Sense-making from Distributed and Mobile Sensing Data: A Middleware Perspective Santanu Sarma, Nalini Venkatasubramanian, and Nikil Dutt Department of Computer Science, University of California, Irvine, USA [santanus, nalini, dutt]@ics.uci.edu Abstract This paper presents a scalable and collaborative mobile crowdsens- ing framework for efﬁcient collective understanding of users, con- texts, and their environments. Collaborative mobile crowdsensing enables information to be gathered and shared by users who are directly involved (participatory sensing) or integrated seamlessly as needed (opportunistic sensing) through user mobile platforms. To address the scalability needs of the mobile ecosystem, we ad- ditionally employ compressive sensing techniques for approximate gathering and processing of sensor data - this requires new mech- anisms for sensor data collection, tunable approximate processing, and mobile networking architecture, to create a compressive col- laborative mobile crowdsensing platform called SenseDroid. The proposed framework is build using a multi-tired hierarchical archi- tecture to sense spatial variations of a parameter of interest, per- ceive spatio-temporal ﬁelds, and enable energy efﬁcient local mo- bile sensing with a small number of measurements. This approxi- mate, yet tunable approach combines different sensing approaches opportunistically while trading scalability (and coverage) for data accuracy (and energy efﬁciency). In this paper we propose and dis- cuss the framework and the challenges associated with compres- sive and collaborative mobile sensing for multi-tired hierarchical mobile network architecture for emerging mobile collaborative ap- plications. General Terms Mobile Sensing, Distributed Middleware, Internet- of-Things Keywords Compressive Sensing, Participatory Sensing, Collabo- rative Sensing, Sensor Network, Mobile Phone Sensing. 1. Introduction The proliferation of device platforms and mobile applications (12 billion devices and over 3 million apps by 2017) has changed how humans inter- act. These new interactions enable the mobile device/user to be an active participant in the collection, sharing and dissemination of information- end platforms/users capture and process local context and communicate this in- formation to other platforms/services using heterogeneous connectivities. We envision that the next generation mobile ecosystem will be far more sophisticated, complex and diverse than what is currently used. With each Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for proﬁt or commercial advantage and that copies bear this notice and the full citation on the ﬁrst page. Copyrights for components of this work owned by others than ACM must be honored. Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior speciﬁc permission and/or a fee. Request permissions from Permissions@acm.org. DAC’14, June 01 - 05 2014, San Francisco, CA, USA. Copyright c  2014 ACM 978-1-4503-2730-5/14/06 ...$15.00. http://dx.doi.org/10.1145/2593069.2596688 passing year, we see increasingly powerful smartphones being manufac- tured, which have a plethora of powerful embedded sensors like micro- phone, camera, digital compass, GPS, accelerometer, temperature sensors and many more [8]. Such platforms also support seamless external sensor connectivity (e.g. on body for health and wellness monitoring) that further equips such devices with rich and unique sensing capabilities. Moreover, the ability to easily program today’s smartphones, enables us to exploit these sensors, in a wide variety of application such as personal safety, emer- gency and calamity response, situation awareness, remote activity moni- toring, transportation and environment monitoring [7, 8]. Furthermore, the ability to share the sensed content with other users and applications has en- abled crowds/humans (and their devices) to become information providers in a crowdsensing ecosystem. Effective use of the sensed data relies on ef- fective ”sensemaking” that transforms the gathered data to meaningful in- formation for improved situational awareness, decision making and control. This paper focuses on enabling such ”sensemaking” from mobile users and devices. Broadly, the emerging ﬁeld of mobile phone sensing or crowdsensing can take multiple forms [7]. In participatory sensing, the user is directly involved in the sensing activity; this burden is alleviated in the opportunistic sensing paradigm by delegating and automating the sensing task to the mobile phone sensing system. In this paper, we argue for a collaborative sensing approach where the users collaborate or cooperate to have better and reliable sensing information and obtain missing sensing information when speciﬁc sensors are not available in their own devices. Collaboration can be useful in generating more accurate and reliable information of spatial ﬁelds distributed across geographical areas and region, e.g., multiple temperature sensor readings in a space would be more reliable than a single reading. We discuss some speciﬁc applications of collaborative sensing and sensemaking in following use case scenarios. Disaster and emergency response: Mobile intelligent networks can play a key role in emergency response, surveillance and security, and battle- ﬁeld operations. Consider a ﬁre scenario where information from in-situ and mobile sensors can help in incident perimeter assessment as well as rapid localization of regions with high impact. Coordination among ﬁre ﬁghters is another important aspect in ﬁre rescue operations. A collaborative mobile crowdsensing framework can be used to coordinate among the ﬁreﬁghters for their own safety and as well as quick evacuation. Collaborative sensing can provide situation awareness of different users in a facility during the rescue operation. Based on the situations, rescue operations can be coordi- nated more effectively to reduce response time and save precious lives. Personal health monitoring and wellness: Mobile phone sensing has the potential to continuously collect/sense data for health and wellness anal- ysis. UbiFit Garden [3] is a mobile phone sensing system jointly developed by Intel and University of Washington, which uses small inexpensive on- body sensors and mobile phones along with machine learning techniques for activity modeling to infer people’s activities throughout everyday life. In [12], stress level of mobile user was measured using mobile phones, while [11] explored the use of smartphones in predicting the mode of the users. This can be extended to a family or a group of related people to jointly infer their moods, and exercise routines, exposures to pollutants etc. to ﬁnd com- bined stress quotient. The same can also be used to achieve a family health indicator.