Heng-Ming Chen

Intelligent feature extraction and classification of anuran vocalizations

Abstract An intelligent identification system for mixed anuran vocalizations is developed in this work to provide the public to easily consult online. The raw mixed anuran vocalization samples are first filtered by noise removal, high frequency compensation, and discrete wavelet transform techniques in order. An adaptive end-point detection segmentation algorithm is proposed to effectively separate the individual syllables from the noise. Six features, including spectral centroid, signal bandwidth, spectral roll-off, threshold-crossing rate, spectral flatness, and average energy, are extracted and served as the input parameters of the classifier. Meanwhile, a decision tree is constructed based on several parameters obtained during sample collection in order to narrow the scope of identification targets. Then fast learning neural-networks are employed to classify the anuran species based on feature set chosen by wrapper feature selection method. A series of experiments were conducted to measure the outcome performance of the proposed work. Experimental results exhibit that the recognition rate of the proposed identification system can achieve up to 93.4%. The effectiveness of the proposed identification system for anuran vocalizations is thus verified.

An adaptive multimedia streaming dissemination system for vehicular networks

An adaptive seamless streaming dissemination system for vehicular networks is presented in this work. An adaptive streaming system is established at each local server to prefetch and buffer stream data. The adaptive streaming system computes the parts of prefetched stream data for each user and stores them temporarily at the local server, based on current situation of the users and the environments where they are located. Thus, users can download the prefetched stream data from the local servers instead of from the Internet directly, meaning that the video playing problem caused by network congestion can be avoided. Several techniques such as stream data prefetching, stream data forwarding, and adaptive dynamic decoding were utilized for enhancing the adaptability of different users and environments and achieving the best transmission efficiency. Fuzzy logic inference systems are utilized to determine if a roadside base station or a vehicle can be chosen to transfer stream data for users. Considering the uneven deployment of BSs and vehicles, a bandwidth reservation mechanism for premium users was proposed to ensure the QoS of the stream data premium users received. A series of simulations were conducted, with the experimental results verifying the effectiveness and feasibility of the proposed work.

A self-healing clustering algorithm for underwater sensor networks

Underwater wireless sensor networks (UWSNs) is a novel networking paradigm to explore aqueous environments. The characteristics of mobile UWSNs, such as low communication bandwidth, large propagation delay, floating node mobility, and high error probability, are significantly different from terrestrial wireless sensor networks. Energy-efficient communication protocols are thus urgently demanded in mobile UWSNs. In this paper, we develop a novel clustering algorithm that combines the ideas of energy-efficient cluster-based routing and application-specific data aggregation to achieve good performance in terms of system lifetime, and application-perceived quality. The proposed clustering technique organizes sensor nodes into direction-sensitive clusters, with one node acting as the head of each cluster, in order to fit the unique characteristic of up/down transmission direction in UWSNs. Meanwhile, the concept of self-healing is adopted to avoid excessively frequent re-clustering owing to the disruption of individual clusters. The self-healing mechanism significantly enhances the robustness of clustered UWSNs. The experimental results verify the effectiveness and feasibility of the proposed algorithm.

Applications of machine learning techniques to a sensor-network-based prosthesis training system

Abstract In the past, the utilization of the limb prosthesis has improved the daily life of amputees or patients with movement disorders. However, a leg-amputee has to take a series of training after wearing a limb prosthesis, and the training results determine whether a patient can use the limb prosthesis correctly in her/his daily life. Limb prosthesis vendors thus desire to offer the leg-amputee a complete and well-organized training process, but they often fail to do so owing to the factors such as the limited support of human resource and financial condition of the amputee. This work proposes a prosthesis training system that the amputees can borrow or buy from the limb prosthesis vendors and train themselves at home. Instant feedback messages provided by the prosthesis training system are used to correct their walking postures during the self-training process. An embedded chip is used as a core to establish a body area sensor network for the prosthesis training system. RFID readers and tags are employed to acquire the 3D positioning information of the amputees limbs in this work to assist in diagnosing the amputees walking problem. A series of simulations were conducted and the simulation results exhibit the effectiveness and practicability of the proposed prosthesis training system.

Application of cellular automata and type-2 fuzzy logic to dynamic vehicle path planning

Abstract Nowadays, most road navigation systems’ planning of optimal routes is conducted by the On Board Unit (OBU). If drivers want to obtain information about the real-time road conditions, a Traffic Message Channel (TMC) module is also needed. However, this module can only provide the current road conditions, as opposed to actually planning appropriate routes for users. In this work, the concept of cellular automata is used to collect real-time road conditions and derive the appropriate paths for users. Notably, type-2 fuzzy logic is adopted for path analysis for each cell established in the cellular automata algorithm. Besides establishing the optimal routes, our model is expected to be able to automatically meet the personal demands of all drivers, achieve load balancing between all road sections to avoid the problem of traffic jams, and allow drivers to enjoy better driving experiences. A series of simulations were conducted to compare the proposed approach with the well-known A* Search algorithm and the latest state-of-the-art path planning algorithm found in the literature. The experimental results demonstrate that the proposed approach is scalable in terms of the turnaround times for individual users. The practicality and feasibility of applying the proposed approach in the real-time environment is thus justified.

An Intelligent Energy Management Mechanism for Electric Vehicles

ABSTRACT Electric vehicles (EVs) have become increasingly popular all over the world in recent years. Many countries have been offering reward policies and facilitating the establishment of EV charging stations and battery exchange stations to encourage use of these vehicles by the public. However, in terms of electricity demand, the rapid establishment of EV charging stations and battery exchange stations may lead to significant increases in peak loads, the contracted capacities, and basic electricity charges. In this work, an intelligent EV energy management mechanism is proposed to make use of scheduling systems for the charging stations in order to determine when to store electricity in batteries according to the real-time electricity price and the recharging requirements of EVs. Meanwhile, a recharging suggestion module is presented in this work for locating the most suitable charging station or battery exchange station for an EV according to the available information on hand. When an EV cannot reach any charging station because it is running out of electric power, a mobile CV management module is used to assist the EV to find a suitable mobile CV for recharging. Notably, a well-known machine learning technique, multiobjective particle swarm optimization, was employed in this work to assist in solving the multiobjective optimization problems during the design of an energy management mechanism. The experimental results show that the proposed mechanism can balance the loading of battery charging and exchange stations, and lower the load peak to keep electricity cost down. Meanwhile, the recharging suggestion module can decrease the driving distance of EVs for finding the charging stations, as well as decreasing the waiting time wasted while charging. The mobile CV management module, for its part, can effectively prevent EVs from becoming stranded on the road because they have run out of electricity.

Application of fast learning neural-networks to identification of mixed anuran vocalizations

The proposed identification system for mixed anuran vocalizations is to provide the public to easily consult online. The raw mixed anuran vocalization samples are first filtered by noise removal, high frequency compensation, and discrete wavelet transform techniques in order. An adaptive end-point detection segmentation algorithm is proposed to effectively separate the individual syllables from the noise. Six features, including spectral centroid, signal bandwidth, spectral roll-off, threshold-crossing rate, spectral flatness, and average energy, are extracted and serve as the input parameters of wrapper feature selection method. Meanwhile, a decision tree is constructed based on several parameters obtained during sample collection in order to narrow the scope of identification targets. Then fast learning neural-networks are employed to classify the anuran species. Experimental results exhibit that the recognition rate of the proposed identification system can achieve up to 93.3%. The effectiveness of the proposed identification system for anuran vocalizations is thus verified.

A prediction-based joint bandwidth allocation scheme for heterogeneous wireless networks

With advanced network technologies in recent years, people may connect with different types of networks anytime, anywhere. Since wireless network resource distribution is an important issue, we propose a user mobility prediction algorithm, which considers the coverage of different types of base stations and varied mobility of pedestrians, vehicles, and mass transportation. In addition, a novel bandwidth utilization optimization technique is employed in this work to allocate bandwidth more efficiently. Hybrid genetic algorithm, which combines Genetic Algorithm and the local search to improve the frequency of finding Pareto set, is adopted to realize the optimization problem. The performance of our algorithm is compared to two other state-of-the art approaches in the literature. The simulation results show that our algorithms can achieve desirable performance in terms of network utilization, throughput, and QoS quality in the heterogeneous wireless networks.

A group intelligence-based asynchronous argumentation learning-assistance platform

Structured argumentation support environments have been built and used in scientific discourse in the literature. However, to the best our knowledge, there is no research work in the literature examining whether student’s knowledge has grown during learning activities with asynchronous argumentation. In this work, an intelligent computer-supported collaborative argumentation-based learning platform that detects whether the learners address the expected discussion issues is proposed. After each learner presents an argument, a term weighting method is adopted to derive input parameters of a one-class support vector machines classifier which determines if the learners’ arguments are related to the discussion topics. Notably, a peer review mechanism is established to improve the quality of the classifier. Besides, a feedback module is used to issue feedback messages to the learners if the learners have gone off on a tangent. The experimental results revealed that the students were benefited by the proposed learning-assistance platform.

A self-adaptive joint bandwidth allocation scheme for heterogeneous wireless networks

Our approach coordinately manages radio resources with multiple radio access technologies in an optimum way. A mobility prediction module and a remaining bandwidth estimation module are established at each BS as shown in the lower part of the above figure. The global radio resource manager located at the upper part of this figure consists of a bandwidth utilization optimization module. A mobile host (MH) collects required parameters and the latest MHs state is immediately updated at the BS the MH resides in when a MHs state changes. The mobility prediction module will check with a pre-built lookup table to determine whether the handoff will occur. If the handoff is predicted to occur, the MH sends out a bandwidth requirement to a nearest BS, and the remaining bandwidth estimation module in the target BS will be activated to determine the amount of remaining bandwidth that can be used by the handoffs. In case some BS finds that its bandwidth is going to be inadequate, a bandwidth adjustment coordinator located at the BS will request the common bandwidth utilization optimization module at its upper level to reallocate the bandwidth for the BSs/APs of different RATs. The results of the bandwidth reallocation will then be sent back to each BS to reassign the bandwidth to the MHs. A mobility prediction module is proposed to predict the mobility pattern of mobile host.A lookup table records the probability for the possible handoffs moving into the coverage of the BS.Bandwidth is estimated for the possible arriving MHs outside the coverage of the BS.The optimization of utilization and fairness for the bandwidth allocation are addressed.Hybrid Genetic Algorithm is employed to achieve real-time computation requirement. Recently, people have been able to connect with different types of networks anytime, anywhere using advanced network technologies. In order to properly distribute wireless network resources among different clients, this work proposed a user mobility prediction algorithm, which takes the coverage of different kinds of base stations, and the volatile mobility of pedestrians, vehicles, and mass transportation, into consideration. In addition, a novel bandwidth utilization optimization technique is proposed in the algorithm to allocate bandwidth more efficiently. The Hybrid Genetic Algorithm, which combines the Genetic Algorithm and local searches to improve the frequency of finding a Pareto set, is used to realize the optimization problem as well. Compared with our previous work and the other four methods in the literature, the simulation results show that our proposed work can achieve desirable performance by network utilization, throughput, and QoS quality in the heterogeneous wireless networks.