One-to-One and One-to-Multiple User-Task Mapping in Task-Abundant Mobile Crowd Sensing

August 2021

Mentor: Ting Li


Mobile crowd sensing (MCS) is a promising sensing paradigm that effectively gathers and evaluates data collected from the real world. It relies on sensors in mobile devices that are commonly used among people, such as cameras and GPS in smartphones and wearables.[1] MCS can be utilized in a variety of applications, including public transportation information collecting, environmental monitoring, and indoor localization.[1] With a large number of users and tasks, one of the fundamental problems is how to allocate these tasks to each user with the consideration of distance traveled by all users, maximum distance traveled by each user, the number of total users recruited, and time sensitivity for each task. Thus, this study investigates the task allocation in task-abundant MCS. In particular, there are two scenarios studied: one-to-one mapping (each user can only fulfill one task) and one-to multiple mapping (each user can fulfill multiple tasks) between users and tasks. This research uses real-world data from the D4D dataset, containing 50,000 users’ phone call records and cell phone tower locations. The algorithms used to match tasks to users are bipartite matching through repeated augmenting paths and maximum flow (one-to-one mapping) and greedy algorithms (one-to-multiple mapping). The results show that 1. bipartite matching algorithms (one-to-one) present lower task completeness rates but better (smaller) distance/user value; 2. there is a trade-off existing between distance traveled and user participation in one-to-multiple mapping; 3. selecting the minimum distance algorithm may lead to smaller loss compared to the minimum user algorithm, given the same weighting for 1 km traveled and one user recruited.

[1] X. Li and X. Zhang, "Multi-Task Allocation Under Time Constraints in Mobile Crowdsensing," in IEEE Transactions on Mobile Computing, vol. 20, no. 4, pp. 1494-1510, 1 April 2021.

POTS Modification on FAU Membrane for Oil/Water Separation

August 2019

PI: Aisheng Huang


Oil spillage has caused serious environmental issues around the world. Therefore, separating oil and water has become a vital concern, and effective ways for separation need to be developed. Existing traditional ways, including gravity separation, skimming, flotation, absorption and electrocoagulation, are finding difficult to achieve low costs and high efficiency.[1][4] Therefore, interests in hydrophobic material have raised among scientists to separate oil/water as it can only allow oil to pass through but not water. Numbers of coatings or membranes are found out to be effective for oil/water separation, including MFI and COF. [2][4] However, costs of synthesis, separation efficiency and flux have the potential to be further improved. The FAU zeolite membrane can bear high pressure and is suitable for post-synthesis to change its properties. [3] The stainless-steel wire net is a promising material with low cost and large pores. The large pores can provide a high flux for oil/water separation, whereas the low cost is crucial for industrial application. Thus, an interest has been drawn to a stainless-steel wire net supported FAU zeolite membrane for oil/water separation. However, FAU zeolite membrane is a hydrophilic material. In order to change FAU zeolite membrane from hydrophilic to hydrophobic, post-synthesis may be a feasible solution. Perfluorooctyltriethoxysilane (POTS) is a chemical compound that is largely used to modify the surface of a material to gain a hydrophobic property, thus it may be able to modify the FAU zeolite membrane through post-synthesis to achieve a hydrophobic property without changing FAU crystalline structure.[4] The hydrophobic property can therefore allow FAU membrane to separate oil/water, hopefully remains the advantages of FAU membrane and stainless-steel wire net. The results show that the POTS has successfully presented on the FAU membrane, resulting a hydrophobic membrane. The separation efficiencies of FAU membrane with POTS modification on stainless-steel wire net to separate dichloromethane/water and petroleum ether/water are 98.3% and 98.6%; and the flux for the separations are 9.2e+04 Lm^(-2)h^(-1) and 1.1e+05 Lm^(-2)h^(-1) respectively.

[1] X. Du, S.J. You, X.H. Wang, Q.R. Wang, J.D. Lu, Switchable and simultaneous oil/water separation induced by prewetting with a superamphiphilic self-cleaning mesh, Chem. Eng. J. 313 (2017) 398–403.
[2] Liu, Ruochen, et al. “Hydrophilicity-Controlled MFI-Type Zeolite-Coated Mesh for Oil/Water Separation.” Separation and Purification Technology, vol. 195, 2018, pp. 163–169., doi:10.1016/j.seppur.2017.11.064.
[3] Huang, Aisheng, et al. “Seeding-Free Synthesis of Dense Zeolite FAU Membranes on 3- Aminopropyltriethoxysilane-Functionalized Alumina Supports.” Journal of Membrane Science, vol. 389, no. 272, ser. 279, 2012. 279.
[4] Jiang, Yunzhe, et al. “Stainless-Steel-Net-Supported Superhydrophobic COF Coating for Oil/Water Separation.” Journal of Membrane Science, vol. 587, no. 117, ser. 177, 2019. 177.

Designer ☆ Yijun Liu

Developer ☆ Yijun Liu

Please use your laptop to access this site.