Software-defined networking (SDN) is an emerging architecture used in many applications because of its flexible architecture. It is expected to become an essential enabler for the Internet of Things (IoTs). It decouples the control plane from the data plane, and the controller manages the whole underlying network. SDN has been used in wireless sensor networks (WSNs) for routing. The SDN controller uses some algorithms to calculate the routing path; however, none of these algorithms have enough ability to obtain the optimized routing path. Therefore, reinforcement learning (RL) is a helpful technique to select the best routing path. In this article, we optimize the routing path of SDWSN through RL. A reward function is proposed that includes all required metrics regarding energy efficiency and network Quality-of-Service (QoS). The agent gets the reward and takes the next action based on the reward received, while the SDWSN controller improves the routing path based on the previous experience. However, the whole network is also controlled remotely through the Web. The performance of the RL-based SDWSN is compared with SDN-based techniques, including traditional SDN and energy-aware SDN (EASDN), QR-SDN, TIDE and non SDN-based techniques, such as $Q$ -learning and RL-based routing (RLBR). The proposed RL-based SDWSN outperforms in terms of lifetime from 8% to 33% and packet delivery ratio (PDR) from 2% to 24%. It is envisioned that this work will help the engineers for achieving the desired WSN performance through efficient routing.

360-degree video streaming is expected to grow as the next disruptive innovation due to the ultra-high network bandwidth (60–100 Mbps for 6k streaming), ultra-high storage capacity, and ultra-high computation requirements. Video consumers are more interested in the immersive experience instead of conventional broadband televisions. The visible area (known as user’s viewport) of the video is displayed through Head-Mounted Display (HMD) with a very high frame rate and high resolution. Delivering the whole 360-degree frames in ultra-high-resolution to the end-user significantly adds pressure to the service providers’ overall intention. This paper surveys 360-degree video streaming by focusing on different paradigms from capturing to display. It overviews different projections, compression, and streaming techniques that either incorporate the visual features or spherical characteristics of 360-degree video. Next, the latest ongoing standardization efforts for enhanced degree-of-freedom immersive experience are presented. Furthermore, several 360-degree audio technologies and a wide range of immersive applications are consequently deliberated. Finally, some significant research challenges and implications in the immersive multimedia environment are presented and explained in detail.

Reinforcement learning (RL) is an unsupervised learning technique used in many real-time applications. The essence of RL is a decision-making problem. In RL, the agent constantly interacts with the environment and selects the next action according to previous feedback in terms of reward. In this paper, RL trains Software-Defined Wireless Sensor Networks (SDWSNs) controller to optimize the routing paths. We combine RL and SDN, where RL is applied to the SDN controller to generate the routing tables. We also propose four different reward functions for optimization of the network performance. RL-based SDWSN improves network performance by 23% to 30% in terms of lifetime compared with RL-based routing techniques. RL-based SDWSN performs well because it can intelligently learn the routing path at the controller. In addition, it has a faster network convergence rate than RL-based WSN.

A wireless sensor network (WSN) has achieved significant importance in tracking different physical or environmental conditions using wireless sensor nodes. Such types of networks are used in various applications including smart cities, smart building, military target tracking and surveillance, natural disaster relief, and smart homes. However, the limited power capacity of sensor nodes is considered a major issue that hampers the performance of a WSN. A plethora of research has been conducted to reduce the energy consumption of sensor nodes in traditional WSN, however the limited functional capability of such networks is the main constraint in designing sophisticated and dynamic solutions. Given this, software defined networking (SDN) has revolutionized traditional networks by providing a programmable and flexible framework. Therefore, SDN concepts can be utilized in designing energy-efficient WSN solutions. In this paper, we exploit SDN capabilities to conserve energy consumption in a traditional WSN. To achieve this, an energy-aware multihop routing protocol (named EASDN) is proposed for software defined wireless sensor network (SDWSN). The proposed protocol is evaluated in a real environment. For this purpose, a test bed is developed using Raspberry Pi. The experimental results show that the proposed algorithm exhibits promising results in terms of network lifetime, average energy consumption, the packet delivery ratio, and average delay in comparison to an existing energy efficient routing protocol for SDWSN and a traditional source routing algorithm.

Identification of plant disease is usually done through visual inspection or during laboratory examination which causes delays resulting in yield loss by the time identification is complete. On the other hand, complex deep learning models perform the task with reasonable performance but due to their large size and high computational requirements, they are not suited to mobile and handheld devices. Our proposed approach contributes automated identification of plant diseases which follows a sequence of steps involving pre-processing, segmentation of diseased leaf area, calculation of features based on the Gray-Level Co-occurrence Matrix (GLCM), feature selection and classification. In this study, six color features and twenty-two texture features have been calculated. Support vector machines is used to perform one-vs-one classification of plant disease. The proposed model of disease identification provides an accuracy of 98.79% with a standard deviation of 0.57 on 10-fold cross-validation. The accuracy on a self-collected dataset is 82.47% for disease identification and 91.40% for healthy and diseased classification. The reported performance measures are better or comparable to the existing approaches and highest among the feature-based methods, presenting it as the most suitable method to automated leaf-based plant disease identification. This prototype system can be extended by adding more disease categories or targeting specific crop or disease categories.