Vijay Raman

Performance Tradeoffs Among Percolation-Based Broadcast Protocols in Wireless Sensor Networks

Broadcast of information in wireless sensor networks is an important operation, e.g., for code updates, queries, membership information, etc. In this paper, we analyze and experimentally compare the performance of vanilla versions of several well-known broadcast mechanisms namely, flooding, site percolation, bond percolation, and modified bond percolation. We carry out our comparison for different network topologies: random, grid, and clustered. Our analysis is performed at the link layer level using a MAC-independent propagation model based on real experiments from the literature. Our main metrics are bandwidth, energy usage, and broadcast latency. Our analytical and experimental results show that, given a desired high reliability for all topologies, flooding and site percolation has the lowest latency; but flooding consumes the most energy per broadcast compared to site percolation. For dense networks, modified bond percolation further lowers energy consumption compared to site percolation, while basic bond percolation leads to a latency increase. For sparse networks, results are similar to a dense network except that site percolation consumes lower energy than modified bond percolation. We briefly discuss implications for different broadcast applications.

A Practical Approach for Providing QoS in Multichannel Ad-Hoc Networks Using Spectrum Width Adaptation

Multichannel wireless networks provide the flexibility to utilize the available spectrum efficiently for achieving improved system performance in terms of throughput and spectral efficiency. However, there has been no practical means for provisioning quality of service (QoS) in multichannel wireless networks. While previous proposals providing signaling and adaptation mechanisms for QoS, they support only fixed-width channels, restricting system performance in networks supporting variable-width channels. In this paper, we propose a distributed mechanism for provisioning QoS by adapting the channel widths in a multichannel, ad-hoc network. Our algorithm builds upon the well-known ETT routing metric to incorporate bandwidth adaptability. We also propose mechanisms for performing admission control and congestion control jointly in a multihop setting. We demonstrate the performance of our algorithm using a modified AODV routing protocol through extensive simulations. Our simulations results show that our proposed approach can achieve up to twice the spectral efficiency and data rates when compared to the greedy approach. Furthermore, our results show that our proposed approach scales well as the network density increases.

SHORT: a static-hybrid approach for routing real time applications over multichannel, multihop wireless networks

Many of the existing multichannel wireless network implementations rely on channel switching capability of the wireless radios to ensure network connectivity. However, due to both software and hardware restrictions switching channels incur a significant delay, which can be prohibitive for many delay sensitive, real time applications, such as VoIP and interactive gaming. The situation can be worse in the case of a multihop network, as every node along the traffic path may require a channel switch that adds up to the overall end-to-end delay. This motivates the need for efficient routing strategies that can make use of the flexibilities of a multichannel network while favoring delay sensitive applications by routing them on low delay paths. In this paper, we propose SHORT, a Static-Hybrid approach for rOuting Real Time applications over multichannel, multihop wireless networks, which ensures low delay paths for delay sensitive applications. Using measurements on a real multichannel testbed, we show that our protocol can provide significantly low delay multihop paths for delay sensitive applications (eg., VoIP) without degrading the throughput performance of non-delay sensitive, best effort traffic, such as TCP that may co-exist in a network.

Performance tradeoffs among percolation-based broadcast protocols in wireless sensor networks

Broadcast of information in wireless sensor networks is an important operation, e.g., for code updates, queries, membership information, etc. In this paper, we analyze and experimentally compare the performance of vanilla versions of several well-known broadcast mechanisms namely, flooding, site percolation, bond percolation, and modified bond percolation. We carry out our comparison for different network topologies: random, grid, and clustered. Our analysis is performed at the link layer level using a MAC-independent propagation model based on real experiments from the literature. Our main metrics are bandwidth, energy usage, and broadcast latency. Our analytical and experimental results show that, given a desired high reliability for all topologies, flooding and site percolation has the lowest latency; but flooding consumes the most energy per broadcast compared to site percolation. For dense networks, modified bond percolation further lowers energy consumption compared to site percolation, while basic bond percolation leads to a latency increase. For sparse networks, results are similar to a dense network except that site percolation consumes lower energy than modified bond percolation. We briefly discuss implications for different broadcast applications.

Signaling Reduction in Idle Mode for Inter-Technology Mobility

With the proliferation of air interface technologies, there are increasingly multi-mode user equipment devices (UEs) that can operate in more than one technology, although not at the same time. To save battery life, the UEs will camp in only one technology at any point in time even when there is coverage of multiple technologies. One key requirement is to limit the uplink signaling when UEs capable of operating in multiple radio access technologies perform inter-technology handovers in idle mode. In this paper, we propose a scheme that reduces this signaling by sending an inter-technology update to the network only when the UE loses coverage of the previously used technology. We also analyze the performance benefits of this scheme with that of the approach of camping on the preferred technology whenever coverage of that technology is available.

Interference aware channel allocation in a multichannel, multi-interface wireless network

In this article, we present an interference-aware channel allocation algorithm for a multichannel, multi-interface wireless network. The channel allocation algorithm minimizes the effect of adjacent and co-channel channel interference both within a node (due to multiple interfaces) and across nodes. Furthermore, our allocation makes use of all the available channels for improving spatial reuse in the network. We demonstrate our algorithm using a multichannel testbed, and bring out the performance benefit of our algorithm compared to another algorithm that utilizes only orthogonal channels.