Urban forests can maintain urban ecological balance and improve environmental quality, but it is difficult to identify such forests accurately due to its complex and fragmented features. This study aims to develop a deep-learning network to extract the urban forest spatial distribution from high-spatial resolution image, like from Chinese Gaofen-2 (GF-2) image. Based on the GF-2 surface reflectance image and urban forest samples known in prior, this study firstly create a U-Net network to train and generate a predictive model to identify the urban forest, and then used the trained model to get the spatial distribution of urban forest in the Beibei district. Results showed that the U-Net Network can predict urban forests distribution accurately and rapidly.

Taking the Dexing Copper Mine in Jiangxi Province, China as an study area, we used the long-term sequence summer Landsat images in 2002-2019 to investigate the variation of vegetation growth status and their ecological restoration effects. According to the specific situation of the study area, the Green-Red Normalized Difference Vegetation Index (GRNDVI) calculated from remote sensing data was used to analyse the growth of mine vegetation and the dynamic change in the whole mining area. Moreover, the annual growth changes were compared with normal vegetation growth. The CV method, Hurst method, and Sen+Mann-Kendall method were combined used to evaluate the intensity of vegetation growth, change patterns and change sustainability analysis to obtain the overall growth and change of vegetation and then predict the vegetation growth trend in the study area. The results show that this method can assess accurately the vegetation growth trend in the study area.

Effectiveness of MapReduce as a big data processing framework depends on efficiencies of scale for both map and reduce phases. While most map tasks are preemptive and parallelizable, the reduce tasks typically are not easily decomposed and often become a bottleneck due to constraints of data locality and task complexity. By assuming that reduce tasks are non-parallelizable, we study offline scheduling of minimizing makespan and minimizing total completion time, respectively. Both preemptive and non-preemptive reduce tasks are considered. On makespan minimization, for preemptive version we design an algorithm and prove its optimality, for non-preemptive version we design an approximation algorithm with the worst ratio of 3/2-1/2h where h is the number of machines. On total complete time minimization, for non-preemptive version we devise an approximation algorithm with worst case ratio of 2-1/h, and for preemptive version we devise a heuristic. We confirm that our algorithms outperform state-of-art schedulers through experiments.

One type of distributed systems is the client/server system consist of clients and servers. In order to improve the performance of such a system, client assignment strategy plays an important role. There are two criteria to evaluate the load on the servers - total load and load balance. The total load increases when the load balance decreases, vice versa. It has been proved that finding the best client assignment is NP-hard. In this paper, we propose a new model for the client assignment problem and design an algorithm based on Semidefinite programming (SDP). Our method has a (relaxed) performance ratio 0.87 when only 2 servers exist. In general case, our method becomes a heuristic, and the ratio of each iteration is 0.87. We are the first one to give these bounds. Our simulation results are compared with the state-of-art client assignment method, and our strategy outperforms it in terms of running time while keeps the load in similar level.

In this paper, we present a disruption-tolerant message forwarding algorithm for optimizing the message delivery ratio of an ad-hoc network in presence of multiple unknown probabilistic influences that are cyclic in nature. Our model assumes each node moves independently from other nodes in the network, and each node's mobility behavior, although not deterministic, can be considered as random variables associated with an unknown sequence of probability distributions that follows a cyclic pattern over time. Based on those assumptions, our algorithm enables any node in the network to estimate the cycle lengths of hidden events that cause variations in each node's mobility behavior using its known mobility history as well as the history information from all peer nodes at its current landmark. Subsequently using such estimates, each node can predict other nodes' most likely trajectories in future and make optimized routing decisions based on these predictions. In this paper, we also report results from experiments that measure the effectiveness of our proposed algorithm by comparing its message delivery ratio with those of two baselines on networks of different sizes.

When a large amount of sensors are randomly deployed into a field, how can we make a sleep/activate schedule for sensors to maximize the lifetime of target coverage in the field? This is a well-known problem, called Maximum Lifetime Coverage Problem (MLCP), which has been studied extensively in the literature. It is a long-standing open problem whether MLCP has a polynomial-time constant-approximation. The best-known approximation algorithm has performance ratio 1 + ln n where n is the number of sensors in the network, which was given by Berman et. al [1]. In their work, MLCP is reduced to Minimum Weight Sensor Coverage Problem (MWSCP) which is to find the minimum total weight of sensors to cover a given area or a given set of targets with a given set of weighted sensors. In this paper, we present a polynomial-time (4 + ∈)-approximation algorithm for MWSCP and hence we obtain a polynomial-time (4 + ξ)-approximation algorithm for MLCP, where ∈ >; 0, ξ >; 0.

Directional antennas can divide the transmission range into several sectors. Thus, through switching off sectors in unnecessary directions in wireless networks, we can save bandwidth and energy consumption. In this paper, we will study a directional virtual backbone (VB) in the network where directional antennas are used. When constructing a VB, we will take routing and broadcasting into account since they are two common operations in wireless networks. Hence, we will study a VB with guaranteed routing costs, named α Minimum rOuting Cost Directional VB (α-MOC-DVB). Besides the properties of regular VBs, α-MOC-DVB also has a special constraint - for any pair of nodes, there exists at least one path all intermediate directions on which must belong to α-MOC-DVB and the number of intermediate directions on the path is smaller than α times that on the shortest path. We prove that construction of a minimum α-MOC-DVB is an NP-hard problem in a general directed graph. A heuristic algorithm is proposed and theoretical analysis is also discussed in the paper. Extensive simulations demonstrate that our α-MOC-DVB is much more efficient in the sense of VB size and routing costs compared to other VBs.

It is well-known that the application of directional antennas can help conserve bandwidth and energy consumption in wireless networks. Thus, to achieve efficiency in wireless networks, we study a special virtual backbone (VB) using directional antennas, requiring that from one node to any other node in the network, there exists at least one directional shortest path all of whose intermediate directions should belong to the VB, named as Minimum rOuting Cost Directional VB (MOC-DVB). In addition, VB has been well studied in Unit Disk Graph (UDG). However, radio wave based communications in wireless networks may be interrupted by obstacles (e.g., buildings and mountains). Thus, in this paper, we model a network as a general directed graph. We prove that construction of a minimum MOC-DVB is an NP-hard problem in a general directed graph and in term of the size of MOC-DVB, there exists an unreachable lower bound of the polynomial-time selected MOC-DVB. Therefore, we propose a distributed approximation algorithm for constructing MOC-DVB with approximation ratio of 1 + ln K + 2ln δD, where K is the number of antennas on each node and δDis the maximum direction degree in the network. Extensive simulations demonstrate that our constructed MOC-DVB is much more efficient in the sense of MOC-DVB size and routing cost compared to other VBs.

Topology control is one vital factor to a wireless network's efficiency. A Connected Dominating Set (CDS) can be a useful basis of a backbone topology construction. In this paper, a special CDS, named \alpha Minimum rOuting Cost CDS (\alpha-MOC-CDS), will be studied to improve the performance of CDS based broadcasting and routing. In this paper, we prove that construction of a minimum \alpha-MOC-CDS is NP-hard in a general graph and we propose a heuristic algorithm for construction of \alpha-MOC-CDS.

In this paper, we will study a special Connected Dominating Set (CDS) problem - between any two nodes in a network, there exists at least one shortest path, all of whose intermediate nodes should be included in a special CDS, named Minimum rOuting Cost CDS (MOC-CDS). Therefore, routing by MOC-CDS can guarantee that each routing path between any pair of nodes is also the shortest path in the network. Thus, energy consumption and delivery delay can be reduced greatly. CDS has been studied extensively in Unit Disk Graph (UDG) or Disk Graph (DG). However, nodes in networks may have different transmission ranges and some communications may be obstructed by obstacles. Therefore, we model network as a bidirectional general graph in this paper. We prove that constructing a minimum MOC-CDS in general graph is NPhard. We also prove that there does not exist a polynomial-time approximation algorithm for constructing a minimum MOCCDS with performance ratio plnδ, where p is an arbitrary positive number (p <; 1) and δ is the maximum node degree in network. We propose a distributed heuristic algorithm (called as FlagContest) for constructing MOC-CDS with performance ratio (1 - ln2) + 2lnδ. Through extensive simulations, we show that the results of FlagContest is within the upper bound proved in this paper. Simulations also demonstrate that the average length of routing paths through MOC-CDS reduces greatly compared to regular CDSs.

Using directional antennas to conserve bandwidth and energy consumption in ad hoc wireless networks has attracted much attention of the research community in recent years. However, limited research has focused on applying directional antennas to broadcasting. We devise a link reduction (LR) based broadcasting protocol for ad hoc wireless networks using directional antennas. LR outperforms most existing omnidirectional and directional broadcasting schemes in the sense that its normalized transmission cost and redundancy are significantly reduced, and consequently it is more bandwidth and energy- efficient. Based on 2-hop neighborhood information, LR relies on no location or angle-of-arrival (AOA) information. LR is a localized protocol and it achieves full delivery in ideal networks. Simulation is conducted in ideal networks where packet collision, channel contention, or node mobility is avoided.