Search

This paper presents a supply-insensitive low-power quadrature local oscillation (LO) generator based on the current-reused voltage-controlled oscillator (VCO) and divide-by-two frequency divider (FD). By utilizing the mutual power-supply rejection (M-PSR) technique, a voltage regulator is embedded into the FD to provide a supply-insensitive voltage for VCO. Meanwhile, the feedback loop provides a supply-insensitive current and time constant compensation for FD. The M-PSR technique does not require extra voltage headroom, thus the proposed quadrature LO generator can work under the supply of 0.9 V. A prototype of the proposed quadrature LO generator is implemented in a 28-nm CMOS process. Measurement results indicate that the LO generator achieves a phase noise of -110.5 dBc/Hz@1MHz at the carrier of 5.9 GHz and the power consumption is 1.8 mW. This work has a competitive figure of merit in supply sensitivity or in area (FOMss or FOMa respectively) among the state-of-the-art works. Ministry of Education (MOE) Published version This work was supported in part by Singapore Ministry of Education Academic Research Fund Tier 2 under Grant MOE2019-T2-1-114, and in part by China Scholarship Council.

With the popularization of the cloud, an increasing number of users and developers need to combine multiple different services to satisfy their requirements in the cloud. Consequently, the question of how to combine these services that are published by different service providers in the cloud has become an active focus of research in service-oriented cloud systems. Service composition is the core technology of service-oriented cloud systems.

A growing number of mobile services demand intensive computation resources and high energy consumption. Traditional cloud-based task offloading incurs excessive transmission delay. With mobile edge computing, we can offload tasks to the nearby edge servers, thus mitigating long data transmission latency. Current task offloading and resource allocation approaches ignore the extra overhead brought by container instance startup. This naturally leads to a long overall latency and high energy consumption. Motivated by the limitations of existing literature, we address task offloading and resource allocation problem considering container instance startup time. We first examine the startup latency of containers started from different types of images, and present its impact on overall performance. Then, we formulate the problem as optimization problem. The problem is hard to address because mutual effects of offloading decision, resource allocation decision and container instance startup contribute to the overall performance. We resort to Gibbs sampling to decouple offloading decision and resource allocation. Finally, taking into account all potential conditions brought up by container instance startup, an optimal computation resource allocation and uplink power assignment scheme are derived through bisection, and Karush-Kuhn-Tucher (KKT) conditions. Comprehensive evaluations are conducted to verify the performance of the proposed approach, and demonstrate its superiority over current notable solutions.

This paper reports a cross-coupled pair regeneration based programmable gain amplifier (CCPR- PGA) with an improved total harmonic distortion (THD) performance. By utilizing the ping-pong technique, two identical conventional CCPR-PGAs are operating alternatively. These two CCPR-PGAs connect to the output of the overall PGA only at their hold state, as a result, the ripples generated during the processes of reset, sample and regeneration are effectively suppressed. Simulated in 28-nm CMOS technology, the proposed PGA achieves a gain range of 45.4 dB with a gain error of ± 0.3 dB. The bandwidth of the proposed PGA reaches 40 MHz with a clock frequency of 100 MHz. Moreover, the THD performance is improved by 25.8 dB at 1 MHz and 17.6 dB at 10 MHz when compared to the conventional structure. The overall PGA consumes 460 μΑ current from a supply voltage of 0.9 V.

Graph neural networks (GNN) have achieved state-of-the-art performance on various industrial tasks. However, the poor efficiency of GNN inference and frequent Out-of-Memory (OOM) problem limit the successful application of GNN on edge computing platforms. To tackle these problems, a feature decomposition approach is proposed for memory efficiency optimization of GNN inference. The proposed approach could achieve outstanding optimization on various GNN models, covering a wide range of datasets, which speeds up the inference by up to 3×. Furthermore, the proposed feature decomposition could significantly reduce the peak memory usage (up to 5× in memory efficiency improvement) and mitigate OOM problems during GNN inference.

With continuous exploitation of sub-6GHz wireless communication standards and the advent of 5G and 6G, the demand for faster speed and wider coverage keeps evolving. While it is expected to utilize higher working frequency to provide larger bandwidth, lower noise Figure (NF) is also preferred to achieve better sensitivity or wider coverage. A straightforward solution to achieve a low NF is to employ LNA-first architecture; this configuration helps to achieve a sub-2dB NF for frequencies up to 6GHz. However, the Gm-cell does consume extra power [1], [2]. Meanwhile, noise cancellation techniques are widely used to improve the noise performance [3] –[5]. Nevertheless, due to the input parasitic capacitance, the common-source/mixer-first hybrid architecture cannot guarantee a low NF for frequencies above 3GHz [3]. The 1.8dB-NF 5GHz receiver eliminates the noise contribution of its input series inductor but requires a large power consumption $(\sim 100$ mW) [4]. Similar power consumption is needed for the frequencytranslational quadrature-hybrid receiver [5].

Cognitive services have revolutionized the way we live, work and interact with the world. In recent years, deep neural networks have become the mainstream approach in cognitive service, and mobile edge computing facilitates a variety of cognitive services for users by offloading computation tasks from resource-limited mobile devices to relatively wealthy edge servers. Combining the two to provide users with a higher quality of cognitive service is an issue worth researching. However, many related studies are not easy to provide fast responses because in these systems, edge servers are only used to pre-process data, and the cloud server is used to perform tasks. In this paper, we aim to study deploying deep neural network models on edge servers to provide fast services. However, a single edge server collects only a small amount of data, which results in low inference accuracy. To address this problem, we propose a cloud and edge collaboration framework. The key idea of the proposed framework is to use a cloud model to assist in training an edge model to improve the latter's inference accuracy and enable the latter to provide fast response and high-performance cognitive service. Experimental results demonstrate the effectiveness of our proposed framework.

A wideband variable-gain amplifier (VGA) with a negative exponential generation using 40 nm CMOS technology is reported. By compensating a single-branch negative exponential generator (NEG) which features a composite of dual Taylor series, the proposed negative exponential generation further extends the dB-linear range. The measurement results show the overall VGA achieves a dB-linear range of 51 dB (-34 ~ 17dB) with a gain error less than ± 1 dB. In addition, the bandwidth is around 7 GHz under different gain settings. The core circuit draws 24.6 mA current from a 1.1 V power supply (excluding the output buffer) and occupies an active area of 0.038 mm2. Agency for Science, Technology and Research (A*STAR) Accepted version This research is supported by A*STAR under its RIE2020 Advanced Manufacturing and Engineering (AME) Industry Alignment Fund - Pre Positioning (IAF-PP) (Award A19D6a0053).

Predicting stock trend by using machining learning is a hot research issue today. However, due to the non linearity and instability of the stock data, it is still very difficult to predict the stock trend with high accuracy. In order to improve the accuracy, most researchers focus on the models selection and features construction. A variety of feature construction methods have been proposed. However, not all features constructed in those paper are equally useful. Further more, many features of significant importance may not be selected in prediction. In order to improve the accuracy of stock trend prediction, this paper will focus on the features selection problem. Most feature selection methods employed in the stock trend prediction are based on filtration methods. Wrapper methods are rarely used. Compared with filtration methods, wrapper methods have better stability and accuracy. In this paper, we propose a feature selection algorithm by extending genetic algorithm (GA). Experiments are conducted on real-world stock price data set. The experiment results show that our GA-based feature selection algorithm is better in both stability and performance.

Hierarchical Task Network (HTN) planning is an AI planning technique, which can be employed to implement service composition. Current HTN-based service composition systems fail to solve the problem comprehensively for only considering functional properties and ignoring the constraints of QoS. In this paper, we address this issue and solve the problem by exploiting the HTN planner JSHOP2. We implement an automatic service composition system by extending JSHOP2 with the consideration of both functional and non-functional properties. Furthermore, we conduct experiments on the system. Experiments show the effectiveness of our system.

Effective virtual machine placement algorithms can improve network resource utilization in cloud data centers. In order to design a more effective VM placement algorithm, we usually rely on a large-scale experimental platform to evaluate and verify its performance. However, large-scale experiment may have a huge impact on data center network, which makes it very unlikely to happen. Therefore, researchers need an experiment platform that can support large-scale environment to meet the above requirements. In order to evaluate the performance of virtual machine deployment algorithm, this paper first designs and implements a cloud data center network experiment system, WebCloudSim, which can support the joint experimental verification of real environment and simulation environment. Then, based on WebCloudSim, the paper implements three classic algorithms for virtual machine deployment, and verifies and analyzes the results. The experimental results show that WebCloudSim can effectively support the performance evaluation of virtual machine placement algorithm.

With the development of mobile Internet, mobile service is emerging one after another, and the problem of information overload is becoming ever more serious. As an important tool to alleviate information overload, mobile service recommendation has attracted more and more attention. However, traditional recommendation algorithms always recommend popular services to users, which result into a rich-get-richer problem and become a barrier for the unpopular services to startup and growth. In order to promote the healthy development of the service ecosystem, it is necessary to guarantee the fairness of unpopular services. To address this problem, this paper proposes a fairness-aware mobile service recommendation method (FMSR), which gives a relatively fair recommendation opportunity for unpopular services. FMSR makes a tradeoff between recommendation accuracy and fairness, and can recommend popular services and unpopular services respectively. For unpopular services, we design a fair efficiency function and use combinatorial optimization techniques to achieve recommendations. For popular services, bias matrix factorization is utilized to implement recommendations. Experimental results based on real-world demonstrate that FMSR significantly improve the fairness of mobile service recommendation in the evolving mobile service ecosystem.

The proliferation of real-time applications, such as online gaming, video chatting, brings unprecedented pressure for multi-cloud application providers to secure the good quality of experience. These applications are designed for cooperative group users/members/clients and are expected to have low data transmission latency due to frequent interactions. We use total data transmission latency (TDTL) to represent the group latency in this paper. However, existing approaches perform poorly since they often ignore the feature of the cooperative group scenario and cannot provide a flexible rental cost model for application providers. Minimizing TDTL is challenging due to the difficulties in 1) Finding an effective method to transmit data through multi-cloud datacenters, 2) Taking the cooperative group scenario into consideration, 3) Providing a flexible transmission rental cost model. To this end, we propose CMIP, an innovative approach that aims to minimize the TDTL by renting virtual machines of well-connected datacenters as the proxy for the cooperative group scenario in multi-cloud providers. First, a weighted graph is constructed to illustrate the multi-cloud network. Second, we define a path latency model and a rental cost model for the cooperative group. After that, Yen’s algorithm and mixed integer programming are adopted to optimize the TDTL with rental cost constraint. To study the performance of CMIP, we conduct extensive evaluations using multiple real data traces and compare it with related approaches. The comprehensive evaluation analysis shows that the CMIP achieves much lower total latency and is more flexible than other approaches.

As service computing becomes increasingly prevalent, the number of web services grows rapidly. It becomes very important to recommend suitable, personalized web services to users. Collaborative Filtering based on Quality of Service (QoS) has been widely used for service recommendation, and variety of factors such as location, environment are taken into account to improve the accuracy of recommendation. However, temporal influences, which is one of key factors affecting the QoS, are not fully considered by the investigators. In this paper, we propose a novel temporal influences-aware collaborative filtering method which designs an enhanced temporal influences-aware similarity measurement to predict QoS values. Finally, we conduct a series of experiments to evaluate the effectiveness of our method, and results show that our method outperforms other state-of-the-art methods.

As communication and sensing capabilities of mobile devices increase, mobile applications are becoming increasingly complex. The ability of computation offloading, which is one of the main features of mobile edge computing gains relevance as a technique to improve the battery lifetime of mobile devices and increase the performance of applications. In this paper, we describe the offloading system model and present an innovative architecture, called "MVR", contributing to computation offloading in mobile edge computing.

Alcohol is one of the major threats to health in United States. With the emerging of next-generation sequencing technology, the association between alcohol preference and the variants and expression of genes has been investigated. However, the roles of long non-coding RNAs (lncRNA) in alcohol preference remains unclear. In this study, we identified 37 novel lncRNAs that differentially expressed across alcohol preferring (P) and non-preferring (NP) rats. The functional study on these lncRNAs demonstrates that they are associated with gene regulation, as well as neural functions. This suggests that these lncRNAs may contribute to the alcohol preference behaviors.

This poster addresses the problem of access point ranking for cloudlet placement in edge computing environment. Access point ranking is an important step for cloudlets placement. Currently, there is no method that can rank the access point effectively for ignoring the connection features of the access points. To address the problem, we propose an adaptive and comprehensive access point ranking method by investigating the connection features of the access points. We introduce the system model and the technical detail of the proposed access point ranking method in this paper. The experimental results validate the effectiveness of our proposed access point ranking method.

The widespread utilization of cloud computing services has brought in the emergence of cloud service reliability as an important issue for both cloud providers and users. To enhance cloud service reliability and reduce the subsequent losses, the future status of virtual machines should be monitored in real time and predicted before they crash. However, most existing methods ignore the following two characteristics of actual cloud environment, and will result in bad performance of status prediction: 1. cloud environment is dynamically changing, 2. cloud environment consists of many heterogeneous physical and virtual machines. In this paper, we investigate the predictive power of collected data from cloud environment, and propose a simple yet general machine learning model StaP to predict multiple machine status. We introduce the motivation, the model development and optimization of the proposed StaP. The experimental results validated the effectiveness of the proposed StaP.

Trust model has been suggested as an effective security mechanism in distributed network environment. Considerable researches have been done on trust evaluation and trust prediction. Traditional methods take the historical behavior data into consideration to predict the trust value of the network entity. However, the context of the network entity is seldom taken into account. It is obvious that the trust of the same network entity could be different in different contexts. To improve the accuracy of trust prediction, we propose a context-aware trust prediction method in this paper. Firstly, behavioral data normalization is conducted to allow for a uniform measurement of behavior property value independent of units. Secondly, the context properties that have impact on the behavior properties are extracted by the correlation analysis. Thirdly, irrelevant historical records are filtered out based on the context similarity. Finally, we predict the trust value for the network entity based on the results of aforesaid steps. The simulation results show that the proposed method outperforms the existing method in prediction accuracy.

Personalized service recommendation becomes increasingly essential because of the growing number of services. To enhance the performance of personalized service recommendation in collaborative tagging systems, not only tag information but also time and social relations information should be considered. In this paper, we propose a hybrid method aiming at taking advantage of tag, time and users' social relations information for a preferable service recommendation. We first improve a simple tag-based recommendation method by a time-decay function. Then we develop a temporal social-based recommendation method which analyzes user familiarity and user preference similarity between friends. Based on these two steps we integrate them as a temporal tag-and social-based (TTS) recommendation algorithm. Experiment results indicate that our method outperforms general tag-based and social-based recommendation methods.

Many factors affect the time cost of Cloud computing tasks. One of the most serious factors is data transmission latency, which reduces the efficiency of Cloud computing. Existing notable schemes ignore the communication cost among virtual machines (VMs) in the MapReduce environment. In this paper, we propose a VM placement approach to reduce data transmission latency with the communication cost among VMs. We first construct bipartite graph and classify VMs as two groups according to their transmission latency with data nodes. Then we propose two VM placement optimization algorithms to minimize the total data transmission latency (TDTL) and the maximum data transmission latency (MDTL) in the MapReduce environment. Finally, we place VMs for Reduce phase. The evaluation results show that our approach reduces the average data transmission latency by 26.3% compared with other approaches.

Exiting schemes rarely take into account coordinated problem among multiple virtual machines (VMs) which collectively complete a task, since if these VMs are uncoordinated, the execution results of a task are not correct. To solve the problem, we first exploit a proactive prediction scheme to predict the VM's status. Then, if VM's status is deteriorating, coordinated checkpoint is adopted to suspend the current task to search an optimal target host. Finally, we introduce an efficient heuristic algorithm to solve the optimal target host selection problem. The experimental results demonstrate the efficiency and effectiveness of our proposed approach.

Currently, many researchers play more attention on cloud data center network and hope to verify their achievements. However, existing cloud data center simulation tools are difficult to provide an online cloud data center simulation. In this paper, we design an online cloud data center simulation system based on CloudSim and some extensions. This system support Browser/Server mode and some simple configuration can simplify the simulation process.

To maximize the profit of cloud service provider, we present a dynamic virtual resource renting method under SLA (service level agreement) constraints. According to the price distribution and current task emergency, the method attempts to adjust acceptable price of each virtual resource type at different price interval. If there is price acceptable resource, we choose to rent the most profitable one. Otherwise, if SLA permits, we even suspend the task and restart it when price falls. Partial experimental result in simulation environment is also presented.