This paper proposes a technique for 360-degree panoramic video generating with multiple actual and virtual cameras in real-time. Our method acquires multiple images consisting of real images from physical cameras and virtual images from virtual cameras by GPU-based texture and encoding. Then, our method generates 360-degree panoramic images by ORB feature detection and matching to virtual and real images, automatic homography estimation based on the random absent consensus algorithm, multi-band blending and produces high-resolution video stream by Nvidia encoder based Hardware accelerated video encoding. From experimental results, we verified that our method takes at least 45fps at 2K resolution and 33fps at 4K resolution when transmitting the head mounted display stereo, enabling real-time 360-degree panoramic video generation as well as transmission.

Wireless Rechargeable Sensors Network (WRSN) has recently emerged as a promising solution to solve the energy limitation of WRSN. This study considers large-scale WRSNs, where many Mobile Chargers (MCs) are placed to ensure the target coverage and connectivity. We propose a distributed charging algorithm that allows MCs to decide their optimal charging path and charging time. Our proposal is based on the Q-learning technique, where every MC maintains a Q-table that measures the goodness of actions. According to the evaluation, the proposed charging scheme extended the network lifetime by 3.52 times in average and 5.06 times in the best case.

In IoT networks, mitigation of channel access contention is one of the crucial requirements due to a massive number of devices connecting to the shared communication channel. In order to satisfy this requirement, IEEE 802.11ah introduces the restricted access window (RAW) mechanism which divides wireless stations (STAs) into several STA groups and allocates a dedicated RAW slot to each STA group. Since STA grouping plays an important role in this mechanism, several grouping methods have been proposed. However, the IoT networks have various network non-uniformity such as in STA deployment. Although the existing method cannot handle this non-uniform characteristic enough to maintain fairness among STAs. We take up non-uniform STA deployment, heterogeneous traffic patterns and multiple data rates at STAs as the potential network non-uniformity. This paper proposes a sector-based grouping method with adjustable central angles, which can handle all three factors of the network non-uniformity. The performance for the proposed method is evaluated through computer simulation to show that the proposed method can improve the network performance even when STAs with various traffic patterns and data rates are deployed non-uniformly.

In the era of 5G and beyond, advanced network technologies will be adopted along with providing high, diverse Quality of Service (QoS) levels of network services. In such a context, it is necessary to have a more flexible and responsive network provisioning system than the existing ones in 4G. This paper addresses the issue of fine-grained QoS provisioning with payment, which benefits the users and providers. As a solution, we propose Software Defined Networking (SDN)-based QoS provisioning with IOTA micropayments (SQI), which allocates the guaranteed QoS service on a per-network flow or per-user basis after confirmed payment. Different from the previous works, SQI can provide both bandwidths and delay guaranteed services, which leverage the SDN technology. Additionally, SQI can support concurrently multiple network flows, e.g., from different clients. Besides, SQI adopts the (public) IOTA Tangle 1.5, which is more lightweight and faster than the previous IOTA versions. Moreover, SQI includes a private IOTA Tangle that can be tunable to speed up the transaction process. We have implemented SQI using Mininet-WiFi, the POX SDN controller, and IOTA 1.5 and evaluated SQI in various scenarios. The evaluation results show that SQI can provide expected delay and bandwidth (with micropayments) to one and multiple flows. We have also compared the processing request duration of the public IOTA and private ones to reveal the appealing feature of private IOTA.

The Piggyback Network has been proposed as one of the technologies to enable Beyond 5G society. The Piggyback Network provides a high-speed data transfer system with frequency radio, such as millimeter-wave (mmW) links and Store-Carry-Forward (SCF) based communication among multiple autonomous mobilities. Communicating with mmW enables high-speed and large-capacity data transfers. Autonomous mobilities include service robots, cars, and unmanned aerial vehicles (UAV). Simple calculations demonstrated the end-to-end throughput performance when a single mobility unit transfers a single piece of data. This paper considers expansion to handle multiple mobilities and data. To realize this data transfer system, we require data delivery assignment to autonomous mobility and route optimization. We propose local search algorithm for minimizing cost value based on the total time required for data to be delivered to the destination. This paper evaluates the data transfer performance of the Piggyback Network by UAVs in a large-scale field. If multiple data and delivery routes are appropriately assigned to multiple mobilities that can communicate at 10Gbps and move at 100km/h, the Piggyback Network can transfer 100GB of data within approximately 2 min, which corresponds to a throughput of approximately 3000Mbps. If the data size is 1000GB and mobilities can move at 10Gbps, data transfer is completed in approximately 4 min, which achieves approximately 30000Mbps, faster than 5G and long term evolution (LTE) networks.