BEHAVE Behavior-Aware, Intelligent, and Fair Resource Management for Heterogeneous Edge-IoT Systems

Abstract : Data-driven approaches are envisioned to build future Edge-IoT systems that satisfy IoT devices demands for edge resources. However, significant challenges and technical barriers exist which complicate resource management of such systems. IoT devices can demonstrate a wide range of behaviors in the devices resource demand that are extremely difficult to manage. In addition, the management of resources fairly and efficiently by the edge in such a setting is a challenging task. In this paper, we develop a novel data-driven resource management framework named BEHAVE that intelligently and fairly allocates edge resources to IoT devices with consideration of their behavior of resource demand (BRD). BEHAVE aims to holistically address the management technical barriers by 1) building an efficient scheme for modeling and assessment of the BRD of IoT devices based on their resource requests and resource usage; 2) expanding a new Rational, Fair, and Truthful Resource Allocation (RFTA) model that binds the devices BRD and resource allocation to achieve fair allocation and encourage truthfulness in resource demand; and 3) developing an enhanced deep reinforcement learning (EDRL) scheme to achieve the RFTA goals. The evaluation results demonstrate BEHAVE's capability to analyze the IoT devices BRD and adjust its resource management policy accordingly.

 EXISTING SYSTEM :

 ? Their solution addresses very important issues in this area, such as the existence of multiple IoT standards, the heterogeneity of IoT devices and non-interoperability of existing solutions. ? Their solution overcomes some drawbacks of existing designs such as interoperability and scalability. ? It monitors the real-time location of workers, their activity, as well as the existing temperature and lighting in the workplace according to the external weather conditions. ? This has been the case with this new proposal, where an existing energy efficiency system has been improved notably with the application of CAFCLA and GECA, reducing the data traffic that enters the Cloud, especially lessening the amount of information that needs to be transmitted from the ZigBee beacons to the locating engine running in the Cloud.

 DISADVANTAGE :

 ? The work in tackled the problem of admission control and resource allocation with optimization of the devices’ utility using Lyapunov dynamic stochastic optimization approach. ? Thus, DRL is exploited in BEHAVE to generate resource allocation actions as it is capable of modeling complex problems such as resource allocation in Edge-IoT. ? The identification task is formulated as an unsupervised one-class classification (OCC) problem to detect the outliers of the BRD baseline without any labeling overhead. ? The allocation problem is casted as behavior oriented allocation with the goal of maximizing devices’ gain constrained by their budgets. ? EDRL scheme with novel accurate value function approximation for solving dimensionality problem of DRL.

 PROPOSED SYSTEM :

 • Several authors have proposed solutions to motivate, educate and raise awareness among users about the importance of smart energy consumption. • In this regard, serious games are one of the solutions proposed to improve user behaviour and these have been presented and analyzed in numerous research. • Due to the wide variety of the proposed solutions, and above all, the multiple application areas, the state of the art presented here includes the most outstanding works in three different areas, Smart Grids, Industrial and commercial environments and Domestic environments. • The proposed architecture is experimental and is based on embedded devices.

 ADVANTAGE :

 ? Some resource allocation proposals aim to optimize performance objectives such as power efficiency, delay, and computation rate. ? It targets optimization of the IoT system performance with consideration of devices’ BRD and devices’ budgets. ? Our system aims to balance the tradeoff between performance and complexity. ? ON We evaluate the performance of BEHAVE in terms of the detection accuracy of the irregular BRD, detection overhead, IoT devices’ gain, fairness, variance of edge server load and converge ? To evaluate the performance of BEHAVE with respect to resource allocation, we consider the DQN scheme proposed in, DRLRA scheme in , and the optimal exhaustive search for comparison.