Performance of all-directional transmission and reception algorithms in wireless ad hoc networks with directional antennas

摘要

In this paper, we present simulation results of an all directional transmission and reception algorithm, proposed in Z. Zhang (2005), (named as LiSL/d for a link scheduler) in mobile wireless ad hoc networks with directional antennas. The all-directional algorithms are designed mainly for military applications and exploit the beamforming capabilities of smart antennas to tailor resource access according to the services desired for individual traffic flows while limiting interference, probability of detection and jamming in the network. Specifically, the newly proposed protocol offers four significant advantages: (1) it assumes only directional transmission/reception (no omni-directional or omni-capability is assumed at all), (2) it is distributed, that is, it relies on local information only, (3) it allocates slots to different links dynamically based on demand, (4) power control is easily carried out during neighbor discovery, reservation as well as data transmission period with very little overhead. To evaluate the performance of the proposed algorithms against with IEEE 802.11 omni protocol and the directional virtual carrier sensing (DVCS) protocol, extensive simulations have been conducted in QualNet for the Lakehurst scenario and some random network topologies. Simulation results show that without jamming, both LiSL/d and DVCS improve the network performance by about 3 times and LiSL/d also outperforms DVCS. When there is jamming, depending on the power level of jamming nodes, the percentage of packet received under LiSL/d can be 7 times higher than that of DVCS and IEEE 802.11 with omni-directional antennas.