Research On Key Technologies Of Data Transmission In Mobile Sensor Networks For Collaborative Emergency Action

Mobile sensor networks are one kind of wireless sensor networks, where sensor nodes can move. Sensor nodes are often deployed on mobile carriers, and can communicate with each other on the move, which can be used for mobile networking and data transmission. Since mobile sensor networks do not require preset infrastructure and have strong mobility, mobile sensor networks can be widely applied to disaster relief, battlefield reconnaissance and surveillance, counter-terrorism and stability maintenance, ecology and natural environmental monitoring and emergency situations, accordingly become a hot challenging research field, which attracts plenty of attentions from the industry and academic community.Data transmission is mainly used to solve the problem of information exchange between mobile sensor nodes, and is an important research direction of mobile sensor networks. Due to nodal mobility, the relationship between nodes in mobile sensor networks may dynamically change, which results in the evolution of data transfer mode from simply forwarding of the nodes in traditional static wireless sensor networks to “store-carry-forward”, and brings great challenge for the corresponding data transmission technology. Typically, the data transmission in mobile sensor networks is mainly based on historical interaction information, which incurs high data transmission delay and energy consumption. Little work considers improving the efficiency of data transmission through the use of location information. In fact, we can use the static sensor selfpositioning technology to accurately locate the position of a mobile sensor node, and select suitable forwarding nodes through some strategies to improve the efficiency of unicast and multicast routing in mobile sensor networks based on the location information. Therfore, this thesis is committed to collaborative emergency environment, and studies three key problems in data transmission, including designing an efficient localization algorithm, an efficient unicast routing protocol and an efficient multicast routing protocol. The main work of this thesis includes the following:1. Proposed a mobile sensor node localization algorithm based on RSSI pairs delay. For the existing problems of node localization algorithms, this paper presents a pair of RSSI-based mobile sensor node localization algorithm. This paper uses pairs of static sensor nodes with known locations to perceive the mobile node signal strength within a short time slot. According to the probability of all pairs of nodes in the size of the signal strength of the relationship between perception, this paper constructs a planning problem on mobile nodes’ locations, and obtains real-time locations of the mobile nodes by the least squares method. The proposed method use the probability of the mobile node achieving static nodess’ RSSI receiver pairs as the magnitude relationship between the positioning basis, reducing the error node RSSI absolute distance measurement and the non-symmetries of RSSI. Extensive experimental results show that the method proposed in this paper has more flexibility and higher accuracy compared to other similar methods.2. Proposed a geographic unicast routing protocol called Geo Rescue based on the prediction and a perception-based energy-aware transmission control strategy. When nodes in mobile sensor networks move dynamicly, there may be errors in choosing nexthop nodes for existing location-based unicast routing algorithms. To address this issue, this paper exploits the current positions and moving directions of the nodes to compute the locations of nodes in the next moment, and proposes a geographic unicast routing algorithm based on prediction. In addition, this paper analyzes the "hops disaster" problem caused by simple position prediction algorithms, and proposes an energy-aware transmission control strategy from the viewpoint of energy savings. This strategy makes further restrictions on the data forwarding conditions, to control node s’ energy consumption and to extend the network lifetime. The experimental results show that, the proposed algorithm outperforms existing algorithms in delivery rate, routing delay, forward hops, node load, and so on, and especially improves network energy consumption significantly.3. Proposed a multicast routing protocol MPGP in mobile sensor networks based on location pre-judgment. The connection between the nodes in mobile sensor networks hasintermittent connectivity and long delays, and location service is a very important service in mobile sensor networks. Many existing studies do not take full advantage of the benefits brought by the location service, and do not fully exploit the individual mobility and the group mobility. To solve these problems, this paper exploits the provided location service, for the application of group mobility model in the network, makes full use of network nodes’ individual mobility and group mobility, and proposes a location pre-judgment-based multicast routing protocol called MPGP. Simulation results show that, compared with existing multicast routing protocols MCopy and DF, MPGP protocol can significantly reduce the multicast delay while requires a small number of forwarding.