The flexible nature of elastic optical networks (EONs) effectively uses spectral resources for optical communication by allocating the minimum required bandwidth to network connections. Since the energy consumption of such networks scales with the magnitude of bandwidth demand, addressing the issue of energy wastage is important. This fact has a profound impact on the design of efficient schemes for energy aware optical networks, and adaptivity arises as one of the most important properties of these networks. Learning Automata are Artificial Intelligence tools that have been used in networking algorithms, when adaptivity to the characteristics of the network environment can result in significantly improved network performance. In this work, a new adaptive power-aware algorithm is introduced, which selectively switches off bandwidth-variable optical transponders (BVTs) under low utilization conditions, to achieve energy efficiency. A novel adaptive scheme, which makes use of Learning Automata to significantly reduce the total energy consumption, while at the same time avoiding the onset of congestion, is proposed. The proposed scheme monitors network congestion, in terms of Bandwidth Blocking Probability (BBP), and the learning mechanism finds the optimal amount of energy-saving so that congestion is avoided, while at the same time significant energy savings are achieved. The proposed Learning Energy-Saving Algorithm (LESA) is evaluated via extensive simulation results, which indicate that it achieves an energy saving of up to 50%, compared to other energy efficient solutions.INDEX TERMS Adaptivity, elastic optical networks, energy-efficiency, learning automata, metropolitan networks.

Low Power Wide Area Networks have emerged as a leading communications technology in the field of Internet of Things sensor and monitoring networks. In such networks, uplink traffic is characterized as a combination of periodic data reports and event-triggered alarm reports. When an many devices detect an event in a short timespan, a burst of concurrent transmissions can occur, leading to a surge of collisions, and thus severe data delivery performance degradation. In this paper, a hybrid random/scheduled access strategy is proposed for mitigating the impact of traffic-triggering events on network performance. Under periodic report traffic the LoRaWAN standard Class A protocol is in effect, but after an event a TDMA scheme is applied. Three implementations of this strategy are described. The first is a pair of novel MACs for LoRaWAN, allowing (a) synchronization of end devices with the network using the event detection as a crude synchronization point, and (b) the dynamic scheduling of groups of devices. The other two implementations build upon a single-hop and a two-hop previously proposed LoRaWAN-based wake-up architectures, respectively. The above approaches are validated and studied through extensive simulation. The results show improved packet delivery ratio over the Class A MAC. The effect is more prominent as the event propagation velocity increases. The proposed approach also surpasses LoRaWAN in energy per delivered bit for high event propagation velocities. Finally, the novel protocol has a lower hardware and deployment complexity than the wake up radio based alternatives, at the cost of higher energy consumption.

No abstract

Summary Elastic optical networks (EON) have emerged as a solution to the growing needs of the future internet, by allowing for greater flexibility, spectrum efficiency, and scalability, when compared to WDM solutions. EONs achieve those improvements through finer spectrum allocation granularity. However, due to the continuity and contiguity constrains, distant connections that are routed through multiple hops suffer from increased bandwidth blocking probability (BBP), while more direct connections are easier to form. This paper proposes HopWindows, a novel method that strategically allocates bandwidth to connections based on their hop distance. This new algorithm applies masks that control the range of frequency slots (FSs) allocated to each n‐hop connection. Furthermore, a new network metric is introduced, the normalized bandwidth blocking probability (normalized BBP). Utilization of this metric ensures increased fairness to distant connections. Extended simulation results are presented which indicate that the proposed HopWindows method achieves a superior performance over the well‐known FirstFit algorithm. The proposed algorithm may achieve a decrease in bandwidth blocking probability of up to 50%.

In recent years, the Internet of Things (IoT) is growing rapidly and gaining ground in a variety of fields. Such fields are environmental disasters, such as forest fires, that are becoming more common because of the environmental crisis and there is a need to properly manage them. Therefore, utilizing IoT for event detection and monitoring is an effective solution. A technique for monitoring such events over a large area is proposed in this research. This work makes use of the Long-Range Wide Area Network (LoRaWAN) protocol, which is capable to connect low-power devices distributed on large geographical areas. A learning-automata-based hybrid MAC model is suggested to reduce the transmission delay, when a small part of the network produces event packets stemming from an event occurrence that is related to environmental monitoring applications, such as events related to forest fires. The proposed hybrid MAC is evaluated via simulation, which indicates that it achieves significantly higher performance in terms of packet delay, when compared to traditional LoRaWAN schemes.