Advanced Adaptive Protocols for Next Generation Wireless Networks

Supported by: Department of Defense (DoD)



 
Project
Overview

 
BWN-Lab
Personnel

 
Project
Description

 
Publications

 
BWN Projects
Project Description


Architecture for Next Generation Wireless Networks (NGWN)

The architecture of NGWN should have  the following characteristics.
Economical: The architecture should use as much of the existing infrastructure as possible and minimize the use of new infrastructures. This will ensure economical and speedy deployment.

Scalable:  The architecture should be able to integrate any number of wireless systems of both existing and future service providers who may not have service level agreement (SLA) among them.
Reliable: The architecture should be robust enough to provide fault tolerance.
Secure: The architecture should provide security and privacy equivalent to the existing wireless networks.
Seamless Mobility Support: The architecture should support seamless mobility management to eliminate connection interruption and QoS degradation during inter-system roaming.

A novel  architecture called AMC (Architecture for ubiquitous Mobile Communications)  is proposed that integrates the heterogeneous wireless systems of  NGWN. AMC eliminates the need for direct service level agreement (SLA) among service providers by using a third party, Network Inter-operating Agent (NIA). Instead of developing new infrastructures, AMC extends the existing infrastructure to integrate the heterogeneous wireless systems. We also designed the  authentication and accounting mechanisms for AMC.  AMC supports seamless intra-system and inter-system roaming in NGWN. Finally, a hierarchical structure for AMC is proposed to realize the integration of heterogeneous wireless systems around the globe.

Related work:

Ian F. Akyildiz, S. Mohanty and J. Xie, ``AMC: an architecture for ubiquitous mobile communications in next generation heterogeneous wireless systems," submitted for publications.


Adaptive Medium Access Control for NGWN


NG wireless networks, impose challenges due to their architectural heterogeneity in terms of different access schemes, resource allocation techniques as well as diverse quality of service requirements. These heterogeneities must be captured and handled dynamically as mobile terminals roam between different wireless architectures. However, to address these challenges, the existing proposals require either a significant modification in the network structure and in base stations or a completely new architecture, which lead to integration problems in terms of implementation costs, scalability and backward compatibility. Thus, the integration of the existing medium access schemes, e.g., CSMA, TDMA and CDMA, dictates an adaptive and seamless medium access control (MAC) layer that can achieve high network utilization and meet diverse QoS requirements.

Related work:

M. C. Vuran and I.F. Akyildiz, ``A-MAC: Adaptive Medium Access Control for Next Generation Wireless Terminals,'' submitted for publication.


Cross-Layer Mobility Management Framework for NGWN

In NGWN  users will encounter both intra- and inter-system handoffs. It is essential that the applications running on the mobile terminals remain unaware of users' movement; both intra- and inter-system, to ensure uninterrupted services with minimum QoS degradation to the mobile users. This can be achieved by reducing the handoff failure probability and by restricting the handoff latency and bit errors to the values that are tolerable by the applications. This asks for the design of efficient mobility management protocols for NGWN. We advocate that efficient intra- and inter-system handoff protocols should have the following characteristics to support seamless roaming in NGWN.
Minimum Handover Latency: The  protocols should introduce only minimum latency to the ongoing communications.

Low Packet Loss:  Packet loss during handoff should be minimized.

Limited Handover Failure:  Handover failure should be limited to a desired value.


Mobility management protocols operating from different layers of the network protocol stack (e.g., link layer, network layer, transport layer, and application layer) are proposed to provide seamless services to the mobile users. However, none of them alone are capable of supporting seamless mobility to the mobile users in NGWN. Therefore, recently the design of cross-layer mobility management protocols has gained significant attention. However, there exist no work that defines how the layers should interact to provide efficient mobility management. Therefore, we plan to investigate the issue of cross-layer mobility management and develop  a generic framework for cross-layer mobility management.



Resource and Handoff Management in NGWN

Handoff management plays an important role in the integration of heterogeneous wireless networks. Horizontal handoff (intra-system handoff) occurs when a mobile terminal (MT) is moving out of the coverage area of a base station (BS) into the coverage area of another BS within the same system. Vertical handoff (inter-system handoff) is defined as handoff between BSs that are using different wireless network technologies. It usually occurs in wireless overlay networks where the coverage areas of different tiers of networks are overlapping to each other.

For horizontal handoff, the choice of the ``best'' BS is purely based on the signal strength an MT receives from neighboring BSs. However, in wireless overlay networks where an MT is reachable via multiple networks, the choice of the ``best'' network for vertical handoff places a new challenge. We propose a hybrid control scheme for vertical handoff decision in heterogeneous wireless overlay networks. The objective of the proposed scheme is to provide satisfactory overall performance of the whole system as well as take into account the user preferences which are specified in the distributed user profiles.



Related work:

I. F. Akyildiz, J. Xie, and S. Mohanty,  ``A Survey on Mobility Management in Next Generation All-IP Based Wireless Systems,'' IEEE Wireless Communications Magazine, vol. 11,  no. 4, pp. 16-28, Aug. 2004.


Adaptive Transport Layer Protocols for NGWN
The NGWN objectives pose several challenges for the development of efficient high performance transport layer solution for the realization of the NGWN as follows:
Heterogeneous Wireless Architectures: The wireless systems that will be part of the NGWN have different characteristics summarized as follows:

  • Access Delay: While the one-way access delay in the wireless link may not be significant in WLAN, a typical round-trip time (RTT) varies between few hundred milliseconds and one second in 3G links. The access delay is much higher in satellite links, which have high propagation delay up to 270 ms.
  • Link Errors: The packet loss rates vary from very low levels in near-wireless environments such as WLAN, 3G pico-cells to higher than 1% in macro-cellular environments and satellite networks. The mobility related packet losses, i.e. blackouts due to handoff or signal loss, amplify the degradation.
  • Mobility Pattern: The mobility rate may increase the number of blackouts due to handoff and hence decrease the transport efficiency. During a connection period, almost no handoff is experienced in the global coverage, whereas frequent handoffs may take place in pico-cellular environments.
Heterogeneous Service Demands: The services that will be provided by the NGWN vary from high rate reliable data to real-time multimedia such as live video streaming. While the former requires 100% reliable transport, the latter, instead, needs timely delivery and smooth rate variation. For multimedia traffic delivery, UDP-based streaming without any transmission rate control may also lead to unfairness to the TCP sources and further result in a congestion collapse.



Related work:

O. B. Akan and I.F. Akyildiz, ``ATL: Adaptive Transport Layer for Next Generation Wireless Internet,''  IEEE Journal on Selected Areas in Communications (JSAC), vol. 22, no. 5, pp. 802-817, June  2004.


Location Management in NGWN

To support global roaming, NG wireless systems requires the integration and interoperation of heterogeneous mobility management techniques. Mobility in a hierarchical structure or multi-layered environment should be supported. When the service areas of heterogeneous wireless networks overlap, mobile terminals are reachable via multiple access networks. Under this heterogeneous environment, new challenges for intelligent location management techniques arise.

We propose a new architecture for location management in NG heterogeneous wireless overlay networks. With the Internet as the common backbone network for signaling message transmission, this architecture is more cost efficient and can be built in a hierarchical structure to make it more scalable. Three location management techniques are designed under the proposed architecture. Under the proposed schemes, each mobile terminals updates its location in one individual network at any time and has user preference call delivery support.

Related work:

J. Xie and I.F. Akyildiz, ``Location Management in Next Generation Heterogeneous Overlay Networks,'' submitted for publication.


Integration of 3G Networks and WLANs

Wireless LAN (WLAN) access technology has strong potential to provide a perfect broadband complement to Third Generation (3G) Wireless Systems. This has raised much interest in the integration and inter-operation of 3G and WLAN. Out of several challenges involved in the integration of 3G and WLAN, mobility management is one of the key subjects in order to support global roaming of mobile users between these two systems in an efficient way. We propose a novel architecture using the Network Inter-operating Agent (NIA) and Integration Gateway (IG) to support seamless roaming between 3G and WLAN. The proposed architecture has several advantages such as roaming support between 3G and WLANs of various providers with/without direct service level agreement between them, scalability, and easy implementation. Moreover, protocols for inter-system handover (ISHO) between 3G and WLAN are designed for mobility management in this new architecture.

Related work:

S. Mohanty and I.F. Akyildiz, ``A New Architecture for 3G and WLAN  Integration and Inter-System Handover Management,'' submitted for publication.

S. Mohanty and I.F. Akyildiz, ``LAMP: Link-Layer Assisted Mobility Management Protocol for Seamless Inter-System Roaming Between 3G and WLAN,'' submitted for publication.


Velocity Estimation in NGWN
In NGWN intelligent techniques  are required to eliminate the QoS degradation and forced termination that may occur when enough resources are not available to  accommodate the handoff requests. Information about  the position and  velocity of the mobile users can be used to accurately predict their trajectories, which can be used to estimate the handoff instances. Then resources for handoff requests can be reserved in advance. This will eliminate handoff failure and the associated QoS degradation. Moreover,  in the hierarchical cellular architecture, which is being proposed for future wirless networks, the knowledge of the users' velocity can be used to assign slow moving users to micro/pico and fast moving users to macro cells. As a result, the handoff rate for the  fast moving users can be reduced. This will increase the system capacity and reduce the number of dropped calls, and hence, will enable the network to provide an  improved quality of service  using its limited resource.

It is desirable to have an  algorithm to estimate the mobile velocity to the desired level of accuracy. The level of accuracy depends on its use. For example, for assignment of  users to a macro, micro, or pico cell, a coarse estimation of velocityclassifying it to slow, medium or fast is sufficient. On the other hand, an accurate velocity estimation is required to predict the trajectory of the mobile users. The estimation accuracy of the algorithm should remain the same over a large velocity range i.e., for  low, medium and high velocity values. It is required that the accuracy of velocity estimationto be independent of the fading types and frequency selectivity of the channel. Moreover, the algorithm should not be computationally extensive. Also, the algorithm should be able to track the change in  velocity accurately, especially when the velocity  is used for tracking the mobile users. To the best of our knowledge none of the existing velocity estimation techniques  satisfy all these requirements simultaneously. Hence, there is a need to find a velocity estimation technique having all these properties. A novel velocity estimation algorithm is proposed that uses the power spectral density of the received signal envelope to estimate user velocity. The proposed algorithm addresses the above issues.

Related work:

S. Mohanty and I.F. Akyildiz, ``An Accurate Velocity Estimation Algorithm for Resource Management in Next Generation Wireless Systems,'' to appear in IEEE Trans. on Wireless Communications.