In the present paper, a call admission control scheme that can learn from the network environment and user behavior is developed for code division multiple access (CDMA) cellular networks that handle both voice and da...
详细信息
In the present paper, a call admission control scheme that can learn from the network environment and user behavior is developed for code division multiple access (CDMA) cellular networks that handle both voice and data services. The idea is built upon a novel learning control architecture with only a single module instead of two or three modules in adaptive critic designs (ACDs). The use of adaptive critic approach for call admission control in wireless cellular networks is new. The call admission controller can perform learning in real-time as well as in offline environments and the controller improves its performance as it gains more experience. Another important contribution in the present work is the choice of utility function for the present self-learning control approach which makes the present learning process much more efficient than existing learning control methods. The performance of our algorithm will be shown through computer simulation and compared with existing algorithms.
Although reconfigurable flight control has been well demonstrated on fixed-wing aircraft using existing control surface redundancies, such failure accommodations were widely believed to be impossible for single main r...
详细信息
Although reconfigurable flight control has been well demonstrated on fixed-wing aircraft using existing control surface redundancies, such failure accommodations were widely believed to be impossible for single main rotor helicopters. In fact, on any such existing helicopter, the failure of a main rotor actuator is catastrophic, resulting in a complete loss of control. An actuator geometry that enables reconfigurable helicopter flight control is presented, and, from this, reconfigurable flight-control methods are developed to accommodate main rotor actuator failures. The geometry provides control axes coupling that is key to solving the reconfigurable flight-control problem. Such coupling enables the development of a swashplate reconfiguration strategy, where any two of the three control axes can be controlled at a time, despite a failure in any one of the main rotor actuators. Of most interest is the particular swashplate reconfiguration solution where attitude control (pitch and roll) is retained, whereas vertical control is sacrificed. Vertical control is then achieved by one of two methods: flight to a longitudinal velocity that supports the desired vertical velocity or the use of rotor speed to change the main rotor's thrust to perform limited closed-loop vertical control. These methods are combined to form a robust reconfigurable flight-control method. All of these methods are independent of the underlying flight-control system. Such independence is demonstrated by evaluation of the reconfigurable flight-control methods for two different flight-control systems: a classic proportional-integral-derivative controller and neural dynamic programming, a neural network based controller. All designs are tested by the use of a sophisticated nonlinear validated model of the Apache helicopter.
暂无评论