Adaptive Medium Access Control for Internet-of-Things Enabled Mobile Ad Hoc Networks
MetadataShow full item record
An Internet-of-Things (IoT) enabled mobile ad hoc network (MANET) is a self organized distributed wireless network, in which nodes can randomly move making the network traffic load vary with time. A medium access control (MAC) protocol, as a most important mechanism of radio resource management, is required in MANETs to coordinate nodes’ access to the wireless channel in a distributed way to satisfy their quality of service (QoS) requirements. However, the distinctive characteristics of IoT-enabled MANETs, i.e., distributed network operation, varying network traffic load, heterogeneous QoS demands, and increased interference level with a large number of nodes and extended communication distances, pose technical challenges on MAC. An efficient MAC solution should achieve consistently maximal QoS performance by adapting to the network traffic load variations, and be scalable to an increasing number of nodes in a multi-hop communication environment. In this thesis, we develop comprehensive adaptive MAC solutions for an IoT-enabled MANET with the consideration of different network characteristics. First, an adaptive MAC solution is proposed for a fully-connected network, supporting homogeneous best-effort data traffic. Based on the detection of current network traffic load condition, nodes can make a switching decision between IEEE 802.11 distributed coordination function (DCF) and dynamic time division multiple access (D-TDMA), when the network traffic load reaches a threshold, referred to as MAC switching point. The adaptive MAC solution determines the MAC switching point in an analytically tractable way to achieve consistently high network performance by adapting to the varying network traffic load. Second, when heterogeneous services are supported in the network, we propose an adaptive hybrid MAC scheme, in which a hybrid superframe structure is designed to accommodate the channel access from delay-sensitive voice traffic using time division multiple access (TDMA) and from best-effort data traffic using truncated carrier sense multiple access with collision avoidance (T-CSMA/CA). According to instantaneous voice and data traffic load conditions, the MAC exploits voice traffic multiplexing to increase the voice capacity by adaptively allocating TDMA time slots to active voice nodes, and maximizes the aggregate data throughput by adjusting the optimal contention window size for each data node. Lastly, we develop a scalable token-based adaptive MAC scheme for a two-hop MANET with an increasing number of nodes. In the network, nodes are partitioned into different one-hop node groups, and a TDMA-based superframe structure is proposed to allocate different TDMA time durations to different node groups to overcome the hidden terminal problem. A probabilistic token passing scheme is adopted for packet transmissions within different node groups, forming different token rings. An average end-to-end delay optimization framework is established to derive the set of optimal MAC parameters for a varying network load condition. With the optimal MAC design, the proposed adaptive MAC scheme achieves consistently minimal average end-to-end delay in an IoT-based two-hop environment with a high network traffic load. This research on adaptive MAC provides some insights in MAC design for performance improvement in different IoT-based network environments with different QoS requirements.