In a typical modern computer networks the presence of mixed traffic, and there is a steady increase in the share multimedia traffic [1, 2, 3]. Each type of traffic has its own requirements for quality of service (QoS) . With regard, to that we formulate the basic requirements for QoS for voice and video. In [5, 6] outlines the parameters of QoS for voice.Voice quality, three factors affect the quality of QoS : packet loss, delay and delay variation. Packet loss causes short-term gaps in conversation.
Next Generation Network (NGN) which is a packet-based network can provide services including Telecommunication Services and can also make use of multiple broadband, QoS enabled transport technologies and in which service- related functions are independent from underlying transport-related technologies. NGN gives revolution in the mobile networks, the capability to assure the seamless mobility with end-to-end QoS present an essential criterion of the success in the NGN. Because of high demand of multimedia facilities trafficengineering has become a necessary characteristic, trafficengineering in telecommunication depends on special performance parameters. It offers options to choose best route for data routing while effectively utilizing network resources.
 With the evolution of mobile devices and networks, and the fast developing trend of mobile Internet access rich, multiplayer gaming usingmobile devices, similar to system based Internet games which have potential and interest. Such the current client-server architecture for system based Internet games where most of the storage and lumber of the game lies with the client device which will not work with mobile devices processing mobile gaming to downloaded single player games, or very light non-interactive versions of Internet games. In this paper we will study a cloud server based approach, termed Cloud Mobile Gaming where the burden of executing the gaming engines is put on cloud servers, and the mobile devices just communicate with the users' gaming commands to the servers. We analyse the factors affecting the quality of user experience using the CMG approach, including the game generate, video encoding & decoding factors, and the conditions of the wi-fi network. Based on the analysis, there was development of a model for Mobile Gaming User Experience (MGUE), and develop a prototype for real-time measurement of MGUE that can be used in real networks. We approved MGUE model using control subjective testing and by using to characterize user experience achievable using the CMG approach in wireless networks.
ternational Telecommunications Union to meet the peak speed requirement of 1 Gbps for stationary connections and 100 Mbps for a mobile connection . The 4G service is designed to offer a faster and secured all-IP, roaming mobile broadband solution. The user can experience high speed network applications, including on-demand High Definition IP Television (HD IP-TV), Voice/Video-over-IP services, on-demand gaming, and high-speed internet access by using Smartphone. But the main challenge will remain with the power consump- tion of both the Smartphone and the BS.
In order to illustrate this fact, an important work presented by Gianoli (Amaldi et al., 2013), described an alternative method to solve the problem here discussed motivated by the fact that solutions of the original MILP formulation are extremely time consuming and capable of draining all computer resources such as memory. He used in his article network instances from the SNDlib (Orlowski et al., 2007) and in some cases it was impossible to find a result due to out of memory. For network Diyuan-30, for example, with 11 nodes and 42 links, solution could not be reached after more than 20 hours.
which enables MSs to share content downloaded from BS to other MSs nearby who request the same content through local WiFi, thus reducing 3G/4G link traffic. In , a peer-to-peer application for live streaming of video content encoded at multiple bit rates is presented, which exploits an ICN API using a Java implementation running on plain laptops to simplify the application de- velopment. In the application, peers are a small set of neighboring mobile cellular devices and they have access to a remote cellular network through the cellular inter- face and a local full mesh one hop network through a proximity channel, which enables cooperative download- ing of a live video stream. The quality of video playback is increased by the design of cooperation, during which the peer can redistribute the chunks downloaded through cellular interface on the proximity interface to request peers in a multicast fashion, and the part is also cached in the CCN content store in order to be shared with other peers requesting it on the proximity interface. Multicasting, in-network caching and multi-rate encod- ing approach with dependent substreams are utilized to improve the peer to peer sharing. As an extension of , by deploying the proof-of-concept application on commercial Android devices, the limitations of An- droid devices is dealed with, and the quality of video playback streaming is improved for collaborating de- vices that exploit all the available radio access tech- nologies .
The implementation of ant colony algorithm was carried out on a network simulation that uses Mininet emulator. Several simulation scenarios with varies scenarios had been conducted. The experiment recorded completion time, the throughput, and trafficload between two or more internet connection lines when load balancing algorithm is applied. By using the ant colony, the value of throughput was greater than Round-Robin. The test results are intended to maximize the speed. CPU utilization generated by the ant colony optimization algorithm is evenly distributed because it has a less CPU utilization difference value for each server. Ant colony optimizationload balancing algorithm will stabilize the overloaded server, so it can be concluded that the use of this algorithm will be very helpful because this algorithm has a mechanism that is fitting in order to look for the shortest path.
Abstract— In this paper analysis and performance of OLSR and TORA routing protocols is done. We used OPNET Simulation tool we created a network containing 50 mobile nodes with data rate 1 Mbps and 2 Mbps with transmission power 0.005 watts and buffer size 256000 bits each node moves randomly in the network and simulation time was 1500 sec. TORA and OLSR routing protocols were compared in terms of Traffic Received, Traffic Sent, NetworkLoad, Retransmission Attempts and Throughput. According to the resulted performance OLSR performed better then TORA in both 1 Mbps and 2 Mbps
On large service provider networks, no other technology can engineer traffic as efficiently as MPLS does. We can share the load on the links by changing the metrics used by the routing protocols such as OSPF, EIGRP, IS-IS etc. but that method is not practical in large ISP environments. MPLS TE very conveniently uses the under-utilized links for carrying traffic and using existing network resources. MPLS TE creates tunnels to carry traffic and path of these tunnels can be explicitly assigned. MPLS facilitates important user’s data traffic such as voice, video and bank transactions with dedicated tunnels for them to avoid any unnecessary delay. In case of a link or node failure along the primary tunnel’s path,
MPLS enhance the performance of the network by using signaling protocols for trafficengineering. Through the signaling protocol, trafficengineering selects the network paths for forwarding the packets to the routers in a balanced manner. This paper explains the study of performance analysis of Constraint-Based routed LDP signaling protocol and Resource Reservation Protocol -TE signaling protocol. This paper has demonstrated that the MPLS system using CR-LDP TE signal convention has a visible execution favorable position contrasted with the MPLS system using RSVP TE signal convention as far as the quantity of got voice packets and the quantity of kept up calls with both GSM and PCM codecs. This is for the most part because of the poor adaptability of RSVP convention came about because of the additional activity prerequisites for intermittent refreshment of movement, high LSP disappointment recuperation movement and RSVP messages to keep up the positions in all LSR.
In chunk-match technique Rabin fingerprints are being generated over the payload of a packet. When a fingerprint matches a specified constant Y (by performing a modulo operation over Y ), that fingerprint constitutes a boundary. The data between two boundaries are the chunks. Chunk identifiers are then being generated through SHA-1 hashes. While this approach is similar to  and , we have made a number modifications. The chunk-match algorithm is performed per packet and every packet generates at least one chunk. The algorithm is therefore used for any kind of traffic (TCP, UDP etc.) and the chunk sizes are small enough to identify similar redundancy as in max-match. Also, instead of sending the SHA-1 hash (20 bytes) for every chunk, we use the following approach: i) if redundancy spans over two or more chunks, a single hash value is generated for the maximum matching region (multi resolution chunking). ii) for the encoding, we perform the location-length encoding policy, therefore a pointer to the location is send and the size of the maximum matching region (send 8 Bytes). This resolves one of the drawbacks of this methodology, as noted in , that a 20 bytes SHA-1 per chunk would add a significant overhead.
Based on the assignment of flag, the loads in the distributed system are balanced. For instance, if the new task arrives, it will search for the resource needed to process the task. Before assigning the task to the particular resource, it is necessary to check the flag of the particular resource. If the flag is 1, then the resource is already processing the tasks for a certain interval of time. If the flag is 0, then assign the task to the particular resource. If more than one task requests the same set of resources, then the data traffic occurs. When the traffic occurs, the task in the queue will get hassled. The presence of hassles is identified based on the occurrence of data traffic in the distributed system. If new task is successfully done with the assigned resource, then increase the value of flag. The flag value increases based on the number of successful tasks processing in the particular resource is done. The possibility of assigning the new task with the resource R i determined as,
Congestion can be divided into two categories: Congestion due to uneven traffic distribution and uniform and persistent congestion. The former can be addressed by Traffic Engineer- ing (TE), the later requires capacity expansion or admission control. TE can be applied if traffic is not mapped efficiently to the available network resources, i.e. there are network parts that have to sustain more than the allowable load despite there is ample network capacity. These congested areas are usually referred to as “hot spots”. Generally, ISP networks are well dimensioned and have sufficient capacity. For example, Sprint, one of the largest (tier-1) ISPs in the USA, over-provision their network by maintaining the maximum utilisation of any links to be below 50%. The average link utilisation is between 20 - 25%  . In situations where capacity is available, trafficengineering is an efficient alternative to capacity expansion.
We have presented an approach based on trafficengineering and routing constraint to provide quality of service of VoIP in wireless mesh network. Moreover, because of the wireless interference, looking for a feasible route to accommodate an incoming call can be computationally hard. We have used the independent set to avoid the critical links. We have also optimized a cost function according to routing metrics such as hop count, links criticality and the networkload in order to find an optimal path in the network. Our modelling improves performance significantly compared to naive methods.
This In recent years, the use of the Internet as communi- cation infrastructure for different telecommunication applications has been growing significantly. Because band- width is one of the most important requirements of these applications, network hardware should support band- width management techniques. Trafficengineering (TE) is a bandwidth management technique that considers different objectives such as maximum throughput, mini- mum congestion and load balancing in the network. TE puts the traffic where network bandwidth is available . TE with the objective of load balancing can reduce maximum link utilization (MLU) and increase bandwidth efficiency (BWE). Because considerable delay may oc- cur at congested links, reduction of end to end delay can be achieved as a side result of load balancing.
There exist three major types of traffic control modes; Chaotic, Pre-timed and adaptive traffic systems. In the chaotic system, the signal timings are fixed through all periods of the day. Chaotic system, currently applied in Amravati road network, is the most primitive type of control systems, because it does not consider the balance in the demand-supply allocation problem. Pre-timed systems are more advanced, where adequate time-plan is developed, in order to assign different signal light timings to different periods of the day based on the demand pattern. Each period of the day characterized by an average demand, is given different signalization settings. The adaptive traffic control system is state-dependent. It can instantly adjust the timing intervals rendered to a specified control point according to the fluctuating demand crossing that point. Selection of the appropriate control system that best optimizes the vehicular flow in the network is major problem in trafficengineering.
In the MPLS framework, given a set of paths, there are several critical questions to answer: (1) How to allocate traffic along these paths. To our best knowledge, this is not resolved before, except LP solutions; (2) How to select and update the path set. Pre-selected paths like k-shortest-paths that are used by some existing schemes may be able to achieve close-to-optimal MLU in some cases, it may not work for all topologies. It’s important to select path dynamically based on network status; (3) How to adapt to traffic dynamics. In this thesis, we propose a trafficengineering scheme named TEA and provide solutions to the three questions in an efficient way. In TEA, we propose a traffic allocation algorithm named BALANCE which is executed by each ingress router in a distributed way and only needs the traffic vector and bottleneck links information as an input. For path sharing between different IE pairs, BALANCE uses a set of criteria to determine which IE pair should transmit how much traffic to which shared bottleneck link. From the simulations, we show that BALANCE gives a better traffic distribution than others including OSPF and random distribution. With this initial traffic distribution, the whole approach can be more efficient and converge faster than others using random starting distributions. To balance load further, TEA uses alternative paths to reroute a minimum subset of selected flows to bring down MLU. In TEA, rerouted flows are selected using a threshold-based approach; Traffic dynamics can be accommodated by adapting the threshold and we propose a binary search algorithm that can converge fast to find an appropriate threshold value. We show that TEA can converge in a constant number of round trip times.
Ant as a sole proprietor has a very restricted potency. But as a member of a decisive colony, it becomes one robust agent, working for the growth in the colony. Every Ant lives for the colony and thrive only as a part of it. Each Ant is able to interface, grasp, cooperate, and all in all they are capable to evolve themselves and colonies on a large area. Ants manage such great affluence by enlarging the number of individuals and being unusually well planned. The self - organizing principles they are using allow a highly coordinated behavior of the colony. Ants are able to react to notable catalyst," signals that trigger a genetically conceal response. The results of these reply’s can act as current important stimuli for both the insect that created them and for the different insects in the colony. This procedure is known as Stigmergy. Stigmergy is defined as the process of implied relay in a self-organizing budding system where its sole parts interface with one other by changing their neighboring domain.
Reena Dadhich (2011) represents a paper in on VANETs vehicular ad-hoc networks have been recently attracting an increasing attention from both research and industry communities. VANET technology is distinguished from mobile ad hoc networks (MANET) and wireless sensor networks (WSN) by large scale deployed autonomous nodes with abundant exterior assisted information, high mobility with an organized with constrained pattern, change in frequency, topology leading to frequent network fragmentation with varying drivers behaviour factors. This paper also introduces the realistic vehicular mobility model and evaluates the performance of following routing protocols: AODV, DSR and TORA. It also introduce the different highway scenarios, characterized by the mobility, load and size of the network also be simulated. Result indicates that the reactive routing protocol performance which is suitable for VANET scenarios in term of packet delivery ratio, routing load and end to end delay 
Many approaches have been proposed to improve the energy efficiency of DCNs. These techniques can usually be classified into two categories: The first type of technique is designing new topologies that use fewer network devices while guaran- teeing similar performance and connectivity, such as the flatted butterfly proposed by Abts et al.  or PCube , a server- centric network topology for data centers, which can vary the bandwidth availability according to traffic demands. The second type of technique is finding optimization methods for current DCNs. The most representative work in this category is ElasticTree , which is a network-wide power manager that can dynamically adjust a set of active network elements to satisfy variable data center traffic loads. Shang et al.  considered saving energy from a routing perspective, routing flows with as few network devices as possible. Mahadevan et al.  discussed how to reduce the network operational power in large-scale systems and data centers.  studied the problem of incorporating rate adaptation into data center networks to achieve energy efficiency. Vasic et al.  devel- oped a new energy saving scheme that is based on identifying and using energy-critical paths. Recently, Wang et al.  proposed CARPO, a correlation-aware power optimization algorithm that dynamically consolidates traffic flows onto a small set of links and switches and shuts down unused network devices. Zhang et al.  proposed a hierarchical model to optimize the power in DCNs and proposed some simple heuristics for the model. In , the authors considered integrating VM assignment and trafficengineering to improve the energy efficiency in data center networks. To the best of our knowledge, the present paper is the first paper to address the power efficiency of DCNs from a comprehensive point of view, leveraging an integration of many useful properties that can be utilized in data centers.