Blind separation of instantaneous mixture of sources via anindependent component analysis

In this paper, we introduce a procedure for separating a multivariate distribution into nearly independent components based on minimizing a criterion defined in terms of the Kullback-Leibner distance. By replacing the unknown density with a kernel estimate, we derive useful forms of this criterion when only a sample from that distribution is available. We also compute the gradient and Hessian of our criteria for use in an iterative minimization. Setting this gradient to zero yields a set of separating functions similar to the ones considered in the source separation problem, except that here, these functions are adapted to the observed data. Finally, some simulations are given, illustrating the good performance of the method     

Temporal Evolution of Normal Hot Spots in Current-Driven Superconducting Films

In addition to the critical current I_c (T), which generates Phase-Slip Centers (PSCs), thin superconducting films possess a well-defined second limiting current I_h, or current intensity able to maintain a preestablished hot spot. By pulsing step functions of the current and monitoring the voltage response on the nanosecond scale, we have determined (T I_c) and (T I_h). From a dynamic study of the two main modes of dissipation in YBCO and Nb films, it is concluded that PSCs are stable structures in current-biased bridges. In contrast, hot spots grow at a constant rate of a few tens of meters per second, determined by the thermal diffusivity of the material and by its bolometric response time. On reducing the current from I_h, the so-called healing length, or minimum normal length, was found, of the order of 0.2 m in YBCO and 2 m in Nb. In summary, the experiment provides three independent measurements (PSC nucleation time, velocity of growth, and minimum length) for only two parameters (D and ).     

Neuro-fuzzy MIMO nonlinear control for ceramic roller kiln

Artificial neural networks (ANNs) and neuro-fuzzy systems (NFSs) have been widely used in modeling and control of many practical industrial nonlinear processes. However, most of them have concentrated on single-output systems only. In this paper, we present a comparative study using ANNs and co-active neuro-fuzzy inference system (CANFIS) in modeling a real, complicated multi-input–multi-output (MIMO) nonlinear temperature process of roller kiln used in ceramic tile manufacturing line. Using this study, we prove that CANFIS is better suited for modeling the temperature process in control phase. After that, a neural network (NN) controller has been developed to control the above mentioned temperature process due to a feedback control diagram. The designed controller performance is tested by a Visual C++ project and the resulting numerical data shows that this controller can work accurately and reliably when the roller kiln set-point temperature set changes.     

Performance analysis of the IEEE 802.16e power management for the initiations of awakening

To increase battery life in IEEE 802.16e systems, it is essential to efficiently manage energy in mobile stations. The sleep-mode operation in power management helps to increase the life of a station by saving energy consumption. In power management, there are two important performance metrics energy consumption and the response delay of awakening medium access control (MAC) service data unit (SDU). While in a base station (BS) initiation of awakening, the two performance metrics should be simultaneously considered, in a mobile subscriber station (MSS) initiation of awakening, the response delay is not considered because it is self-operational. There performance metrics are affected by the minimum sleep interval (Tmin), the maximum sleep interval (Tmax), and the average interarrival time of awakening MAC SDUs (TI) during sleep-mode operation. Therefore, it is imperative to evaluate the two initiations of awakening depending on TI. To reach a fuller understanding of the performance, this paper shows an analytical mode and simulations results for the standard sleep-mode operation in the IEEE 802.16e MAC.     

Dynamic Resource Allocation for Beyond 3G Cellular Networks

The Multicarrier CDMA Transmission Techniques for Integrated Broadband Cellular Systems (MATRICE) project addresses a candidate solution for a Beyond 3G (B3G) air-interface based on Multi-Carrier Code Division Multiple Access (MC-CDMA). It investigates dynamic resource allocation strategies at the Medium Access Control (MAC) layer to support the transport of Internet Protocol (IP) packets over the air-interface in a cost effective manner and maximise the cell capacity with a target QoS. A candidate Dynamic Resource Allocation (DRA) protocol architecture is proposed that is based on cross-layer signalling to provide reactive resource allocation according to the fast channel and traffic variations. In-line with B3G expectations, the proposed DRA handles a very large number of users with inherent flexibility and granularity necessary to support heterogeneous traffic, and still with limited complexity. Thanks to the modular architecture of the DRA, various scheduling policies are investigated and compared in terms of capacity and reactivity to the system environment. Simulation results have shown that the MATRICE system has the potential to deliver broadband heterogenous services in a cost-effective manner, and emerge as a propespective candidate air-interface for B3G cellular networks.     

REINFORCEMENT LEARNING FOR POMDP USING STATE CLASSIFICATION

Reinforcement learning (RL) has been widely used to solve problems with a little feedback from environment. Q learning can solve Markov decision processes (MDPs) quite well. For partially observable Markov decision processes (POMDPs), a recurrent neural network (RNN) can be used to approximate Q values. However, learning time for these problems is typically very long. We present a new combination of RL and RNN to find a good policy for POMDPs in a shorter learning time. This method contains two phases: firstly, state space is divided into two groups (fully observable state group and hidden state group); secondly, a Q value table is used to store values of fully observable states and an RNN is used to approximate values for hidden states. Results of experiments in two grid world problems show that the proposed method enables an agent to acquire a policy with better learning performance compared to the method using only a RNN.     

EXPERIMENTS WITH ONLINE REINFORCEMENT LEARNING IN REAL-TIME STRATEGY GAMES

Real-time strategy (RTS) games provide a challenging platform to implement online reinforcement learning (RL) techniques in a real application. Computer, as one game player, monitors opponents’ (human or other computers) strategies and then updates its own policy using RL methods. In this article, we first examine the suitability of applying the online RL in various computer games. Reinforcement learning application depends on both RL complexity and the game features. We then propose a multi-layer framework for implementing online RL in an RTS game. The framework significantly reduces RL computational complexity by decomposing the state space in a hierarchical manner. We implement an RTS game—Tank General—and perform a thorough test on the proposed framework. We consider three typical profiles of RTS game players and compare two basic RL techniques applied in the game. The results show the effectiveness of our proposed framework and shed light on relevant issues in using online RL in RTS games.     

Machine learning techniques for anti-money laundering (AML) solutions in suspicious transaction detection: a review

Money laundering has been affecting the global economy for many years. Large sums of money are laundered every year, posing a threat to the global economy and its security. Money laundering encompasses illegal activities that are used to make illegally acquired funds appear legal and legitimate. This paper aims to provide a comprehensive survey of machine learning algorithms and methods applied to detect suspicious transactions. In particular, solutions of anti-money laundering typologies, link analysis, behavioural modelling, risk scoring, anomaly detection, and geographic capability have been identified and analysed. Key steps of data preparation, data transformation, and data analytics techniques have been discussed; existing machine learning algorithms and methods described in the literature have been categorised, summarised, and compared. Finally, what techniques were lacking or under-addressed in the existing research has been elaborated with the purpose of pinpointing future research directions.