Efficient Media Access Using Adaptive Array Antennas

Wireless ad hoc networks have traditionally been assumed to use omni-directional antennas. However, it has been recently recognized that the use of advanced antenna technologies could help alleviate the capacity limitations of multi-hop wireless networks, stemming from interference generated in the shared medium [1]. Adaptive array antennas have the ability to automatically respond to an unknown interference environment, in real time, by steering nulls and reducing side lobe levels in the direction of interference, while retaining some desired signal beam characteristics. However, real-time beamforming usually requires sophisticated DSP algorithms that use extra synchronization information, like training sequences, and might be quite “costly” for small wireless terminals. What is more, simply equipping each node with an adaptive antenna is not merely sufficient to improve performance. In this paper, we present a cross-layer approach that enables nodes in an ad hoc network to adapt the radiation pattern of a fully adaptive array antenna using information already available or easily obtainable at the MAC layer. We propose a protocol, called ADAPT, that extends the basic mechanisms of 802.11, and adjusts the antenna pattern, on a perpacket basis, using only MAC layer information (semi-static beamforming). Additionally, we use analytical methods to assess the effect on performance due to adapting the antenna pattern on a per-packet basis, instead of real-time, in the presence of node mobility and contenting neighboring communications. Finally, we use ns-2 simulations to extensively compare the performance of our adaptive antenna configuration against typical omnidirectional and directional configurations. Our protocol is shown to achieve up to a 3× improvement over the omni-directional case and up to a 50% improvement over the directional one, in most scenarios considered.