Improving High Quality And Low Latency Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

The challenging issue in MANETs is to achieve efficient video streaming with low latency constraints. The dynamic change in topology poses a challenging problem of these networks. Loss of Packets and Delay is the most considerable factor in real time video streaming. Video Streaming in real time requires exact techniques that can overcome delay and loss of packets in the unreliable networks. Real Time Transport protocol(RTP) provides end to end transmission of multimedia data and build on top of UDP. Along with ABCD protocol and Multiple Description Coding,RTP reduces delay further and produces high quality video. Additional implementation of Multicast Routing Technique deliver video contents to active and interested group of receivers. It avoids the unnecessary usage of network resource and overload. RTCP Monitor provides long term feedbacks among active mobile mesh nodes. The proposed techniques seems to enhance the performance of video quality and delay factor.


In WMNs, signals are routed optimally and nodes can automatically join and leave the network at any time. Furthermore, networks can be established instantly virtually anywhere, even in places with no fixed infrastructure.Mobile Ad-hoc Network (MANET) is a special type of WMN. In MANETs, each node forwards data for other nodes dynamically to form multihopping communication. However, there is no gateway in charge of authentication or security services in MANETs. Compared to the low-mobility nodes of WMNs, nodes are mobile in MANETs. Therefore, nodes are normally lack of sufficient energy.In addition, due to the mobility nature of devices, the topology may change rapidly.

Mobile Adhoc network is a dynamic network of mobile devices connected by wireless links.Generally,wireless links have high data loss rate compared with wired networks.Streaming video in the presence of wireless links can be a great concern here because it may results in error due to huge data loss.Using the radio waves as a transmission medium makes Mobile Adhoc networks susceptible to interference.Radio waves are affected by environmental factors which disturbs signal,block its path and adds noise,echo in it.Noise and echo produces errors whereas blocking of radio waves disconnect transmission of video.The bandwidth of wireless links is lower comparable with wired networks.

Error resilience encoding is a proactive error correction technique.It should be done at source side before packet loss occurs.It enhances the robustness of compressed video in case of packet loss.

Mobile Mesh Topology is the preferred form of topology used to establish communication among end to end mobile nodes. Real Time Video Streaming is the most challenging issue in unreliable networks. Without preexisting infrastructure, MANET should offer a set of properties such as flexibility, efficiency, ease of use and robustness. When TCP/IP protocols are used in real time multimedia applications,it introduces certain amount of delay to establish and terminate communication.

Procedural steps are followed for transferring packets from one node to another.Reliable transmission is guaranteed using TCP,but it is not suitable for real time multimedia application because of the delay factor.Multimedia data should be delivered within limited amount of time in real time system based applications.

IEEE 802.11 standard is specifically designed for wireless LAN communications.It operates on two operational modes such as Infrastructure based and Infrastructureless or adhoc based.Mobile Ad hoc network is based on second operational mode.There is no Access point in adhoc mode and each mobile node can act as a Access point to other in a mesh topology.In a Adhoc netork,Mobile nodes emits radio waves and establish communication with one another.

Realtime Transport protocol provides end to end transmission of real time video and better suited for real time applications.Structured protocols are not suited for Mobile Adhoc Networks.


Important parameters of ad hoc networks such as node mobility was considered. Content routing/delivery protocol inherently designed for the ad hoc wireless case with broadcast property. A MAC Layer modification for ABCD protocol was designed. The goal of our modified MAC layer is to make the nodes able to read messages. It is meant to make the nodes able to gather information about the availability of resource in set of their neighbours.ABCD Protocol aim to increase the quality of the received video stream by maximizing the number of received descriptions and the signal-to-noise ratio while minimizing the number of hops between each node and its source of the stream. Cross-layer design framework (ABCD) was designed in order to integrate congestion-distortion scheduling into ad-hoc networks.P2P does not take into account of parameters as node mobility, link quality and node density. To assure diversity and robustness, Multiple Description Coding was used to split the stream and routed each sub stream separately.

Multiple Description Video Multicast in MANET's was designed. It addresses the problem of video multicast for heterogeneous destinations.MD video are encoded and transmitted over different paths to each destination node. In Sequential Multiple Disjoint Multicast Routing, all destination nodes must be on the first multicast tree to receive the first video description and next description should be on the following multicast trees. It increases the number of assigned descriptions to each node. It outperforms Serial MDMTR in terms of user satisfaction, overhead,the ratio of bad frames and the number of bad periods.

The concept of .Congestion-Distortion Optimized Video Transmission over Ad hoc Networks was proposed. It is one of the models used to reduce delay and produce high quality video. An optimal routing algorithm was proposed to minimize congestion by optimally distributing traffic over multiple paths. Video Distortion model for live video streaming in wireless ad hoc networks was designed. This model takes into account of both encoder distortion and packet loss due to network congestion. Multiple routes achieving higher bandwidth must be chosen at the network model. This model helps to distribute traffic wisely over the network and reduces congestion. Distortion model is used in predicting Distortion rate and increases error resilience.

Analysis of Error propagation due to frame losses in distributed video coding had been undergone and makes it possible to identify loss of frames and retransmit it. Using rate-distortion functions, we analyze the impact of a frame loss on the average distortion of a group of pictures depending on the position of the lost frame within the GOP, as well as the level of motion in each frame and the quantization errors in the key frames and the Wyner-Ziv frames. This theoretical analysis is further confirmed by a practical implementation of the DVC framework using different configurations of frame losses. . Distributed Video Coding (DVC) is a promising paradigm in video coding that allows moving the computation burden from the encoder to the decoder by performing intraframe coding at the encoder and inter-frame decoding at the decoder.

Broadcasting is mostly used operation in mobile ad hoc networks for spreading data and control messages in many applications.A network backbone is constructed for efficient broadcasting to avoid the problem caused by simple blind flooding. Forwarding nodes forward data to the entire network.

Cross-layer design breaks away from traditional network design where each layer of the protocol stack operates independently. It explores the potential synergies of exchanging information between different layers to support real-time video streaming. In this new approach information is exchanged between different layers of the protocol stack, and end-to-end performance is optimized by adapting to this information at each protocol layer. We discuss key parameters used in the cross layer information exchange along with the associated cross-layer adaptation. Substantial performance gains through this cross-layer design are demonstrated for video streaming. While cross-layering provides significant performance advantages, it can also greatly increase complexity, which can make it more difficult to obtain design insights.

Destination Sequenced Distance Vector(DSDV) protocol are also called table-driven protocols since the routing table will be updated for each change in link states in a network and routes are discovered using information stord in routing tables.It is a proactive unicast mobile ad hoc networking protocol.In routing tables of DSDV,an entry stores the next hop towards a destination.

Dynamic Source Routing(DSR) protocol is an efficient reactive routing protocol designed for wireless multihop ad hoc networks of mobile nodes.It allows the network to be completely self-organizing and self-configuring without the need for any existing network infrastructure or administration. A node only tries to discover a route to destination and there is no currently known route.A complete sequence of intermediate nodes from a source to destination will be determined at a source node and all packets transmitted by a source node to a destination follow the same path.


System Architecture

For attaining successful and efficient delivery of video, Realtime Transport Protocol was proposed. In many of the existing system, the process of transmitting video was slow and missed the timing for delivery. In the proposed system, Real Time Control Protocol continuosly monitors the flow of packets from source to destination. Loss of packets is identified using sequence numbering of packets. Clients allow to stream video without delay. The proposed work will be better suited even in dynamic changes of topology and further reduces delay. It provides Congestion free streaming. Multicast mechanism used to reach only the specific destinations instead of flooding.


RTCP Monitor


S1 S2


Multicast Routing







Real Time Transport protocol has been specifically designed for transferring multimedia data such as video over Mobile Ad hoc networks. RTP runs on the top of transport (UDP) and network (IP) protocols to send RTP data. It provides real time delivery of video. RTP is mainly designed for multicast transmissions which do not fit well with the connection oriented TCP. It supports Multicast and Unicast mechanisms. RTP offers end to end transport services for data with real-time characteristics.

In order to establish an RTP session, an application uses IP and Port Addresses. An RTP session is characterized by the following parameters:

IP address of participants:

This can be either a multicast IP address, which corresponds to the multicast session of the participants group, or a set of unicast addresses.

RTP port

The port number used by all participants in the session for sending data.


Both Sender and receiver uses Timestamps for providing synchronization of signals.

Sequential Numbering of packets

RTP packets are sequentially numbered upon transmission These sequence numbers are also used in order to detect packet losses.


RTP is used for multicast communication, so it manages an RTP data packet contains the identity of the sender of the information.The session group can easily identify which member of the session transmits data. The sender's identity is provided in the source identification field.


RTP Control Protocol (RTCP) used to monitor the quality of service and to convey information between participants of an ongoing session. RTCP (Real Time Control Protocol) is the control protocol for RTP (Real Time Protocol). The idea behind the control protocol RTCP is that applications that have recently transmitted multimedia data generate a sender report which is sent to all the participants in the RTP session. This report includes counters for the packet data and the bytes sent, and the receivers can use them to estimate the actual data transmission rate. RTCP packets report the transmission quality for each separate session.

An RTCP sender report contains an indication of the actual time and an RTP timestamp which can be used for synchronizing multiple data flows at the receiver. RTCP messages include a source description (SDES) which contains relevant information. Such a body of information is the canonical name, a globally unique identification code of the session participant.

RTCP Packet Types




Sender report


Receiver report


Source description





Modification of MDC and ABCD protocol in RTP

Multiple Descriptions Coding (MDC) has emerged as a promising technique to enhance the error resilience of a video transmission system. The source is encoded into one or more correlated coded representations called descriptions, which are transmitted over separate channels.. In the presence of losses, when one description is lost, an acceptable quality can be achieved without making use of retransmission mechanisms such as the Selective Repeat, Go-back-N and Stop and Wait.. This allows exploiting path diversity on communication networks. Transmission of the descriptions over multiple diverse paths can compensate the dynamic and unpredictable nature of the communication medium, different paths might have different error characteristics, introduce different delays and so on. It has been shown that MDC combined with multipath routing performs significantly better over diverse networks than the single path and single description coding (SDC), especially at low rates and under delay constraints. In Multiple Description Coding,video input is encoded in multiple number of independent streams referred to as descriptions.Each description provides basic quality.Additional description improve high quality of video.It improves robustness over a MANET by sending each description on a different and independent channel.

The simplest MDC architecture is shown in figure . The encoder creates two descriptions which are sent separately across the wireless channel as stream s1 and s2. Three scenarios are possible: both descriptions are received by the MDC decoder or either one of the two descriptions is missing. The decoder has three decoders, each corresponding to three scenarios. This force the encoder to consider explicitly that the decoder may be in one of three states, even though the encoder cannot know which of the three states the decoder is in. The central decoder receives both descriptions. Increasing the number of descriptions increases the probability that at least one one description reaches the decoder. However, it also increases the coding redundancy and the decoder complexity.

MDC is an attractive coding approach as it provides error resilience and scalability using only part of the data sent to the decoder and no need to employ no priority -enabled transmission mechanisms in the network. MDC is especially advantageous in short-delay media streaming scenarios such as video conferencing and when broadcast over error-prone channels where it provides acceptable reconstruction quality in case of packet loss.A Broadcast Content Delivery Protocol designed mainly to adapt Multiple Description Coding in wireless Adhoc environments.It forms a Multicast tree for every description across adhoc network region .Every node in the network is connected by wireless links in a mesh topology.Nodes can freely move within

Automated Connective Multicast Maintenance Algorithm

We want to introduce a fully operational cross-layer protocol, capable to offer complete coverage of an ad-hoc wireless network of mobile devices with a real-time videostream. We called this protocol ABCD: A Broadcast Content Delivery Protocol.Our protocolmust jointly select both server and router, an approach that showed a significant gain in video quality [Mao et al., 2007]. To the nodes' benefit, we aim to increase the quality of the received video stream by maximising the number of received descriptions and the signal-to-noise ratio on the last link, while minimizing the number of hops between each node and its source of the stream. However, to the network's benefit, we also aim to minimise the number of messages sent forstreaming and set-up: we try to minimise the number of nodes transmitting the stream, and to maximise the information a node can retrieve without making any request. Let us focus on this latter point: nodes should be able to gather information without making explicit requests. One obvious solution to this problem is that they

should be able to read other nodes' messages and extract information from there.


Input: source s, receiver r, Mobile mesh node m, forwarder f.

Multiple Description Coding




enque(packet) {

if(packet.isVideo() ) {

insertToQueue(packet, head)


else {

insertToQueue(packet, tail)


if(queue.size() > limit ) {




Source Node

Procedure source_handling_packet(p){

if source attached S1,S2{


s.update_ src_routing-table(p);

Procedure send_multicast_data_packet(p)

s.sends _S1,S2 to r; }

Forming Multicast tree

Procedure mesh_handling_packet {

if m=s {

m.update_src_ mcast_table(p); }

if m=f && m=r {

m.status= forwarder_receiver; }

else if m=f {

m.status=forwarder; }

else if m=r {

m.status= leaf_receiver; }


Data Packet Forwarding by Nodes

Procedure send_multicast_data_packet();

Procedure recv_multicast_data_packet(p){

if m.status= forwarder_receiver||forwarder {



Receiver Node

Procedure recv_multicast_data_packet(p)

if m.status= forwarder_receiver||leaf_receiver {

m. recvd p; }


In routing layer, multicast refers to the delivery of the same message originating from a given source to a group of nodes in the network. If the set of destinations is reduced to one node, the communication becomes point-to-point unicast, while it becomes a broadcast operation if the destination set consists of all the other nodes. Multicast is one of the most primitive group capabilities of any message passing network. Thus, multicast routing is the term used to describe the finding of paths from source node to multiple destination nodes in a network.It scales to a larger receiver population but it may not have prior knowledge of a receiver's identity or prior knowledge of the number of receivers. Multicast uses network infrastructure efficiently by requiring the source to send a packet only once, even if i t needs to be delivered to a large number of receivers.


Real Time Video Streaming should overcome the limitation of packet loss and delay that make the delivery of video on MANET's challenging .The combination of Multiple Description Coding with ABCD protocol and in RTP provides the splitting of main streams into substreams and forms a automated mutitree mesh overlays to traverse each description from source to destination sideImplementation of RTP protocol with these techniques further enhanced the performance by delivering video-on-time without loss of packets and better suited for real time video streaming applications. Multicast Routing along with the above techniques and protocol guarantees real time video streaming to multiple recipients at a time.

In real time video streaming,there are some other challenging issues such as multipath fading,collision and interference which have to be taken into consideration.Solving these issues further reduces delay and enhances quality of video by avoiding loss of packets.