HK1120934B

HK1120934B - Method and system for power over ethernet

Info

Publication number: HK1120934B
Application number: HK08112356.5A
Authority: HK
Inventors: 韦尔.威廉.戴博; 明夏因．希赫
Original assignee: 美国博通公司
Priority date: 2006-12-19
Filing date: 2008-11-11
Publication date: 2012-11-16

Description

Power over ethernet system and method

Technical Field

The present invention relates to methods for network cabling systems, and more particularly, to a system and method for discovering cable types for power over ethernet (PoE) applications.

Background

The IEEE 802.3af PoE standard provides a framework for transferring power from Power Sourcing Equipment (PSE) to Powered Devices (PDs) over ethernet cabling. In this PoE process, first an efficient device detection is performed. This detection process identifies whether the power device is connected to an active device to ensure that power is not being supplied to the non-PoE capable device.

After finding a valid PD, the PSE optionally performs power classification. IEEE 802.3af defines five power classes for PD devices. Completion of the power classification process may enable the PSE to manage the power to be transferred to the various PDs connected to the PSE. If a particular power level is identified as being available for a particular PD, then the PSE may allocate the appropriate power for that PD. If no power allocation is performed, then a default classification may be used, where the PSE will provide the full 15.4W of power to a particular port.

For efficient operation of a PSE, it is important to manage the budget of power allocated to each PD connected to it. In high coverage (Broad Reach) applications of PoE, where the PD is connected to a PSE using an ethernet cable greater than 100 meters (e.g., 300- > 500 meters), management of the power budget is even more important. Typically, the total amount of power that can be allocated to each PD is limited by the PSE capacity. Therefore, a mechanism is needed such that the PSE can identify the exact amount of power each port should budget.

Disclosure of Invention

A system and/or method for controlling power delivered to a powered device, substantially as shown in and/or described in connection with at least one of the figures, as set forth more completely in the claims.

According to one aspect of the invention, a power over ethernet system comprises:

the power consumption equipment detection assembly is used for detecting the existence of power consumption equipment, and the power consumption equipment is connected with the power supply equipment through an Ethernet cable;

a cable detection component for measuring electrical characteristics of the Ethernet cable; and

a power controller for controlling power distribution to the power source equipment ports based on the type of the Ethernet cable indicated by the measured electrical characteristic.

Preferably, the cable detection assembly measures an insertion loss of the ethernet cable.

Preferably, the cable detection component measures crosstalk of the ethernet cable.

Preferably, the cable detection assembly measures the length of the ethernet cable.

Preferably, the power controller controls power distribution based on the type and length of the ethernet cable.

Preferably, the power controller controls the power distribution based on a resistance of the ethernet cable determined using the type of the ethernet cable.

Preferably, the power controller controls power distribution based on authentication of the ethernet cable.

Preferably, the power controller identifies a power budget allocated to the port.

According to one aspect of the invention, a power over ethernet method comprises:

determining a type of the Ethernet cable based on the measured electrical characteristic upon connecting the powered device to the power device port via the Ethernet cable; and

allocating a power budget to the power source equipment port, the allocated power budget being based on the determined type of the Ethernet cable.

Preferably, the determining comprises measuring an insertion loss of the ethernet cable.

Preferably, the determining comprises measuring crosstalk of the ethernet cable.

Preferably, the type determination includes determining whether the ethernet cable is a type 3 ethernet cable.

Preferably, the allocated power budget is based on the determined type and the determined length of the ethernet cable.

Preferably, the method further comprises determining a resistance of the ethernet cable using the determined type of the ethernet cable and the measured length.

determining the type of the Ethernet cable connecting the power supply equipment and the electric equipment based on the measured electrical characteristics of the Ethernet cable; and

determining whether to provide power to the powered device based on the determined type.

Preferably, the power determination is for ethernet cabling greater than 100 meters.

Preferably, the power determination is for the use of category 3 cabling in a power over ethernet add-on application.

Preferably, the power determination is based on the identified length of the ethernet cable.

Drawings

In order to illustrate the manner in which the above-recited and other advantages and features of the invention are obtained, a more particular description of the invention briefly described above will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings. Understanding that these drawings depict only typical embodiments of the invention and are not therefore to be considered to be limiting of its scope, the invention will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:

FIG. 1 shows a schematic diagram of an embodiment of a Power over Ethernet (PoE) system;

FIGS. 2A and 2B show a block circuit diagram modeled on the PoE system;

FIG. 3 shows a flow chart of PoE processing;

FIG. 4 shows a schematic diagram of a connector in the middle of a media dependent interface link (media dependent interface link);

FIG. 5 shows a cable pair that may be shortened on both the line side and the transceiver side of a transformer;

FIG. 6 shows a schematic diagram of a PoE system that may implement the transfer of cable characterization information from a PHY to a PSE;

FIG. 7 shows a flow diagram of a process of transferring cable characterization information from a PHY to a PSE;

FIG. 8 shows a schematic diagram of an embodiment of insertion loss measurements for category 3 and 5 cables;

FIG. 9 shows a schematic diagram of near-end crosstalk testing for category 3 and category 5 cables.

Detailed Description

Various embodiments of the invention are discussed in detail below. It is to be understood that the specific embodiments have been discussed for illustrative purposes only. It will be apparent to those skilled in the art that other elements and configurations may be used without departing from the spirit and scope of the invention.

Fig. 1 illustrates an embodiment of a power over ethernet (PoE) system. As shown, the PoE system includes a Power Sourcing Equipment (PSE)120 that delivers power to a Powered Device (PD) 140. Power transferred by the PSE to the PD is provided by applying a voltage across a center leg of a transformer that is connected to Transmit (TX) and Receive (RX) pairs carried within the ethernet cable. Two TX and RX pairs are used for data communication between ethernet PHYs 110 and 130.

As further shown in fig. 1, the PD 140 includes an 802.3af module 142. The module includes electronics that enable the PD 140 to communicate with the PSE 120 in accordance with the IEEE 802.3af standard. The PD 140 may also include a pulse-width modulation (PWM) DC: DC controller 144 for controlling a power FET 146 to provide constant power to a load 150. Generally, there are two types of loads: a purely resistive load (e.g., a lamp) and a constant power load supplied by a DC: DC power controller. The present application relates generally to constant power loads supplied by DC to DC power controllers.

The power transfer from the PSE 120 to the load 150 may be modeled by the circuit in fig. 2A. As shown, the power supply provides a voltage V to the circuit_PSEThe circuit comprises a first pair of parallel resistors (R)₁、R₂) Load resistance R_LOADAnd a second parallel resistor pair (R)₃、R₄). Here, a first pair of parallel resistors R₁、R₂Representing the resistance of the TX line pair, and a second parallel resistance pair R₃、R₄Representing the resistance of the RX line pair.

Resistance R₁、R₂、R₃And R₄Is determined by the type and length of the ethernet cable. In particular, the resistance R₁、R₂、R₃And R₄With some resistance/length determined by the ethernet cable type (e.g., category 3, 5, 6, etc.). For example, for a type 3 Ethernet cable, resistor R₁、R₂、R₃And R₄Is about 0.2 omega/meter. Thus, for a 100 meter class 3 Ethernet cable, the resistance R₁、R₂、R₃And R₄Each of which has a resistance value of 20 omega. In this example, the parallel resistor R₁、R₂Is equivalent to 10 omega, and the parallel resistor R₃、R₄The resistance of (c) also corresponds to 10 Ω. Overall, the total resistance (R) of the Ethernet cable_cable) The value is 10 Ω +10 Ω to 20 Ω. The simplified PoE circuit model can include a single cable resistance value R shown in fig. 2B_cable。

In the IEEE 802.3af standard, a PSE may optionally perform a classification step that identifies the power classification of a PD. Table 1 below shows the 5 PD classes supported by the IEEE 802.3af standard.

Categories	Use of	Minimum output power of PSE	Maximum input power of PD
				0	By default	15.4W	0.44 to 12.95W
1	Optionally	4.0W	0.44 to 3.84W
				2	Optionally	7.0W	3.84 to 6.49W
3	Optionally	15.4W	6.49 to 12.95W
				4	Reservation	Operate as class 0	Reservation

TABLE 1

As shown, the class 0 (default) and class 3 PD classifications specify a minimum output power of 15.4W for the PSE. For lower power PDs such as class 1 and class 2 devices, the minimum output power of the PSE is specified to be 4.0W and 7.0W, respectively. Alternatively, however, the identification of the correct PD power class allows the PSE to only budget the power required at each port. This effectively increases the ability of the PSE to provide power to a group of PDs connected thereto.

It is a feature of the present invention that measurements of one or more characteristics of the ethernet cable can be used to affect the operation of the PoE system. In one embodiment, the measured characteristics may be used to identify the type and/or length of the ethernet cable. The identified type and/or length of the ethernet cable may then be used to estimate the resistance of the ethernet cable. The resistance estimate of the ethernet cable is then used to estimate the power loss in the cable that affects the power budget allocated to a particular PSE port.

The general process of the present invention can be understood with reference to the flow chart of fig. 3. As shown, the process begins at step 302, where one or more characteristics of the Ethernet cable are measured. In one embodiment, this measurement step may be performed as part of a PHY analysis of the electrical performance of the ethernet cable. For example, the measurement step may be performed as part of an echo cancellation convergence process (echo cancellation convergence process) performed by the PHY.

In one embodiment, one or more characteristics of those ethernet cables that may be measured at step 302 are those that may enable the PoE system to better estimate the resistance of the ethernet cables. Here, the estimation of the actual cable resistance may enable the PoE system to estimate the actual power loss of the cable. In one embodiment, the PHY may be designed to measure the insertion loss, crosstalk, and length characteristics of the determinable ethernet cable.

After completing the measurement of one or more characteristics of the ethernet cable in step 304, the PoE system can then determine the type and length of the ethernet cable. In one embodiment, the cable type of the ethernet may be determined based on the measured insertion loss, crosstalk, and length of the ethernet cable. These measurements of the ethernet cable may cause the PoE system to determine, for example, whether the ethernet cable is a category 3, 5, 6, or 7 ethernet cable.

It should be appreciated that different cable types have different associated resistance values. For example, the resistance of a type 3 Ethernet cable is approximately 0.2 Ω/meter, while the resistance of a type 5 Ethernet cable is approximately 0.1 Ω/meter. Once the type and length of the ethernet cable is determined at step 304, the PoE system can then determine its impact on the PoE system at step 306.

The specific impact of the cable type and length information on the PoE system, which may vary from application to application, will be described in detail below. It is a feature of the invention herein that the PoE system can use the type and length information of the cable during dynamic setup or operation. For example, the identified type and length of the ethernet cable may be used to diagnose the ethernet cable, determine whether power may be supplied to the PD, determine to adjust the power budget for a particular PSE port, and so forth.

To describe the various ways in which the type and length of an identified ethernet cable affects a PoE system, consider a first application related to a conventional PoE system, which is supported by the IEEE 802.3af specification. In this application, the determination of the type and length of the cable can be used to identify the resistance R_cable(see FIG. 2B).

In the circuit model of FIG. 2B, where PD includes a DC to DC converter, load R_LReceivable constantConstant power P_LAnd a visible voltage V_LAt its input. Due to P of the load_LIs stationary, P_L＝I*V_LWhere I is the current through the entire circuit. The power loss of the cable may be P_loss＝I²*R_cable。

In specifying a minimum output power of 15.4W for a PSE, the IEEE 802.3af specification assumes that the PD is connected to a PSE using a 100m class 3 cable. The resistance of a 100 meter category 3 cable is about 20 omega. When the current limit is 350mA, the power loss P due to the worst case_loss＝(350mA)²20 Ω ═ 2.45W. The worst case power loss of 2.45W is the difference between the minimum output power of the PSE and the maximum power drawn by the PD (i.e., 15.4W-12.95W — 2.45W).

In accordance with the present invention, the worst case power budget allocated to a PSE port can be adjusted based on the determination of the ethernet cable type. In particular, the identification of the ethernet cable type may result in an accurate power loss estimate without other additional information of the PoE system. For example, assume that the measurement profile indicates that the PD is connected to the PSE using a category 5 cable rather than using a category 3 cable. Even using the worst case assumption of a cable length of 100m and a current of 350mA, the resistance of the cable would be estimated to be 10 Ω for a category 5 cable, rather than 20 Ω for a category 3 cable. This determination of a half reduction in resistance will result in a half reduction in power loss to 1.225W, and accordingly the saved 1.225W power can then be used to reduce the power budget allocated to the port, thus effectively increasing the capacity of the PSE.

By combining the determination of the length of the cable with its type determination, a more accurate power loss estimate can be obtained. With the additional circuit length information, the resistance of the cable can be further reduced from the worst case 100 m. For example, it is assumed that the type of the cable is determined as category 5, and the length of the cable is further determined as 50 m. In this example, the resistance of the category 5 cable can be further reduced by half to 5 Ω. The 50m class 5 cable then has a power loss of Pl_oss＝(350mA)²*5Ω＝0.6125W. Accordingly, the saved power of 2.45W-0.6125W-1.8375 may then be used to reduce the power budget allocated to that port. It will be appreciated that determination of the length of the individual cables may provide the power savings advantages described above. While conventional systems have contemplated using cable length determinations in typical PoE applications (e.g., below 100 meters), using cable length determinations in PoE applications greater than 100 meters is a distinguishing feature of the present invention.

In the above examples, the determination of the cable type alone or in combination with the determination of the ethernet cable length may be used to reduce the power budget allocated to the PSE port. The identification of the cable type provides a significant advantage over the identification of the circuit length alone. It is desirable that these advantages be obtained without any other information of the system. More detailed power loss calculations may be generated if other information is available to the system.

The voltage drop across the cable (voltage drop) may be defined as V_PSE-V_L＝I*R_cable. This equation can be used to obtain the allowed voltage V at PD by performing the following calculation_L：

V_PSE-V_L＝I*R_cable

V_PSE-V_L＝(P_L/V_L)*R_cable

V_PSE*V_L-V_L ²＝P_L*R_cable

V_L ²-V_PSE*V_L+P_L*R_cable＝0

V_L＝[V_PSE+/-SQRT(V_PSE ²-(4*P_L*R_cable))]/2

If V_PSEKnown as 48V, P_LIs 12.95W (maximum power allowed by all PDs), and R_cableDetermined as 5 omega (resistance of category 5 cable of 50 meters), followed by V_L＝(48+/-SQRT(48²-4 × 12.95 × 5))/2 ═ (48+/-45.22)/24 ═ 46.61V. V can then be used_PSE-V_L＝I*R_cableThe current was calculated such that 48V-46.61V ═ I × 5 Ω gave I ═ 0.278A. The total power output of the PSE is 12.95W plus the power loss in the cable. The power loss in the cable in this case is I²*R_cable＝(0.278A)²5 Ω ═ 0.39W. In this example, the total power of the PSE ports is 12.95W +0.39W — 13.34W. The saved power budget is 15.4W-13.34W-2.06W.

As further described in this example, the worst case estimate of the IEEE 802.3af standard for a 100m class 3 cable (with a worst case cable resistance of 20 Ω) would result in unnecessary waste of the power budget allocated to the port. When summing up the ports of all PSEs, the waste in power budget will unnecessarily reduce the actual power capacity of the PSEs.

A second application of the present principles is applicable to PoE + applications as supported by the future IEEE 802.3at specification. PoE + applications can be designed to support higher power PDs and assume the use of type 5 or higher ethernet cables. A PD of up to 30W may be considered a two-pair PoE + system, while a PD of up to 56W may be considered a four-pair PoE + system. It will be appreciated that the same principle can be applied to two or four pair systems. Generally, higher power PDs with PoE + can be supported, enabling e.g. WiMAX transmitters, pan-tilt-zoom cameras, video phones, thin computers (thin clients).

In this application, the principles of the present invention may be used, first of all, as a diagnostic tool for verifying an Ethernet cable connected to a PSE port. In one embodiment, the diagnostic tool will identify the ethernet cable type and use that identification to determine how to control the PoE + PD device.

In one embodiment, if the ethernet cable is determined to be a type 3 cable, then the PSE may refuse to provide power to the PoE + PD devices of that port. In another embodiment, the diagnostic tool may be used to extend the potential applications of PoE + PSE. For example, even if the diagnostic tool has determined that the PoE + PD device is connected to the PSE using a class 3 cable, the diagnostic tool will continue to determine whether the PoE + PD device is still drawing power over the class 3 cable. For example, a diagnostic tool may be used to verify the port to determine if it is compatible with PoE + PD devices, even if the port is connected to the PSE via a class 3 cable. This verification may be based on the actual characteristics of the cable (e.g., length), rather than simply depending on the type of cable (e.g., category 3, 5, etc.).

Even though the resistance of the category 3 cable is about twice that of the category 5 cable, the category 3 cable may be used in PoE + applications in some cases. The PSE may use a length, V, of e.g. a category 3 cable_PSE、V_LAnd PoE + PD power make an informed decision as to whether to apply power to a particular port and how much power budget to allocate to that port. Effectively, this intelligent decision making enables the PSE to identify additional port placements that could benefit from PoE +, without relying on placing too wide restrictions on the characteristics of the installed ethernet cable.

For example, consider a case where V_PSEIs 50V, P_LIs 15W, and R_cableDetermined as 15 Ω (resistance of 75 meters of category 3 cable). From this set of operating parameters, V can be calculated_L＝(50+/-SQRT(50²-4 x 15))/2 ═ (50+/-40)/2 ═ 45V. V can then be used_PSE-V_L＝I*R_cableThe current was calculated such that 50V-45V ═ I × 15 Ω gave I ═ 0.333A. The power loss in the cable is I²*R_cable＝(0.333A)²15 Ω ═ 1.66W. In this example, the total power budget of the PSE port is 15W +1.66W — 16.66W. When the set of operating conditions is allowable operating conditions for a class 3 cable, the PSE may choose to provide power to PoE + PD on the class 3 cable.

In another case, if V_PSEIs 50V, P_LIs 20W, and R_cableDetermined as 20 omega (resistance of 100 meters of category 3 cable) and then V can be calculated_L＝(50+/-SQRT(50²-4*20*20))/2＝(50+/-30)/2-40V. V can then be used_PSE-V_L＝I*R_cableThe current was calculated such that 50V-40V ═ I × 20 Ω gave I ═ 0.5A. Irrespective of power loss (I)²*R_cable＝(0.5A)²20 Ω -5W) is acceptable, the current I of 500mA has been higher than the limit current 350mA of category 3 cable. In this case, the PSE may choose not to provide power to PoE + PD over class 3 cable.

In yet another embodiment, assume P_LIs 15W, and R_cableDetermined as 20 Ω (resistance of 100 meters of category 3 cable) and V_LKnown as 43V. It should be appreciated that V may be communicated using a variety of communication schemes, such as some forms of layer 2 communication_LFrom the PD to the PSE. In this case, I ═ P may be used_L/V_LThe current was calculated at 15W/43V-0.349A. In this case, the PSE may choose to provide power to PoE + PD over class 3 cable.

As these examples show, the PSE can make an informed decision on whether to provide power to PoE + PD over class 3 cable. This dynamic procedure is significant and the classification of the entire class 3 arrangement cannot be categorically excluded from the supported PoE + PD. Although the above provides only a few examples, it should be appreciated that the PoE + system can detect potential category 3 cabling using any amount of information available to it. In general, the more information that is available, the more authentication that a category 3 cable is available for PoE⁺The greater the likelihood of use.

Conventional PoE + cabling based on category 5 cables may also benefit from the principles of the present invention. This is especially true when considering the power budget allocated to the PoE + PSE ports.

For a conventional 802.3af installation, the worst-case power loss of the cable is P_loss＝(350mA)²20 Ω ═ 2.45W. This worst case power loss is based on a 350mA current limit per PD due to cable and patch panel limitations and the 20 Ω resistance of a class 3 ethernet cable. In a double current PoE application, for example, the power loss of the class 5 cable may be P_loss’＝(700mA)²*10Ω＝4.9W＝2*P_loss. As the brief calculation shows, the power loss per meter in PoE + cabling can be twice that of conventional 802.3af cabling, even reducing the cable resistance by 50%. For this reason, identification of category 5 cable lengths can lead to more significant power budget savings in reducing the worst case power loss for a port. For example, if the length of the cable is determined to be 25m, then the power loss at a current of 700mA may be calculated to be 1.225W. This is significantly lower than the power loss of 4.9W when assuming a length of 100 meters for a category 5 cable. Of course, when using V as described above with respect to_PSE、P_LAnd R_cableThe estimated power loss in this cable can be further reduced when estimating the actual current.

In addition, power loss calculations may also benefit from the cable type information obtained from PoE + deployment. Here, a determination is made whether the ethernet type is better than a category 5 cable (e.g., category 6 or 7 cable) will be used to reduce the resistance estimate of the cable, thus further reducing the estimated power loss.

A third application of the present principles is PoE high coverage (PoE-BR) applications. In PoE-BR applications, the PD may be connected to a PD with an ethernet cable above 100 m. For example, a PoE-BR application may be defined to support distances up to 500m or more.

In PoE-BR applications, determining the ethernet cable type would provide a simple benefit in extending existing PoE applications. Consider, for example, a worst-case 802.3af application where power is provided by a PD on a category 3 cable that exceeds 100 m. In this worst case application, the resistance of the cable is about 20 Ω. If a category 5 cable is used instead, a lower resistance category 5 cable may allow for a longer length, consistent with a resistance of 20 Ω. For example, assume that the worst case category 5 cable includes a connector located in the middle of a Media Dependent Interface (MDI) link. As shown in fig. 4, a connector located in the middle of a Media Dependent Interface (MDI) link may be introduced into the middle of the MDI link through the contents of the cross-connect system, a wall outlet, or the like. In this case, the resistance of the ethernet cable may be about 12.5 Ω. From this estimate, the length of category 5 cables can be extended to 100m 20/12.5 to 160m, with a resistance of 20 Ω. Thus, even without any operational information of the PoE system, simple identification of the cable type can result in providing power to PDs at distances greater than 100 m.

Typically, an increase in the distance between the PSE and the PD (e.g. up to 500m) gives rise to a larger range of potential operation in PoE-BR systems. This operating range makes it more difficult to provide system specifications using worst-case operating parameters. For example, assume that the PoE-BR specification supports cable type 3. When in this case, the resistance of the cable may be specified as 20 Ω -100 Ω. Clearly, in identifying a power budget as listed in Table 1, it is impractical to assume a 100 Ω worst case cable resistance. The category 5 cable specification also needs to be tolerated since the resistance of the cable is specified as 10 Ω -50 Ω.

It is therefore a feature of the present invention that a PD in a PoE-BR application can be powered based at least in part on a particular port arrangement. For example, assume V_PSEAt 51V, the PD will consume a constant 12.95W and the voltage of the PD is 37V, in which case the current I ═ P can be calculated as_L/V_L12.95W/37V-0.34A. The maximum resistance R of the cable can then be calculated in this way_cable＝(V_PSE-V_L)/I＝(51V-37V)/0.34A＝41Ω。

Having a maximum resistance R_cableThe PoE-BR system can then determine whether a particular port is accommodating such a deployment, 41 Ω. For example, if it is determined that a category 3 cable is used, then a PD up to 205 meters in distance may be powered. Similarly, if it is determined that a category 5 cable is used, then a PD up to 410 meters in distance may be powered.

The cable length information may be used to determine the power loss of the cable. For example, if a category 5 cable is determined to be 400 meters, assuming a resistivity of 10 Ω/meter, the resistance of the cable is about 40 Ω. The power loss P can then be calculated in this way_loss＝(340mA)²40 Ω -4.62W. Then the total power budget for that port is 12.95W+4.62W＝17.57W。

As is known from the above, the power budget of a port may vary greatly due to the range of distances served by a PoE-BR application. For example, if a 120 meter category 5 cable is used, the resistance of the cable is about 12 Ω. The power loss P can be calculated in this way_loss＝(340mA)²12 Ω ═ 1.39W. The total power budget of the port is 12.95W +1.39W — 14.34W. In both cases, the 3.23W (i.e., 17.57W-14.34W) difference in power budget indicates the benefit of observing the type and/or length of cable rather than relying on the basic worst case assumption.

Due to the large range of cable resistance in PoE-BR applications, the minimum voltage of the PD can be reduced relative to conventional 802.3af PoE. For example, assume that the minimum voltage of PD is as low as 30V. This 30V value can be used to verify the deployment of a given port when the cable type and length are known. It should be appreciated that PD has higher requirements for the boot-up voltage than the lowest voltage. During power-up, it may occur that the PD is unable to draw full power, so that the voltage of the PD is almost the same as the PSE.

Suppose V_PSE＝50V，P_L12.95W, and R_cable45 Ω (450 meters of resistance of category 5 cable). For this set of operating parameters, V may be calculated_L＝(50+/-SQRT(50²-4 × 12.95 × 45))/2 ═ (48+/-30)/2 ═ 30.5V. In calculating V_LThe PoE-BR system can then determine the calculated voltage V from the lowest voltage_LWhether it is available. In this case, V_L30.5V is above the minimum limit, so the PoE-BR system can authenticate the port under these operating conditions. With respect to the power budget allocated to the port, the PoE-BR system can use V_PSE-V_L＝I*R_cableThe current was calculated so that 50V-30.5V × 45 Ω gave I ═ 0.433A. The power loss in the cable can be calculated as I²*R_cable＝(0.433A)²45 Ω -8.44W. Then, the total power budget of the PSE port in this example is 12.95W +8.44W — 21.39W.

Using the principles of the present invention, the excessive negative effects of using worst-case resistance in a PoE-BR link can be minimized. First, the power budget allocated to a particular port is saved, thus increasing the overall capacity of the PSE. Second, when using a worst case estimate of cable resistance, the PSE can verify excluded port placements.

As can be seen from the above, measuring one or more characteristics of the ethernet cable allows the PoE system to estimate the resistance of the ethernet cable and ultimately the actual power loss of the ethernet cable. To facilitate such estimation, the PoE system can measure such characteristics of the ethernet cable as insertion loss, crosstalk, length, etc. Measurements of insertion loss, crosstalk, length of an ethernet cable are examples of characteristics used to estimate cable resistance and, in turn, power loss in the cable.

In general, different cable types meet their own standards defining insertion loss over a range of frequencies. The electrical signals transmitted over the cable are attenuated differently depending on the type of cable. Insertion loss is a function of frequency and cable length and is well defined for each cable type. To determine the cable type, the PoE system can send one, multiple, or continuous pulses having predetermined frequency components into the cable. At the receiving end, the PoE system can measure attenuation magnitude (attenuation) and phase distortion, and then combine this information with the cable length to determine the cable type.

In one embodiment, the link partner (link partner) may be turned off and the cable pair disconnected from the line side or the opposite (transceiver) side of the transformer. In this case almost all the incidental pulses may be retro-reflected to the transmitting end with the same polarity and these pulses will experience an insertion loss corresponding to twice the cable length. Fig. 8 shows an example of the insertion loss that can be used to measure 100m category 3 and category 5 cables.

In another embodiment, the link partner may be turned off and the cable pair shortened from the line side or the opposite (transceiver) side of the transformer. As shown in FIG. 5, where A + is shorter than A-. In this case, almost all the incidental pulses may be retro-reflected to the transmitting end with opposite polarity, and these pulses will experience an insertion loss corresponding to twice the cable length.

In another embodiment, the link partner may be closed and the two cable pairs disconnected and shorted from the other pair to form one loop (e.g., A + shorted to B + and A-shorted to B-). This can be done on the line side or the opposite (transceiver) side of the transformer. In this case almost all of the accompanying pulses can be sent back to the different pairs of transmitting ends and these pulses will experience insertion losses corresponding to twice the cable length.

In another embodiment, the link partner may be temporarily turned on to send a predetermined pulse. In this case, these pulses will experience insertion losses corresponding to the length of the cable.

Crosstalk is similar to insertion loss, with different cable types meeting their own defined crosstalk standards over a range of frequencies. The electrical signals transmitted over the cable inject different noise into adjacent pairs depending on the type of cable. Crosstalk is a function of frequency and cable length and is well defined for each cable type. To determine the cable type, the system may send one, multiple, or continuous pulses having predetermined frequency components into the cable. At the receiving end, the system can measure the attenuation magnitude and phase distortion, and then combine this information with the cable length to determine the cable type.

There are two types of crosstalk: near-end crosstalk (NEXT) and far-end crosstalk (FEXT). For NEXT, noise injection from one or more local transmitters, and for FEXT noise injection from one or more remote transmitters, whether NEXT or FEXT or a combination thereof, can be used to determine the cable type. Fig. 9 shows an example that can be used to measure category 3 cables and category 5 cables.

In one embodiment, the cable length may be determined directly using a Time Domain Reflectometer (TDR). In an alternative embodiment, the cable length may be determined indirectly based on data generated during the measurement of the insertion loss using the round trip of the injected signal. Here, the time interval for transmitting and receiving the above-mentioned pulses is linearly proportional to the cable length. The cable length is calculated by multiplying the propagation speed and the time interval and then dividing by 2 to calculate the round trip delay.

As described above, various cable characteristics may be used to determine the cable type and, in turn, the resistance and power loss of the cable. It should be appreciated that other features than those described above may be used to enable the PoE system to determine the resistance and power loss of the cable. Regardless of the measurement data used, it is significant that the PoE system can use this data to dynamically adjust certain aspects of the configuration or operation of the PoE system. As indicated above, this feature of the present invention may be used in a variety of applications.

Fig. 6 illustrates an embodiment of a PoE environment 600 in which the principles of the present invention may be implemented. As shown, environment 600 includes PHYs 630-1 through 630-N, each of which may be coupled to Ethernet switch 620. Each PHY may include one or more Ethernet transceivers, only one of which is shown wired to PHY 630-N. Each PHY may also be connected to CPU 610, with only a single connection from CPU 610 to PHY 630-N shown for simplicity. In one embodiment, CPU 610 is integrated with Ethernet switch 620 and PHYs 630-1 through 630-N on a single chip. In another embodiment, the Ethernet switch 620 and the PHYs 630-1 to 630-N are integrated on a single chip and separate from the CPU 610 and may communicate with the CPU 610 through a serial interface. Also shown in PoE environment 600 is PSE 640 providing power through the central leg of the transformer shown. As shown, PSE 640 is connected to CPU 610 through an optical isolator 650 for isolation demarcation.

For a description of the operation of PoE environment 600 in carrying out the principles of the present invention, reference is made to the flow chart illustrated in fig. 7. As shown, the flow of FIG. 7 begins at step 702, where a transceiver in PHY 630-N measures a line characteristic of an Ethernet cable connected to PHY 630-N. In one embodiment, the measurements may be used to determine the pair of insertion loss, crosstalk, and cable length employed during the echo cancellation convergence process performed by the echo canceller module controlled by CPU 610. In step 704, the line characteristic measurements obtained by the transceiver are sent to the CPU 610.

Next, in step 706, CPU 610 uses the line characteristic measurement data to determine the cable type and length. In one embodiment, the cable type and length information is then provided to PSE 640 in step 708. Here, it should be noted that the PSE may be arranged to determine the type and length of the cable using line characteristic measurement data.

Regardless of the cable type and length determined, PSE 640 may utilize it to determine its impact on PoE system configuration and/or operation at step 710. This impact determination may take into account the cable type and length, and thus the cable's resistance and other PoE system parameters such as V_PSE、P_L、V_LAnd the like. It should be appreciated that this impact analysis may be implemented by any system component that may be used to diagnose the ethernet cable, determine whether power may be supplied to the PD, determine to adjust the power budget for a particular PSE port, etc. In general, this impact analysis may be based on one or more parameters, such as cable resistance, cable current, V, which may be communicated, discovered, or assumed by the appropriate system components_PSE、P_L、V_L. For example, one or more parameters may be based on a system specification (e.g., IEEE 802.3af), using measurement data to obtain one or more calculated values (e.g., cable resistance from a determined cable type and length), or received from other system components with information about the parameters (e.g., V for PD delivery to PSE)_L)。

Various aspects of the invention will become apparent to those skilled in the art upon review of the foregoing description. While various salient features of the invention are disclosed above, it will be apparent to those skilled in the art from this disclosure that the invention may be implemented or carried out in various ways, and thus the foregoing description should not be considered as excluding other embodiments. Also, it is to be understood that the phraseology and terminology used herein is for the purpose of description and should not be regarded as limiting.

Claims

1. A power over ethernet system, comprising:

a cable detection assembly for measuring electrical characteristics of the Ethernet cable including insertion loss, crosstalk, and length of the Ethernet cable; and

a power controller for determining the type of the Ethernet cable based on the measured electrical characteristics, and controlling power distribution to the power supply equipment ports based on the type and length of the Ethernet cable.

2. A power over ethernet system according to claim 1, wherein said measurements of insertion loss, cross-talk, length of the ethernet cable are used to estimate cable resistance and thus power loss in the cable.

3. A power over ethernet system according to claim 1, wherein the system transmits one, more or consecutive pulses having predetermined frequency components into the cable and measures the attenuation magnitude and phase distortion at the receiving end, combining this measurement information with the cable length to determine the cable type.

4. A power over ethernet system according to claim 1, characterized in that the cable length is determined directly using time domain reflectometry, TDR.

5. A power over ethernet system according to claim 1, wherein the cable length is determined indirectly based on data generated during the measurement of the insertion loss using the round trip of the injected signal.

6. A power over ethernet system according to claim 1, wherein said power controller controls said power distribution based on a resistance of said ethernet cable determined using said type of said ethernet cable.

7. A power over ethernet method, comprising:

upon connecting a powered device to a power device port through an Ethernet cable, determining a type of the Ethernet cable based on measured electrical characteristics, including insertion loss, crosstalk, and length of the Ethernet cable; and

allocating a power budget to the power source device port, the allocated power budget being based on the determined type and length of the Ethernet cable.

8. A method according to claim 7, characterized in that the system sends one, several or consecutive pulses with predetermined frequency components into the cable and measures the attenuation level and phase distortion at the receiving end, combining this measurement information with the cable length to determine the cable type.

9. A power over ethernet method, comprising:

determining a type of Ethernet cable connecting a power sourcing equipment and a powered device based on measured electrical characteristics of the Ethernet cable, including insertion loss, crosstalk, and length of the Ethernet cable; and

determining whether to provide power to the powered device based on the determined type and length.

10. The method of claim 9, wherein the power determination is for ethernet cabling greater than 100 meters.