BACKGROUND Through last couple of years, cellular systems have becomes one of the most popular wirelesses Communication mediums. But in the hot spot region use of WLAN has enlarged due to its high data rate and low installation cost. The recent increase in the worldwide usage of networking technologies has posed several questions and brought new opportunities to the research and academic community. The developments in these technologies have occurred independent from one another. This has limited the interoperability of various systems. On the other hand, due to e-commerce and other interesting new applications of the networking technologies, the user is attracted to subscribing more than one type of services, such as wireless Internet, multimedia, global positioning systems, etc. However, there is a need of integration of these technologies from an inter-operability point of view. This need seems to be best met by defining international common goals that could be used as guidelines for designer and vendors of such technologies.
INTRODUCTION In this paper, we present an integration of GPRS/3G/4G and WLAN in connection with Multi protocol label switching (MPLS). The mobility is managed by hierarchical mobile IPv6 (HMIPv6). The aim of this paper is to look at seamless mobility with low signaling overhead and provision for optimal quality of service (QoS)
Wireless networks are projected to integrate not only the services (to provide multimedia), but also encompass an integration of technologies. The technology integration has two aspects, namely, the integration of the same technology from different parts of the world, and secondly, the integration of different technologies in the same country.(3G) wireless systems are working diligently to realize the integration of various Code Division Multiple Access (CDMA) standards. Similarly, IP services with CDMA, integration of CDMA air-interface with other wireless access networks, such as wireless LANs, wireless ATM, fixed wireless broadband Internet access are all proceeding at a rapid pace. In other words, a user with a terminal and a single subscription may have access to any or all services that multimedia mobile data networks have to offer.
Technically, there are many ways in which these integration goals can be achieved. One (almost traditional) approach is to have one-to-one interface between any two different technologies at a fixed point and allow all traffic to pass though this point for a 'protocol conversion'. An example of this is the use of ATM Adaptation Layers (AALs) to carry IP and telephone traffics over an ATM backbone. AAL provides a separate interface between each type of higher layer above the backbone network. All traffic is switched end-to-end via the ATM switches in this case.
A second approach of the technology integration is to allow all traffics share the network resources and incorporate some sort of multiple label switching. This approach places the design complexity in a switching device with a signaling additive, and liberates the traffic from delays for protocol conversion. An example of this approach is the Internet's multiple protocol labels switching (MPLS). This approach is already proving to be agreeable among the research and industrial community and has the exclusive benefit of reducing processing over packets of data - much desirable for multimedia traffic. The input and output links to the switch, are shared by all traffic types. When a packet enters the switch, a label-processing block first determines the packet type (IP, ATM, Wireless, Etc.). The label processor then directs the packet to one of the planes labeled as Wireless, IP and ATM depending upon the label on the packet. Each one of these planes performs routing according to a different principle. For example, ATM routing is based on virtual circuit identifier or virtual path identifier (VCI/VPI). Similarly, IP routing is of store-and-forward nature, while a wireless network may provide a mixture of the two. In this way, the switching and routing delay experienced by a packet in the switch depends on its network type.
Future research may have to expand the scope of MPLS concept to bring it at the access node of a network. Let us look at the limitations of Wireless Networks
DESIGN LIMITATIONS OF A WIRELESS NETWORK From an engineering point of view, the ultimate design goal of any network is the resource utilization efficiency. Resource is anything that has a cost associated with it. For wireless networks, bandwidth is the most precious resource. Not only is the bandwidth tightly regulated by respective governments, but it is also consumed more easily due to unreliable wireless channel. In a cable channel, the signal energy is directed along with the cable. The wireless signal, however, faces many channel impairments, such as attenuation, multipath, interference from other users, and interference from natural sources of noise. The design problems are worsened due to a multitude of access technologies to be interfaced with Third and higher generation (3G+) wireless networks. The 3G+ wireless networks are expected to integrate services and technology on a global level. They are to provide the multiple QoS of an ATM network with the flexibility of an IP network. Due to the datagram nature of the IP networks and limited radio resource of a wireless channel, the issue of guaranteeing the Quality of Service (QoS) automatically requires efficient call control and resource management mechanisms.
QUALITY OF SERVICE (QOS) DEFINITION In many parts of the world, (such as the United States of America), the developments in wireless and fixed networks have occurred independent of each other. This independence (read non-cooperation) is to such an extent that there are compatibility and interoperability problems even among various systems using the same technologies. This problem is naturally worsened when multimedia is introduced in 3G+ wireless systems (same applies to fixed networks, such as ATM and IP networks). The CDMA, the Internet and ATM standards all have different definitions of QoS. The objective of all networks is to provide multimedia services to a non-technical end-user. Exploiting this common goal, there is a need to investigate the possibility of defining QoS that could be mapped from one network to another without sacrificing the quality or the resources, be it from loss/delay probabilities point of view, bandwidth guarantees, or congestion control mechanisms. Various researchers have tried to find a common ground among different networks, yet there is no general definition of QoS available in the literature. There is a need for comprehensive examination of the motivations behind these localized definitions of QoS to come up with some definitions applicable to any multimedia network. This task might lay the foundations for much of the future research work.
DEFINING UNIVERSAL GOALS OF ANY NETWORKING TECHNOLOGY As noted above, there have been many independent developments in all areas of networking. One such example is the multitude of radio access networks available today in some countries. Indoor wireless LANs, outdoor cellular networks, fixed Internet access (wireless local loops or WLLs), mobile Internet access, wireless packet data and a large number of systems falling under the Personal Communications Services (PCS) offer private and public networking services . These various systems sometimes differ in many ways, such as, the radio frequency spectrum, the scope of operation, the type of QoS provided and even the data rates. Obviously one type can't be considered to compensate for all; their mere existence is their justification. But, for a user who needs subscription to all or most services, the existence of so many different technologies creates a dilemma as to what type of subscription can meet maximum demands. It is perceived by industry and researchers that the future networks will have the provision that subscription to any service anywhere in the world will provide a user with all the services from a single point of service (POS). This is a very powerful concept and has been the mission of many visionaries in research and standardization agencies. Defining universal goals by standardization agencies could go a long way in bringing the integration of these various technologies. This is a humongous task and there are no guidelines to follow to begin with. The idea is of defining a constitution of network technologies and the technologies can be considered as laws based on this constitution.
EFFICIENT CALL CONTROL FOR WIRELESS NETWORKS It should be obvious by now that the logical unification of QoS definitions and categorization of various radio access technologies would result in wireless mobile networks provisioning a diverse database of users. For successful resource allocation, call progression and completion, each type of user application may require a different set of control parameters. Thus, there is a need of universal call acceptance and control procedures with common parameters among various networks. For example, each network should be able to implement a very general mechanism with parameters that have different values for different service types. This concept, very similar to the original ATM UNI specifications, can help achieve the universal connectivity on global bases. No doubt that the realization of UMTS is a step in that direction, the actual call control mechanisms in the specifications are inherited from earlier wireless standards. This has left a void for implementers of 3G technologies that directly translate to interesting research issues for the researchers. Call control mechanisms are required to be defined at the call origination and hand-off points. Novel architectures, such as the use of multiple-label switching at the UNI and base stations could provide a ubiquitous solution to technology integration.
QOS MAINTENANCE DURING DIFFERENT PHASES OF A CALL This issue is the crux of integration of technologies providing multiple QoS services. Regardless of technologies incorporated, the final goal of any multimedia network (wired or wireless) is to satisfy the end-user. That means several things from a network design point of view, including,
(i) Availability of services with differing QoS attributes.
(ii) Making certain that enough resources are available before a call is admitted to a network, and most importantly,
(iii) Maintaining the promised QoS during call progression.
The last item is perhaps the most difficult task - and is expected to be the continuing subject of much of the future research in wireless networks. This problem is especially worsened in network architectures under focus for next generation of multimedia services due to many reasons. Some reasons are enumerated below:
(a) In any IP network (wireless or fixed) committing to a QoS is still an unresolved issue. This is due to the connectionless nature of the IP protocol. Since the early 1990's there have been several approaches used by IETF to resolve this issue . But due to extremely heterogeneous nature of the Internet, existence of firewalls coupled with feedback congestion control of TCP protocol, none of the proposals has been a decisive winner.
(b) The wireless channel is not only highly statistical; its statistical properties fluctuate depending on a myriad of factors
(c) Specific issues due to mobility, such as handoff, make a permanent QoS solution impossible at this time.
Here is an interesting example of such problem: Suppose that a wireless multimedia terminal (WMT) has a packet that is scheduled in cell i to be transmitted before time TF. Before the packet is actually transmitted, the WMT moves to another cell, say cell j. In cell j a new scheduling time is allocated, say, T*F. Since there is high probability that T*F > TF, the QoS promised to the WMT will suffer unless a QoS maintenance algorithm is available with the handoff procedure to assure the QoS is met.
The above issues are expected to shape the research in multimedia networking of the near future, which in turn will change the way we live. With all challenges come opportunities unique to them. The integration of networking technologies is no exception and brings home several interesting issues to be researched. CONVERGENCE OF WLAN AND 3G MOBILE There's a cold, hard inevitability about the convergence of WLAN and 3G mobile, but the business case that will drive this integration forward has split the wireless industry - with VOIP the primary cause of the rift, according to a new report from the subscription research service Unstrung Insider.
"The major issue dominating the sector is how, and whether or not, VOIP calls over wireless LAN should be integrated into the mobile network call model," "Offloading calls to VOIP has its attractions, but operators are lukewarm on subsidizing handsets that could be used to bypass their networks."
The resolution of this issue is critical to the ongoing integration of multiple radio access networks, including WiMax, into a single converged core network. "Rather than straightforward VOIP offload, converged applications and rich-call services - data services over an IMS core - represent the best long-term opportunity for 802.11-to-3G convergence,".
0 comments:
Post a Comment