Adaptive universal multiple access

Details Adaptive universal multiple access Adaptive universal multiple access

1999-10-29

2001-06-26

Vu, Huy D. (Department: 2733)

Multiplex communications

Channel assignment techniques

Only active channels transmitted

C370S437000, C370S465000

Reexamination Certificate

active

06252881

ABSTRACT:

This application claims the benefit of Disclosure Document No. S00613, entitled “Adaptive Universal Multiple Access,” submitted by Stanislav I. Samoylenko on Nov. 4, 1997 with a PTO stamped date of receipt of Nov. 12, 1997.
The teachings of U.S. Pat. No. 5,576,702, entitled “Method and Apparatus For Fault Tolerant Transmission In Multi-Channel Networks,” issued Nov. 19, 1996 to the inventor of the present application, are also hereby incorporated by reference.
FIELD OF THE INVENTION
The present invention generally relates to a multi-station data network where multiple devices share some common resources, such as communication media and data-processing systems, switching systems. More precisely, the present invention is directed to an apparatus and method for collision-free multiple access in data networks which also adapts to varying traffic, varying number of active stations, variable channel distances, and variable transmission speed in the channels. Potential applications of the present invention include cabled, wired or wireless networks for collision-free, real-time, multimedia applications in the following areas: Ethernet LANs, wireless LANs and MANs, intra-chip communications, packet data on cellular networks, packet voice on LANs and MANs, interactive TV home-access networks, high-speed Internet access networks, mobile Internet access via cellular networks, and network computers.
BACKGROUND OF THE INVENTION
A multi-station data network includes a plurality of stations connected to a common medium for communication. Communication networks usually have many variable parameters, such as changing station activity, varying average load, and bursty traffic. Additionally, from time to time the network topology may be changed. For example, new modes and channels can appear, some nodes can be switched off, some channels can be interrupted, and channel speed or length may change.
Communication networks commonly operate under changeable conditions. For example, network load, nodes activity, network topology, inter-node distances, channel speed and other network parameters typically change with time. Bursty traffic is transmitted in many networks. While some nodes have long sequences of packets other nodes have no messages at all. New stations or hubs can be added in the network. For example, some nodes can be permanently or temporarily switched off, channel or node failures can change network topology, etc. Network load and traffic are, in general, random variables whose average value depends on the time of day, day of week, etc. Even in stable networks, it is desirable to use adaptive mechanisms at the time of network initialization to implement plug-and-play technology or in order to simplify network management.
To coordinate transmission over a common medium by the stations of the network, communications on the network typically follow a prescribed multiple access technique or protocol. Such protocols determine the sequence of actions to be performed by each station to avoid or reduce the impact of interference arising from transmission of other stations.
Multi-station networks may have different topologies. Common topologies include the multi-hub network shown in FIG.
1
.
FIG. 1
is a prior art network topology comprising a plurality of hubs
10
, a plurality of stations
15
and a plurality of full duplex channels
11
for inter-node communication.
FIG. 2
shows a prior art head-end networks topology comprising a head end
20
, two unidirectional wired or wireless channels for transmitting from the head-end to station
21
, and from the stations to the head-end
22
.
FIG. 3
shows a prior art wired or wireless bus-topology network, comprising a set of stations
35
, interconnected each other by a bi-directional channel
30
. Data transmission in the networks shown in
FIGS. 1-3
is coordinated by a multiple access technique or protocol which determines the sequence of actions to be performed by each node to avoid or reduce the impact of interference due to transmission by other nodes.
Conventional communication protocols and multiple access techniques have the common drawbacks that they do not provide an efficient means for collision-free data transmission with automatic adaptation to changing network conditions. This reduces the potential network efficiency because in common network situations there is random network traffic and node activity, variable distances between nodes, and variable data transmission speed.
Networks with a ring topology commonly use a token-ring multiple access. Networks based on a bus topology sometime use a token-bus multiple access. Token access techniques for networks with bus topology (such as that described by IEEE standard 802.4) or ring topology (such as that described by IEEE standard 802.5 or FDDI) are collision free but they require transmission of control information (tokens) from one network station to another. When the token is distorted a complex token-recovery procedure has to be performed, what is the major limitation of token access techniques. Access management procedures in local area networks with token access are comparatively complicated and implementation of such networks requires high investments.
To avoid collisions in high-speed local area networks the Demand Priority Protocol (DPP) was proposed for the IEEE 802.12 standard. This protocol is applicable in networks with star or tree topology consisting of one or more interconnected hubs and a plurality of network stations connected to the hubs. The hubs perform a round-robin poll to all connected stations or lower-level hubs.
While the DPP protocol is collision-free, one drawback of the DPP protocol is that the implementation of a polling algorithm tends to cause a loss of network throughput. This is because in large multi-level tree networks the protocol tends to be complex and requires a correspondingly complex network management system. Moreover, the DPP algorithm cannot be implemented on a conventional network with a bus topology. The polling technique of the DPP protocol requires all network nodes to be polled in each polling cycle. A major drawback of the DPP algorithm is that it does not automatically adapt for changes in node activity, network distances, or speed of transmission.
Networks which rely on the bus topology commonly use the Carrier Sense Multiple Access with Collision Detection (CSMA/CD) protocol. An example of the CSMA/CD protocol is disclosed in U.S. Pat. No. 4,063,220 issued on Dec. 13, 1988 to Metcalfe, et al. In accordance with the CSMA/CD protocol, whenever any network station has a packet for transmission, it senses the communication medium, and if the medium is quiescent, the packet is impressed on the network. If the medium is engaged, the station waits and transmits its packet when it detects that the medium is idle. In view of the signal propagation delay in the medium, it is possible that two or more stations may start transmitting on the medium almost simultaneously. These transmissions may become scrambled and a collision may appear in the network. Any packets that have collided must be retransmitted. While the CSMA/CD protocol does facilitate the successful transmission of data between network stations, a considerable amount of time and medium throughput is lost whenever a collision occurs. Collision probability increases with traffic load, transmission rate or network length. An increased collision probability results in a longer packet transmission delay. Consequently, the CSMA/CD protocol can not be used for real-time transmission of multimedia information. Moreover, the CSMA/CD protocol does not have any means for adaptation to network traffic, node activity, network distances or speeds of transmission.
One technique to improve the efficiency of CSMA/CD is disclosed in U.S. Pat. No. 4,628,311, issued Dec. 12, 1986 to Milling. The technique of U.S. Pat. No. 4,628,311 assigns to each station an access window within a predetermined response period after the medium becomes idle. To distribute transmission capacity among the stations th

Affiliated with

Also associated with

Rating

Adaptive universal multiple access does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Adaptive universal multiple access, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive universal multiple access will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2438322