Review of Mobile Presence Services in Social Network Applications


AbstractionToday broad scopes of nomadic devices are available. As the use and engineering of nomadic devices additions, societal web applications are going popular. An indispensable constituent of this societal web application is nomadic presence service, which maintains each nomadic user’s presence information. The presence information means the current position, GPS location of nomadic user and besides updates the user’s online friend’s information. As the update occur often, figure of message distributed by presence waiter may take to scalability job.To specify this job, scalable waiter architecture called PresenceCloud is defined.In cosmopolitan nomadic telecommunication system ( UMTS ) , the presence service is performed by IP multimedia nucleus web subsystem ( IMS ) .A weak consistent scheme is used to cut down the presentment traffic. Presentment clip means clip needed to advise the presence information to other users. This paper makes a reappraisal on work outing buddy list hunt job.The consequence shows that PresenceCloud achieves public presentation addition in the hunt cost without compromising hunt satisfaction.

Keywords—social web ; nomadic presence service ; presence information ;

I. Introduction

Mobile computer science is human-computer by which a computing machine is expected to be transported during normal use. This involved nomadic communicating, hardware etc. Chiefly nomadic calculating describes about communicating among client to server, client to other terminal user, waiter to server etc.

Today as the engineering and use of cyberspace additions, societal web application/services ( illustrations are Facebook [ 1 ] , Twitter [ 2 ] , buddycloud [ 3 ] , Google latitude [ 4 ] , foursquare [ 5 ] ) go popular in worldwide. Many communications like confab, online concern, informations sharing etc can be done utilizing societal web application/services. Hence, it is inevitable that societal web services will be the following coevals of nomadic cyberspace application. Mobile presence service is one of the of import component in societal web application. Mobile presence service means maintain up-to day of the month presence information of all nomadic user. Presence information includes location, position, activity etc about the nomadic users. Each nomadic user has a list of friend’s list which we can name as buddy list. On the other manus, we can state that buddy list contains the contact information of the user he/she needs to pass on. Each nomadic user in the societal web application gets information about their on-line friends/buddies via the presence information. The nomadic user’s position is broadcast automatically to the buddy list whenever the user moves from one position to another. So presentment is an of import portion in societal web application. Most presence services use server bunch engineering, to maximise hunt velocity and minimise the presentment clip. The waiter involved in this communicating, keeps up-to day of the month path of all these presence information sing the nomadic user. There are so many issues in nomadic computer science. Some are deficient bandwidth, require careful usage of practical private web, cell phone signals may do wellness jobs, signal jobs during going, more power ingestion during less signal strength etc.

In cosmopolitan nomadic telecommunication system ( UMTS ) , the presence service is performed by IP multimedia nucleus web subsystem ( IMS ) .Here a weak consistent scheme is used to cut down the presentment traffic. Hence presentment clip can be cut downing. Presentment clip means clip needed to advise the presence information to other users.

Instantaneous message provides communicating ( online confab ) between two or more users over cyberspace. Microsoft courier, yahoo courier are some of the illustrations for the instant message. Overview of the available characteristics, maps, architecture, protocols used in these popular populace IM web are described. Besides discuss advantages and disadvantages of these IM webs.

In societal web application an efficient and scalable server-to-server sheathing architecture called PresenceCloud is used to better the efficiency of nomadic presence services for large-scale societal web. A quorum-based server-to-server architecture is used for efficient buddy list seeking. The comparing consequence shows that PresenceCloud achieves public presentation additions without giving hunt satisfaction. Search satisfaction is defined as the clip it takes to seek for the geting user’s buddy list.

II. Different engineerings

This subdivision describes assorted method or engineerings used in different systems. Here, foremost describes about different characteristics and maps supported by three most popular IM systems: AIM, Microsoft MSN and Yahoo! Messenger. Second describes, in cosmopolitan nomadic telecommunication system ( UMTS ) web architecture, to cut down the presentment traffic a decrepit consistent strategy called delayed update can be used. Third describes about chord which provide a scalable peer-to-peer search service for internet application. Finally, an efficient and scalable waiter architecture called PresenceCloud is defined.

In [ 6 ] describes different maps and characteristics of three IM systems such as AIM, Microsoft MSN and Yahoo Messenger. Instant messaging can be specify a public confab etc. In extra some provide file transportation, webcam use, voice confabs session etc. Here the writer comparing three IM systems ( AIM, Microsoft MSN and Yahoo Messenger ) .All of these IM systems uses client-server architecture. Supporting immense figure of user is an of import issue in these IM systems. Two methods are available: 1 is symmetric and other is asymmetric. Each waiter performs indistinguishable maps in symmetric method and in asymmetric method each waiter execute peculiar activity assigned to it like log in, detecting other users, maintain chat room etc.

For normal operation AIM uses client-server architecture and for voice confabs session uses peer-to-peer attack. Therefore two clients communicate straight without utilizing a confab room. For normal and voice confabs YMSG uses client-server architecture. Through centralised voice chat server YMSG voice traffic is routed. One advantage of utilizing this attack is that within the same confab room it can back up multiple users and each user can stipulate their ain voice specification. For normal operation MSN uses client-server architecture and for voice confabs session uses peer-to-peer approach.MSN voice chat session is limited to two individuals. Most of the IM systems have techniques for keeping list of friends. They are called buddy list, allow list and block list. Buddy list contain peoples that a user demand to pass on and supervise their presence. Block list contains peoples that a user demand to barricade communicating with them, it’s besides called black list. Allow list contains people that a user needs to pass on. MSN, AIM and YMSG contains buddy list and block list.MSN and AIM besides contain allow list.

Many web protocols like TCP and IP have used binary representation of informations in byte order. Application bed protocol ( HTTP and SMTP ) have used text-based attack. The advantage of utilizing bye order is that it supports efficient usage of infinite in the web. The advantage of text-based attack is that the representation is closer to human position information and debugging is easier.AIM and YMSG utilizations binary representation for their header.AIM contains two-level binary representation called FLAP and SNAC packages. Fixed length heading and variable length day of the month in FLAP packages. SNAC packages are subtype of FLAP packages that contains fixes length Fieldss followed by a variable information constituent.

Fig.1.AIM FLAP and SNAC package format

YMSG has individual degree construction for fixed Fieldss followed by variable length

Fig.2.YMSG package format

One of the of import job in IM suppliers are user send informations at inordinate rate, inundation caused in the web because of useless traffic.TCP provide some protection mechanism against this through congestion control.AIM has a complex algorithm for solve this, but it has different rate ( rates are based on a clip window ) bounds based on message type. The user will be warned, if the client exceeds the rate and if this continues, the waiter will get down dropping message and will unplug the client. YMSG has a bound of three IM per seconds. Another method to minimise the burden is by acquiring rid of idle clients. Each system in IM maintains a keep-alive pulse message: the connexion may be terminated if the client does non supply pulse message or response to the question. In AIM the client must direct this keep-alive message in every minute to the waiter. YMSG contains two types of pulse petition: a primary and a secondary.MSN has client and waiter pulse message.

In [ 7 ] paper, to cut down the presentment traffic a decrepit consistent scheme is used. In IP multimedia nucleus web subsystem ( IMS ) , the presence waiter is responsible for advising an authorised spectator of the updated presence information. Presence service provides service to entree other user’s presence information such as the user position, activities and so on. In cosmopolitan nomadic telecommunication system ( UMTS ) web architecture, a user with user equipment entree presence service in IMS. When the user provides information to the presence waiter, so user play the function as presentity and the user accesses other user’s information so it act as a spectator. In IMS, control signaling is carried out by call session control map ( CSCF ) .The presence service process is defined by 3rdcoevals partnership undertaking ( 3GPP )

The presence waiter instantly notifies other users in the contact, when the presence information of a user alterations. If the updates occur more often, so the figure of presentment messages increases. To cut down the presentment traffic a decrepit consistent strategy called delayed update can be used. In delayed update, when the presence waiter receives the updated presence information, the presence waiter starts a delayed timer with a period T. This period is called delayed threshold. Within the period T, if the presence information is updated, so old information is replaced by the new information. When the timer expires, the presence waiter notifies the spectator of the presence information. Therefore, the presentments for the updates in T are saved.

In [ 8 ] mentioned about chord which provide a scalable peer-to-peer search service for internet application. A cardinal job that challenges peer-to-peer application is turn uping the node that store a peculiar information point. This paper presents a solution for this by showing chord. The chord provides one operation: given a key, it maps the cardinal onto a node.In peer-to-peer system, the application is distributed without any cardinal control, where the package running equivalent in functionality at each node. The corresponding node is responsible for hive awaying a value associated with the key. Chord uses hashing to delegate keys to allow chord nodes. Each chord node needs merely a few other nodes routing information. Because the routing tabular array is distributed to other nodes. If N-node presented in a system, so each node maintain information merely approximately O ( log N ) other nodes. The chord maintains its routing information as the nodes articulations and leaves a system.

In [ 9 ] paper, an efficient and scalable waiter architecture called PresenceCloud is defined. PresenceCloud solve the scalability job called buddy list hunt job. Buddy list hunt job can be defined as the scalability job occurs when presence service is overloaded with buddy hunt message. The PresenceCloud can be used to expeditiously question the system for buddy list hunt. In PresenceCloud design, it consists of three constituents: PresenceCloud waiter sheathing, one-hop caching scheme and directed buddy hunt. When the nomadic user get into the PresenceCloud, the user authenticate to the nomadic presence service. Then the nomadic user opens a TCP connexion to one of the presence waiter. The nomadic user sends a petition to the presence waiter for buddy list hunt. PresenceCloud returns the presence information of the brothers. PresenceCloud is based on grid-quorum system ( size is vn?vn. When a presence waiter joins into the PresenceCloud, it gets an Idaho in the grid and obtains its presence waiter list. Presence server node can take one column and one row of entries. This entry is called the presence waiter list.

In the fig: 3, the figure of presence waiter n=9.The PresenceCloud is arranged in grid-quorum with v9?v9 size. The presence waiter node 8 has presence server list { 2,5,7,9 } and 3 has { 1,2,6,9 } .

PresenceCloud uses a hoarding scheme to retroflex presence information of each user to better the efficiency of the searching operation. Each presence server node maintains presence information of the affiliated users. When the neighbour establishes a connexion to presence waiter, the cache is updated. When a presence waiter receives a query/request from the nomadic user, it can react non merely from the lucifers from its ain list, but besides provide lucifers from its caches.

Fig:3 PresenceCloud Server Overlay

The figure of presentment messages increases when the nomadic user often changes its presence information. This buddy list hunt job can be solved by utilizing an algorithm called directed buddy hunt algorithm. This algorithm uses bipartisan sheathing and one-hop caching scheme. This helps PresenceCloud to supply fleet responses for big figure of nomadic user. The algorithm is defined in [ 9 ] .

III. Comparison

In mesh based design, all the presence information at each node is replicated to other nodes. Hence its hunt cost is merely one message. In distributed hash tabular array ( chord based ) , no reproduction ; presence information of a user is merely stored in one presence waiter. Here, each brother must be searched one by one. In PresenceCloud, its hunts buddy list from ain node list and cache list. Besides notifies other users about the new presence information.

Simulation consequence is mentioned in [ 9 ] .This paper shows graphical representation of their consequence. Findingss from that graphical representation are:

( I ) As the reaching rate of nomadic user additions, the figure of presence message is really low in PresenceCloud. Hence PresenceCloud outperforms all other design.

( two ) As the reaching rate of nomadic user additions, the mean seeking messages/user is low in PresenceCloud. That means PresenceCloud requires least message transmittal.

( three ) Average message transmittal of PresenceCloud additions bit by bit with the figure of presence waiter.

( four ) Average message transmittal of PresenceCloud is non impacted by the figure of brothers.

When analysing hunt satisfaction in waiter architecture, the findings are ;

( I ) Buddy seeking latency grows with the figure of presence waiter in PresenceCloud. Here mesh-based design performs better.

( two ) When the figure of brothers additions, buddy hunt latency is really low in mesh-based design. but suffer to a great extent communicating cost.


From the analysis and survey, it’s found that PresenceCloud achieves low hunt latency & A ; shows better public presentation than others. PresenceCloud solve brother list hunt job. Overall, PresenceCloud support a scalable nomadic presence service in big graduated table societal web services.


[ 1 ] Facebook, hypertext transfer protocol: //

[ 2 ] Twitter, hypertext transfer protocol: //

[ 3 ] Buddycloud, hypertext transfer protocol: //

[ 4 ] Google latitude, hypertext transfer protocol: //

[ 5 ] Foursquare, hypertext transfer protocol: //

[ 6 ] R. B. Jennings, E. M. Nahum, D. P. Olshefski, D. Saha, Z.-Y. Shae, and C. Waters, ”A survey of cyberspace blink of an eye messaging and confab protocol, ” IEEE Network, 2006. .

[ 7 ] W.-E. Chen, Y.-B. Lin and R.-H. Liou, ”A weakly consistent strategy for ims presence service, ”IEEE Transactions on Wireless

Communicationss, 2009.

[ 8 ] I. Stoica, R. Morris, D. Karger, M. F. Kaashoek, and H. Balakrishnan, ”Chord: A scalable peer-to-peer search service for cyberspace, ”IEEE/ACM Tran. on Networking, 2003.

[ 9 ] Chi-Jen Wu, Jan-Ming Ho “A Scalable Server Architecture for Mobile Presence Services in Social Network Applications, ” IEEE minutess on nomadic computer science, 2013