Multiplex communications – Communication techniques for information carried in plural... – Adaptive
Reexamination Certificate
1998-06-02
2002-08-27
Olms, Douglas (Department: 2661)
Multiplex communications
Communication techniques for information carried in plural...
Adaptive
C370S252000, C370S348000, C370S401000, C370S444000, C370S450000, C370S459000
Reexamination Certificate
active
06442174
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates in general to a method for saving consumed power in a network system, more specifically, to a method for controlling network nodes to be operated in a power-saving transmission protocol by utilizing the auto-negotiation function provided by a typical fast Ethernet controller when the network system is idle, thereby reducing the overall power consumption of the network system.
2. Description of the Related Art
An Ethernet is one of the most widely employed local area networks (LAN). In addition, an Ethernet. is similar to the IEEE 802.3 standard proposed by the Institute for Electrical and Electronic Engineers (IEEE), except for the packet format. The access control in the Ethernet is mainly achieved by Carrier Sense Multiple Access/Collision Detection (CSMA/CD).
FIG. 1
(Prior Art) illustrates the network topology of an example of the Ethernet. As shown in
FIG. 1
, Ethernet
10
minimally comprises four network nodes, which are denoted by N
1
, N
2
, N
3
and N
4
, respectively. Fundamentally, the Ethernet does not comply with master-slave architecture; in other words, any network node in the Ethernet owns the same right to access the data transmitted therein. When one of these network nodes attempts to send data (packets), this network node should first detect the status of the Ethernet
10
and judge whether any other network node is sending data or not. This can prevent the collision of the packets sent from different sources. In an Ethernet, the detection scheme which depends on the absence or presence of carrier signals, is called Carrier Sense (CS). The feature that every packet sent to the Ethernet
10
can be accessed by all network nodes in Ethernet
10
is called Multiple Access (MA). In the practical accessing process, each network node can determine the source and destination network nodes of a data packet according to a source node code and a destination node code therein. In addition, each of these nodes is capable of detecting data collision in Ethernet
10
. This feature is called Collision Detection (CD). When data collision occurs, the network nodes that are currently transmitting packets temporarily cease transmission. Transmission can resume after a time period, which is randomly specified by the network nodes. According to the above description, all network nodes in Ethernet
10
have equal access.
The Ethernet protocol can be divided into various physical-layer protocols, depending on the data transmission speed and the transmission media. For example, some popular physical-layer protocols are 10 BaseT, 10 Base2, 10 Base5 and 100 BaseT. Among these notations, the first number before “Base” represents the operating speed of the Ethernet implementations in megahertz. For example, “10” represents the 10 Mbps (megabytes per second) and “100” represents the 100 Mbps. In addition, the last character or number denotes the transmission media employed in this Ethernet implementation. For example, “T” represents the twisted pair like the telephone wiring lines, “2” represents a thin coaxial cable like RG-58A/U, and “5” represents a think coaxial cable like RG-8. Currently the Ethernet implementations 10 BaseT and 100 BaseT are most widely used. In addition, 100 BaseT can be further divided into 100 BaseTX, 100 BaseT4 and 100 BaseTF. 100 BaseTX employs UTP
5
(Unshield Twisted Pair, No. 5) wires, 100 BaseT4 employs UTP
3
or UTP
4
, and 100 BaseTF employs optical fibers.
In fact, the 10 BaseT and 100 BaseTX physical-layer protocols not only differ in the transmission media, but also in the signal waveform transmitted in the wiring.
FIG. 2A
(Prior Art) illustrates a typical signal waveform diagram in the wiring of the Ethernet implemented by the 10 BaseT physical-layer protocol. As shown in
FIG. 2A
, two regions of the signal waveform in the wiring of 10 BaseT can be distinguished. One is for the data transmission period, which is denoted by “Tx,” and another is for the network idle period, which is denoted by “Idle.” Data transmitted in the data transmission period Tx may be data packets or control signals for some network protocols. On the other hand, the network idle period “Idle” contains no waveform.
In addition, the signal waveform in the 100 BaseT implementation is different from that in the 10 BaseT implementation.
FIG. 2B
(Prior Art) illustrates a typical signal waveform diagram in the wiring of the Ethernet implemented by 100 BaseTX physical-layer protocol. As shown in
FIG. 2B
, the wiring in the 100 BaseTX contains waveforms in both the data transmission period and the network idle period. This is due to the scrambling process adopted by the 100 BaseT implementation before transmitting data. Typically, the 100 BaseT implementation may scramble four-bit digital data into five-bit scrambled data for practical transmission. Therefore, the original digital data in the network idle period, which should be all zeros, may be scrambled into a digital code containing information “1,” thereby producing a signal waveform in the wiring in response to the information “1.”
On the other hand, manufacturers usually incorporate the 10 BaseT implementation into 100 BaseTX Ethernet cards to ensure the compatibility of 10 BaseT and 100 BaseTx in the same wiring network. In other words, an Ethernet card having the 10 BaseT and 100 BaseTX implementations can automatically adjust the settings according to the practical situation. However, since the transmission performance of the 100 BaseTX implementation is superior to that of the 10 BaseT implementation, the 10 BaseT/100 BaseTX network card gives top priority to the 100 BaseTX physical-layer protocol. Therefore, if a 10 BaseT/100 BaseTX network card is trying to setup the connection with another network node supporting 100 BaseTX, the 100 BaseTX physical-layer protocol will be chosen for the network operation. In addition, it is apparent that the old-fashioned 10 BaseT network card cannot transmit data at an operating speed of 100 Mbps. Therefore, if the 10 BaseT/100 BaseTX network card is trying to setup the connection with another network node only supporting 10 BaseT, the 10 BaseT physical-layer protocol will be chosen for the network operation. Deciding the appropriate Ethernet physical-layer protocol for the network setup is achieved by the auto-negotiation function.
The above description is an overview of the Ethernet application to date. It is readily found that network cards supporting both of 100 BaseTX and 10 BaseT will be widely adopted in the market since their superior performance and compatibility. However, the 100 BaseTX implementation still has some shortcomings. Referring to
FIG. 2B
, signal waveforms are always present in the wiring of the Ethernet using 100 BaseTX, during both the data transmission period and the network idle period. In other words, the network system will always consume power at any time. Usually, there is plenty of network traffic in the daytime, but relatively little or none at night. Clearly, the network wastes power during such idle periods. Therefore, the present invention addresses the problem of power consumption in the Ethernet using 100 BaseTX.
SUMMARY OF THE INVENTION
Therefore, an object of the present invention is to provide a method for saving power in a network system, such as the 100 BaseTX network that would otherwise waste a great deal of power during idle periods. In addition, this method provides an automatic switching function, which can switch to a physical-layer protocol that uses less power when the network system is in an idle condition.
The present invention achieves the above-indicated objects by providing a method for saving power in a network system. It is assumed that the network system is capable of executing at least two transmission protocols, such as 10 BaseT and 100 BaseTX in the modern Ethernet system. First, the network nodes may continuously detect the status of the network system and determine if it is in an idle period. If the idle period exceeds a pre-determin
Hom Shick
Ladas & Parry
Olms Douglas
Winbond Electronics Corp.
LandOfFree
Method for saving power in a network system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for saving power in a network system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for saving power in a network system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2934428