Social-Aware Data Collection Scheme Through Opportunistic Communication in Vehicular Mobile Networks

Zhipeng Tang, Anfeng Liu, Changqin Huang
2016 IEEE Access  
To enable the intelligent management of Smart City and improve overall social welfare, it is desirable for the status of infrastructures detected and reported by intelligent devices embedded in them to be forwarded to the data centers. Using "SCmules" such as taxis, to opportunistically communicate with intelligent devices and collect data from the sparse networks formed by them in the process of moving is an economical and effective way to achieve this goal. In this paper, the social welfare
more » ... ta collection paradigm SWDCP-SCmules data collection framework is proposed to collect data generated by intelligent devices and forward them to data centers, in which "SCmules" are data transmitters picking up data from nearby intelligent devices and then store-carry-forwarding them to nearby data centers via short-range wireless connections in the process of moving. Because of the storage limitations, "SCmules" need to weigh the value of data and select some less valuable data to discard when necessary. To quantify the value of data and find a well-performed selection strategy, the concept of priority is introduced to the SWDCP-SCmules scheme, and then, the simulated annealing for priority assignment SA-PA algorithm is proposed to guide the priority assignment. The SA-PA algorithm is a universal algorithm that can improve the performance of SWDCP-SCmules scheme by finding better priority assignment with respect to various optimization targets, such as maximizing collection rate or minimizing redundancy rate, in which priority assignment problem is converted into an optimization problem and simulated annealing is used to optimize the priority assignment. From the perspective of machine learning, the process of optimization is equal to automatically learn socialaware patterns from past GPS trajectory data. Experiments based on real GPS trajectory data of taxis in Beijing are conducted to show the effectiveness and efficiency of SWDCP-SCmules scheme and SA-PA algorithm. INDEX TERMS Vehicular mobile networks, social welfare, data collection, opportunistic communication, oblivious data mules, simulated annealing, machine learning. 2169-3536 2016 IEEE. Translations and content mining are permitted for academic research only. Personal use is also permitted, but republication/redistribution requires IEEE permission. See http://www.ieee.org/publications_standards/publications/rights/index.html for more information. VOLUME 4, 2016 Z. Tang et al.: Social-Aware Data Collection Scheme Through Opportunistic Communication VOLUME 4, 2016 where P is priority assignment and T is time interval. S, H, V and Q are known elements of C. As a result, the simplified notation of J C , J R and J S are listed below: J C (P, T) = C (P, T) J R (P, T) = 1 − R (P, T) J S (P, T) = λ 1 C (P, T) + λ 2 (1 − R (P, T)) (λ 1 + λ 2 = 1, λ 1 ≥ 0 and λ 2 ≥ 0) In summary, the formal description of the problem solved by SA-PA algorithm can be stated as follow: Given S, H, V, Q and the value of D (V i , t x ) where V i ∈ V and t x ∈ [ t 1 , t 2 ), find a P best which can maximize J (P best , [ t 2 , t 3 )).
doi:10.1109/access.2016.2611863 fatcat:5ly5fbl3drbm7oyrvqfuwc6pue