Technical Program

Paper:

FP1-L3.5

Session:

Ad hoc Networks

Time:

Friday, January 12, 14:00 -16:00

Title:

Efficient Geometric Routing in Ad-hoc Wireless Networks

Authors:

Van M. Chhieng, Ryan H. Choi, Raymond K. Wong, , University of New South Wales, AU

Abstract:

Most geometric routing algorithms for ad-hoc wireless networks rely on greedy forwarding strategies to deliver messages from their sources to destinations. However, there is no guarantee that paths produced by these protocols are optimal. In this paper, we present a Path Regression approach which improves the qualities of paths produced by all geometric routing protocols. Furthermore, we show the use of Path Regression in

RFR which progressively looks for an optimal path that can only be produced by DSR. Extensive experiments show that the proposed algorithm out-performs other approaches such as GOAFR+ by a significant margin.

Paper:

SA1-L2.4

Session:

Autonomic Communications I

Time:

Saturday, January 13, 9:00 -10:20

Title:

An Immunologically-inspired Adaptation Mechanism for Evolvable Network Applications

Authors:

Chonho Lee and Junichi Suzuki, University of Massachusetts, Boston, US

Abstract:

Large-scale network applications are expected to be more autonomous and adaptive to dynamic changes in the network to improve user experience, expand applicationsĄŻ operational longevity and reduce maintenance cost. Based on the observation that various biological systems have already met the requirements (i.e., autonomy and adaptability), this paper describes a biologically-inspired framework, called iNet, to design autonomous and adaptive network applications. iNet is designed after the mechanisms behind how the immune system works. iNet models a set of environment conditions (e.g., network traffic and resource availability) as an antigen and a behavior of network applications (e.g., migration and reproduction) as an antibody. iNet allows network applications to autonomously sense its surrounding environment conditions (i.e., antigens) and adaptively invoke a behavior (i.e., antibody) suitable for the conditions. The configuration of antibodies evolves via genetic operations (e.g., mutation and crossover). Simulation results show that iNet allows agents to autonomously adapt to changing environment conditions by invoking their behaviors suitable for the current environment condition and evolving their antibody configurations.