Multiplex communications – Pathfinding or routing – Combined circuit switching and packet switching
Reexamination Certificate
1999-05-25
2003-06-03
Vanderpuye, Ken (Department: 2661)
Multiplex communications
Pathfinding or routing
Combined circuit switching and packet switching
C370S401000
Reexamination Certificate
active
06574218
ABSTRACT:
BACKGROUND OF THE INVENTION
1. The Field of the Invention
This invention relates generally to data transfer over a network or mixed networks. More specifically, this invention relates to acceptable multimedia data streaming over one or more combined networks in the presence of reduced bandwidth and less reliable network paths.
2. Relevant Technology
The public switched telephone network (PSTN) is designed to carry voice traffic as inexpensively as possible. Until about the end of the 1970's, the PSTN was an entirely analog communication system throughout the world. With the advent of digital computers, it became very desirable to provide a means through which computers could exchange digital information. Special signal transforming devices, called modems, were created, allowing digital devices to communicate over analog communication channels.
Since the end of the 1970's, the core of the PSTN in the United States and other industrialized countries has been completely digital. Still, for cost reasons, most users have an analog connection to their telephone company's digital central office. Since the bandwidth of this analog connection, again for cost reasons, is limited to about 3000 Hz, and since the signal-to-noise ratio is slightly over 30 dB, it follows that according to Shannon's theory, the maximum speed at which information can be exchanged is slightly over 56,000 bits/second. This maximum speed is presently achieved. See International Telecommunications Union, Telecommunication Standardization Sector (ITU-T) Recommendation V.90, Geneva, Switzerland (1998). Modems able to communicate at speeds up to 56,000 b/s are widely available in the marketplace with the largest modem vendor being the assignee, 3Com Corporation.
Until the 1990's, almost 99% of the traffic over the PSTN was voice traffic. The fact that the PSTN is poorly suited to carry data traffic was therefore of little concern since only small portion of the PSTN traffic was data. More recently, the Internet is causing an unprecedented data communication revolution. Currently, about 15% of the total PSTN traffic in the United States is data traffic. This figure is rapidly increasing and is expected to increase in the next several years to about 90%. The Internet continues to grow exponentially and the growth rate shows no sign of slowing. Additionally, the vast majority of present users are connected to the Internet using a V.34 (up to 33,600 bits/second) or V.90 (up to 56,000 b/s in the downstream direction) protocol.
Those familiar with communication network architectures appreciate that the PSTN is a circuit-switched network. A circuit-switched network is one in which the communicating entities are interconnected via a circuit or direct line dedicated interface. A circuit-switched network offers low bandwidth, but high reliability. The high reliability is largely due to the direct dedicated coupling of the communicating entities. Modems used in circuit-switched applications can communicate digital information at a low probability of bit error.
In contrast to the dedicated direct interface of a circuit-switched network, other topologies exist such as a packet-switched network. In a packet-switched network, rather than establishing a dedicated direct connection between the communicating entities, data information packets are addressed and delivered into the network. Routing entities within the network then examine the packet addressing associated with the data information packets and route the packets toward their destination. Additionally, while the PSTN was originally optimized to carry voice traffic, packet-switched networks such as the Internet are optimized to carry non real-time data traffic. Packet networks offer high bandwidth, but do not provide the necessary Quality of Service (QoS) for multimedia communications. Such a low QoS is primarily due to the fact that data information is delivered into a connectionless network where packets may be delivered late, out of order or even lost within the system, unlike in a circuit-switched environment where a direct connection is established between the communicating entities. In a packet-switched network bandwidth is measured in bits/second, as is common in computer networks.
A significant impediment to reliable transmission of multimedia over packet networks is packet delay, reordering, or loss. The most significant of these is packet loss, meaning the concept of bit error rate is meaningless in a packet-switched network. Packets may be lost for a variety of reasons, namely:
congestion of routers and gateways, which leads to a packet being discarded;
delays in packet transmission, which may cause a packet to arrive too late at the receiver to be played back in real-time;
heavy loading of the workstations, leading to scheduling difficulties in real-time multitasking operating systems.
To combat the realities of lost packets in a non-real-time system, there exist retransmission protocols such as TCP that facilitate recovery of lost packets. TCP operates by sending a positive acknowledgement only when a packet is received both in an expected sequence and within a designated time-out period. Furthermore, packets are often re-transmitted due to excessive delay, even though they may not be lost. Such unnecessary retransmission not only increases the overhead, but may be counter-productive in an attempt to maximize bandwidth. Multimedia, especially video, requires significant bandwidth. Unnecessary retransmission of packets can easily cause congestion resulting in exacerbated packet loss. It is widely recognized that TCP is not well-suited to real-time multimedia packet transfers especially those real-time “streaming” types of transfers.
Several approaches for multimedia streaming from a server to a client have been attempted. According to one approach, an entire multimedia file is downloaded using the existing protocols (such as TCP) from the server to the client and then, at the client, the file is played back locally. The shortcomings of such an approach are apparent in that only relatively small multimedia files may be downloaded, otherwise the client has to wait for a long time before the start of playback.
In a second approach, multimedia information is streamed immediately to the client without any re-transmission being preformed to recover lost packets. Such an approach eliminates the delay associated in the first approach, however, quality suffers dramatically as a result of packet loss. In general, packet loss can be between 3% and in some extreme cases up to 25%. Any prior success of either of the aforementioned approaches has been shortlived and are not presently commercially viable. Therefore, other approaches continue to be actively investigated.
One alternate approach is described in the “Real Time Video and Audio in the World Wide Web”, by Z. Chen, S.-M. Tan, R. H. Campbell and Y. Li, published in the Fourth Int. WWW conference, 1995. That approach recognizes that not all packets have inherent equal value in a multimedia stream, for example, some packets are more important to an individual perceiving the multimedia data stream than others. In the above approach, a receiver detects which packets are lost and only requests re-transmission of the more important packets. In such an approach, the client also maintains control of the bit rate of the streaming based on the packet loss rate and any re-transmission requests. Such an approach is still not very efficient when the packet loss rate is high. It can lead to congestion and unacceptably low quality.
An additional alternative scheme is described in U.S. Pat. No. 5,768,527, assigned to Motorola Inc. of Schaumburg, Ill. While that patent takes into consideration the low bandwidth provided by dial-up modems as a result of the fundamental limitations of the PSTN, its fundamental disadvantage is that the QoS manager is situated at the client and is responsible for the QoS over both the packet network and the low-speed access link. Thus, from the client's point of view, the overa
3Com Corporation
Vanderpuye Ken
Workman & Nydegger & Seeley
LandOfFree
Method and system for spatially disjoint joint source and... 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 and system for spatially disjoint joint source and..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for spatially disjoint joint source and... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3151494