Showing papers on "Packet loss published in 1987"

On Packet Switches with Infinite Storage

Abstract: Most prior work on congestion in datagram systems focuses on buffer management. We find it illuminating to consider the case of a packet switch with infinite storage. Such a packet switch can never run out of buffers. It can, however, still become congested. The meaning of congestion in an infinite-storage system is explored. We demonstrate the unexpected result that a datagram network with infinite storage, first-in, first-out queueing, at least two packet switches, and a finite packet lifetime will, under overload, drop all packets. By attacking the problem of congestion for the infinite-storage case, we discover new solutions applicable to switches with finite storage.

The packer filter: an efficient mechanism for user-level network code

Abstract: Code to implement network protocols can be either inside the kernel of an operating system or in user-level processes. Kernel-resident code is hard to develop, debug, and maintain, but user-level implementations typically incur significant overhead and perform poorly.The performance of user-level network code depends on the mechanism used to demultiplex received packets. Demultiplexing in a user-level process increases the rate of context switches and system calls, resulting in poor performance. Demultiplexing in the kernel eliminates unnecessary overhead.This paper describes the packet filter, a kernel-resident, protocol-independent packet demultiplexer. Individual user processes have great flexibility in selecting which packets they will receive. Protocol implementations using the packet filter perform quite well, and have been in production use for several years.

Packet switching system arranged for congestion control

Abstract: A method for controlling congestion in a packet switching network uses a packet dropping algorithm to determine when to drop a marked packet wherever the network is congested at any point along the path being traversed by the marked packet.

Packet switching system arranged for congestion control through bandwidth management

Abstract: In a packet switching network, packet monitoring and marking algorithms are used for determining which data packets, received from a customer by an access node, are being transmitted at an excessive transmission rate and accordingly are marked. Additionally every packet from a special customer can be marked. Along in the network, marked packets are dropped where the network is congested along the path being traversed by the data packets.

Industrial control communication network and method

Abstract: A network consists of a programmable controller coupled to several sensors by an interface circuit. A common communication protocol is used to exchange messages containing commands and data between the devices coupled to the network. A protocol message packet has a header with fields for a task command, sensor identification, device status information and error codes. The header contains the same fields whether the message packet is for the interface circuit or one of the sensors connected to it. The headers for message packets going to and from the programmable controller and the interface circuit also have the same fields although the contents of the fields may vary depending upon the direction of the message packet. The message packet may also contain several data blocks each specifying a separate operation for the sensor to perform.

Method for error free message reception

Abstract: A method for ensuring accurate reception of digitally encoded messages that are transmitted over a broadcast network to a plurality of receivers. Each message is divided into packets and the set of packets comprising each message is cyclically retransmitted for a selected number of transmission cycles. Typically, from two to five transmission cycles are sufficient to ensure accurate reception of each message. Each packet includes all or part of a digitally encoded message, a header, and a cyclic redundancy check (CRC) code (or other error code). Each header identifies the message to which the packet belongs and the position of the packet within the message. Each receiver in the network stores an address in an internal memory, and is capable of comparing a transmitted address (in the header of one or more of the packets in each message) with its stored address and accepting only those packets having an appropriate address. Each accepted packet is checked for errors using the transmitted error code (which preferably is a CRC code). If no errors are detected, the packet number of the packet is added to a list of correctly received packets for the message. If an error is detected, the packet number of the packet is not added to this list. Upon retransmission of the same message, the receiver checks the list against the header of each retransmitted packet of the message, and accepts only those packets which were not previously received without error. In a preferred embodiment, the messages are transmitted in the SCA band of an FM channel. Each receiver preferably includes an appropriately programmed computer in which the signal processing operations comprising the invention are performed in software.

Time stamp and packet virtual sequence numbering for reconstructing information signals from packets

Abstract: Unwanted distortion in signals being reconstructed from packets is substantially eliminated by employing a virtual sequence packet numbering scheme in conjunction with a time stamp value. A virtual sequence number which accompanies the packet identifies the location of the packet in an information spurt. If the packet is the initial packet in the information spurt, the time stamp value is used to "build out" the delay experienced by the packet to a fixed overall value. Subsequent packets in the information spurt are concatenated to the information spurt. A packet following one or more lost packets is also identified by the packet virual sequence number and is also treated as an initial packet.

Packet flow control method

Abstract: A packet flow control method where delay data are added into an initial packet as it traverses a packet switching network, and where the receiver of the initial packet, rather than the sender, establishes a window size based on such delay data to be used for the duration of a packet connection through the network The delay data allow for the calculation of an average rather than an instantaneous network delay such that the flow control mechanism is not dependent on the magnitude of network congestion that happens to be present when the connection is first established Since the receiver determines the window size, the flow control mechanism is put in place as an integral part of the initial packet exchange used to establish the two-way packet connection rather than requiring an additional packet communication to the receiver after a window size calculation by the sender

Packet switching network

Abstract: Packet switching network with first (DN) and second (RN) cascaded parts including first and second switching modules respectively In the second switching modules the path selection is controlled by routing information contained in the packets In the first switching modules this selection is performed without using routing information only for a path set up packet, whilst for the following packets use is made of routing information on the route followed by the path set up packet Each module decides to multiplex an input stream on an output only when a calculated traffic load is smaller than a limit value This load is calculated from traffic load parameters contained in the path set up packet

Performance boundaries for prioritized multiplexing systems

Abstract: Systems in which many data sources are multiplexed over a single communication channel are considered. Data from all the sources are generated in fixed-length packets and are stored in a common buffer with finite capacity. Packets that overflowed or were removed from the buffer prior to transmission are lost. The system performance measure is the set of packet loss probabilities associated with the sources. Queueing disciplines vary depending on the stringency of prioritization and the utilization of the system resources. The set of all possible performances is characterized as we span the set of all queueing disciplines. Whether a given performance is possible can be deduced. Strategies that achieve the minimum overall loss probability are identified. The extreme disciplines are specified, and their performances are calculable by means of a given algorithm.

Queueing Analysis For Packet Switched Video

Abstract: Packet switching of variable bit-rate real-time video sources is a means for the efficient sharing of communication resources, while maintaining a uniform picture quality Performance analyses for the statistical multiplexing of such video sources are required as a first step towards assessing the feasibility of packet switched video This paper extends our earlier work in modelling video sources which have been coded using inter-frame coding schemes, and in carrying out buffer queueing analyses for the multiplexing of several such sources Our previous models and analysis were suitable for relatively uniform activity scenes Here we consider models and queueing analysis for more realistic scenes with multiple activity levels where the coder output bit-rates may change violently We present correlated Markov source models for the corresponding sources, and using a flow-equivalent queueing analysis, obtain common buffer queue distributions and probabilities of packet loss Our results demonstrate efficient resource sharing of packetized video on a single link, due to the smoothing effect of multiplexing several variable-rate video sources

Perceived degradation effects in packet speech systems

Abstract: A packet speech system's sound quality is degraded when packets do not arrive at their destination in a timely fashion. This can happen when packets are lost in a network or when network delay is excessive. Simulation results are reported which measure the effects of such losses and delays on sound quality and intelligibility. Thirty human subjects were asked to listen to the output of a simulated packet speech system in which some packets are missing. Sound quality and intelligibility were measured as a function of packet loss and packet length.

Improvement to digital packet switching networks

Abstract: In a packet switching network, voice packets are made to include an EC field whose contents indicates whether the corresponding packet is eligible for clipping if required in a node queue within the network. Should clipping of a non eligible packet be required, then another bit field would be set in the following packet on same link to limit any possible clipping of successive packets an same link.

Packet-speech multiplexer

Abstract: The paper attempts to clarify some issues concerning the use, and usefulness, of the packet-speech multiplexer. An accurate mathematical analysis is used to find the trade-off between delay and packet loss. The subjective effect of delay and packet loss is determined using data from our own, and published, conversation tests. It is found that the multiplexer will operate most efficiently if queueing delay is limited, with a large delay limit for heavy traffic conditions and a small delay limit for light traffic conditions. The improvement in TASI (time assignment speech interpolation) advantage gained through allowing delay is small, and it is concluded that the packet-speech multiplexer is unlikely to be used purely for its buffering facility. However, its ability to multiplex speech channels where the information rate is continually varying may be of significance for the next generation of speech coders.