Re-OPSEC: Real time opportunistic scheduler framework for energy aware mobile crowdsensing

Assignment of Sensing Tasks to IoT Devices: Exploitation of a Social Network of Objects

“…7: Comparison of the CS-ME algorithm with an equivalent scenario where a constant sampling frequency is assigned to each device taking part in the crowdsensing and with the Re-OPSEC algorithm [11] we consider here is the one where MSF was set to 0.067 Hz, which is compared with the case of crowdsourcing with an equally set sampling frequency (ESF) of 0.067/20 Hz for each node (the number of considered devices was constant and equal to 20, all with the battery fully charged). We further compared the algorithm with the Re-OPSEC algorithm proposed in [11]. According to this approach, a single sensing task is assigned to the most convenient sensing nodes at each time period, i.e.…”

“…Fair resource allocation mechanisms can be applied to overcome this problem. Some examples are provided by the studies presented in [9][10][11][12]. The Contextaware Mobile Sensor Data EngiNe (C-MOSDEN), a scalable energy-efficient data analytics platform for on-demand distributed mobile crowd, is proposed in [9].…”

“…selecting the most appropriate communication channel, or sampling rate. A similar approach is used in [10], where [26] No activity based participation Lathia et al [27] No context-based Goldman et al [28] No smart city Wang et al [29] No community based Eisenman et al [30] No social vectors based Perera et al [9] Yes location and activity-aware Skorin-Kapov et al [10] Yes cloud-based quality-driven management Bradai et al [11] Yes expected opportunity based Hu et al [12] Partially application-oriented service collaboration Capponi et al [33] Partially based on sensing potential and data collection utility particular attention is put on satisfying global sensing coverage requirements and avoiding redundant sensory activities. In [33], a distributed framework for data collection is proposed to minimize the cost of sensing by estimating the sensing potential and the data collection utility.…”

“…The latter is computed by accounting the sensing interest of the involved applications. Energy efficiency and battery saving are the main objectives of Re-OPSEC [11], where tasks are assigned to smartphones according to their expected opportunity of succeeding in the sensing task, and their residual battery charge. In [12], the concept of social vector, where node attributes are defined, is introduced.…”

“…Within this context, one of the main issues is related to the fact that most of the involved objects are characterized by limited amounts of resources, such as energy, processing power, storage capacity, and transmission bandwidth. Therefore, resource management is of paramount importance, as it is confirmed by the high number of different approaches proposed in the literature to address this issue [9][10][11][12]. Another key issue is related to the involvement of the devices that should perform the sensing tasks, which requires a description of the available capabilities and the way to transfer this information to either the community of involved objects (for distributed approaches) or to the central controller.…”