Multiplex communications – Channel assignment techniques – Combining or distributing information via time channels...
Reexamination Certificate
1998-03-20
2001-06-26
Chin, Wellington (Department: 2664)
Multiplex communications
Channel assignment techniques
Combining or distributing information via time channels...
C370S503000
Reexamination Certificate
active
06252884
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the management of computer devices within a wireless network, where the computer devices are apt to appear and disappear on a regular basis.
2. Description of the Prior Art
Wireless networks connecting computer devices are a necessary component of distributed, collaborative, and portable applications. However wireless networks pose a unique set of software problems to be solved. Traditional wired networks are relatively static. When a new computer or branch of the network is added, it can be assumed that it will exist for a lengthy period of time. In a dynamic, volatile user environment such as emergency medical care, people and computer components can be arriving and departing on a minute by minute basis. If the computer components are running applications that wish to share data, or if the computers wish to share applications themselves, it is necessary to automatically manage the appearance and disappearance of network connections.
Unfortunately, existing networking methodologies are primarily designed to assume that the various computer devices are static—e.g., that they will always be in range and connected to the network. Of course, this may not always be the case, especially in certain applications. For example, with respect to the emergency medical care environment mentioned previously, medical technicians and medical vehicles may each be outfitted with computer devices for collecting and sharing medical information, etc. Because such computer devices may be brought into use with each other in a dynamic fashion, it cannot be assumed that any of these devices will be present, nor in any specific configuration. Therefore, prior art wireless network maintenance systems are simply unsuited to such environments.
There is therefore a significant need in the art for a system for dynamically managing a wireless network of computer devices, so as to ensure that each device will properly be connected to the network.
SUMMARY OF THE INVENTION
In the present invention, each computer that is to participate in the dynamic network continuously broadcasts its address to any other computer within range of the wireless network hardware. To minimize the overhead on the available communications bandwidth, this broadcast only contains a number identifying this message as an address broadcast and another number representing the address of the sending machine. This message must be sent as often as the network is expected to change (for example, once per minute for a highly dynamic network).
When a computer receives a broadcast message from a machine it is not currently connected to, it can then use any standard communications protocol (i.e., TCP/IP) to establish a connection to the broadcasting machine. Depending on the application requirements, a set of rules might be consulted to decide whether to connect to a particular machine or not. Once the connection is established, a message is sent to the broadcasting machine notifying it of the new connection. This allows for either client/server, peer-to-peer, or other communications strategies to be implemented, depending on the application.
Upon establishing a new connection between a pair of computers, a data synchronization protocol is employed to exchange data, applications, or configure services. Each machine in the newly connected pair sends a set of messages to the other. Each of these messages contains an identifier for the data object, application, or service the computer can supply along with the status of the data object, application, or service (the status could be the last date and time a particular data object was updated, for example). When a machine receives a data synchronization message, it can look at the identifier and status to decide whether to send a request to its partner machine. For example, if it notices that a data object it has interest in has a more recent update time stamp, it can request a new copy of the data object. The number and type of data synchronization messages and the response to those messages can vary to satisfy specific application requirements.
In a mobile environment, it is likely that a computer could move out of communication contact for a brief time (seconds or minutes) and then come back into range. To avoid having many disconnects, reconnects, and data synchronizations happening, a connection degradation strategy is used. When a connection is first established or when any data is received from a connection (including a broadcast message), that connection is marked as LIVE. At regular timed intervals, all the connections a machine has are downgraded one level. From LIVE, a connection moves to STALE; from STALE, to DEAD; and from DEAD, to DISCONNECTED. The amount of time between downgrades should be closely tied to the broadcast message rate. A simple implementation is to downgrade before each broadcast, if it can be assumed that each machine participating in the dynamic network is broadcasting at the same rate. If an application attempts to send data on a connection that is any state other than LIVE, the data is queued up and not transmitted. As soon as the connection becomes LIVE again, any queued data can be sent according to any scheduling rules that are in place. Once a connection has reached the DISCONNECTED state, any termination required by the underlying communications protocol can be done. No data can be sent to or received from the machine at the other end of a DISCONNECTED connection until a broadcast message has caused a reconnect and data synchronization to happen.
REFERENCES:
patent: 5088094 (1992-02-01), Grauel et al.
patent: 5732360 (1998-03-01), Jarett et al.
patent: 5737319 (1998-04-01), Croslin et al.
patent: 6034966 (2000-03-01), Ota
patent: 6049535 (2000-04-01), Ozukturk et al.
Chin Wellington
Jones Premell
NCR Corporation
Needle & Rosenberg P.C.
LandOfFree
Dynamic configuration of wireless networks does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Dynamic configuration of wireless networks, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic configuration of wireless networks will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2502510