ADAI and Adaptive PSO-Based Resource Allocation for Wireless Sensor Networks

Abstract

Resource allocation in the Internet of Things (IoT) applications for Wireless Sensor Networks (WSNs) is a challenging problem that requires tasks processing from the appropriate sensor nodes without compromising the Quality-of-Service (QoS). Due to heterogeneity in sensors, the inter-cluster and intra-cluster cooperative communication between sensor nodes hinders the overall resource allocation of the network in terms of energy consumption and response time. Therefore, this paper establishes a multi-agent clustering WSN model, i.e., Adaptive Distributed Artificial Intelligence (ADAI) technique with a hierarchical resource allocation strategy to address the issue of resource allocation in these types of network. For the inter-cluster power allocation, we are considering QoS and energy consumption factors with DAI. Moreover, for intra-cluster resource allocation, this paper introduces Adaptive Particle Swarm Optimization (APSO), which uses its objective functions as the node distance and respective energy loads. The mathematical analysis and simulation results validate the propose method in terms of energy consumption and response time of the network.

Back Propagation Neural Network Based Cluster Head Identification in MIMO Sensor Networks for Intelligent Transportation Systems

“…It can be clearly seen that the computation time almost gets optimized from 0.3 to 0.5 processing power (normalized to 1). The proposed method is shown in green color which is compared with the Adaptive Distributed Arti- ficial Intelligence (ADAI) [50] and standard protocol technique IEEE 802.11, which initially gives higher computation time due the improper distribution of loads in nodes. Although there is a difference of 10-15ms with respect to the 0.3-0.5 processing power, the method will significantly reduce the computation time for any given larger network.…”

Energy-Efficient Resource Allocation Strategy in Massive IoT for Industrial 6G Applications

“…the comparison of optimal allocation of resources is shown with the other existing optimization methods, where the normalized value of power is presented with respect to number of simulations. The optimal allocation becomes less as the number of simulation increases due to the utilization of resources is maximum at the beginning of the process and our proposed hybrid NN optimization scheme outperforms other two methods: APSO [34] and Deep Neural Network (DNN) [36]. The efficiency of correlation or the achieved correlated information value after clustering is presented in Fig.…”

“…In the simulated representation of channel use, the number of bits that can be transferred by each channel varies with respect to signal to noise ratio (SNR). For the comparison purpose, we have taken three methods: conventional (power is distributed equally to nodes), hierarchical maximum likelihood (HML) and advanced particle swarm optimization (APSO) [34]. Among these three, the channel capacity using APSO is good in lower SNR compare to proposed scheme but for higher value of SNR, proposed method outperforms all the other techniques in terms of bits per channel use.…”

“…The mathematical expression of the correlation between the multivariate distributions of different observations is [34]:…”

Adaptive Particle Swarm Optimisation based Energy Efficient Dynamic Correlation Behavior of Secondary Nodes in Cognitive Radio Sensor Networks

“…The authors in [29] have explored the interference power in the case of dynamic spectrum access in a cooperative communication scenario. The authors in [30] studied the location of the FC in a cooperative communication system, considering the different power allocations to make the FC location closer to dense clusters.…”