Search for: All records

WiFi has emerged as a pivotal technology for delivering Quality of Experience (QoE) to mobile devices. Unfortunately, exploding numbers of competing devices, potential encroachment by cellular technology, and dramatic increases in content richness deliver a more variable QoE than desired. Moreover, such variance tends to occur both across time and space making it a difficult problem to debug. Existing active approaches tend to be expensive or impractical while existing passive approaches tend to suffer from accuracy issues. In our paper, we propose a novel passive client-side approach that provides an efficient and accurate characterization by taking advantage of the properties of Frame Aggregation (FA) and Block Acknowledgements (BA). We show in the paper that one can accurately derive important metrics such as airtime and throughput with only a minimal amount of observed BAs. We show through extensive experiments the validity of our approach and conduct validation studies in the dense environment of a campus tailgate. 

WiFi serves as one of the key mechanisms for wireless access for mobile devices whether at home, on travel, or during normal day-to- day activities. Unfortunately, the perceived high bandwidth and low cost of WiFi is often tempered with varying degrees of quality. Compounding this further, existing techniques for assessing network performance are often expensive in terms of time, bandwidth, and energy making them ill-suited for widespread, longitudinal deployment. In this paper, we propose Fast Mobile Network Characterization (FMNC) to address this shortcoming. FMNC uses sliced, structured, and reordered packet sequences along with an awareness of frame aggregation to rapidly characterize available bandwidth. FMNC does this within the context of a single HTTP GET, consuming less than 100 KB on the downlink with resolution of the path characteristics typically occurring in under 250 ms. We demonstrate the performance of FMNC through extensive lab experiments under a variety of configuration scenarios.