This letter studies the tracking error in Multi-input Multi-output Feedback Error Learning (MIMO-FEL) system having insufficient excitation. It is shown that the error converges to zero exponentially even if the reference signal lacks the persistently excitation (PE) condition. Furthermore, by making full use of this fast convergence, we estimate the plant parameter while in operation based on frequency response. Simulation results show the effectiveness of the proposed method compared to a conventional approach.

View full abstract

In a real-time system, when the execution of a task is preempted by another task, the interrupted task falls into a blocked state. Since its re-execution begins from the interrupted point generally, the task's timer containing the remaining time until its completion should be maintained in the blocked state. This is the reason for introducing the notion of memorable events in this paper. We present a new timed discrete event model (TDEM) that-adds the memorable events to the TDEM framework of Brandin and Wonham (1994). Using supervisory control theory upon the proposed TDEM, we analyze the schedulability of preemptable periodic and sporadic tasks executing on a uniprocessor.

View full abstract

A novel electronically tunable high input impedance voltage-mode multifunction filter with single inputs and three outputs employing two single-output-operational transconductance amplifiers, one differential difference current conveyor and two capacitors is proposed. The presented filter can be realized the highpass, bandpass and lowpass functions, simultaneously. The input of the filter exhibits high input impedance so that the synthesized filter can be cascaded without additional buffers. The circuit needs no any external resistors and employs two grounded capacitors, which is suitable for integrated circuit implementation.

View full abstract

In this paper, we suggest that all pairings are in a group from an abstract angle. Based on the results, some new pairings with the short Miller loop are constructed for great efficiency. It is possible that our observation can be applied into other aspects of pairing-based cryptosystems.

View full abstract

A novel automatic image annotation (AIA) scheme is proposed based on multiple-instance learning (MIL). For a given concept, manifold ranking (MR) is first employed to MIL (referred as MR-MIL) for effectively mining the positive instances (i. e. regions in images) embedded in the positive bags (i. e. images). With the mined positive instances, the semantic model of the concept is built by the probabilistic output of SVM classifier. The experimental results reveal that high annotation accuracy can be achieved at region-level.

View full abstract