Multiplex communications – Channel assignment techniques – Carrier sense multiple access
Reexamination Certificate
1998-07-10
2002-11-19
Olms, Douglas (Department: 2661)
Multiplex communications
Channel assignment techniques
Carrier sense multiple access
C370S462000
Reexamination Certificate
active
06483846
ABSTRACT:
A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by any one of the patent disclosure, as it appears in the Patent and Trademark Office patent files or records, but otherwise reserves all copyright rights whatsoever.
FIELD OF THE INVENTION
The invention is related to network data communications, and in particular to a middleware approach to implementation of real-time Ethernet.
BACKGROUND OF THE INVENTION
Computer networks have become widely popular throughout business and industry. They may be used to link multiple computers within one location or across multiple sites.
The network provides a communication channel for the transmission of data, or traffic, from one computer to another. Network uses are boundless and may include simple data or file transfers, remote audio or video, multimedia conferencing, industrial control and more.
Perhaps the most popular network protocol is Ethernet, a local area network specification for high-speed terminal to computer communications or computer to computer file transfers. The Ethernet communication protocol permits and accommodates data transfers across a data communication channel or bus, typically a twisted pair or coaxial cable.
The Ethernet communication protocol was standardized as the IEEE 802.3 standard for communications over a local area network (LAN). This protocol incorporates a 1-persistent, Carrier Sense Multiple Access with Collision Detection (CSMA/CD) protocol meaning that one or more nodes of a shared network may monitor the channel if they have a data packet to transmit, and transmit that packet immediately upon detecting the channel to be idle.
A “collision” of data packets may occur if two or more nodes begin transmitting simultaneously on the network. Colliding nodes will detect such a collision of data and terminate their transmission, waiting a randomly-determined time period before attempting transmission again. Under current standards, a failure will be generated after a node makes sixteen unsuccessful attempts to transmit its data packet without collision.
Under lightly-loaded conditions, collisions are infrequent and resolution is rapid. However, heavy loading may lead to indeterminate access time. While some applications may be relatively insensitive to collisions and their resultant delays on data transfer, other applications may be time sensitive such that collisions of data packets are undesirable or even intolerable. Examples of such time-sensitive or real-time applications may include remote video or control of industrial process equipment. The requirement for some applications to circumvent collisions and guarantee successful transmission and reception has led to various improvements to Ethernet.
One Ethernet improvement is a token-based protocol standardized under IEEE 802.4 (Token bus) or 802.5 (Token ring). The primary difference between these two standards is in the network topology each is designed to address. Token bus addresses a network in which the nodes form a logical ring; Token ring addresses a network in which the nodes form a physical ring.
Token-based protocols generate a “token” which is passed to every node along the network. These protocols permit data transmission only when the node is in possession of the token, and each node is given a fixed amount of time to transmit data. This transmission time is further divided into multiple segments or timers relating to different priority levels. These priority levels may be assigned to different data streams depending upon their criticality and time sensitivity. Nodes may only transmit data of a given priority level during its respective timer. Under this approach, real-time data may be assured a fraction of the bandwidth free of collision. However, some of these token-based protocols may allow a given node only its fixed share of bandwidth regardless of whether other nodes make full or even any use of their bandwidth.
Improvements on these token-based protocols have also been proposed. As an example, an academic prototype has been proposed for a software-oriented real-time Ethernet implemented on a UNIX platform utilizing a token-based protocol. (see Chitra Venkatramani, “The Design, Implementation and Evaluation of RETHER: A Real-Time Ethernet Protocol,” Ph.D. Dissertation, State University of New York, January 1997) RETHER, however, only provides for non-real time traffic when there is no more real time traffic to be sent by any node. Depending on the type of traffic on the network, this led to low network throughput and utilization due to token passing overhead for non-real time traffic, and did not support hard real time traffic.
Another prior solution is hardware based. Under this approach, data packet collisions are avoided through hardware. These hardware-based solutions may be necessary for certain critical real-time applications such as aviation, to meet stringent performance and reliability requirements. However, such solutions are proprietary and vendor-dependent, making them difficult and expensive to implement. Hardware-based solutions may be incompatible with many existing Ethernet networks, requiring costly and complicated modifications. In addition, although these hardware solutions prevent collisions, they do not offer scheduling of real-time traffic in an entire system. Both solutions also require modification of existing hardware or software.
Accordingly, there exists a need for an efficient deterministic service to prevent collisions of and guarantee real-time traffic over Ethernet that can be implemented on existing Ethernet networks and is compatible with a wide variety of commercial-off-the-shelf (COTS) hardware and applications. Such a solution is needed for process control networks, time sensitive multimedia and Internet applications.
SUMMARY OF THE INVENTION
A middleware approach to implementation of real-time Ethernet provides deterministic, i.e. predictable, communication services for both real time and non-real time traffic over a conventional Ethernet network having a plurality of nodes desiring to transmit packets of data. The middleware comprises computer software residing above a network interface device and the device driver, yet below the system transport services and/or user applications. The invention provides a Middleware Real-Time Ethernet or MRTE which does not require modification to existing hardware that implements Ethernet.
In one embodiment, Ethernet bandwidth is divided into cycles. During each cycle, a first time interval is provided for real time data packet traffic using a deterministic scheduling protocol such as by passing a token, such that no collisions can occur. During a second time interval, the standard carrier sense, multiple access, collision detect Ethernet protocol is used for non-real time traffic. By using these two time intervals, bandwidth is shared between real time and non-real time traffic, ensuring that both will receive desired bandwidth.
In one embodiment, separate queues are used for deterministic scheduling to determine the order of packet queuing and transmission on each node such that (1) real-time traffic can be guaranteed once admitted for transmission service, (2) non-real-time traffic can be served, and (3) the Ethernet bandwidth utilization can be optimized.
Quality of Service, QoS, enables making on-line tradeoffs between network bandwidth availability and network transmission quality. Examples of QoS include (1) degree of packet collisions when Ethernet is shared by soft- or non-real-time traffics during certain time slots and (2) amount of end-to-end packet transmission latency.
When QoS is used, periodic data, such as video at 30 frames per second may be given a priority or criticality, and a cumulative loss factor, e.g. up to four frames in a row may be discarded. If there is sufficient bandwidth remaining after higher priority tasks or data streams are handled, the video will be accepted to the real time queue with at least five f
Chen Donghui
Huang Jiandong
Honeywell Inc.
Olms Douglas
Sam Phirin
Schwegman - Lundberg Woessner - Kluth
LandOfFree
Middleware-based real-time communication 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 Middleware-based real-time communication system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Middleware-based real-time communication system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2986193