Joint Optimization of Computation Offloading, Data Compression, Energy Harvesting, and Application Scenarios in Fog Computing

Abstract

Fog computing is considered to be an effective method to solve the problem of high latency and high energy consumption of IoT devices. A suitable computation offloading strategy can provide a low offloading cost to the user device. Most researches on computation offloading in fog computing focus on one or two targets to improve system performance, however, the actual system needs to meet a comprehensive demand. Therefore, the joint optimization of multi-objective in multiple scenarios is a very meaningful problem. Inspired by this, the paper highlights the joint optimization research for fog computing, which proposes a Joint Computation offloading, Data compression, Energy harvesting, and Application scenarios (JCDEA) algorithm. The related mathematical model is constructed and the cost expressions of local computing, fog computing, and cloud computing are derived. Through the proposed algorithm, solving the computation offloading strategy is transformed into solving the minimum cost and is simplified by controlling strategy factors. Moreover, five simulation experiments are conducted and the meaningful conclusions are drawn, which contain that (1) the cost of fog computing is lower than that of local and cloud computing in most time slots and cloud computing can compensate for fog computing in complex environments; (2) the cost increases approximately linear with the amount of offloaded data; (3) the number of user devices and the compression ratio affect the fog-to-cloud ratio (FCR), while the FCR affects the cost; and (4) the related offloading strategy distribution and the cost are obtained for different scenarios. The JCDEA algorithm always outperforms than that of the random selection algorithm in all scenarios.

Randomization-Based Dynamic Programming Offloading Algorithm for Mobile Fog Computing

“…e task offloaded to the fog will be redistributed by the algorithm, and some will be further offloaded to the cloud server. According to [9,11,14,15], it is known that the offloading backhaul data is extremely small, much smaller than of the offloaded, so the return cost is chosen to be ignored. Let the transmission rate between users and F-AP be B F ; then, the transmission time to the fog is…”

“…Moreover, the processing speed of the local device, the fog server, and the cloud server is set to f L � 2.1 × 10 8 cycle/s, f F � 5 × 10 9 cycle/s, and f C � 10 × 10 9 cycle/s [9], respectively. e wireless transmission rate between the local and F-AP is 15 Mbps and from the fog to cloud is 40 Mbps [15]. Besides, ϕ � 100cycle/s, T limit � 1200ms, H � 500, and c � 1J/s are further set.…”

Energy Coherent Fog Networks Using Multi-Sink Wireless Sensor Networks

“…Biswas et al [80] focused on reducing the energy loss problem and designing an energy-efficient data transfer scenario between IoT devices and clouds. Bai et al [81] suggested a joint computation offloading, data compression, energy harvesting, and application scenarios algorithm for FC. Maher et al [82] introduced a data backup system based on multicloud and FC.…”

Impact of Artificial Noise on Security Capability of Energy Harvesting Overlay Networks

“…Among these solutions, harvesting energy from RF signals neither demands complex energy harvesting equipments nor depends time-variant energy resources. Such advantages of this energy harvesting solution induce it to become a bright candidate deployed in small-size users in 5G mobile communications or IoT to provision energy, linger the life-time, and enhance energy efficiency [6][7][8]. Currently, this solution can be carried out through SWIPT (simultaneous wireless information and power transfer) [9][10][11] or relaying communications [12][13][14].…”