Research On Wireless Cooperative Communication Mechanism For Robot Rescue Communications

With the utilization of various of wireless access technologies, the heterogeneity of communication network in robot rescue-a classical multi-agent system-becomes more prominent. In recent years, the cooperative communication mechanism is turning into an important research direction of wireless heterogeneous networks. Under the premise that the complex robot rescue communications environment suffers from limited resources and dynamic changes, this thesis utilizes the mechanism of distributed optimization, cooperative relay, game theory and coordinative scheduling to study some key issues of robot rescue communications, such as dynamic spectrum sensing, cooperative relay optimization, optimal power control, and distributed queue scheduling. A more complete robot rescue communications mechanism which focuses on distributed resource optimization, information fusion, relay collaboration and quality of service assurance is established. The main research work of this paper is as follows:An intelligent cognitive cooperative communication system model is built for robot rescue communications:each node in the model shares its cognitive local information with other nodes, and the model realizes accurate global information with the inclusion of an information fusion mechanism. Each user dynamically obtains the network source according to the network state, selecting the most optimized relay for cooperative data transmission; each user realizes optimal power allocation with gaming theory based on self-interest, thus guaranteeing the real-time performance and equity of communications through staging coordination scheduling, and improving the cooperative transmission efficiency and network throughput of the robot rescue communications. All of the above work together to provide an effective communication mechanism for robot rescue that is service reliable, energy efficient, and has a vast coverage area.The research on the utility of frequency spectrum resources in robot rescue communication mainly focuses on sensing available authorized frequency spectrum sources with cognitive radio technology. Recognizing the contradiction between the performance of spectrum sensing and energy assumption, a distributed cooperatively-optimized frequency spectrum sensing algorithm is put forward. Dual thresholds are set to divide the cognitive users into two parts, a trusted group and an incompletely trusted group; a sub-gradient method is taken to execute distributed cooperative optimization for the utility function in spectrum sensing; the energy detection threshold is dynamically regulated to improve the detection accuracy of the incompletely trusted cognitive users in dynamic network environments. Cooperative partners are selected according to the convergence characteristic in the optimization process, and in order to achieve the spectrum sensing result through weighting fusion. Simulation results show that the proposed method enhances both the accuracy and speed of spectrum sensing and reduces the overhead of the cognitive radio networks at the same time.To deal with the requirement of high bandwidth load and the huge volume of data transmission in robot rescue communications, the cooperative relay mechanism is involved to form several relay nodes into the virtual antenna array to assist the communication between the source node and the destination node. A channel gain equivalence-based chance relay selection algorithm is put forward to obtain the optimized space subset gain and user subset gain; the possibility of instantaneous bit errors is utilized to acquire equivalent channel gain, which is the foundation for the selection and management of relay nodes, thus avoiding the case that the communication performance could decrease when involvement relay nodes possesses bad channel qualities, and minimizing data transmission delay and improving the channel volume of the heterogeneous network. Additionally, in order to lower the energy consumption of multi source users and relay users, a utility function is constructed and virtual price of relay users is set; a multi-master multi-slave two-level distributed gaming algorithm is constructed according to the self-interest of users, and virtual price balance is achieved through non-cooperative gaming; this optimizes power management under the guarantee of data flux, and reduces the link disturbance among nodes, extend living span of networks, thus realizing the power union optimization of source nodes and relay nodes.Recognizing the abruptness and locality of data in robot rescue communication, a staging cooperative dispatch based distributed fair queue algorithm is put forward to ensure real-time performance and equity in data transmission. After building the model for the staging cooperative dispatch, we involve a compensation mechanism for data transmission and divide the data transmission process into three stages; the transmission, compensation, and dormancy periods. Time limit constraint functions are designed according to the parameters of the scheduling model to limit the executive time length in each stage; cooperative dispatch rules are regulated by each node according to the dispatch model; and time slot ratios are set according to time limit functions to compensate transmission nodes with large time delays, thus ensuring the real-time performance of data transmission and achieving the equity of data transmission in robot rescue transmissions.