Collaborative Sensor Networking Towards Real-Time Acoustical Beamforming in Free-Space and Limited Reverberance.
Pierpaolo Bergamo, Shadnaz Asgari, Hanbiao Wang, Daniela Maniezzo, Len Yip, Ralph E. Hudson, Kung Yao, and Deborah Estrin, IEEE Transactions on Mobile Computing, Vol. 3, No. 3, July-September 2004
Beamforming and localization are two interlinking problems and various algorithms have been proposed to tackle each problem individually and jointly. The goal of source localization is to estimate the positions of either a fixed or moving source using a passive and stationary sensor network. Beamforming can be used to determine the direction-of-arrivals and the locations of one or more acoustic sources. In this paper, the authors show that sophisticated realtime acoustical beamforming operation for source localization can be achieved by appropriately using novel array and signal processing algorithms implemented on very low-cost commercial-off-the-shelf platforms. They first present an approximate maximum-likelihood method for DoA estimation, and then show that the least-square method applied to the estimated angels can be used for source localization in free space. By using a novel virtual array model, source localization in the reverberant case is also feasible. The experimental testbed, which is built by handheld, battery-powered Compaq iPAQ devices, is also reported. Extensive results on subarray configurations and measurements for free-space and controlled limited reverberant scenarios are given.
Link-Level Measurements from an 802.11b Mesh Network
Daniel Aguayo, John Bicket, Sanjit Biswas, Glenn Judd, and Robert Morris, Proceedings of ACM SIGCOMM, Portland, Oregon, USA, August 2004
In this paper, the authors provide a measurement study of packet loss on a 38-node urban IEEE 802.11b-based mesh network. The paper reports several observations. The distribution of inter-node loss rates is relatively uniform over the entire range of loss rates. Links with intermediate loss levels are the common case. Most links have relatively stable loss rates from one second to the next, though a small percentage has very bursty losses at that time scale. There is no clear distinction between working and non-working links. Link distance and signal-to-noise ratio do have an effect on loss rates, but the correlation is weak. Experiments using a hardware channel emulator suggest that an important cause of intermediate loss rates is multi-path fading due to reflections in the radio environment rather than attenuation or interference. The patterns and causes of packet loss are important in the design of routing and link-layer protocols.
Comparison of Routing Metrics for Static Multi-Hop Wireless Networks
Richard Draves, Jitendra Padhye, and Brian Zill, Proceedings of ACM SIGCOMM, Portland, Oregon, USA, August 2004
Most current ad hoc routing protocols select paths that minimize hop count. However, minimal hop count paths can have poor performance in static ad-hoc networks since they tend to include long wireless links that are slow or lossy, thus leading to poor throughput. Therefore, a routing algorithm can select better paths by explicitly taking the quality of wireless links into account. In this paper, the authors examined three candidate link-quality metrics for ad-hoc routing, i.e., expected transmission count (ETX), per-hop round trip time (RTT), and per-hop packet pair delay (PktPair), and compare them with minimal hop-count routing in the context of a DSR-based routing protocol. The results are based on measurements on a 23-node static ad-hoc network in an office environment. The empirical results show that, in scenarios with static nodes, the ETX metric outperforms hop-count though it uses longer paths. The one-hop RTT and one-hop PktPair metrics perform poorly, because their load-sensitivity leads to self-interference. However, in a scenario involving a mobile sender, minimum hop-count routing performs considerably better than link-quality routing because the link-quality metrics do not react sufficiently quickly to fast topology change.