Q-learning has been primarily used as one of the reinforcement learning (RL) techniques to find the optimal routing path in wireless sensor networks (WSNs). However, for the centralized RL-based routing protocols with a large state space and action space, the baseline Q-learning used to implement these protocols suffers from degradation in the convergence speed, network lifetime, and network energy consumption due to the large number of learning episodes required to learn the optimal routing path. To overcome these limitations, an efficient model-free RL-based technique called Least-Square Policy Iteration (LSPI) is proposed to optimize the network lifetime and energy consumption in WSNs. The resulting designed protocol is a Centralized Routing Protocol for Lifetime and Energy Optimization with a Genetic Algorithm (GA) and LSPI (CRPLEOGALSPI). Simulation results show that the CRPLEOGALSPI has improved performance in network lifetime and energy consumption compared to an existing Centralized Routing Protocol for Lifetime Optimization with GA and Q-learning (CRPLOGARL). This is because the CRPLEOGALSPI chooses a routing path in a given state considering all the possible routing paths, and it is not sensitive to the learning rate. Moreover, while the CRPLOGARL evaluates the optimal policy from the Q-values, the CRPLEOGALSPI updates the Q-values based on the most updated information regarding the network dynamics using weighted functions.

No abstract

This paper presents the development and integration of a power control algorithm into the User Association Algorithm with Optimal Bandwidth Allocation (UAAOBA) to form a Hybrid Algorithm for User Association and Resource Allocation (HAUARA). The power control algorithm updates the transmit power of the Base Stations (BSs) towards a minimum transmit power that satisfies the minimum data rate requirement (1 Gbps) of the User Equipment UEs. The power update is achieved using the Newton Rhapson’s method and it adapts the transmit powers of the BSs to the number of their connected UEs. The developed HAUARA provides an optimal solution for user associations, bandwidth allocation, and transmit powers to UEs concurrently. This maximizes the network energy efficiency by coordinating the load fairness of the network while guaranteeing the quality of service requirement of the UEs. The network energy efficiency performance of the developed HAUARA is compared with that of the UAAOBA. The results show that the developed algorithm has network energy efficiency improvement of 12.36%, 10.58%, and 13.44% with respect to UAAOBA for increase number of macro BS antennas, pico BSs, and femto BSs, respectively. Also, the network load balancing performance of the developed HAUARA is compared with that of the UAAOBA. The results show that the developed algorithm has network load balancing improvement of 12.62%, 10.04%, and 10.34% with respect to UAAOBA for increase number of macro BS antennas, pico BSs, and femto BSs, respectively. This implies that the developed algorithm outperforms the UAAOBA in terms of network energy efficiency and load balancing.

Aims: To assess the traditional birth attendants' (TBAs') knowledge of HIV and prevention of mother to child transmission (PMTCT) of HIV and their practice of PMTCT of HIV services delivery.

Full duplex (FD) and Device-to-Device (D2D) communication are two revolutionary protocols that have enabled better spectrum utilization and more reliable data delivery in wireless networks.  In addition, stochastic geometry tools have become necessary to characterize the randomness in the present networks with respect to the irregular architecture and the competing access schemes. This work analyses the performance of a mobile network comprising nodes which are randomly distributed in a square area, which are equipped with FD radios, and can communicate using D2D. The base station (BS) nodes and user nodes in the network are modelled as points of a homogenous binomial point process (BPP) and a homogeneous Poisson point process (PPP) respectively. The network area is tessellated into cells using Voronoi diagrams which approximates to a nearest BS-to-user node association policy. The user nodes can cache popular file objects which are available in a centralized server in the network and other nodes in proximity can request for such objects and receive them using D2D. Using well known distance distribution expressions and stochastic geometry analysis, the distribution of the signal-to-interference ratio (SIR), the D2D and FD collaboration probabilities and the average coverage probability are derived. It is shown that a network-wide quality of service is maintained without additional spectrum utilization when the user nodes can be intelligently tuned to transmit and receive using FD and/or D2D modes. Keywords— Device-to-Device Communication, Full Duplex Communication, stochastic geometry analysis, Voronoi diagrams, Distance    Distributions