Trail is a common outdoor recreation facility. The construction of trail often changes space structures of the nature and affect the vegetation of the trailside. Many studies indicated that the impact of outdoor recreation directly and indirectly change the plant species coverage and composition. A few new trails were recently established in Panping-Shan, Kaohsiung, and there is a need of environmental data for the management . However, the vegetation structure along the trail and the degree of trail construction impact were unclear. This study sampled vegetation and environmental factors along the trails in Panping-Shan, Kaohsiung. The trail in Panping-Shan has three patterns, footpath, plank and macadam. 35 large plots ¡]10*10m2¡^and 132 small plots ¡]1*1m2¡^ have been established along the trail. The data were analyzed with Two-way indicator species analysis ¡]TWINSPAN¡^and Canonical correspondence analysis ¡]CCA¡^ to seek the correlation of the vegetation composition, species distribution and the trailside environment gradient. The results show that the dominant species in Panping-Shan are Broussonetia papyrifera and Macarange tanarius, and the vegetation could be divided into four subtypes according to the indicator species; the understory vegetation could be divided into six types by TWINSPAN. The CCA result shows that the canopy coverage and the edaphic factors, soil depth and soil moisture, are the major environment factors influencing the understory vegetation. Between the different patterns of trail, there were no significant difference, but the environment of the footpath is more heterogeneous. It is suspected that the current light recreation activities and the unique geological structures made no obvious understory vegetation difference between the trailside and the forest interior.

In this thesis, we introduce a novel model based on the idea of autoencoders. Different from a classic autoencoder which reconstructs its own inputs through a neural network, our model is closer to Kernel Canonical Correlation Analysis (KCCA) and reconstructs input data from another data set, where these two data sets should have some, perhaps non-linear, dependence. Our model extends traditional KCCA in a way that the non-linearity of the data is learned through optimizing a kernel function by a neural network. In one of the novelties of this thesis, we do not optimize our kernel based upon some prediction error metric, as is classical in autoencoders. Rather, we optimize our kernel to maximize the "coherence" of the underlying low-dimensional hidden layers. This idea makes our method faithful to the classic interpretation of linear Canonical Correlation Analysis (CCA). As far we are aware, our method, which we call a Kernel Coherence Encoder (KCE), is the only extent approach that uses the flexibility of a neural network while maintaining the theoretical properties of classic KCCA. In another one of the novelties of our approach, we leverage a modified version of classic coherence which is far more stable in the presence of high-dimensional data to address computational and robustness issues in the implementation of a coherence based deep learning KCCA.

Abstract Patterns of plant species composition and their relationships to environmental factors were investigated in Jhuokou River basin. 102, 20 ¡Ñ 20 m plots with woody stems ¡Ù1.0 cm diameter at breast height (DBH) data and 12 environmental variables were analysed by Two Way Indicator Species Analysis (TWINSPAN), Detrended Correspondence Analysis (DCA) and Canonical Correspondence Analysis (CCA) to classify the vegetation types and determine the significant environmental variables that affect the distribution of vegetation. Classification and regression tree (CART) were then used to perform vegetation classification tree based on these significant variables. The vegetation classification result showed that 102 sampling plots can be classified into 9 vegetation types : 1. Daphniphyllum hlaucescens subsp. oldhamii - Cyclobalanopsis morii vegetation type ; 2. Neolitsea acuminatissima - Cyclobalanopsis morii vegetation type ; 3. Adinandra formosana - Lithocarpus lepidocarpus - Machilus thunbergii vegetation type ; 4. Elaeocarpus japonicus - Castanopsis cuspidate - Machilus thunbergii vegetation type ; 5. Ardisia quinquegona - Tricalysia dubia - Beilschmiedia erythrophloia vegetation type ; 6. Schefflera octophylla - Helicia formosana - Beilschmiedia erythrophloia vegetation type ; 7. Castanopsis formosana - Mallotus paniculatus - Schefflera octophylla vegetation type ; 8. Cyclobalanopsis glauca - Glochidion rubrum - Sapindus mukorossii vegetation type ; 9. Champereia manillana - Kleinhovia hospita - Murraya vegetation type. DCA and CCA distinguished 8 significant environmental variables from 12 measured variables. Altitude and warmth index were the most important variables in 8 significant environmental variables, but were highly correlated. When used vegetation classification tree to predict the position of the reference vegetation alliance, average accuracy was 56.9 %. The results indicated that the current data was still insufficient to predict the vegetation type at alliances level with environmental variables.

No description available.

Weathered waterborne treated wood is believed to behave differently than new wood during service regarding the loss of its metallic-based preservatives. Also, weathered preserved wood should be separated from the unpreserved wood upon recycle. The first objective of this dissertation was to evaluate losses from weathered CCA-treated wood samples at different retention levels under normal field conditions and to compare leaching to new ACQ (as alternative to CCA). Results showed that arsenic leached at a higher rate than chromium and copper in all CCA treated wood samples, while copper leached the highest from the ACQ sample. Overall results suggest that the leaching rate of metals on a percent basis from in-service pressure treated wood may increase as the wood weathers; however due to lower retention levels of the metals in the wood as it ages the yearly mass of metals lost would be at similar or at lower quantities in comparison to new treated wood. The second objective was to evaluate the use of automated X-ray fluorescence (XRF) systems for identifying and removing As-based and Cu-based treated wood within the recovered wood waste stream. A full-scale online automated XRF-detection, conveyance and diversion system was used for experimentation. At the different applied feeding rates and belt speeds, online sorting efficiencies of waste wood by XRF technology were high (>70% for both treated wood and untreated wood). The incorrectly diverted pieces of wood were attributed to deficiencies in the wood conveyance systems and not to deficiencies in the XRF-based detection. Online sorting was shown to sort wood which would meet the residential soil cleanup target levels in Florida when an infeed is composed of 5% of treated wood pieces. Comparisons with other sorting methods show that XRF technology can potentially fulfill the need for cost-effective processing at large wood recycling facilities (> 30 tons per day). Management of weathered CCA-preserved wood, due to its continuity of leaching metals and the need to remove it upon the recycle of wood, will likely continue until complete banning and removal from the environment, a process that may extend up to the next century.

Increasing volumes of Chromated Copper Arsenate (CCA) treated wood materials are being removed from service, either as a result of reaching the end of their service life or being replaced with non-treated wood materials. These materials, while not regulated as hazardous wastes, have the potential to leach toxic metals into the environment. Incineration and landfilling are not ideal disposal methods as incineration could result in the volatilization of carcinogenic arsenic and chromium, while landfilling requires the waste to be sorted and placed in appropriately lined landfills to prevent contaminant release. Viable disposal methods must be considered to manage the significant waste stream of CCA-treated wood. This study investigates pyrolysis and electrokinetics (EK) as possible remediation techniques for the treatment of CCA-treated wood waste. A unique bench pyrolysis reactor was designed and optimized over three pyrolysis temperatures for retention of CCA metals within the char and bio-oil pyrolysis products. A batch pyrolysis system was constructed to produce large quantities of pyrolysis char for use in EK experiments. Mass balances were performed across both bench and batch pyrolysis systems, with metal distribution trends ascertained. EK experiments were performed on the char generated during batch pyrolysis as well as CCA-treated wood waste to evaluate the efficiency of coupling the two technologies. These EK studies were performed using pH regulation at the cathode and applying constant current. Distribution of the CCA metals post-EK experiment was determined and removal efficiencies were calculated. The mobility of the CCA metals was evaluated during all phases of this investigation using Toxicity Characteristic Leaching Procedure (TCLP), Synthetic Precipitation Leaching Procedure (SPLP), and Deionized Water Leaches to determine the likelihood of metal leaching under different environmental conditions.

Information protection is a key issue for the acceptance and adoption of pervasive computing systems where various portable devices such as smart phones, Personal Digital Assistants (PDAs) and laptop computers are being used to share information and to access digital resources via wireless connection to the Internet. Because these are resources constrained devices and highly mobile, changes in the environmental context or device context can affect the security of the system a great deal. A proper security mechanism must be put in place which is able to cope with changing environmental and system context. Usage CONtrol (UCON) model is the latest major enhancement of the traditional access control models which enables mutability of subject and object attributes, and continuity of control on usage of resources. In UCON, access permission decision is based on three factors: authorisations, obligations and conditions. While authorisations and obligations are requirements that must be fulfilled by the subject and the object, conditions are subject and object independent requirements that must be satisfied by the environment. As a consequence, access permission may be revoked (and the access stopped) as a result of changes in the environment regardless of whether the authorisations and obligations requirements are met. This constitutes a major shortcoming of the UCON model in pervasive computing systems which constantly strive to adapt to environmental changes so as to minimise disruptions to the user. We propose a Context-Aware and Adaptive Usage Control (CA-UCON) model which extends the traditional UCON model to enable adaptation to environmental changes in the aim of preserving continuity of access. Indeed, when the authorisation and obligations requirements are fulfilled by the subject and object, and the conditions requirements fail due to changes in the environmental or the system context, our proposed model CA-UCON triggers specific actions in order to adapt to the new situation, so as to ensure continuity of usage. We then propose an architecture of CA-UCON model, presenting its various components. In this model, we integrated the adaptation decision with usage decision architecture, the comprehensive definition of each components and reveals the functions performed by each components in the architecture are presented. We also propose a novel computational model of our CA-UCON architecture. This model is formally specified as a finite state machine. It demonstrates how the access request of the subject is handled in CA-UCON model, including detail with regards to revoking of access and actions undertaken due to context changes. The extension of the original UCON architecture can be understood from this model. The formal specification of the CA-UCON is presented utilising the Calculus of Context-aware Ambients (CCA). This mathematical notation is considered suitable for modelling mobile and context-aware systems and has been preferred over alternatives for the following reasons: (i) Mobility and Context awareness are primitive constructs in CCA; (ii) A system's properties can be formally analysed; (iii) Most importantly, CCA specifications are executable allowing early validation of system properties and accelerated development of prototypes. For evaluation of CA-UCON model, a real-world case study of a ubiquitous learning (u-learning) system is selected. We propose a CA-UCON model for the u-learning system. This model is then formalised in CCA and the resultant specification is executed and analysed using an execution environment of CCA. Finally, we investigate the enforcement approaches for CA-UCON model. We present the CA-UCON reference monitor architecture with its components. We then proceed to demonstrate three types of enforcement architectures of the CA-UCON model: centralised architecture, distributed architecture and hybrid architecture. These are discussed in detail, including the analysis of their merits and drawbacks.

Smart home systems (SHS) have become an increasingly important technology in modern life. Apart from safety, security, convenience and entertainment, they offer significant potential benefits for the elderly, disabled and others who cannot live independently. Furthermore, smart homes are environmentally friendly. SHS functionality is based on perceiving residents’ needs and desires, then offering services accordingly. In order to be smart, homes have to be equipped with sensors, actuators and intelligent devices and appliances, as well as connectivity and control mechanisms. A typical SHS comprises heterogeneous services and appliances that are designed by many different developers and which may meet for the first time in the home network. The heterogeneous nature of the systems, in addition to the dynamic environment in which they are deployed, exposes them to undesirable interactions between services, known as Feature Interaction (FI). Another reason for FI is the divergence between the policies, needs and desires of different residents. Proposed approaches to FI detection and resolution should take these different types of interaction into account. Negotiation is an effective mechanism to address FI, as conflicting features can then negotiate with each other to reach a compromise agreement. The ultimate goal of this study is to develop an Agent-Based Negotiation Approach (ABNA) to detect and resolve feature interaction in a SHS. A smart home architecture incorporating the components of the ABNA has been proposed. The backbone of the proposed approach is a hierarchy in which features are organised according to their importance in terms of their functional contribution to the overall service. Thus, features are categorised according to their priority, those which are essential for the service to function having the highest priority. An agent model of the ABNA is proposed and comprehensive definitions of its components are presented. A computational model of the system also has been proposed which is used to explain the behaviour of different components when a proposal to perform a task is raised. To clarify the system requirements and also to aid the design and implementation of its properties, a formal specification of the ABNA is presented using the mathematical notations of Calculus of Context-aware Ambient (CCA), then in order to evaluate the approach a case study is reported, involving two services within the SHS: ventilation and air conditioning. For the purpose of evaluation, the execution environment of CCA is utilised to execute and analyse the ABNA.

Wireless local area networks (WLANs) and wireless sensor networks (WSNs) technologies have been comprehensively developed and deployed during recent years. Since commercial WLAN and WSN products share the same free of license frequency band, the low power, low rate ZigBee based WSNs are vulnerable to the interference from Wi-Fi WLANs. Therefore, it is important to evaluate the performance of ZigBee WSNs that are subjected to interference generated by collocated Wi-Fi WLANs and to design effective counter-measuring techniques should performance improvement is needed. In this research, a versatile testbed for conducting various experiments is established and the ZigBee system’s performance with different clear channel assessment (CCA) modes and energy detection (ED) thresholds are evaluated through extensive experimental measurements in the testbed. It can be concluded from the results that CCA has significant impact on ZigBee’s performance. An existing theoretical analysis approach that is based on the collision time model between ZigBee and Wi-Fi packets is suitably modified to provide analytical evaluation means of the cases we examined. In order to mitigate the interference from the collocated Wi-Fi system, a novel and effective interference-aware adaptive CCA (IAACCA) scheme is proposed and implemented as firmware flashed into Crossbow motes. Experiments confirmed the ability of IAACCA to countermeasure effectively interference generated by Wi-Fi and thus improve the performance of ZigBee WSNs. Finally, a thorough statistical analysis is performed to understand the factors impacting the performance of ZigBee system and is used to further verify our experimental methods.

Wireless local area networks (WLANs) and wireless sensor networks (WSNs) technologies have been comprehensively developed and deployed during recent years. Since commercial WLAN and WSN products share the same free of license frequency band, the low power, low rate ZigBee based WSNs are vulnerable to the interference from Wi-Fi WLANs. Therefore, it is important to evaluate the performance of ZigBee WSNs that are subjected to interference generated by collocated Wi-Fi WLANs and to design effective counter-measuring techniques should performance improvement is needed. In this research, a versatile testbed for conducting various experiments is established and the ZigBee system’s performance with different clear channel assessment (CCA) modes and energy detection (ED) thresholds are evaluated through extensive experimental measurements in the testbed. It can be concluded from the results that CCA has significant impact on ZigBee’s performance. An existing theoretical analysis approach that is based on the collision time model between ZigBee and Wi-Fi packets is suitably modified to provide analytical evaluation means of the cases we examined. In order to mitigate the interference from the collocated Wi-Fi system, a novel and effective interference-aware adaptive CCA (IAACCA) scheme is proposed and implemented as firmware flashed into Crossbow motes. Experiments confirmed the ability of IAACCA to countermeasure effectively interference generated by Wi-Fi and thus improve the performance of ZigBee WSNs. Finally, a thorough statistical analysis is performed to understand the factors impacting the performance of ZigBee system and is used to further verify our experimental methods.