Pralhad Deshpande

Bartendr: a practical approach to energy-aware cellular data scheduling

Cellular radios consume more power and suffer reduced data rate when the signal is weak. According to our measurements, the communication energy per bit can be as much as 6x higher when the signal is weak than when it is strong. To realize energy savings, applications must preferentially communicate when the signal is strong, either by deferring non-urgent communication or by advancing anticipated communication to coincide with periods of strong signal. Allowing applications to perform such scheduling requires predicting signal strength, so that opportunities for energy-efficient communication can be anticipated. Furthermore, such prediction must be performed at little energy cost. In this paper, we make several contributions towards a practical system for energy-aware cellular data scheduling called Bartendr. First, we establish, via measurements, the relationship between signal strength and power consumption. Second, we show that location alone is not sufficient to predict signal strength and motivate the use of tracks to enable effective prediction. Finally, we develop energy-aware scheduling algorithms for different workloads - syncing and streaming - and evaluate these via simulation driven by traces obtained during actual drives, demonstrating energy savings of up to 60%. Our experiments have been performed on four cellular networks across two large metropolitan areas, one in India and the other in the U.S.

Predictive methods for improved vehicular WiFi access

With the proliferation of WiFi technology, many WiFi networks are accessible from vehicles on the road making vehicular WiFi access realistic. However, several challenges exist: long latency to establish connection to a WiFi access point (AP), lossy link performance, and frequent disconnections due to mobility. We argue that people drive on familiar routes frequently, and thus the mobility and connectivity related information along their drives can be predicted with good accuracy using historical information - such as GPS tracks with timestamps, RF fingerprints, and link and network-layer addresses of visible APs. We exploit such information to develop new handoff and data transfer strategies. The handoff strategy reduces the connection establishment latency and also uses pre-scripted handoffs triggered by change in vehicle location. The data transfer strategy speeds up download performance by using prefetching on the APs yet to be encountered. Experimental performance evaluation reveals that the predictability of mobility and connectivity is high enough to be useful in such protocols. In our experiments with a vehicular client accessing road-side APs, the handoff strategy improves download performance by roughly a factor of 2 relative to the state-of-the-art. The data transfer strategy further improves this performance by another factor of 2.5.

Performance comparison of 3G and metro-scale WiFi for vehicular network access

We perform a head-to-head comparison of the performance characteristics of a 3G network operated by a nation-wide provider and a metro-scale WiFi network operated by a commercial ISP, from the perspective of vehicular network access. Our experience shows that over a wide geographic region and under vehicular mobility, these networks exhibit very different throughput and coverage characteristics. WiFi has frequent disconnections even in a commercially operated, metro-scale deployment; but when connected, indeed delivers high throughputs even in a mobile scenario. The 3G network offers similar or lower throughputs in general, but provides excellent coverage and less throughput variability. The two network characteristics are often complementary. It is conceivable that these properties can be judiciously exploited for a hybrid network design where 3G data can be offloaded to WiFi for better performance and to reduce 3G network congestion and to lower costs.

Drive-By Localization of Roadside WiFi Networks

We use a steerable beam directional antenna mounted on a moving vehicle to localize roadside WiFi access points (APs), located outdoors or inside buildings. Localizing APs is an important step towards understanding the topologies and network characteristics of large scale WiFi networks that are deployed in a chaotic fashion in urban areas. The idea is to estimate the angle of arrival of frames transmitted from the AP using signal strength information on different directional beams of the antenna - as the beam continuously rotates while the vehicle is moving. This information together with the GPS locations of the vehicle are used in a triangulation approach to localize the APs. We show how this method must be extended using a clustering approach to account for multi-path reflections in cluttered environments. Our technique is completely passive requiring minimum effort beyond driving the vehicle around in the neighborhood where the APs need to be localized, and is able to improve the localization accuracy by an order of magnitude compared with trilateration approaches using omnidirectional antennas, and by a factor of two relative to other known techniques using directional antennas.

Moving bits from 3G to metro-scale WiFi for vehicular network access: An integrated transport layer solution

We investigate a transport layer protocol design that integrates 3G and WiFi networks, specifically targeting vehicular mobility. The goal is to move load from the expensive 3G network to the less expensive WiFi network without hurting the user experience. As the test platform we choose a nationwide 3G network and a commercially operated metro-scale WiFi network. We exploit the often complementary characteristics of these networks for a hybrid design at the transport layer. To this end, we modify the stock Linux SCTP implementation to support ‘striping’ across multiple interfaces and the ability to handle frequent path failures and recovery in a seamless fashion. Instead of simply striping data over two network connections, we develop a utility and cost-based formulation that decides the right amount of load that can be put on the 3G network to maximize the users benefit. We develop and experiment with a transport level scheduler to do this. We call the new SCTP design as oSCTP, meaning ‘SCTP to be used for offloading.’ We demonstrate the effectiveness of oSCTP and show that it is able to deliver superior network throughput and user experience, while significantly reducing the load on the 3G network.

BRAVE: bit-rate adaptation in vehicular environments

Rate selection in a wireless network is the problem of estimating the current channel conditions and determining the best physical layer bit rate for the outgoing frames in order to maximize the current throughput. All rate adaptation algorithms in literature arrive at an estimate of the current channel conditions by considering the recent history often in the order of seconds. In vehicular WiFi access networks, the constantly changing wireless channel conditions make the channel history quickly irrelevant. We develop BRAVE - an SNR-based rate adaptation algorithm, which only considers short history (500 ms) to make rate selection decisions. We show that a coarse-grained training approach is sufficient to estimate the SNR thresholds for rate selection as opposed to previous approaches that train on a per environment or a per AP basis. We study three frame-based rate adaptation algorithms and a popular SNR-based rate adaptation algorithm along with BRAVE and highlight their shortcomings in the rapidly changing vehicular WiFi access environment. In order to compare the algorithms under repeatable channel conditions, we also develop and use a novel emulation methodology where a software radio-based programmable noise generator is used to emulate varying link quality under vehicular mobility. We show that BRAVE performs significantly better than several prominent frame-based and the SNR-based rate adaptation algorithms.

M4M: A model for enabling social network based sharing in the Internet of Things

The true potential of the Internet of Things (IoT) will be realized only when devices are able to harness the collective capabilities of a wide range of peer-devices. In this paper, we propose a novel model where friends in a social network can share device capabilities with their peers in an access controlled manner. We develop a theoretical model of such a peer-to-peer network in which devices can search for remote capabilities, and elaborate on the trade-offs of different algorithms in terms of capability searching and execution. We study various types of social network models to understand the degree of sharing in such networks. Our results show that Barabási-Albert graphs that approximate most real world social networks have a high degree of sharing, validating the utility of our social network based model. We also propose an algorithm that takes any given network and augments it to increase the level of sharing by intelligently suggesting friendship recommendations among vertices. Finally, we describe the prototype implementation of an android mobile application that uses facebook APIs to allow smartphones share GPS and camera capabilities with other devices owned by friends.

MRMV: design and evaluation of a multi-radio multi-vehicle system for metro-WiFi access

In this work we design and evaluate the MRMV (Multi-Radio Multi-Vehicle) system for vehicular WiFi access in the 2.4 GHz band. The design essentially centers around the vehicular client that accesses typical metro-WiFi networks (V2I) under vehicular mobility. MRMV has two features that makes it unique -- i) it uses multiple WiFi interfaces that intelligently associates to different APs to mask handoff latencies, (ii) it is able to use other MRMV clients as relays (V2V) and is thus able to avoid coverage holes. The V2V link uses 900 MHz interfaces for interference avoidance and better ranges. We provide extensive performance evaluation of the MRMV system using a large scale metro-WiFi deployment. We show both connectivity (periods of non-zero throughput) and median throughputs improve substantially over default cases. Overall performance numbers indicate that the MRMV system can be an excellent platform for offloading data from cellular data networks onto unlicensed bands for ubiquitous and high throughput vehicular connectivity.