Search

Recently, with the rapid expansion of cities, optimizing taxi driving routes for improving taxi revenue efficiency has become the core issue of taxi system. However, most current research focuses on increasing platform revenue instead of improving drivers’ revenue in a centralized dispatch taxi system just like DiDi, which results in a slower driver income growth and greater difficulties for recruiting drivers. To solve this problem, we propose a strategy of deep reinforcement learning based on driver mode. Firstly, the sequence selection process of drivers is modeled as markov decision-making process in driver mode. Then, we propose a learning scheme based on deep Q network to optimize the driver's decision-making strategy. We know that the real selection of historical taxi drivers is very helpful to the selection of current taxi drivers, so we choose the historical record of the current location as the edge data to update the edge network. Finally, we used a real data set generated by more than 1,400 taxis in Changsha. The simulation experiments show that our scheme reduced cruising time of taxis and improved the driver's income by 4-5%. The carbon emissions are obviously reduced by saving almost 6% fuel consumption, which contributes significantly to green mobility.

Concurrent transmission (CT) has been widely adopted to optimize the throughput of various data transmissions in wireless networks, such as bulk data dissemination and high-rate data collection. In CT, besides the possible data frame collision at receivers, we observe that acknowledgment frame (ACK) collision at senders can also significantly diminish concurrency opportunities. In this article, to avoid the potential ACK collision in CT, we propose ALIGNER which develops a new transmission pattern to coordinate concurrent senders in a distributed manner. The key idea is to align the silent periods of concurrent transmitters. To achieve this goal, we align the end of data frames concurrently transmitted by several senders. Therefore, the potentially arriving ACKs can avoid a collision with ongoing data transmissions because the concurrent senders are in a listening state to wait for receivers’ ACKs for a short and fixed period. ALIGNER can be applied for both deterministic and opportunistic forwarding protocols. It optionally uses a random back-off and slotted ACK mechanism to avoid a potential collision among simultaneously arrived ACKs in opportunistic forwarding. In addition, ALIGNER adopts a tailor-made metrics to analyze the throughput benefit of concurrent transmission for both deterministic and opportunistic data collection protocols. We have implemented ALIGNER in TinyOS and conducted extensive experiments on a real testbed. Experimental results show that ALIGNER can significantly increase the concurrency opportunities in both deterministic (up to 105%) and opportunistic (up to 89.7%) forwarding compared with the state-of-the-art CT methods.

Taxi services play an important role in the public transportation system of large cities. Improving taxi business efficiency is an important societal problem. Most of the recent analytical approaches on this topic only considered how to maximize the pickup chance, energy efficiency, or profit for the immediate next trip when recommending seeking routes, therefore may not be optimal for the overall profit over an extended period of time due to ignoring the destination choice of potential passengers. To tackle this issue, we propose a novel Spatial Network-based Markov Decision Process (SN-MDP) with a rolling horizon configuration to recommend better driving directions. Given a set of historical taxi records and the current status (e.g., road segment and time) of a vacant taxi, we find the best move for this taxi to maximize the profit in the near future. We propose statistical models to estimate the necessary time-variant parameters of SN-MDP from data to avoid competition between drivers. In addition, we take into account fuel cost to assess profit, rather than only income. A case study and several experimental evaluations on a real taxi dataset from a major city in China show that our proposed approach improves the profit efficiency by up to 13.7 percent and outperforms baseline methods in all the time slots.

The communication infrastructures are likely to fail, in the case of disasters like earthquakes and debris flow, resulting in blind areas and the inconvenience of residents’ communication. In this paper, we propose a novel scheme connecting these infrastructure failure areas, namely, an energy-aware device-to-device communication scheme (NEED). Our proposed scheme, taking advantage of clustering technology, connects users within the infrastructure failure areas that often have no direct access to the cellular network. Compared with the clustering used in traditional cases, we add the process of determining candidate cluster heads (CHs) before determining final CHs. Based on location and residual energy, the final CHs are selected in the candidate CHs, and dual CHs in the cluster run alternately to share the communication cost. Besides, a modified ant colony algorithm (MACA) is developed to increase routing efficiency. The simulation results show the effectiveness of our proposed NEED scheme in terms of energy consumption and energy balance, and demonstrate that the scheme significantly extends the lifetime of the whole network.

In the era of big data, practical applications in various domains continually generate large-scale time-series data . Among them, some data show significant or potential periodicity characteristics, such as meteorological and financial data . It is critical to efficiently identify the potential periodic patterns from massive time-series data and provide accurate predictions. In this paper, a Periodicity-based Parallel Time Series Prediction (PPTSP) algorithm for large-scale time-series data is proposed and implemented in the Apache Spark cloud computing environment. To effectively handle the massive historical datasets, a Time Series Data Compression and Abstraction (TSDCA) algorithm is presented, which can reduce the data scale as well as accurately extracting the characteristics. Based on this, we propose a multi-layer time series periodic pattern recognition (MTSPPR) algorithm using the Fourier Spectrum Analysis (FSA) method. In addition, a Periodicity-based Time Series Prediction (PTSP) algorithm is proposed. Data in the subsequent period are predicted based on all previous period models, in which a time attenuation factor is introduced to control the impact of different periods on the prediction results. Moreover, to improve the performance of the proposed algorithms, we propose a parallel solution on the Apache Spark platform, using the Streaming real-time computing module. To efficiently process the large-scale time-series datasets in distributed computing environments , Distributed Streams (DStreams) and Resilient Distributed Datasets (RDDs) are used to store and calculate these datasets. Logical and data dependencies of RDDs in the P-TSDCA, P-MTSPPR, and P-PTSP processes are considered, and the corresponding parallel execution solutions are conducted. Extensive experimental results show that our PPTSP algorithm has significant advantages compared with other algorithms in terms of prediction accuracy and performance.

Gathering extensive public data from the IoT devices scattered throughout a city is a significant challenge in a smart city. In this paper, a novel urban vehicle network we call the location-based urban vehicle network (LUV), is proposed to perform the non-real time data gathering task in smart cities. Different from the most works in-vehicle networks, the empirical research on the real-life 8900 private cars trace data in Changsha, China, compels us to focus on the parked vehicles and the parking places, rather than on the moving cars and the urban roads. The location-based mechanism not only provides more reliable and predictable wireless connections but also dramatically simplifies the system topology. It ensures that the vehicle network deployed on an intricate metropolitan area reaches the desired scale for gathering data.

The development of Internet of Things technology brings new opportunities for the development of edge computing. As an emerging computing model, edge computing makes full use of the equipment resources at the edge of the network and creates a new network computing system at the edge of the network. At the same time, the emergence of edge computing solves the problem of high latency in WAN which cannot be solved for a long time in the field of cloud computing, and brings users with low latency, fast response and good service experience. This article will use the edges computing architecture to construct a multi-layer data collection system. In this system model, sensors upload data to the designated edge nodes for processing, rather than remote cloud computing centers. Data collection and sample training tasks of sensor nodes in different ranges are realized through the design of multi-layer edge nodes. This system reduces the energy consumption of data uploading and the delay in network communication. As a result, it provides a better network experience for the end users. And it tries to solve the problem that the edge node in the edge system cannot satisfy multiple training task requests at the same time.

Taxi drivers always look for strategies to locate passengers quickly and therefore increase their profit margin. In reality, the passenger seeking strategies are mostly empirical and substantially vary among taxi drivers. From the history taxi data, the top performing taxi drivers can earn 25% more than the ones with mediocre seeking strategy in the same period of time. A better strategy not only helps taxi drivers earn more with less effort, but also reduce fuel consumption and carbon emissions. It is interesting to examine the influential factors in passenger seeking strategies and find algorithms to guide taxi drivers to passenger hotspots with the right timing. With the abundant availability of history taxicab traces, the existing methods of doing taxi business have been radically changed. This paper focuses on the problem of mining efficient operation strategies from a large-scale history taxi traces collected over one year. Our approach presents generic insights into the dynamics of taxicab services with the objective of maximizing the profit margins for the concerned parties. We propose important metrics, such as trip frequency, hot spots, and taxi mileage, and provide valuable insights toward more efficient operation strategies. We analyze these metrics using techniques, such as Newton's polynomial interpolation and Gamma distribution, to understand their dynamics. Our strategies use the real taxicab traces from the city of Changsha (P.R.China), may predict the taxi rides at different times by 90.68% per day, and increase the taxi drivers income levels up to 19.38% by controlling appropriate mileage per trip and following the route across more urban hot spots.

The new generation of technologies, represented by cloud computing, artificial intelligence (AI), internet of things (IoT), and mobile internet, is driving the great renovation of human society as well as the rapid development of new economy and new industries. New requirements are then emerged for the software engineering education at this new era. Hunan University (HNU) relies on outstanding platforms, such as the National Demonstration Software College, the National Featured Software Engineering Major, and the National Pilot Zone for Software Talents Education, carrying out researches on teaching mode renovation, course scheme renovation, and practice scheme renovation within a series of teaching renovation projects. As a result, the Innovative Software Talents (IST) education scheme has been established with accountable achievements and it also has been widely introduced, demonstrated, and applied among Chinese universities.

Recently, the combination of between Internet and the traditional business shows a fast development trend, and this further accelerates the fusion of IT technology and business applications. Therefore, multiple new innovative applications emerge for better serving daily life of the public. Facing this new demand and trend, our talent cultivation pattern of traditional software engineering no longer meets the current innovative needs. Therefore, it is urgent to explore the new talent cultivation pattern from the traditional software engineering to the one of internet plus pattern. This paper focuses on the new characteristics of software engineering education under Internet plus environment, and designs innovative methods for improving the talent cultivation of software engineering under the perspective of Internet plus. Furthermore, we build a new talent cultivation framework and propose an innovative talent pattern by analyzing the shortage of existing talent cultivation pattern. We cultivate more innovative software talents by this pattern, and further promote the development of innovation and entrepreneurship.

Big data applications have become increasingly popular with the appearance of cloud computing and green computing. Therefore, internet service providers (ISPs) need to build data centers for data storage and data processing under the cloud service pattern. However, data centers often consume a significant amount of energy and lead to pollutant emissions. In recent years, the high energy consumption and environmental pollution of data centers have become a pressing issue. This paper reviews the progress of energy-saving technologies in high-performance computing, energy conservation technologies for computer rooms and renewable energy applications during the construction and operation of data centers. From multiple perspectives of energy consumption, cost reduction, and environment protection, a comprehensive set of strategies are proposed to maximize data centers’ efficiency and minimize the environmental impact. This paper also provides energy-saving trends for data centers in the future.

It is crucial to provide compatible treatment schemes for a disease according to various symptoms at different stages. However, most classification methods might be ineffective in accurately classifying a disease that holds the characteristics of multiple treatment stages, various symptoms, and multi-pathogenesis. Moreover, there are limited exchanges and cooperative actions in disease diagnoses and treatments between different departments and hospitals. Thus, when new diseases occur with atypical symptoms, inexperienced doctors might have difficulty in identifying them promptly and accurately. Therefore, to maximize the utilization of the advanced medical technology of developed hospitals and the rich medical knowledge of experienced doctors, a Disease Diagnosis and Treatment Recommendation System (DDTRS) is proposed in this paper. First, to effectively identify disease symptoms more accurately, a Density-Peaked Clustering Analysis (DPCA) algorithm is introduced for disease-symptom clustering. In addition, association analyses on Disease-Diagnosis (D-D) rules and Disease-Treatment (D-T) rules are conducted by the Apriori algorithm separately. The appropriate diagnosis and treatment schemes are recommended for patients and inexperienced doctors, even if they are in a limited therapeutic environment. Moreover, to reach the goals of high performance and low latency response, we implement a parallel solution for DDTRS using the Apache Spark cloud platform. Extensive experimental results demonstrate that the proposed DDTRS realizes disease-symptom clustering effectively and derives disease treatment recommendations intelligently and accurately.

Ternary content addressable memory (TCAM) is a popular device for fast regular expression (Regex) matching in networking and security applications. The rapid growth of Regexes necessitates large TCAM memory consumption, which in turn impacts power consumption. Compressing the transition table can cut down TCAM memory consumption, thereby reducing its power consumption. This work identifies the compressibility of transition entries and then proposes a compression scheme. In our scheme, the compression ratio is improved by skillfully assigning identifiers to states in a deterministic finite automaton (DFA). Furthermore, our scheme utilizes the wildcard function and the priority matching mechanism provided in TCAM to exploit the minimum differentiation among a set of source states. A complete implementation of the identifier assignment and transition table compression is presented. Experimental results on real-world Regex sets show that our scheme is significantly more effective, reducing power consumption by 87.4% and memory space by 93.2%, and improving throughput up to 114.7% on average compared to prior work.

Taxi services play an important role in the public transportation system of large cities. Improving taxi business efficiency is an important societal problem since it could improve the income of the drivers and reduce gas emissions and fuel consumption. The recent research on seeking strategies may not be optimal for the overall revenue over an extended period of time as they ignored the important impact of passengers' destinations on future passenger seeking. To address these issues, this paper investigates how to increase the revenue efficiency (revenue per unit time) of taxi drivers, and models the passenger seeking process as a Markov Decision Process (MDP). For each one-hour time slot, we learn a different set of parameters for the MDP from data and find the best move for a vacant taxi to maximize the total revenue in that time slot. A case study and several experimental evaluations on a real dataset from a major city in China show that our proposed approach improves the revenue efficiency of inexperienced drivers by up to 15% and outperforms a baseline method in all the time slots.

Cloud service, which has changed the business computing pattern, redefines the calculation scope of distributed systems by cloud computing approaches. The existing cloud service model provides different users an advanced on-demand computing model, and then achieves resource interaction between cloud service providers and users by cloud computing. However, this model, working as a centralized service strategy of resource allocation, also generates much scattered space debris resulting in a low efficiency of resource scheduling and utilization. To address this issue, we devise an improved model of crowd sourcing service and propose a technique of storage resource allocation and scheduling, where cloud service providers, as a resource agent, only provide leasing storage space for individual applicants instead of providing storage resources directly. This technique achieves allocation and scheduling of individual storage resources dynamically, according to user's expectations and individual storage providers constraints. The experimental results based on a large data set show that the improved model reduces the storage space of cloud service providers by 88.2% and improves storage space utilization rate of individuals by 65.2%, The improved model reduces the user fees by 64.8% and the cloud services development cost by 78.6% than that of existing models, In addition, the improved model and methods provide higher profits and efficiency for cloud service providers.

Recommendation algorithms are widely used to discover interesting content for users from massive data in many fields. However, with more diversification of user requirements, the recommended accuracy and efficiency become a serious concern for improving user satisfaction degree. In this paper, we redefine the concept of content similarity by combining search words with personalized search references and describing their dimensions, then propose the calculation method of content similarity by defining the Hamming distance among current keywords, classified items and historical keywords. Through the pretreatment of support vector data description (SVDD), we may find specific tendency from the personal preference of classified items and present the final recommendation results arranged from high similarity to low one. Simulation experiments show that our proposed approach improves recommendation performance over the other two classical algorithms by an average of 17.2 % and reduces the MAE by 6.3 % on our large-scale dataset. At the same time, our proposed approach has a better performance on recall rate and coverage rate, and user satisfaction degree is also improved at higher extent.

In a secure query scheme over the encrypted cloud data, an authorized cloud user can obtain data files of interest by submitting encrypted query keywords to the cloud server, which performs a certain secure search algorithm and returns back the corresponding data file set. In practice, the returned query results may be incorrect or incomplete due to possible data corruption, software bugs, or intermediate attackers who maliciously tamper with results; moreover, the cloud server may also intentionally omit some qualified results to save computational resources and communication overhead. Thus, a well-functioning secure query system should provide the query results verification mechanism that allows the data user to verify results. In this paper, we design three varigrained and secure query results verification constructions leveraging the Bloom filter and cryptographic hash functions, for a query result set R, by which the data user can verify: (1) the correctness of each data file in R, (2) how many qualified data files are not returned by the cloud, and (3) which qualified data files are not returned by the cloud, respectively. Furthermore, our proposed verification mechanism can be very easily integrated into all secure query schemes for cloud computing. Performance evaluation shows that the proposed schemes are practical and efficient.

Web Service technologies are becoming increasingly popular for integrating systems and services. Web service composition as an important value-added function provides an application foundation for reusing Web services and automating composition. After brief analysis on the service composition concept and the eliciting process of functional requirements, a strategy of Web service composition based software life cycle is presented and the function-driven service composition process is described. This strategy may help effectively composing Web service based on software life cycle and provide a comprehensive support for SOA.