Conclusions and Future Work

In this paper, we argued for potential bandwidth between AP and end-host as an important metric in the process of AP selection. We described a methodology for estimating the potential bandwidth based on delays experienced by beacon frames from an AP. We also presented results from experiments conducted in a low-noise environment, which showed that the proposed approach yields fairly accurate estimates of the bandwidth.

The described work is in progress and is currently looking at the following issues: - In this paper, we showed results from experiments in a noise-free environment. We plan to evaluate our bandwidth estimation scheme in the presence of noise, using controlled experiments. - The issue of how frequently should nodes estimate bandwidth to various APs in range and the duration over which estimations need to be carried out is currently under investigation. - We evaluate our bandwidth estimation with Netgear MA 311 Wireless cards that use the DCF (Distributed Co-ordination Function) mode of operation. Extending our scheme to be able to estimate the bandwidth between PCF-based (Point Co-ordination Function) APs and end-hosts is another interesting future question. - In the Netgear 311 wireless cards, the beacon frames were transmitted with the same priority as the data frames. We wish to consider the case, when beacons are prioritized over other frames. - Finally, our estimation depends on the assumption that time zero at the AP is the time instant when the first beacon frame is scheduled for transmission, as specified in the IEEE 802.11 standard. While we observed this is very likely the case with the Netgear cards we experimented with,different vendors can be expected to implement beaconing differently. Inferring TBTTs by observing inter-beacon times remains a topic for further investigation.