Praveen Gopalakrishnan

Voice capacity of IEEE 802.11b, 802.11a and 802.11g wireless LANs

IEEE 802.11 based wireless local area networks (WLANs) are becoming popular in home, enterprise and public access areas primarily due to their low cost, simplicity of installation and high data rates. While WLANs continue to be predominantly data centric, there is growing interest in using WLANs for voice, especially in enterprise markets. This paper presents new analytical and simulation results for the conversational speech capacity of WLANs and compares the different WLAN technologies in that regard. Specifically, we consider IEEE 802.11b, 802.11a and 802.11g systems in the infrastructure mode and find that voice capacity is a strong function of the channel bandwidth, codec packetization interval, data traffic and the packet size used by data. For the IEEE 802.11g system, we find that capacity depends on the CTS-to-self and RTS-CTS legacy protection mechanisms, with the RTS-CTS mechanism achieving lower capacity. We show that the analytical results are in close agreement with those from simulations and conclude the paper by highlighting some key factors that dictate the capacity of WLANs.

Performance Analysis of Drama: A Distributed Policy-Based System for Manet Management

DRAMA is a distributed policy-based management system designed to manage mobile ad hoc networks (MANETs). Its design philosophy is to create intelligent, self-adaptive policy agents to manage dynamic networks without human intervention. Network management functions are performed in a distributed fashion by these policy agents, rather than being controlled by a centralized management station. Policies are used to control the frequency and content of network management messages exchanged among policy agents in a way that reduces bandwidth usage and increases the utility of management messages. This greatly enhances management efficiency and reduces the bandwidth overhead required for network management. As with any new technology, there is a question about the scalability of this approach. The purpose of the work described in this paper is to study whether the DRAMA policy-based network management system can scale to networks of 500+ nodes. The study uses a novel simulation-based approach to evaluate DRAMA performance when DRAMA is used to manage MANETs of up to 500 nodes. The results confirm that the DRAMA distributed policy-based management paradigm provides superior performance over a centralized management paradigm for MANETs

Improving WLAN voice capacity through dynamic priority access

The most commonly deployed wireless local area network (WLAN) standard, IEEE 802.11b, provides no inherent quality of service (QoS) mechanisms to support delay sensitive applications. In particular, 802.11b WLANs exhibit poor voice capacity. This problem stems from limitations in the mandatory medium access control (MAC) protocol, called the distributed coordination function (DCF). WLANs commonly comprise an access point (AP) that must deliver all downlink packets, yet is afforded no special privileges under the DCF. This disparity leads to high downlink packet delays, which is the limiting factor in WLAN voice capacity. We believe that the main mechanism to improve capacity is to enable the AP to acquire the medium with a higher priority than the stations. This method is espoused in the forthcoming QoS-extension, IEEE 802.11e. In addition to the technique itself, knowing when and how long to apply it is critical for voice capacity improvement. We present such a technique, named aggrementing, for legacy 802.11b APs, as well as a generic approach that relates real-time system performance with the dynamics of such priority medium access techniques. We demonstrate that intelligently applying these techniques can improve WLAN voice capacity by as much as 25%.

Security in an insecure WLAN network

Wireless LANs (WLANs) based on IEEE 802.11 standards are becoming increasingly popular for both business and residential use. However, the very nature of wireless networks exposes the risks and vulnerabilities that a malicious user can exploit and severely compromise the network. In this paper, we provide a design of a secure wireless system (SWS) that can not only detect various attacks and misbehaviors of a 802.11 based wireless networks, but also help recover from them. A prototype implementation is also described.

Dynamic Network Selection using Kernels

We present a new algorithm for vertical handover and dynamic network selection, based on a combination of multi- attribute utility theory, kernel learning and stochastic gradient descent. We show that this new method is able to improve network selection in a non-stationary mobile environment. Furthermore, since the kernel employed is based on the utility functions for attributes such as Availability, Quality and Cost, the kernel regression in fact gives interpretable results. We present simulation results that demonstrate our algorithm being able to dynamically learn utilities and efficiently select networks.

Voice capacity of IEEE 802.11b and 802.11a wireless LANs in the presence of channel errors and different user data rates

IEEE 802.11 based wireless local area networks (WLANs) are becoming popular in home, enterprise and public access areas primarily due to their low cost, simplicity of installation and high data rates. While WLANs continue to be predominantly data centric, there is growing interest in using WLANs for voice, especially in enterprise and hot spot scenarios. The paper presents new results for the conversational speech capacity of 802.11b and 802.11a WLANs by specifically considering channel errors and different user rates. Our results show that both systems suffer a significant loss in capacity for packet error rates (PER) greater than 5%. We also show link adaptation can prove to be invaluable in coping with higher PERs. Finally, we show that even a limited presence of low data rate users severely limits the overall voice capacity of WLANs.

A Generic Solution to Software-in-the-Loop

The low fidelity and speed of traditional simulations have become unacceptable for the complex large-scale networks of today. In this paper we propose alternative techniques and focus on a software-in-the-loop implementation. Software-in-the-loop provides us with the following two-fold advantages: (a) it helps solve traditional simulation problems of model validity and (b) it can be used in the design phase as well as in the testing phase of a project. However, software-in-the-loop brings its own set of challenges, as we discuss in this paper. We will present a generic end-to-end solution that mitigates the challenges of a software-in-the-loop configuration to bring it it to its full potential. The success of our solution will be exemplified by its use in two government funded projects where it was successfully used to analyze scalability performance in one case and to perform unit and integration testing in a second case. The focus of this paper will be on the use of software-in-the-loop versus traditional simulations, discussing the challenges, issues and decision processes involved with the use of software-in-the-loop.

NEDAT - A Network Engineering Design Analytic Toolset to Design and Analyze Large Scale MANETs

Future battlefield networks such as FCS and WIN-T will rely on mobile ad hoc networks (MANETs) to satisfy their communications requirements. Thus there is a critical need for systematic techniques based on formal methodologies to design MANETs to meet mission requirements. However, given the complexity of MANETs, the variety of applications they need to support, and the associated performance measures, it is extremely challenging to perform such an engineering design of MANETs. The challenges stem both from the large number of design choices that can be made for any given mission, and, more importantly, from the lack of a systematic body of principles that a network designer can rely upon to guide these choices. To address this critical need, a joint effort has been initiated between Telcordia and CERDEC to formulate a Network Engineering Design Analytic Toolset (NEDAT) that applies formal network science-based approaches to systematically and accurately design, analyze and predict the performance of MANETs. In this paper, we present the results from the first phase of this joint effort.

INTELiCON: Intelligent Connectivity Framework for the Simultaneous Use of Multiple Interfaces

Widespread deployment of Wi-Fi together with 3G upgrades to cellular networks is rapidly creating areas where multiple wireless IP technologies are accessible. WiMAX will further contribute to this phenomenon. The ability to use diverse access technologies simultaneously creates opportunities to improve application service quality and reliability. In this paper, we present an intelligent connectivity framework (INTELiCON) that allows devices to exploit access diversity in new ways. INTELiCON provides a modular, flexible platform for discovering, selecting, executing and evaluating the most appropriate connectivity strategies. The framework executes connectivity strategies and evaluates their effectiveness while maintaining transparency to the application layer. Furthermore INTELiCON can also dynamically implement new strategies to meet variations in network conditions or application requirements. We present the architectural characteristics of the INTELiCON framework with details of a prototype implementation and show how the framework exploits access diversity to support high-quality streaming video despite limited resources and unreliable links.