Abstract: Performance of transport protocols over a mobile data network is improved by a mobile accelerator, which performs protocol optimizations on-the-fly without requiring any modification to server or client protocol implementations in the operating systems of the server or client. The proposed mobile accelerator is compatible with existing TCP protocol implementations at the server and the client device, without server/client OS modification.

Abstract: An embodiment of the invention includes an efficient error-control system and method for recovering packet losses, especially losses in distributing multicast video over broadband residential networks. Preferably, unlike most existing error-control algorithms designed for Internet multicast, the system and method does not employ substantial feedback suppression. Preferably, the system and method does not employ substantial multicasted retransmission. Preferably, the system and method does not employ substantial parity retransmission. Preferably, the system and method does not employ substantial local loss recovery. The system and method integrates two existing classes of error-control algorithms: Automatic Repeat Request (ARQ) and Forward Error Correction (FEC), to reduce traffic overhead and achieve scalability.

Abstract: In a packet communication system with large bandwidth delay product (BDP) and employing Transmission Control Protocol (TCP), the reported advertised window size as advertised by a receiver is employed only as an indication of window size and not as a throughput limit so that more data is sent than the amount specified by the advertised window size. Since the receiver can process all incoming TCP packets, the receiving buffer is consistently near empty. Since the TCP's advertised window size is not accepted as the absolute amount of buffer space available at the receiver, the sender is not constrained by the absolute value of the receiver's advertised window size and instead can transmit more data than the absolute value of the advertised window, enabling the system to increase the actual window size without modifying the link ends. This improved large-BDP-capable protocol is denoted TCP-SC.

Abstract: Performance of transport protocols over a mobile data network is improved by a mobile accelerator, which performs protocol optimizations on-the-fly without requiring any modification to server or client protocol implementations in the operating systems of the server or client. The proposed mobile accelerator is compatible with existing TCP protocol implementations at the server and the client device, without server/client OS modification.

Abstract: In a peer-to-peer asymmetric communication environment, an adaptive upload Data Rate Control element—called Adaptive-DRC—is used to detect increase in data return-trip time (RTT) caused by the congestion associated with the uploading of data in an uplink which in turn impacts on downloading data on an associated downlink, namely a downlink sharing resources with the uplink. Using Adaptive-DRC, the congestion is detected and thereupon adjustments to are made to allow continued communication. Adaptive-DRC comprises: (i) RTT estimation; (ii) detection of the onset of congestion, herein pre-congestion detection; and (iii) adaptation of the upload data rate limit. Two specific embodiments are contemplated, namely two versions of Adaptive-DRC wherein the difference lies in the upload data rate limit adaptation technique.

Abstract: In a packet communication system with large bandwidth delay product (BDP) and employing Transmission Control Protocol (TCP), the reported advertised window size as advertised by a receiver is employed only as an indication of window size and not as a throughput limit so that more data is sent than the amount specified by the advertised window size. Since the receiver can process all incoming TCP packets, the receiving buffer is consistently near empty. Since the TCP's advertised window size is not accepted as the absolute amount of buffer space available at the receiver, the sender is not constrained by the absolute value of the receiver's advertised window size and instead can transmit more data than the absolute value of the advertised window, enabling the system to increase the actual window size without modifying the link ends. This improved large-BDP-capable protocol is denoted TCP-SC.

Abstract: An embodiment of the invention includes an efficient error-control system and method for recovering packet losses, especially losses in distributing multicast video over broadband residential networks. Preferably, unlike most existing error-control algorithms designed for Internet multicast, the system and method does not employ substantial feedback suppression. Preferably, the system and method does not employ substantial multicasted retransmission. Preferably, the system and method does not employ substantial parity retransmission. Preferably, the system and method does not employ substantial local loss recovery. The system and method integrates two existing classes of error-control algorithms: Automatic Repeat Request (ARQ) and Forward Error Correction (FEC), to reduce traffic overhead and achieve scalability.

Abstract: An embodiment of the invention includes an efficient error-control system and method for recovering packet losses, especially losses in distributing multicast video over broadband residential networks. Preferably, unlike most existing error-control algorithms designed for Internet multicast, the system and method does not employ substantial feedback suppression. Preferably, the system and method does not employ substantial multicasted retransmission. Preferably, the system and method does not employ substantial parity retransmission. Preferably, the system and method does not employ substantial local loss recovery. The system and method integrates two existing classes of error-control algorithms: Automatic Repeat Request (ARQ) and Forward Error Correction (FEC), to reduce traffic overhead and achieve scalability.

Abstract: An embodiment of the invention includes an efficient error-control system and method for recovering packet losses, especially losses in distributing multicast video over broadband residential networks. Preferably, unlike most existing error-control algorithms designed for Internet multicast, the system and method does not employ substantial feedback suppression. Preferably, the system and method does not employ substantial multicasted retransmission. Preferably, the system and method does not employ substantial parity retransmission. Preferably, the system and method does not employ substantial local loss recovery. The system and method integrates two existing classes of error-control algorithms: Automatic Repeat Request (ARQ) and Forward Error Correction (FEC), to reduce traffic overhead and achieve scalability.

Abstract: An embodiment of a system and method for providing video on demand provides pre-scheduled multicasts of videos as well as dynamically initiated transmissions of the front portion of videos. Users may first receive a dynamically initiated front portion of a video and then be merged into a pre-scheduled multicast. Preferably, the dynamically initiated transmission is also a multicast. Preferably, multiple admission controllers and a single server coordinate the dynamically initiated transmissions for any one video. Preferably, interactive controls are supported without requiring extra server-side resources. Preferably, latency is automatically equalized between users admitted via the pre-scheduled and the dynamically initiated transmissions. Preferably, a user receiving a video via a pre-scheduled multicast does not need to change channels to finish receiving the video transmitted.

Abstract: An embodiment of a system and method for providing video on demand provides pre-scheduled multicasts of videos as well as dynamically initiated transmissions of the front portion of videos. Users may first receive a dynamically initiated front portion of a video and then be merged into a pre-scheduled multicast. Preferably, the dynamically initiated transmission is also a multicast. Preferably, multiple admission controllers and a single server coordinate the dynamically initiated transmissions for any one video. Preferably, interactive controls are supported without requiring extra server-side resources. Preferably, latency is automatically equalized between users admitted via the pre-scheduled and the dynamically initiated transmissions. Preferably, a user receiving a video via a pre-scheduled multicast does not need to change channels to finish receiving the video transmitted.

Abstract: An embodiment of the invention includes an efficient error-control system and method for recovering packet losses, especially losses in distributing multicast video over broadband residential networks. Preferably, unlike most existing error-control algorithms designed for Internet multicast, the system and method does not employ substantial feedback suppression. Preferably, the system and method does not employ substantial multicasted retransmission. Preferably, the system and method does not employ substantial parity retransmission. Preferably, the system and method does not employ substantial local loss recovery. The system and method integrates two existing classes of error-control algorithms: Automatic Repeat Request (ARQ) and Forward Error Correction (FEC), to reduce traffic overhead and achieve scalability.