Ioan Sorin Comsa

Dynamic data aggregation protocol based on multiple objective tree in Wireless Sensor Networks

Data aggregation has been widely applied as efficient techniques in order to reduce the data redundancy and the communication load in Wireless Sensor Networks (WSNs). However, for dynamic scenarios, structured protocols may incur high overhead in the construction and the maintenance of the static structure. Without the explicit downstream and upstream relationship of nodes, it is also difficult to obtain high aggregation efficiency by using structure-free protocols. In order to address these aspects, we propose a semi-structured protocol based on the multi-objective tree. The routing scheme can explore the optimal structure by using the Ant Colony Optimization (ACO). Moreover, by using the prediction model for the arriving packets based on the sliding window, the adaptive timing policy can reduce the transmission delay and enhance the aggregation probability. Therefore, the packet transmission converges from both spatial and temporal points of view for the data aggregation procedure. Finally, simulation results validate the feasibility and the high efficiency of the novel protocol when compared with other existing approaches.

Construction of Data Aggregation Tree for Multi-objectives in Wireless Sensor Networks through Jump Particle Swarm Optimization

Abstract As a typical data aggregation technique in wireless sensor networks, the spanning tree has the ability of reducing the data redundancy and therefore decreasing the energy consumption. However, the tree construction normally ignores some other practical application requirements, such as network lifetime, convergence time and communication interference. In this case, the way how to design a tree structure subjected to multi-objectives becomes a crucial task, which is called as multi-objective steiner tree problem (MOSTP). In view of this kind of situation, a multi-objective optimization framework is proposed, and a heuristic algorithm based on jump particle swarm optimization (JPSO) with a specific double layer encoding scheme is introduced to discover Pareto optimal solution. Furthermore, the simulation results validate the feasibility and high efficiency of the novel approach by comparison with other approaches.

Multi Objective Resource Scheduling in LTE Networks Using Reinforcement Learning

The use of the intelligent packet scheduling process is absolutely necessary in order to make the radio resources usage more efficient in recent high-bit-rate demanding radio access technologies such as Long Term Evolution LTE. Packet scheduling procedure works with various dispatching rules with different behaviors. In the literature, the scheduling disciplines are applied for the entire transmission sessions and the scheduler performance strongly depends on the exploited discipline. The method proposed in this paper aims to discuss how a straightforward schedule can be provided within the transmission time interval TTI sub-frame using a mixture of dispatching disciplines per TTI instead of a single rule adopted across the whole transmission. This is to maximize the system throughput while assuring the best user fairness. This requires adopting a policy of how to mix the rules and a refinement procedure to call the best rule each time. Two scheduling policies are proposed for how to mix the rules including use of Q learning algorithm for refining the policies. Simulation results indicate that the proposed methods outperform the existing scheduling techniques by maximizing the system throughput without harming the user fairness performance.

Adaptive data aggregation with probabilistic routing in wireless sensor networks

Periodical extraction of raw sensor readings is one of the most representative and comprehensive applications in Wireless sensor networks. In order to reduce the data redundancy and the communication load, in-network data aggregation is usually applied to merge the packets during the routing process. Aggregation protocols with deterministic routing pre-construct the stationary structure to perform data aggregation. However, the overhead of construction and maintenance always outweighs the benefits of data aggregation under dynamic scenarios. This paper proposes an Adaptive Data Aggregation protocol with Probabilistic Routing for the periodical data collection events. The main idea is to encourage the nodes to use an optimal routing structure for data aggregation with certain probability. The optimal routing structure is defined as a Multi-Objective Steiner Tree, which can be explored and exploited by the routing scheme based on the Ant Colony Optimization and Genetic Algorithm hybrid approach. The probabilistic routing decision ensures the adaptability for some topology transformations. Moreover, by using the prediction model based on the sliding window for future arriving packets, the adaptive timing policy can reduce the transmission delay and can enhance the aggregation probability. Therefore, the packet transmission converges from both spatial and temporal aspects for the data aggregation. Finally, the theoretical analysis and the simulation results validate the feasibility and the high efficiency of the novel protocol when compared with other existing approaches.

Probabilistic Data Aggregation Protocol Based on ACO-GA Hybrid Approach in Wireless Sensor Networks

In Wireless Sensor Networks (WSNs), data aggregation techniques have the ability of reducing the data redundancy and the communication load. The probabilistic aggregation protocols make the dynamic routing decision, the nodes do not have explicit knowledge of downstream and upstream neighbors, and then it is difficult to obtain high aggregation efficiency. In order to address this problem, this paper proposes a new probabilistic aggregation protocol based on Ant Colony Optimization (ACO)-Genetic Algorithm (GA) hybrid approach. The Multi-Objective Steiner Tree (MOST) is defined as the optimal structure for data aggregation, which can be explored and frequently exploited during the routing process. Moreover, by using the prediction model based on the sliding window for the future arriving packets, the adaptive timing policy can adjust the timer interval to reduce the transmission delay and to enhance the aggregation probability. Finally, simulation results validate the feasibility and the high efficiency of the novel protocol when compared against other existing approaches.

Adaptive proportional fair parameterization based LTE scheduling using continuous actor-critic reinforcement learning

Maintaining a desired trade-off performance between system throughput maximization and user fairness satisfaction constitutes a problem that is still far from being solved. In LTE systems, different tradeoff levels can be obtained by using a proper parameterization of the Generalized Proportional Fair (GPF) scheduling rule. Our approach is able to find the best parameterization policy that maximizes the system throughput under different fairness constraints imposed by the scheduler state. The proposed method adapts and refines the policy at each Transmission Time Interval (TTI) by using the Multi-Layer Perceptron Neural Network (MLPNN) as a non-linear function approximation between the continuous scheduler state and the optimal GPF parameter(s). The MLPNN function generalization is trained based on Continuous Actor-Critic Learning Automata Reinforcement Learning (CACLA RL). The double GPF parameterization optimization problem is addressed by using CACLA RL with two continuous actions (CACLA-2). Five reinforcement learning algorithms as simple parameterization techniques are compared against the novel technology. Simulation results indicate that CACLA-2 performs much better than any of other candidates that adjust only one scheduling parameter such as CACLA-1. CACLA-2 outperforms CACLA-1 by reducing the percentage of TTIs when the system is considered unfair. Being able to attenuate the fluctuations of the obtained policy, CACLA-2 achieves enhanced throughput gain when severe changes in the scheduling environment occur, maintaining in the same time the fairness optimality condition.

Backup Path with Energy Prediction Based on Energy-Aware Spanning Tree in Wireless Sensor Networks

In order to decrease the energy consumption and to prolong the network lifetime, the energy-aware spanning tree as a data aggregation technique has been used in wireless sensor networks. Nevertheless, the energy constraint caused by the global reconstruction still severely influences the performance of the sensor system. Our approach aims to reduce the occurrence of the global reconstruction through the backup path. In addition, in order to prevent the redundant paths, a dynamic prediction method is proposed in order to adaptively estimate the possibility of the node failure. The theoretical analysis and simulations demonstrate the efficient performance of the proposed approach.

A scalability hierarchical fault tolerance strategy: Community Fault Tolerance

Most of hierarchical fault tolerance strategies did not pay much attention to scalability of fault tolerance. In distributed system, scalability is a very important feature. To tolerant failures when the scale of the system changing is a normal and important scenario. Especially in nowadays, almost all the cloud computing companies provide their computing services elastically. To add extra devices or remove devices in order to provide different services happens all the time. In such a scenario, it is very important that the fault tolerance strategy is scalable. In this paper, we introduce dynamic programming thoughts to build hierarchical regions as communities for fault tolerance strategy and apply different strategies based on communities instead of a single process. We call this fault tolerance strategy as Community Fault Tolerance. It cannot only reduce the memory overload by eliminating the number of records of messages inside the community region, but also provides a good characteristic of scalability. The scalability property of our strategy makes it handle with the scenario of adding devices or removing devices in the distributed system easily.