Search

Among the challenges in VANETs networks there is power control, which permits to achieve better permformance in terms of Signal to Interference plus Noise Ratio (SINR), Energy Efficiency (EE), Energy Utilization (EU),... Another important point to address is the transmition delay, which we should adapt it to be under an acceptable threshold. For that, we model the network as a Cox Line-Point process, and the transmitter as an M/D/1 queue server. To solve this tradeoff problem, we use machine learning techniques, especialy Q learning algorithm. It is shown , via simulations, that through our algorithm, the vehicular transmitter be able to learn its transmit power in an autonomous way, and achieve better performance for the energy utilization rate, the system waiting time and the area spectral efficiency.

Nowadays, Device to Device (D2D) communication becomes a crucial technology in 5G wireless systems. It is intended to improve the system performance, enhance the user experience and offer a large variety of applications, that is why it attracts more attention. Motivated by the machine learning successful applications to many practical domains, researchers have proposed its application in wireless communication topics, especially in D2D communications. One of the scenarios of D2D communications is underlaid in-band, when each resource block is shared between the D2D and cellular users, thus co-channel interference is a challenging problem. To manage interference mitigation with power allocation technique, we propose, in this paper, a new power control algorithm based on Safe Q-learning algorithm. Our goal is to maximize, in the same time, the throughput of D2D users and the device lifetime metrics, while guaranteeing the required SINR for the cellular communications. It has been demonstrated that through our algorithm, D2D users equipments are able to learn their power in a self-organized manner, in addition to achieving better device lifetime metrics than that based on an enhanced Qlearning algorithm. [ABSTRACT FROM AUTHOR]

The continuous evolution of new wireless technologies has given birth to new application areas for the Internet of Things (IoT) such as Smart Cities, Smart Home, Intelligent Transport Systems, Wireless Sensor Networks, etc. IoT is a network of networks that enables, via small wireless electronic devices called "sensor", to identify and communicate with physical objects. The main IoT objective is to measure and exchange data between physical and virtual worlds. However, the use of these sensors in applications characterized by the coexistence of different traffic loads in the same network could lead to power consumption and latency problems. In this article, we present a new efficient algorithm called (BMPriority-based CSMA/CA) based on message priority and battery energy level of sensor to manage access to the transmission channel. More precisely, our algorithm uses a weighting function to calculate the contention window during which a sensor node must wait before starting its data transmission. Simulation results show that our proposed protocol outperforms S-MAC, IEEE 802.15.4, and ECA-MAC protocols in terms of optimizing Packet Delivery Ratio (PDR), end-to-end latency, and throughput. [ABSTRACT FROM AUTHOR]

One of the toughest challenges encountered in Wireless Sensor Networks (WSNs) is the limited power resources, which directly affects the network services quality and the sensors lifetime. Therefore, the proposed protocols must prolong the network lifetime as long as possible by minimizing the major sources of energy waste duo to the idle listening (no sent or received packets by the sensor). In this paper we propose a new communication protocol based on IEEE 802.15.4 standard called Adaptive Duty-cycle (AD- MAC) that aims at reducing the energy dissipation where the transceiver is waiting potential packets. AD-MAC protocol mainly intends to reduce the power consumed by sensors in Beacon-Enabled mode via a dynamic duty-cycling technique. AD-MAC protocol optimizes the sleep interval between consecutive wake-ups without negatively influencing network performance. This solution uses a new predicting mechanism to estimate the moment when the sensors wake up or sleep according to previous traffic load statistics. Simulation results via NS-2 simulator show that AD-MAC protocol outperforms IEEE 802.15.4 standard in terms of minimizing energy consumption, and Packet Delivery Ratio (PDR). [ABSTRACT FROM AUTHOR]