A Comprehensive Protocol Design for Service Provision of Body Area Network Applications in 5G Networks

• Main Authors: Lin, Chi-Han, 林其翰
• Other Authors: Chen, Wen-Tsuen
• Format: Others
• Language: en_US
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/jw74v4

博士 === 國立清華大學 === 資訊工程學系所 === 107 === The application services of Body Area Networks (BANs) have become more and more popular in recent years. Wearable devices equipped on a human body monitor several physiological data, which are then sent, recorded, and analyzed in remote application services. In this scenario, a smartphone plays an important role to manage these devices and bridge the devices and services. Thus, in this dissertation, the communication from devices to services is divided into two stages, and corresponding methods are respectively designed to enhance the transmission efficiency. The first stage is that the smartphone collects the data from wearable devices. Since a person may equip multiple battery-powered wearable devices (abbreviated as sensors), a smartphone or a hub connecting with these sensors should well schedule the transmissions of sensors to save energy such that the life time of sensors could be lengthened. In other words, an intra-BAN MAC protocol should be designed to avoid signal collisions among sensors. Furthermore, when people meet each other, the signal of a BAN may interfere with those of nearby BANs. Thus, another MAC protocol, namely inter-BAN MAC protocol, should be designed to coordinate BANs to interleave their transmissions. After a hub collects data from sensors, it uploads them to associating Base Station (BS) to perform subsequent data analysis. However, BAN applications may require immediate uploads while detecting high emergency events such as fall, acute heart disease, and epilepsy. These critical tasks should be uploaded as soon as possible, and, hence, the second stage aims to provide an uploading boost protocol for critical tasks. The intra-BAN MAC protocol design aims to alleviate the channel fluctuation that downgrades the transmission rate. To adapt to channel dynamics, a Medium Access Control (MAC) protocol for intra-BAN, called Channel-aware Polling-based MAC protocol (CPMAC), is designed. CPMAC tries to balance the tradeoff between the energy consumption and the probability of completing all transmissions. Simulation results show that, as compared to TDMA-based scheduling and the IEEE 802.15.6 CSMA/CA protocols, CPMAC significantly improves energy efficiency and keeps the latency short. The inter-BAN MAC protocol design takes into account the signal interference from neighboring BANs. Considering the fully distributed nature of BANs, a CSMA/CA-based protocol, called Load-Adaptive MAC Protocol (LAMP), is proposed to automatically configure the system parameter (i.e., contention window size) according to neighbor information and theoretical analysis. LAMP includes an automatic channel switching mechanism to change the operating channel while detecting congestion and a mechanism to rescue sensors that fail to receive the channel switching messages. Additionally, to further reduce the latency of channel access, a Bloom filter-based neighbor information sharing mechanism is designed to rapidly determine whether the current channel is congested. Simulation results show that LAMP outperforms traditional CSMA/CA protocols in terms of throughput and energy consumption up to 42.8% and 50%, respectively. The upload boost protocol is designed under 5G-IoT networks. The idea is to use idle IoT devices of Internet of Things (IoT) in 5G networks to boost the uploading rate. The smartphone divides the data into several segments and sends segments to different IoT devices via the Device-to-Device (D2D) communication of 5G networks. Then, the BS takes advantage of Orthogonal Frequency-Division Multiplexing (OFDM) to simultaneously receive segments from IoT devices and the smartphone. In this dissertation, the problem about the optimal data partitioning and transmission order is formulated. Due to the hardness of the problem, a 1/2-approximation algorithm is proposed. The simulation results shows that the proposed approach can reduce the latency of critical tasks up to 76% comparing with traditional approaches. These three protocols make the service provision of BAN applications become more efficient. This dissertation is concluded by providing some future research issues related to the above protocols.