Electrical computers and digital processing systems: multicomput – Computer-to-computer protocol implementing – Computer-to-computer data transfer regulating
Reexamination Certificate
1997-03-14
2001-09-18
Maung, Zarni (Department: 2152)
Electrical computers and digital processing systems: multicomput
Computer-to-computer protocol implementing
Computer-to-computer data transfer regulating
C709S216000, C709S225000
Reexamination Certificate
active
06292834
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to multimedia communications. More particularly, the present invention relates to the efficient and reliable delivery of multimedia streams over a diverse computer network with dynamically variable bandwidth capacity.
2. Description of the Related Art
With the proliferation of connections to the internet by a rapidly growing number of users, the viability of the internet as a widely accepted medium of communication has increased correspondingly. Bandwidth requirements can vary significantly depending on the type of multimedia data being delivered. For example, a low resolution, low frame rate video telephone call may require only an ISDN connection, while a high resolution video broadcast of a live event to a large group of viewers may require the bandwidth of a T1 connection to the server. Hence, the ability to efficiently deliver multimedia data over a diverse computer network such as the internet is severely limited by the reliability and bandwidth capacity of the network connection.
The first problem is the average transmission capacity. In an ideal packet-based delivery system with an input buffer at a client computer, data packets arrive at the client computer in the same order and at the same interval the packets were sent by the server. In an ideal example, at time t, the client computer receives data packet #1 with a time stamp of 0.0 second. Subsequently, at time t+1 second, data packet #2 with a time stamp of 1.0 second arrives, followed by data packet #3 with a time stamp of 2.0 seconds at time t+2 seconds. As a result, packets arrive at and are consumed by the client computer at the same rate as they were sent.
However in a more realistic example, the network connection may be unable to keep up with the demands of the server/client stream traffic, i.e., the average bandwidth capacity of the network connection may be insufficient. Consequently, data packets will arrive at the client computer later and later in time, causing the input buffer to empty at a faster rate than it can be replenished, and eventually depleting the input buffer. For example, data packet #2 with the time stamp of 1.0 seconds may arrive at time t+1.2 seconds, followed by data packet #3 with the time stamp of 2.0 seconds arriving at time t+2.4 seconds. In other words, the average bandwidth capacity of the network connection is insufficient to support the transmission rate selected by the server/client. This is a first order bandwidth capacity problem.
The second problem is the rate of change of bandwidth capacity over time of the network connection. Since overall traffic within the internet is not constant, and since the internet is packet-switched, the bandwidth capacity provided by the internet for the network connection can vary dynamically over time. Accordingly, if an application is too aggressive in demanding bandwidth, during peak demand periods, the internet may be unable to cope with the peak demand, causing packets to be discarded/lost and requiring retransmission, which further degrades the overall performance of the network connection. This is a second order network bandwidth problem, i.e., changes in the bandwidth capacity over time.
In a real time application, e.g., a video on demand (VOD) application, the discarded/lost packets result in jitter. Jitter is defined as the second order timing difference in the packet arrival times. In the ideal example, where packet #2 and packet #3 arrive at t+1.0 second and t+2.0 seconds, respectively, jitter is zero., because the inter-arrival times, i.e., differences in arrival times, are identical.
However, in a more realistic example, packets #2 and #3 may arrive at t+0.9 second and t+2.1 seconds, with inter-arrival times of 0.9 second and 1.2 seconds, respectively. Although the input buffer of the client computer provides partial relief by buffering the incoming packets and releasing them to applications on the client computer at a less jittery rate, unfortunately, in the real time application, the length of the input buffer has to be kept to a minimum, thereby severely limiting the relief attainable.
In view of the foregoing, there are desired improved techniques for reliable and efficient transmission of multimedia streams to client(s) which efficiently utilize the network resources available over a period of time.
SUMMARY OF THE INVENTION
The present invention provides efficient transmission of multimedia streams from a server to a client computer over a diverse computer network including local area networks (LANs) and wide area networks (WANs) such as the internet. Examples of multimedia streams provided to the client computer include a compressed video stream, a compressed audio stream, and an annotation stream with pointers to textual/graphical data in the form of HTML pages.
In one embodiment, the client computer includes a playout buffer, and the transmission rate is dynamically matched to the available bandwidth capacity of the network connection between the server and the client computer.
If a playtime of the playout buffer, which is one measure of the number of data packets currently in the playout buffer, drops below a dynamically computed Decrease_Bandwidth (DEC_BW) threshold, then the transmission rate is decreased by sending a DEC_BW message to the server.
Conversely, if the number of packets remaining in the playout buffer rises above a dynamically computed Upper Increase_Bandwidth (INC_BW) threshold and does not drop below a Lower INC_BW threshold for at least an INC_BW wait period, then the transmission rate is incremented.
In this embodiment, the transmission rate is selected from among a predetermined set of discrete bandwidth values. However the invention is also applicable to a system in which the transmission rate is selected from within a continuous range of bandwidth values.
In another embodiment, in addition to responding to variations in network connection capacity, the client computer also determines an average client computational capacity. Accordingly, if the average client computational capacity is less than the network capacity, the lower of the two capacities is the determining one, thereby avoiding a playout buffer overrun.
These and other advantages of the present invention will become apparent upon reading the following detailed descriptions and studying the various figures of the drawings.
REFERENCES:
patent: 4862264 (1989-08-01), Wells et al.
patent: 4931950 (1990-06-01), Isle et al.
patent: 5025457 (1991-06-01), Ahmed
patent: 5050161 (1991-09-01), Golestani
patent: 5088107 (1992-02-01), Piasecki et al.
patent: 5119474 (1992-06-01), Beitel et al.
patent: 5208810 (1993-05-01), Park
patent: 5231599 (1993-07-01), Peters et al.
patent: 5274758 (1993-12-01), Beitel et al.
patent: 5313454 (1994-05-01), Bustini et al.
patent: 5359593 (1994-10-01), Derby et al.
patent: 5434848 (1995-07-01), Chimento, Jr. et al.
patent: 5442389 (1995-08-01), Blahut et al.
patent: 5455910 (1995-10-01), Johnson et al.
patent: 5463422 (1995-10-01), Simpson et al.
patent: 5467413 (1995-11-01), Barrett
patent: 5485211 (1996-01-01), Kuzma
patent: 5487167 (1996-01-01), Dinallo et al.
patent: 5490252 (1996-02-01), Macera et al.
patent: 5504744 (1996-04-01), Adams et al.
patent: 5519701 (1996-05-01), Colmant et al.
patent: 5524193 (1996-06-01), Covington et al.
patent: 5533021 (1996-07-01), Branstad et al.
patent: 5537408 (1996-07-01), Branstad et al.
patent: 5543850 (1996-08-01), Pratt et al.
patent: 5544170 (1996-08-01), Kasahara
patent: 5566175 (1996-10-01), Davis
patent: 5574724 (1996-11-01), Bales et al.
patent: 5574861 (1996-11-01), Lorvig et al.
patent: 5577258 (1996-11-01), Cruz et al.
patent: 5583980 (1996-12-01), Anderson
patent: 5594911 (1997-01-01), Cruz et al.
patent: 5600775 (1997-02-01), King et al.
patent: 5602992 (1997-02-01), Danneels
patent: 5623690 (1997-04-01), Palmer et al.
patent: 5633810 (1997-05-01), Mandal et al.
patent: 5633859 (1997-05-01), Jain et al.
Chaddha Navin
de Val David
Klemets Anders Edgar
Ravi Hemanth Srinivas
Cardone Jason D.
Lee & Hayes PLLC
Maung Zarni
Microsoft Corporation
LandOfFree
Dynamic bandwidth selection for efficient transmission of... 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 bandwidth selection for efficient transmission of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Dynamic bandwidth selection for efficient transmission of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2508677