Showing papers on "Packet loss published in 1989"

Models for packet switching of variable-bit-rate video sources

Abstract: The authors extend earlier work (ibid., vol.36, p.834-44, Jul. 1988) in modeling video sources using interframe coding schemes and in carrying out buffer queueing analysis for the multiplexing of several such sources. The previous models and analysis were suitable for relatively uniform activity scenes. Here, models are considered for scenes with multiple activity levels which lead to sudden changes in the coder output bit rates. Such models apply to talker-listener alternating scenes, as well as to situations where there is a mix of dissimilar services, e.g., television and videotelephony. Correlated Markov models for the corresponding sources are given. A flow-equivalent queueing analysis is used to obtain common buffer queue distributions and probabilities of packet loss. The results demonstrate the efficiency of packet video on a single link, due to the smoothing effect of multiplexing several variable-bit-rate video sources. >

Selective recovery of video packet loss using error concealment

Abstract: An efficient recovery method using error concealment is proposed for video packet loss in fast packet switching networks. In this method, the receiver detects the damaged picture area caused by packet loss from the structured picture data received, makes error concealments, notifies the transmitter, and continues decoding. The transmitter, having received the notice, calculates the affected picture area in the local decoded picture and continues encoding without using this affected area. In this selective recovery method, video signals are not stopped even if a long propagation delay exists, no additional information is transmitted to error recovery and conventional coding algorithms can be used. The proposed method is suitable for multipoint communication. Simulation results show the affected picture area is localized for a considerable time attesting to the method's effectiveness. >

Comparison of deflection and store-and-forward techniques in the Manhattan Street and Shuffle-Exchange Networks

Abstract: The Manhattan Street Network (MS-Net) and Shuffle-Exchange Network (SX-Net) are two-connected networks with significantly different topologies. Fixed-size packets are transmitted between nodes in these networks. The nodes are synchronized so that all of the packets that are received by a node within a slot transmission time arrive at a switching point simultaneously. Instead of storing large numbers of packets at intermediate nodes, a deflection strategy similar to hot-potato routing is used. There are characteristics of the MS-Net that make it well suited for deflection routing. With no buffer, 55-70% of the throughput with an infinite number of buffers has been obtained; with a single buffer per node, the throughput increases to 80-90%. With uniform load the throughput does not decrease significantly as the network utilization increases. Therefore, additional flow control mechanisms are not required to achieve the highest network throughput. The SX-Net does not have the above characteristics of the MS-Net. However, deflection routing still provides a significant portion of the available throughput. In the SX-Net, more buffers are required than in the MS-Net, and a flow control mechanism must be used to achieve the greatest throughput. >

Variable bit-rate coding of video signals for ATM networks

Abstract: Statistical characteristics of video signals for video packet coding, are clarified and a variable-bit-rate coding method for asynchronous transfer mode (ATM) networks is described that is capable of compensating for packet loss ATM capabilities are shown to be greatly affected by delay, delay jitter, and packet loss probability Packet loss has the greatest influence on picture quality Packets may be lost either due to random bit error in a cell header or to network control when traffic is congested A layered coding technique using discrete-cosine transform (DCT) coding is presented which is suitable for packet loss compensation The influence of packet loss on picture quality is discussed, and decoded pictures with packet loss are shown The proposed algorithm was verified by computer simulations >

Packet congestion control method and packet switching equipment

Abstract: A packet switching equipment housing therein a plurality of pairs of an input line and an output line is provided with a monitor circuit for monitoring a packet congestion state in the packet switching equipment for each output line. When a packet congestion is detected in association with either one of the output lines, a congestion indicator is added to a packet to be delivered to the output line so as to return the packet as a congestion notice packet to an equipment as the transmission source of the packet; furthermore, the input packet is relayed via the output line to the destination equipment.

Adaptive data link protocol

Abstract: A communication system provides high speed transmission of data over a link, such as a fiber optic link, between a first terminal and a second terminal. The architecture and protocol permits the use of dedicated hardware such as state machines constructed of programmable array logic units, to synchronize the transmission and reception of data packets and the retransmission of designated ones of these packets in the event of a faulty transmission. Packets to be transmitted and received are stored in an array of frames in sub-windows of a memory storage window in each of the termianls, the frame number being equal to the sequence number of the data packet. By embedding sequence and status bits in each packet within control words and bits appended to each packet, the state machine in each terminal can readily track the progress of each packet so as to request acknowledgement of error-free receipt, to send an acknowledgement, to request a retransmission of a packet designated by its serial number and to distinguish a retransmitted packet from an original packet transmitted with error.

Packet voice/data communication system having protocol independent repetitive packet suppression

Abstract: A data communication system includes a repetitive pattern packet suppression technique which suppresses transmission of entire packets in a data stream when a repeating pattern has been established in the previous packet and then is found to repeat throughout the following packets. An expansion part of the technique fills the resulting hole in the data stream with the last pattern from the previously received packet.

Analysis of random drop for gateway congestion control. M.S. Thesis

Abstract: Lately, the growing demand on the Internet has prompted the need for more effective congestion control policies. Currently No Gateway Policy is used to relieve and signal congestion, which leads to unfair service to the individual users and a degradation of overall network performance. Network simulation was used to illustrate the character of Internet congestion and its causes. A newly proposed gateway congestion control policy, called Random Drop, was considered as a promising solution to the pressing problem. Random Drop relieves resource congestion upon buffer overflow by choosing a random packet from the service queue to be dropped. The random choice should result in a drop distribution proportional to the bandwidth distribution among all contending TCP connections, thus applying the necessary fairness. Nonetheless, the simulation experiments demonstrate several shortcomings with this policy. Because Random Drop is a congestion control policy, which is not applied until congestion has already occurred, it usually results in a high drop rate that hurts too many connections including well-behaved ones. Even though the number of packets dropped is different from one connection to another depending on the buffer utilization upon overflow, the TCP recovery overhead is high enough to neutralize these differences, causing unfair congestion penalties. Besides, the drop distribution itself is an inaccurate representation of the average bandwidth distribution, missing much important information about the bandwidth utilization between buffer overflow events. A modification of Random Drop to do congestion avoidance by applying the policy early was also proposed. Early Random Drop has the advantage of avoiding the high drop rate of buffer overflow. The early application of the policy removes the pressure of congestion relief and allows more accurate signaling of congestion. To be used effectively, algorithms for the dynamic adjustment of the parameters of Early Random Drop to suite the current network load must still be developed.

Study of information loss in packet voice systems

Abstract: Once a voice buffer is full, it remains full for a certain period, during which many packets are possibly blocked, resulting in consecutive clippings in voice. The packet loss rate during this period changes slowly and has large fluctuations. It is shown that the temporal behavior of packet loss, especially at high rate, is inherently determined by voice correlation and system capacity and is independent of buffer size. Buffering may reduce the occurrence of short blocking periods associated with low rates packet loss but does not affect long ones associated with high packet loss rates. In fact, increasing the buffer size merely extends nonblocking periods, and thereby reduces the overall average packet loss rate, but packet-loss performance within existing blocking periods is not significantly improved. A simple tool is developed for calculating the boundary performance. It is found that it is possible to design a packet-switched voice system without buffering only at the expense of supporting a fewer number of calls. The issue of voice delay allocation between source and network is discussed, and it is shown that it is more effective to keep the network delay short while extending the source delay. >

Effect of speedup in nonblocking packet switch

Abstract: The nonblocking packet switch under consideration has N inputs and N outputs and operates L times as fast as the input and output trunks. The effect of speedup (L) on packet loss probability and average transmission delay in the case of an arbitrary number L, such that 1 >

Missing packet recovery techniques for low-bit-rate coded speech

Abstract: Since missing-packet recovery techniques for conventional PCM speech are not applicable to packetized speech communication systems with low-bit-rate coding schemes, quality degradation mechanisms are presented for missing-packet recovery techniques. These mechanisms are least significant bit (LSB) dropping, waveform substitution, and odd-even sample-interpolation schemes. A comparison of these techniques in terms of signal-to-noise ratio and perceptual distortion under packet loss conditions shows that the LSB-dropping scheme with embedded coding is the most promising technique for recovering missing packets. >

Congestion control method and apparatus for end-to-end packet communication

Abstract: In a packet switched communications system wherein each destination data terminal sends an acknowledgment packet signalling correct receipt of packets from a source terminal, a congestion detector is provided for detecting a traffic congestion in the system to enable a packet detector. The packet detector, when enabled, detects the receipt of an acknowledgment packet from the destination terminal and stores this packet in a buffer for a specified period of time. The stored packet is then forwarded toward the source terminal upon termination of the specified time period.

Packet recovery and error correction in high-speed wide-area networks

Abstract: The author presents a novel technique for reducing packet loss rate in high-speed wide-area networks in which the BER (bit-error rate) is low. Grouping packets into blocks and adding a packet that computes parity over bits of all packets in a block allow a data recipient to reconstruct any single packet in a block, using the other packets and the block parity packet. The missing packet is identified by observing a sequence-number gap in the stream of incoming packets. Adding another packet containing parity information over the diagonals of a series of blocks allows the decoder to correct a single bit error and reconstruct a missing packet, both occurring in the same block. The performance of the scheme was evaluated using a model of a single-server, discrete-time, finite-capacity queue. It was found that, if the input rate of packets to the queue is such that the packet rejection probability is 10/sup -3/ and below, it is possible to find a proper value of block size for which the decoding yields a substantial reduction in packet loss rate. Further reductions are possible if the server discards not necessarily newly arrived packets but takes into consideration their block affiliations and attempts to distribute the rejected packets among the blocks to maximize the decoding capability. >

Fixed distortion subband coding of images for packet-switched networks

Abstract: A subband image codec is presented that approximately attains a user-prescribed fidelity by allowing the encoder's compression rate to vary. The fixed distortion subband coding (FDSBC) system is suitable for use with future of packet-switched networks. The codec's design is based on an algorithm that allocates distortion among the subbands to minimize channel entropy. By coupling this allocation procedure with judiciously selected subband quantizers, an elementary four-band codec was obtained. Additional four-band structures may be nested in a hierarchical configuration for improved performance. Each of the configurations tested attains mean square distortions within 2.0 dB of the user-specific value over a wide range of distortion for several standard test images. Rate versus mean-square distortion performance rivals that of fixed-rate systems having similar complexity. The encoder's output is formatted to take advantage of prioritized packet networks. Simulations show that FDSBC is robust with respect to packet loss and may be used effectively for progressive transmission applications. >

Computer interconnect system with transmit-abort function

Abstract: A computer interconnect system uses packet data transmission over serial links connecting nodes of a network. The serial links may provide simultaneous dual paths for transmit/receive. An adapter couples a CPU or the like at a node to the serial link. The adapter includes a packet memory for temporarily storing transmit packets and receive packets, along with a port processor for executing the protocol. Packets of data are transferred between the system bus of the CPU and the packet memory by a pair of data movers, one for read and one for write. All of the serial links of the system are connected to a distribution hub which forwards a transmitted packet to a destination node based upon an address sent with the packet. If the path to the destination node is busy, the hub returns a "flow control" signal to the source node, and in response to this signal the transmitted packet is aborted so that time on the network is not wasted by needless transmission that must be discarded.

Congestion free packet network

Abstract: A congestion control scheme for a connection oriented packet network is disclosed. The congestion control scheme eliminates buffer overflow and provides congestion free communication by eliminating the clustering of packets. Clustering is eliminated by imposing a smoothness requirement on the packet stream of each connection at the corresponding source node in the packet network. The smoothness is maintained for each connection over every link traversed by a connection through use of a unique queuing strategy at the nodes wherein packets arriving at a node in an incoming frame are delayed and transmitted from the node in an adjacent outgoing frame.

A congestion control framework for ATM networks

Abstract: A congestion control framework for ATM (asynchronous transfer mode) networks is proposed. Specifically, it is suggested that the network provide two different services: express service and first-class service. Express service is appropriate for real-time applications (e.g., voice and video) whereas first-class service is appropriate for non-real-time applications (e.g., data). To provide such services, the proposed congestion control scheme prevents congestion inside the network by controlling the congestion at two levels: virtual circuit (VC) level and packet level. Express VC traffic is not subject to any flow control. At each intermediate node, transit packets are simply relayed with no traffic control. Various issues related to this scheme are discussed and its performance analyzed. >

Reliable broadcast of information in a wide area network

Abstract: A method and apparatus for creating and managing databases in routers of a routing network. The databases store link state packets, each packet being originated by nodes in the network, and transmitted to other nodes through the network. Each packet contains data identifying its originating node, a sequence number in a linear space indicating its place in the sequence of packets generated by its originating node, and an age value indicating the time remaining before it expires. The contents of the databases are updated by newly received packets. In addition, the nodes themselves are reset if the packets currently in the network have later sequence numbers than new packets. Also, a mechanism is provided to purge the databases of packets from a given router by issuing a purging packet.

Queueing analysis of delay constrained voice traffic in a packet switching system

Abstract: Unlike data traffic, the voice packet stream from a node has very high correlation between consecutive packets. In addition, in order for the speech to be properly reconstructed, a delay constraint must be satisfied. A queueing model that accurately predicts packet loss probabilities for such a system is presented. Analytical results are obtained from an embedded bivariate Markov chain and are validated by a simulation program. Based on this model, the impact of the delay constraint, talkspurt detection thresholds, and packet size on packet loss are studied. Two schemes, named 'instant' and 'random', for discarding late packets are considered. Simulation results show that better performance can be obtained by using the latter scheme. >

Dual-path computer interconnect system with zone manager for packet memory

Abstract: A computer interconnect system uses packet data transmission over serial links connecting nodes of a network. The serial links provide simultaneous dual paths for transmit/receive. An adapter couples a CPU or the like at a node to the serial link. The adapter includes a packet memory for temporarily storing transmit packets and receive packets, along with a port processor for executing the protocol. The packet memory includes two zones in which the received packets are stored interchangably. Packets of data are transferred between the system bus of the CPU and the packet memory by a pair of data movers, one for read and one for write. The packet memory is accessed upon demand by the serial link, the port processor and the data movers, using interleaved cycles. The order of buffering the received packets in the two zones is recorded in a file or silo, and when the packets are transferred to the CPU the packets are accessed by referring to this silo so the order of receipt is maintained.

A high speed transport protocol for datagram/virtual circuit networks

Abstract: We present a design and preliminary analysis of an end-to-end transport protocol that is capable of high throughput consistent with the evolving wideband physical networks based on fiber optic transmission lines and high capacity switches. Unlike the current transport protocols in which changes in control state information are exchanged between the two communicating entities only when some significant event occurs, our protocol exchanges relevant and full state information periodically, routinely and frequently. We show that this results in reducing the complexity of protocol processing by removing many of the procedures required to recover from the inadequacies of the network such as bit-errors, packet loss, out of sequence packets and makes it more amenable to parallel processing. Also, to increase channel utilization in the presence of high speed, long latency networks, and to support datagrams, we propose an efficient implementation of selective repeat method of error control used in our protocol. Thus, we utilize small extra bandwidth to simplify protocol processing; a trade-off that appears proper since electronic speeds for protocol processing are far slower than fiber transmission rates. Our preliminary estimates indicate that 20,000 packets/second can be handled in a completely software implementation on a 10 MIP microprocessor using 8% of its cycles.

Packet video transmission through ATM networks

Abstract: The basic aspects of packet video transmission technology are described, and some new results are presented. The advantages of packet video transmission through ATM (asynchronous transfer mode) networks and items to be solved are discussed from both the network and the user sides. Modeling methods, including burstiness measures are described. Examples of modeling based on the Markov modified Poisson process are shown. Packet loss protection and recovery methods are discussed, and the effects of a layered coding scheme are described. Subjective picture quality based on variable rate video transmission is demonstrated. >

Buffer sizing for synchronous self-routeing broadband packet switches with bursty traffic

Abstract: Previous studies on the performance of synchronous self-routeing packet switches have assumed that the input traffic is random, i.e. there is no correlation between adjacent packet arrivals. This assumption is generally not valid in the data communication environment (e.g. host-to-host communication) where a file transfer usually generates a string of correlated packets. The consequence is that the random traffic assumption greatly underestimates the buffer requirement of the switch. In this paper, we model each input traffic stream as a binary source as a first step to understand the performance of a packet switch in a bursty traffic environment. We found that, given a fixed traffic load (or switch utilization), the required buffer size increases linearly as the burstiness index (the average burst length) of the traffic increases. In addition, the required buffer size is more sensitive to the burstiness of the traffic, when the average traffic load is higher and when the packet loss requirement is more stringent. Initial applications of broadband packet switches are likely to be the interconnections of LANs and hosts. The results of the study indicate that the high burstiness in certain broadband traffic significantly reduces the allowable switch utilization, given a fixed amount of buffers. To increase the switch utilization, an appropriate congestion control mechanism needs to be implemented.

Window-based congestion management in broadband ATM networks: the performance of three access-control policies

Abstract: A comparison is made of the network-level and user-level performances of three proposed access-control mechanisms for foreign congestion management in public broadband packet-switched networks: rate-based windows (leaky buckets), end-to-end-acknowledgement windows, and congestion-feedback acknowledgement windows. Their performance in terms of throughput, delay, and patterns of cell loss was studied using a cell-level simulation of an ATM (asynchronous-transfer mode) network. When assessed at the network level, leaky buckets perform well, whereas neither acknowledgement-based mechanism performs adequately. However, when assessed from the user perspective, leaky buckets also fail to provide satisfactory service. The simulation results reported suggest that further study of intranetwork queuing and rotating mechanisms is required to control congestion within the network and to supply assured quality of service to the end users. >

Congestion control in BBN packet-switched networks

Abstract: We define congestion in a packet network to be the state where performance degrades due to the saturation of a network resource . Resources required by packet networks en compass the communication links between packet switches and the switches' computational resources—processor cycles and buffe r memory . The degradation arises when the network expends resources in doing work i t can't complete . Congestion arises for a number of reasons . The network may simply be underconfigure d for the offered load . Underconfiguration can result from temporary failures elsewhere i n the network, or it may be a busy-hour occurrence, hence not worthy of reconfiguration . In a network whose resources have differen t capacities, when all the capacity of a high speed link is directed over a lower-speed link , that link cannot carry the offered load . Such over-demand can also arise due to the funnel ing together or flows from several incomin g links to an outgoing link . Packet switches should be able to cope gracefully with these situations. They may simply discard the excess packets, relying o n

A class-oriented replacement technique for lost speech packets

Abstract: A replacement technique for lost speech packets is presented. This technique is based on the classification of the packets into four distinct classes: background noise, voiced speech, fricatives, and other. Different encoding schemes and lost packet replacement techniques are used for each class. Results of subjective tests indicate that giving preferential delivery treatment to packets based on class can be used to improve subjective quality. The replacement strategy renders the reconstructed signal indistinguishable from the original utterance for packet loss rates up to 47% for background noise packets, 8% for fricative packets, and 4% for other packets. For voiced packets, the replacement is distinguishable from the original signal even for packet loss rates lower than 5%, but significant improvement may be possible by reducing the memory associated with the voiced speech coding process. >

Packet voice communications over PC-based local area networks

Abstract: Experimental implementations of packet voice communication systems over two types of PC-based local area networks are described. The first is a Proteon proNET token-passing ring network, and the second is an Ethernet network. System configuration, operation, and performance are described for both networks. Models of network performance for estimating the maximum allowable number of active voice stations without incurring intolerable packet loss are presented for each system. The models are defined for systems with and without silence detection. PC-related implementation issues are also discussed. >

Packet transport of VBR interframe DCT compressed digital video on a CSMA/CD LAN

Abstract: The feasibility of providing teleconference/picturephone video services over 10-Mb/s IEEE 802.3 CSMA/CD (carrier sense multiple access with collision detection) local area networks is examined. The system considered is based on variable-bit-rate (VBR) video codecs which produce an interframe discrete cosine transform/differential pulse code modulation (DCT/DPCM) codec output format compatible with the evolving CCITT H.261 standard for px64 kb/s motion video. Transport-level protocol functions (such as delay compensation, flow control, and error control) required to support real-time video at an acceptable quality level are identified and discussed. A simulation model used for assessing the performance of the LAN with several video and data users is described. The model provides quantitative measures of network performance (such as packet loss or delay vs. throughput). Preliminary performance evaluation results demonstrate that, with appropriate transport-level delay compensation, potentially acceptable video and data quality levels can be provided on a well-loaded ( approximately 55%) CSMA/CD LAN. With appropriate transport buffering, acceptably low video packet loss rates can be achieved, although the corresponding subjective quality has yet to be established. >

Delay and throughput for three transmission schemes in packet radio networks

Abstract: The results of a simulation of a fully connect packet communication network are presented. The network employs the slotted ALOHA channel access protocol, and Reed-Solomon coding is used to correct errors and erasures in the received packets. The network has a fixed number of channels available for packet transmissions. In two of the transmission schemes considered, a given packet transmission occurs on a single channel only. One of these employs fixed assignment of terminals to the channels, and the other employs random assignment. A third method permits each transmitter to hop randomly over the set of channels during the transmission of a packet. The terminal transmits on one channel at a time, and it transmits a fixed number of symbols during each dwell interval. Delay and throughput are evaluated for all three transmission methods. The effects of code rate on the performance of the channel-hopping scheme are explored, and the improvement that can be obtained for the channel-hopping scheme by use of incremental-redundancy transmission is examined. >