QoS provisioning in remote monitoring Using SNMP over TCP in multi-hop wireless and wired networks (with Iqbal Asif):

Remote monitoring solutions are prevalent in the financial industry to monitor Financial Transaction Devices (FTD's) such as an Automated Teller Machine (ATM) over highly congested or faulty networks. Fault detection of hardware devices on systems with a diverse software stack is required to be done remotely in a guaranteed delivery mechanism and in a timely fashion. Simple Network Management Protocol (SNMP) has been a standard and is widely used for such applications over UDP for fixed sized and a predefined payload. Such solutions create quality of service (QoS) issues with messages being lost during severe congestion and high network traffic. The message loss can result in, during busy times of the day, a FTD to be down at a remote location for hours. Due to non-receipt of status message by the remote monitoring application, a FTD may stay down for a much longer period than expected resulting financial losses. In this thesis, we propose a solution to this problem which is based on SNMP messaging over TCP. Due to the fact that TCP is congestion-sensitive and UDP is not, and MAC layer back-off algorithms exasperate the delay and loss differently in wired and wireless environment, we conducted a performance analysis of SNMP over several multihop wireless and wired networks. We studied the impact of load and number of hops on the performance of SNMP over TCP and UDP using delay, loss and throughput metrics.

Fault tolerant routing in wireless mesh networks (with Yasir Drabu):

Multi-hop Wireless mesh networks (WMNs) are emerging as a viable alternative solution for cost effective access networks. WMNs can fill the capacity and coverage limitation of traditional cellular, point-to-point wireless systems. In contrast to traditional cellular system, WMNs need only one access point to the wired network, while other access points share a connection over the air. Wireless mesh networks combine several existing technologies and concepts from cellular, ad hoc, and sensor networks to improve network coverage, easy of deployment and better throughput. In this project, we propose and design a fault-tolerant WMN based on a hexagonal topology with multi-radio and directional vectorized antennas. For this model, we introduce an addressing and routing scheme that simplifies the network operations. Further, we extend the routing approach to cope with one or multiple network link failure, as link failure is common in wireless networks. To address this, we exploit the regularity and multipath characteristics of an augmented tri-sectionized hexagonal system to route around link or node failures.

Y. Drabu and H. Peyravi. Fault Tolerant Routing in Tri-Sector Wireless Cellular Mesh Networks. In Proceedings Parallel and Distributed Computer Systems (PDCS), pages 209-216, San Francisco, USA, September 2006.

MAC layer misbehavior in wireless networks (with Jalaa Hoblos):

In wireless networks such as IEEE 802.11, the nodes contend to access the medium through a loosely distributed coordination via contention ( e.g., CSMA/CA) . In these networks the access fairness is provisioned through a uniform random access mechanism that is maintained by an exponential or uniform back-off algorithm. However, nodes can easily gain more access to the medium by manipulating their internal back-off algorithm. A selfish node can reduce its contention window and gain better throughput while other nodes can suffer from media access and hence degradation in their throughput. This misbehavior cannot easily be detected if the base station simply collects samples of back of values. In fact a selfish node can reduces its contention window in an statistical manner and in short run while maintaining the statistical characteristics of its back-off values without being detected in a long term. In this project, we are investigating several statistical techniques to identify a selfish node.

Optimal gateway placement in wireless mesh networks (with Yasir Drabu):

The placement of gateways in  wireless mesh networks has a critical impact on the overall performance in terms of throughput and capacity. Finding optimal locations through search space not only is computationally intractable,but also infeasible due to the dynamic nature of the traffic flows. In this project, we consider the problem of optimally placing a given number of gateway routers in a wireless mesh with the objective to minimize the transmission blocking probability. We are investigating the  performance of spacing the gateways uniformly for an end-to-end performance with uniform loads that are independently distributed across the network. Through simulation and analysis, we are  investigating  the performance gains one can achieve by different placement policies such as uniform placement, random placement, and hot-spot placement for uniform and non-uniform traffic models. For nonuniform link load, we  provide a dynamic programming algorithm for the optimal placement and compare the performance with random and uniform placement.  We are also investigating the effect of different traffic models and applications on the placement algorithms.

Y. Drabu and H. Peyravi. Gateway placement with QoS constraints in wireless mesh networks. In Proceedings of the IEEE International Conference on Networking , pages 46–51, Cancun, Mexico, April 2008. IEEE Computer Society.

Dynamic bandwidth management in IP networks (with Yasir Y. Drabu):

While a significant portion of Internet traffic is still best effort, emerging applications with different quality of service(QoS) requirements (e.g., multimedia) are evolving at a faster pace and require more than best-effort services. The Integrated Services (IntServ) model attempts to achieve an end-to-end per flow QoS through RSVP, however it lacks scalability. The differentiated services (DiffServ) model, on the other hand, attempts to classify packets based on their QoS requirements to receive different services, and it is the focus of recent IETF activity. To ensure QoS, the admission control policy and the packet scheduler should work in tandem to i) control the misbehaved traffic through policing, shaping, metering, pricing techniques, and ii) contain the delay, delay variation, packet loss rate, and bandwidth loss through robust packet scheduling policies. In this project, we investigate an adaptive admission control that reduces bandwidth loss when the traffic is light and enforces a stringent traffic regulation when the traffic load is high. A similar mechanism is used for the packet scheduler that maintains the QoS metrics adaptively within their limited boundaries. Internet2 traffic traces and Opnet synthetic traffic are being used to evaluate the performance of the algorithms.

Y. Drabu and H. Peyravi. An adaptive bandwidth control algorithm for IP routers. In Proceedings International Conference on Communications in Computing (CIC), pages 311–317, Las Vegas, USA, June 2004.

Traffic Management and QoS Provisioning in IP Networks (with Yasir. Drabu):

The objective of this work is to investigate the impact of self-similar traffic on the performance of output buffers in switches and routers. It is a known fact that the superposition of independent alternating renewal processes (flows) can show self-similar characteristics. Since analytical and empirical studies have shown that self-similar traffic can have a detrimental impact on the QoS, finding an effective buffer management algorithm that can manage self-similar traffic has become an important problem in traffic engineering. Optimal resource allocation is directly affected by optimal buffer size and buffer management policy, bandwidth assignment and traffic management. In this project we study the effect of self-similar and bursty traffic on the triggered threshold buffer management algorithms. Besides the second-order self-similar traffic, we are investigating the effects of fractional Brownian motion on active queue management schemes.

Delay and jitter analysis of Generalized TDMA with General Traffic ( with K. Khan):

The TDMA (Time-Division Multiple Access) protocol with its variants has  been widely used and extensively researched in various wire-line and wireless communication networks. Major medium access control (MAC) protocols (e.g.,  random access, fixed and demand assignment protocols), time slot interchange switches, and multiplexers are based on TDMA. The next generation of wireless networks are expected to support multimedia and real-time applications with quality of service (QoS) guaranteed. The General Packet Radio Service (GPRS) that allows continuous wireless connection to data networks is based on TDMA is such an example. The GPRS technology allows mobile phones to be used for sending and receiving data over an Internet Protocol (IP)-based network.Most previously performance analyses of TDMA have been based on the assumption of symmetric and predictable traffic distributions. It has been assumed that either the message inter-arrival time is exponentially distributed and/or the message length is geometrically distributed. While these assumptions provides analytical simplicity to estimate different performance metrics of a communication channel, they may not realistically reflect the behavior of modern traffic which are either correlated and/or bursty in nature. In this research project, first we relax the above assumptions  and present a delay and jitter performance analysis of a generalized TDMA for a general traffic. Second, we present  the analysis for two resource allocation models, namely  consecutive and  interleaved  slot assignment schemes within a frame time.

M. K. Khan and H. Peyravi. Delay and jitter analysis of generalized demand-assignment multiple access (DAMA) protocols with general traffic. In HICSS ’05: Proceedings of the 38th Annual Hawaii International Conference on System Sciences (HICSS’05) - Track 9, page 304.1, Washington, DC, USA, January 2005. IEEE Computer Society.