Ad Hoc Networks Testbed Using a Practice Smart Antenna with IEEE802.15.4 Wireless Modules

Ad Hoc Networks Testbed Using a Practice Smart Antenna with IEEE802.15.4 Wireless Modules

Masahiro Watanabe, Sadao Obana, Takashi Watanabe
DOI: 10.4018/978-1-59904-988-5.ch023
(Individual Chapters)
No Current Special Offers


Recent studies on directional media access protocols (MACs) using smart antennas for wireless ad hoc networks have shown that directional MACs outperform against traditional omini-directional MACs. Those studies evaluate the performance mainly on simulations, where antenna beam is assumed to be ideal, i.e., with neither side-lobes nor back-lobes. Propagation conditions are also assumed to be mathematical model without realistic fading. In this paper, we develop at first a testbed for directional MAC protocols which enables to investigate performance of MAC protocols in the real environment. It incorporates ESPAR as a practical smart antenna, IEEE802.15.4/ZigBee, GPS and gyro modules to allow easy installment of different MAC protocols. To our knowledge, it is the first compact testbed with a practical smart antenna for directional MACs. We implement a directional MAC protocol called SWAMP to evaluate it in the real environment. The empirical discussion based on the experimental results shows that the degradation of the protocol with ideal antennas, and that the protocol still achieves the SDMA effect of spatial reuse and the effect of communication range extension.
Chapter Preview


Directional MAC Protocols

Various MAC protocols using smart antennas or directional antennas, typically referred to as directional MAC protocols, have been proposed for ad hoc networks.

Ko et al. (2000) propose DMAC (Directional MAC) in which all frames are transmitted directionally except for the CTS (Clear To Send). Choudhury et al. (2002) propose MMAC (Multi-hop RTS MAC), which involves the multi-hop RTS (Request To Send) to take advantage of the higher gain obtained by directional antennas. These protocols, however, need various additional mechanisms to provide the location information and to forward the RTS.

In (Fahmy, 2002; Nasipuri, 2002) and (Takai, 2002), RTS is transmitted omni-directionally in order to find the receiver in case location information is not available. Each node estimates the direction of neighboring nodes for pointing the beam with AOA (Angle of Arrival) when it hears any signal. Because these protocols employ at least one omni-directional transmission, it limits the coverage area provided by directional transmissions and do not exploit one of the main benefits of directional antennas, i.e., the increase of the transmission range, either.

Ramanathan (2001) proposes circular directional transmission of periodic hello packets to obtain node information that is located farther away than the omni-directional transmission range. Korakis et al. (2003) proposes circular RTS, which scans all the area around the transmitter to find the addressed receiver and to tackle the deafness and the hidden-terminal problem arisen from directional transmissions. Bandyopadhyay et al. (2001) develops additional frames in order to determine the neighbor topology by recording the angle and signal strength under consideration of propagation conditions. Although these schemes attempt communication range extension, circular transmission increases the delay and incurs large control overhead.

In mobile ad hoc evaluation, Ramanathan (2005) shows a directional MAC testbed and simulation results, but not including actual experimental results. Although Nishida (2005) evaluated AODV (Ad hoc On demand Distance Vector routing) protocol based on IEEE802.11b DCF with omni-directional antennas in the experiment, not used directional antennas.

Complete Chapter List

Search this Book: