₂₉ -Because the status of transmitting packets is unstable, it may be due to insufficient transmission power, so the detector 2 generates an adjustment command and sends it to the energy-saving controller 3, so that the energy-saving controller 3 adjusts the transmission to the port 10a according to the adjustment command The power is increased by one order, CP=CP+1. Then, return to step S22 to continue detecting the data flow.

The present invention controls the opening or closing of the left half and the right half of the process in Fig. 3 respectively by setting the values of _RL and R _H , the process of the left half adjusts the power when the power can be lowered further, and the process of the right half The process is to adjust the power when the power should be increased again, and realize the mechanism of power saving auto-fallback and aggressive. Although the above-mentioned embodiment is illustrated by first entering the automatic lowering process in the left half of FIG. 3 , it is not limited thereto. It is also possible to first enter the right half of the process as shown by the dotted line in FIG. 3 . However, in actual operating conditions, it is possible that the CP value oscillates between the two critical values R _L and R _H. In this embodiment, the number of oscillations of the CP value can be counted again (that is, the CP value changes from increment to decrement and The number of transitions from decreasing to increasing), when the number of oscillations reaches a default value within a predetermined time, the power regulating device 100 can turn off the energy-saving mechanism of the left half, so as to avoid the CP value oscillation caused by reducing the power.

Returning to the calculation part of the error rate in Figure 3, please refer to Figure 1, Figure 4, Figure 5 and the following description. FIG. 4 is a flow chart showing the capability of tracking port 10a for TCP/IP packet transmission (Tx), by analyzing the packets sent out by port 10a (as shown by black dots in FIG. 5 ), to understand the transmission (Tx) of port 10a ability. FIG. 5 is a schematic diagram of packet transmission between the port 10a and the link object B and the port 10b (connected to the link object A).

Step S ₃₁ - In the case of entering the normal working state, the detector 2 initially sets the value of the number of packets Mcnt to 0, the value of the number of errors Ecnt to 0, and starts timing.

Step S ₃₂ - Receive a packet, which may be a packet that port 10a wants to transmit to link object B, or a packet that port 10a receives from link object B and wants to deliver to port 10b (link object A). Detector 2 judges whether the received packet has the characteristics of request and response from the parameters in the packet field? If yes, it belongs to the group to be processed in this embodiment, and proceeds to step _S33 ; if not, ignores this group, and starts from step _S31 .

Step _S33 - judging whether group information has been recorded? If the port 10a transmits a packet (see Figure 5), the buffer (not shown) of the switch 1 will record the packet information; if the port 10a once received a packet (see Figure 7), the buffer of the switch 1 The device will also record the group information. Since FIG. 4 is an example illustrating the method flow of observing the transmission (Tx) capability between the port 10a and the link object B, this step is to determine whether there is any record of the packet transmitted to the link object B in the buffer. If no grouping information has been recorded, then enter the right half of the process - the process of recording grouping information (steps S ₃₄ ~ S ₃₇ ) and then return to step S ₃₂ ; if there is already grouping information recorded, then enter the left half of the process - judgment Flow of whether a transmission error occurs (steps S ₃₈ -S ₄₂ ).

The process of recording group information

Step _S34 - The detector 2 increments the packet number, that is, Mcnt=Mcnt+1, and resets the timer to zero, and then enters step S35.

Step S ₃₅ - the detector 2 then judges whether Mcnt is equal to the preset number of groups Mmax, if yes, proceed to step S ₃₆ ; if not, skip to step S ₃₇ .

Step S ₃₆ - Dividing Mcnt and Ecnt by a value, for example, dividing by 2 together. This step is performed in order to continuously calculate the ratio of Ecnt/Mcnt. The method is not limited to the value of dividing by one, and the value of Mcnt and Ecnt can also be directly set to 0. Then go to step _S37 .

Step S ₃₇ - record the group information of this group, and return to step S ₃₂ after the recording is completed. The grouping information recorded in this step mainly includes the address and port serial number of the destination and source of the grouping, and the storage serial number. In terms of monitoring the transmission (Tx) capability of port 10a, the packet information includes {[IP-A, PortNum-A, IP-B, PortNum-B], KSN-A}, which represents the packet transmission direction as the link object A teleports to link object B. When the record is completed, the detector 2 will determine that "the information of the group has been recorded" in step _S33 when it judges the next group.

The process of judging whether a packet is lost

Step _S38 - When the aforementioned step _S33 judges that the grouping information has been recorded, this step is aimed at the information of the newly received grouping, judging whether the address and port number of the destination end and the source end match the recorded information, in short, It is confirmed whether the packet is a packet transferred between the port 10a and the link partner B. If yes, proceed to the next step S ₃₉ ; if not, it means that the packet is transmitted by a different link, ignore it, and return to step S ₃₂ .

Step _S39 - The detector 2 then judges the transmission direction of the packet. If it is the same as the recorded packet transmission direction, that is, it is transmitted from the link object A to the link object B, it means that the packet is to be sent out to the link object B through port 10a If the packet transmission direction is opposite, it means that _the packet is sent back by link object B and the port 10a is to be delivered to link object A, then proceed to step S ₄₀ .

Step _S40 - compare the response sequence number B_ack of the packet returned by the link object B with the stored sequence number KSN-A of the recorded packet, the stored sequence number is transmitted from port 10b to port 10a earlier and will be sent to the link object The packet sequence number A_seq of the packet of B.

For example, the buffer has recorded a grouping information of (KSN-A, A_ack)=(100, 20), and now a grouping direction is received from the linking object B to the linking object A, and the sequence number of the packet is the same as The response sequence number is (B_seq, B_ack). In this step, it is judged whether the B_ack of the returned packet is greater than the KSN A of the previously recorded packet. If the response sequence number B ack is greater than the saved sequence number KSN-A, for example (B_seq, B_ack) = (20, 101), it means that the packet is delivered successfully, and the link object B requires port 10a to transmit the next packet in the return packet, which also represents the port 10a The transmission (Tx) capability is normal, enter step _S43 to clear the recorded group information, and then return to step _S32 .

If the response sequence number B_ack is less than or equal to the stored sequence number KSN-A, for example (B_seq, B_ack) = (20, 99), it means that the link object B has not received the packet with the sequence number KSN-A sent by the port 10a, and the link object B requires re- The packet is transmitted, so enter step S ₄₃ , the detector 2 increments the error number, that is, Ecnt=Ecnt+1, and then enter step S ₄₃ .

Step _S41 - If it is determined in step _S39 that the packet and the recorded packet have the same transmission direction, then determine whether the packet sequence number A_seq of the packet to be sent to the link object B is equal to the recorded KSN-A. If the packet sequence number A_seq of this packet is equal to the recorded KSN-A, it means that this packet is retransmitted, and enters step _S42 to make Ecnt=Ecnt+1; if not, then this packet is just another one to be sent to the link object B group, ignore it.

The aforementioned step _S34 and step _S42 both affect the error rate Ecnt/Mcnt value shown in FIG. 3 , and are used for calculation after step _S22 of the process shown in FIG. 3 .

In this embodiment, Ecnt/Mcnt is set as an example for illustration. This method needs to continuously calculate the ratio of Ecnt/Mcnt. However, the present invention is not limited thereto, and the method of step S36 may not be used to allow Ecnt/Mcnt to be continuously calculated, but to set the critical value of Ecnt and make the critical value change with the cumulative amount—for example, the aforementioned _RL is set as 3/100,000 means that the number of Ecnt reaches 3 within 100,000 Mcnt, and the error rate reaches the critical value. When 100,000<Mcnt 200,000, the number of Ecnt needs to reach 6, and the error rate reaches the critical value; or use the calculation of the number of errors and set The critical value of the number of errors is used to judge the packet transmission status, and whenever Mcnt reaches Mmax, Ecnt is cleared, and Mcnt and Ecnt are reset to zero.

The foregoing monitoring of the transmission (Tx) capability of the port 10a is based on the premise that the reception (Rx) capability is normal, that is, the transmission (Tx) power setting may be fine, but the reception (Rx) power setting may be problematic. , so as to mistakenly think that there is a problem with the transmission (Tx) capability and adjust the transmission (Tx) power. Therefore, in order to avoid possible problems with the power setting of the receiving (Rx), when the transmitting (Tx) power of the port 10a is adjusted, the receiving (Rx) power can be adjusted at the same time, or can be monitored and adjusted separately.

Please refer to Fig. 1, Fig. 6 and Fig. 7, the receiving (Rx) power of monitoring port 10a is also roughly the same as that of Fig. 4, but the grouping of records is different. In step _S37 , it is recorded with the content of grouping information {[IP-B, PortNum-B, IP-A, PortNum-A], KSN-B} illustrates that the information represents that the packet transmission direction is transmitted from the link object B to the link object A; and in step _S40 is compared For A_ack>KSN-B, in step _S41 compare B seq=KSN-B.

As far as observing the Rx status is concerned, the received packet does not necessarily have to have the characteristics of request and response, and may only have a number for error detection, such as a CRC (Cyclic Redundancy Check) code. In this case, when judging the Rx capability, a method similar to the above can be used to make another judgment, and detect whether the packet damage rate (ie, the CRC error rate) of the received packet is too high.

The present invention can detect its error rate for any protocol with request and response characteristics—for ICMP, the tool program PING can be used to send ICMP packets, and the error rate can be calculated by testing whether the ICMP packet has a response or no response after timeout ; In the case of ARP, the error rate is also calculated by testing whether there is an ARP packet response. The obtained error rate returns to the power adjustment process shown in FIG. 3 for calculation.

To sum up the above, the power adjustment device 100 and the power adjustment method of the present invention can reduce the power as much as possible to save energy while ensuring the quality of transmission and reception, and can avoid data transmission errors caused by power reduction.

But the above-mentioned person is only preferred embodiment of the present invention, when can not limit the scope of the present invention implementation with this, promptly all the simple equivalent changes and modifications that are done according to the patent scope of the present invention and the content of the description of the invention, All still belong to the scope that the patent of the present invention covers.

Claims (10)

1. power of communications control method comprises:

One default power to one communications component is provided;

Utilize this default power to set up linking between this communications component and a linked object;

Utilize a link power to carry out communicating by letter between this communications component and this linked object;

Calculate the error rate when communicating between this communications component and this linked object;

Relatively this error rate and a critical value, and produce a comparative result; And

Determine whether adjusting this link power according to this comparative result;

Wherein this link power is between maximum link power and a min-link power, and should link power before carrying out this set-up procedure more than or equal to this default power.

2. the method for claim 1, wherein utilize this default power to set up this communications component and comprise with the step that links between this linked object:

Detect between this communications component and this linked object and whether link success;

If it is unsuccessful to detect link, then heighten this default power to set up linking between this communications component and this linked object; And

That is utilized when linking successfully between this communications component and this linked object should link power as this by default power.

3. the method for claim 1, wherein relatively the step of this error rate and this critical value comprises:

Judge that whether this error rate is less than a critical value R _L, and produce one first judged result; And

Determine whether downgrading this link power according to this first judged result.

4. method as claimed in claim 3, wherein relatively the step of this error rate and this critical value comprises:

Judge that whether this error rate is greater than a critical value R _H, and produce one second judged result; And

Tie fixed this link power of whether heightening according to this second judged result;

This critical value R wherein _HGreater than this critical value R _L

5. the method for claim 1, the step of wherein calculating this error rate comprises:

When communicating between this communications component and this linked object, detect the number meet a pre-conditioned unit data;

Information according to this unit data determines whether increasing progressively a wrong number; And

Number and this mistake number according to this unit data decide this error rate.

6. method as claimed in claim 5, wherein whether this pre-conditioned this unit data that comprises has request and the characteristic that responds.

7. method as claimed in claim 5, wherein this information comprises the sequence number of this unit data.

8. the method for claim 1, wherein this communications component is an Ethernet port, this unit data is a grouping.

9. a power of communications adjusting device is connected with a port; This device comprises:

Error rate critical value R is preset and stored to the error detector of one link reliability _L, R _H, receive grouping information and mistake in computation rate, and then carry out a following wherein judgement: judge that whether this error rate is less than R from this port _L, judge that whether this error rate is greater than R _HIf error rate is less than R _L, then produce the adjustment instruction that downgrades single order power; If error rate is greater than R _H, then produce the adjustment instruction of heightening single order power and be sent to an energy-saving controller; If error rate is non-less than R _LAlso non-greater than R _H, then do not produce and adjust instruction; And

This energy-saving controller is adjusted instruction according to this, and this port adjustment grouping is transmitted or received power.

10. power of communications adjusting device as claimed in claim 9, wherein, this detector definition error rate equals the ratio of packet count Mcnt and wrong number Ecnt, and the Mcnt value increases progressively with obtaining grouping, and Ecnt increases progressively when the worthwhile grouping information of obtaining shows packet transport error.

Images