Search

Unilateral breast reconstruction is different from bilateral breast reconstruction and breast augmentation, as it requires more consideration of the symmetry between the reconstructed breast and the contralateral breast. However, there are currently no implants specifically targeted to unilateral breast reconstruction, nor is there is objective classification criteria for breast appearance type. A total of 212 breasts from 153 patients were measured. Breast measurements, breast surface curvature distributions, and elliptic Fourier coefficients were used to describe breast features. We objectively classified the appearance of breasts using machine learning and used the resulting cluster centers to construct the most adaptive implants for each type of breast. All of the breasts clustered into 4 types. The implants corresponding to each type of breast were constructed using cluster centers. The resulting cluster centers were then used to choose a suitable implant for patients requiring unilateral breast reconstruction. Contour coefficients were used to evaluate the clustering results, with an average score of 0.53. The two breasts that develop normally in the same person were treated as the same class. The score obtained after statistical classification was 0.47. These results demonstrate that our proposed method can improve the classification of breasts of different shapes. This method provides a foundation for improving the symmetry of unilateral breast reconstruction and the construction and selection of implants.

Memory design continues to scale to smaller area and higher density to meet the industry requirements for high-speed memory performance. The aggressive scaling leads to lithographic difficulties for patterning resolution, 2D CD control and process window. Double patterning technology (DPT) can be employed to overcome lithographic and traditional OPC limitations for low K1 values, but the cost can be high. Inverse Lithography Technology (ILT) has also been demonstrated as capable of extending the use of single patterning to lower K1 cases while achieving acceptable lithographic quality and cost for advance memory semiconductor designs. This paper will present results of single patterning ILT for challenging region of DRAM contact layer. ILT derived rule-based assist features are used for optimal process window. The ILT single patterning solution was developed with process window and sidelobe aware costs functions, which was needed to achieve the lithographic QOR for single patterning. For mask processing turn around time (TAT) and consistency, a pattern-based hierarchy extraction technique was used to generate cell repetition to achieve high compression processing of a full chip flat layout. Along with QOR and TAT data, wafer and simulation results will be shown for single-patterning ILT.

A PWM high-fidelity audio playback system including only a general-purpose embedded microprocessor, switching power transistors and output filter has been designed and measured. The design uses digital signal processing and class D amplifier to amplify audio signals with high efficiency and low distortion. Signal processing uses sample rate conversion, noise shaping and equalization to convert high-fidelity PCM audio input to PWM output. A software approach that can verify the quality of the output before hardware implementation is discussed. Results measured on a prototype board show this approach delivers 200W into a 4 ohm loudspeaker with S3% efficiency, 96 dB SNR and 0.02% THD.

Unmanned aerial vehicle (UAV) provides an alternative way to collect data conveniently, when corresponding sensor (s) is (are) equipped onboard. In this paper, two multispectral cameras were compared mainly based on simulated channel reflectance, which covered visible (i.e., Blue, Green, and Red), near infrared (NIR), and red-edge (RE) regions. The preliminary investigation shows the between-sensor differences in sensor settings, the related differences in channel reflectance, and in the derived vegetation indices. The between-sensor differences in channel reflectance vary with channels. Compared with the Green and NIR channels, the RE and Red channels show more obvious between-sensor difference, which contributes significantly to the differences in the derived vegetation indices. The between-sensor differences in the RE channel are required to be eliminated, because the RE reflectance and the derived indices correspondingly are important in modeling biophysical parameters. Case study shows that the between-sensor differences of vegetation indices were generally more obvious than the reflectance differences of individual channels. Damage assessment for plant over flooded areas was readily done with the help of vegetation indices. Accordingly, the between-sensor differences are likely significant in practical applications. In conclusion, the importance to make consistency among data collections by different sensors (i.e., onboard UAV) is highlighted, and an effective way for between-sensor transformation is required in further investigations.

In this paper, we propose and investigate a low- complexity iterative transmit precoding (TPC) algorithm to enhance the bit error rate (BER) performance for Spatial Modulation (SM) systems over flat-fading multiple-input multiple-output (MIMO) channels. In literature, various TPC techniques have been conceived, but suffer from either over-simplification or lack of generality, thus a low-complexity iterative TPC algorithm is proposed, in which the TPC matrix is diagonal in order to retain the low-complexity benefit of conventional SM systems. Specifically, we investigate a TPC design to maximize the minimum Euclidean distance dmin (max-dmin) between the SM signal points at the receiver side. Different from conventional TPC algorithms, which adjust only two elements in the TPC matrix, all elements of the TPC matrix are iteratively modified until dmin could not be increased. Furthermore, an efficient computing method of dmin is proposed when considering PSK-modulated SM systems. Discussion on complexity analysis and convergence behavior are also provided, which proves the low complexity of the proposed algorithm. Simulation results demonstrate that the proposed TPC algorithm is able to achieve lower BER compared with conventional TPC algorithms.

To tackle the problem of character segmentation for the palm leaf manuscripts, a character segmentation algorithm is proposed in this paper. The whole process of the algorithm consists of two steps, interleaving character segmentation and touching character segmentation. In the first step, the improved Viterbi algorithm and some path filtering rules are used to perform path detecting and path filtering respectively. In the second step, the Concave spot detecting method based segmentation algorithm is proposed to solve the touching character segmentation problem. The experiments in this work base on a text line database obtained by our previous works and the experimental results prove the effectiveness of the proposed algorithm in the character segmentation for the palm leaf manuscripts.

Due to the limited modulation bandwidth of light emitting diodes (LEDs), the imaging optical multiple-input multiple-output (MIMO) technology is applied in visible light communications (VLC) for high data rate transmission. As an imaging lens with a wide angle/field-of-view (FOV) is favorable in the imaging optical MIMO VLC system, a fisheye lens can be utilized to concentrate the light. To avoid inter-user interference among indoor users and satisfy their target bit error rate (BER) requirements with the minimum number of LEDs simultaneously, an interference-free LED allocation scheme for the fisheye lens based imaging optical MIMO VLC system is investigated in this paper. Since the formulated problem is NP-hard, a location-based greedy algorithm is proposed, where the LEDs are allocated to the users sequentially based on their distances to the center of the LED array. Simulation results verify that there exists no interference among all users while the target BER requirements for all users are satisfied with our proposed algorithm.

A low-density spatial downlink reference signal (LDS-RS) design is proposed for frequency-division duplex (FDD) massive full-dimensional multiple-input multiple-output (FD-MIMO) systems. By exploiting the spatial correlation between the channels of different antennas, this scheme can efficiently reduce the downlink RS overhead and therefore enhances the achievable spectral efficiency significantly. Different from classic RS design methods which utilize the correlation of time/frequency domain and arrange scattered RSs in time/frequency resource block, in our design, RSs are transmitted only on part of antennas in low density. Furthermore, the corresponding channel estimation scheme is also investigated for FDD massive FD-MIMO systems utilizing the proposed LDS-RS design. Simulation results have confirmed that the proposed scheme offers an improved spectral efficiency performance when considering the training overhead in FDD systems.

Many emerging motion-related applications, such as virtual reality, decision making, and health monitoring, demand reliability and quick response upon input changes. Motion capture has been a well-researched topic in the past decades with applications in many industries. The ability to capture motion goes hand in hand with real-time capability in a system. This paper gives an overview on real-time motion capture, including frameworks, recording devices, and algorithms. Using real-time motion capture as a tool to solve application problems is also described.

Nowadays, with the development of micro-electro- mechanical technologies, sweep coverage are more and more popular in wireless sensor networks, which is also applied widely in other scenarios, such as message ferrying and data routing in the ad-hoc network. In order to reduce the sweep cycle and the number of required mobile sensors, we propose the \emph{Distance-Sensitive-Route-Scheduling} (DSRS) problem, which is the first to consider the effect of sensing range. We prove that DSRS is NP-complete, and consider two different scenarios: the single kissing-point case and the general case. The former case requires a mobile sensor to change its moving direction after visiting a target. Correspondingly, we propose an approximation ROSE to schedule the routes of mobile sensors efficiently. For the latter general case, we present another approximation G- ROSE based on ROSE. We further characterize the non- locality property and design a distributed sweep algorithm D-ROSE, cooperating sensors to guarantee the required sweep requirements with the best effort. Our algorithms is scalable to different sweep coverage problems involving route schedules. We compare our algorithms with several previous algorithms, and the simulation results show that our algorithms greatly outperform other works especially with large sensing range, which can be improved up to $45\%$.

Sweep coverage plays an important role in many applications like data gathering, sensing coverage, and devices control, etc. In this paper, we deal with sweep coverage problem with multiple mobile sensors to periodically cover n targets in the surveillance region. We propose three constant-factor approximations, namely CycleSplit, CoCycle, and AugPrim, to minimize the longest trajectory length of mobile sensors under different scenarios respectively. CycleSplit deals with non-cooperative sweep coverage problem (NCSC), in which each mobile sensor works independently along different trajectory cycle. It has an approximation ratio of (5-\frac{2}{n-m+1}), which improves the best known approximation ratio of 5. CoCycle solves the cooperative sweep coverage problem (CSC), in which multiple sensors work together to cover $n$ targets. Its approximation ratio is 4, which outperforms the state-of-art best approximation ratio of 6. AugPrim is also a 6-approximation. It computes the sensor route for multi-sink sweep coverage problem (MSSC), where each sensor has to transmit the gathered data to one of the given sinks in each detection period. It implements the modified Prim algorithm and augment paths in turn according to the sink positions. AugPrim is the first approximation for sweep coverage problem with sinks. Finally, we provide various numerical experiments and comparisons with several previous work to validate the efficiency of our design.

In this paper, by exploiting the special features of temporal correlations of dynamic sparse channels that path delays change slowly over time but path gains evolve faster, we propose the structured matching pursuit (SMP) algorithm to realize the reconstruction of dynamic sparse channels. Specifically, the SMP algorithm divides the path delays of dynamic sparse channels into two different parts to be considered separately, i.e., the common channel taps and the dynamic channel taps. Based on this separation, the proposed SMP algorithm simultaneously detects the common channel taps of dynamic sparse channels in all time slots at first, and then tracks the dynamic channel taps in each single time slot individually. Theoretical analysis of the proposed SMP algorithm provides a guarantee that the common channel taps can be successfully detected with a high probability, and the reconstruction distortion of dynamic sparse channels is linearly upper bounded by the noise power. Simulation results demonstrate that the proposed SMP algorithm has excellent reconstruction performance with competitive computational complexity compared with conventional reconstruction algorithms., Comment: Conference: 2015 IEEE Global Communications Conference: Wireless Communications - Wireless Communications

In order to decrease the sweep cycle and the number of mobile sensors required, we propose a route scheduling problem in this paper which is the first to consider the effect of sensing range. We prove that the Distance-Sensitive-Route Scheduling(DSRS) problem is NP-hard, and consider two different scenarios: the single kissing-point case and the general case. For different cases, We propose three corresponding approximation algorithms ROSE, G-ROSE, D-ROSE.

This paper considers the problem of tracking a dynamic sparse channel in a broadband wireless communication system. A probabilistic signal model is firstly proposed to describe the special features of temporal correlations of dynamic sparse channels: path delays change slowly over time, while path gains evolve faster. Based on such temporal correlations, we then propose the differential orthogonal matching pursuit (D-OMP) algorithm to track a dynamic sparse channel in a sequential way by updating the small channel variation over time. Compared with other channel tracking algorithms, simulation results demonstrate that the proposed D-OMP algorithm can track dynamic sparse channels faster with improved accuracy., Comment: Conference: Milcom 2015 Track 1 - Waveforms and Signal Processing - IEEE Military Communications Conference 2015

In wireless sensor networks (WSNs), there is no predefined infrastructure. Nodes need to frequently flood messages to discover routes, which badly decreases the network performance. To overcome such drawbacks, WSNs are often grouped into several disjointed clusters, each with a representative cluster head (CH) in charge of the routing process. In order to further improve the efficiency of WSNs, it is crucial to find a cluster partition with minimum number of clusters and the distance between each node to its corresponding CH can be bounded by a constant number of hops. Finding such a partition is defined as minimum $d$-hop cluster head set ($d$-MCHS) problem, which is proved to be NP-hard. In this paper, we propose a distributed approximation algorithm, named $d^2$-Cluster, to address $d$-MCHS problem and prove that the approximation ratio of $d^2$-Cluster under unit disk graph (UDG) is a constant factor $\lambda$ which is related to $d$. To the best of our knowledge, it is the first constant approximation ratio for $d$-MDS problem in UDG.

This paper addresses the problem that how to estimate the channel practically for the large-scale multiple antenna systems. Based on the uniform linear array, we reformulate the signal model for channel estimation into a transform-domain (TD) representation. Then the TD filtering schemes are investigated. In particular, we propose a sequence-based filtering scheme, which is asymptotically optimal in the sense of noise cancelation as the antenna number grows large. The channel estimation scheme here we proposed has much lower complexity than the traditional minimum mean square error (MMSE) estimator and does not depend on channel statistical information. Simulation results validate the effectiveness of the proposed scheme.

Recently, compressive sensing (CS) algorithms have been adopted to improve the accuracy of channel estimation for wide-band wireless communication systems. Since the wireless channel is usually time and frequency doubly selective fading, the performance of CS based channel estimation is limited by the estimation accuracy of the channel's sparsity level. In this paper, an adaptive subspace pursuit (ASP) algorithm is proposed which can search the sparsity level and update channel estimation result dynamically. The advantage of the proposed ASP method is that it does not need sparsity level as a priori information. Simulation results demonstrate that the proposed method outperforms conventional methods and the latest CS based methods.

To deploy Map-Reduce as a data processing service over cloud computing, we must provide necessary security mechanisms to protect customers confidential data processed. In this paper, we present Map-Reduce based framework which provides strong security and privacy guarantees for distributed computations on sensitive data. The framework is a novel integration of access control via attribute-based encryption, and privacy-preserving aggregate computation via homomorphic encryption. Data providers control the security policy for their sensitive data. Users without security expertise can perform computations on the data, but the framework confines these computations, preventing information leakage beyond the data provider's policy. Our prototype implementation demonstrates the flexibility of the framework on several case studies. It was proved more efficient than fully homomorphic encryption.

Recently, compressive sensing (CS) methods have been considered to improve the accuracy of channel estimation. Since the accurate estimation of the channel relies on the size of the observations, the fixed size of the observations is usually the trade-off between the channel estimation accuracy and the length of the channel delay to handle. In this paper, an adaptive compressive sensing (A-CS) algorithm is proposed to obtain the entire inter-block-interference (IBI)-free observations which can be detected dynamically. The A-CS method can utilize the entire IBI-free observations effectively. Simulation results demonstrate that the proposed method outperforms the state of art solutions whose performances are limited by the fixed size of the IBI-free region.

This paper proposes a lightweight routing protocol which can satisfy the city taxi-call application. The proposed routing protocol predicts and updates the routing information and uses the greedy algorithm based on hop distance to find routes to destination quickly. The proposed routing protocol performs well in taxi-call applications. We do comprehensive simulations to evaluate the proposed protocol by varying the node number and node velocity. The result proves its efficiency.

As the go as the vigorous development of Chinese nuclear power industry, nuclear power technology is increasingly mature. During Nuclear power plant construction preparation period, a large number of production materials need to be standardized and need professional management, which are still needing to be solved. This paper uses ECP system (take equipment management as the center of production materials platform, the taking Equipment management as the Center of production and goods Platform) to introduce a new management and utilization modes of modern production materials.

Proportional fair (PF) scheduling algorithm has been extensively utilized in many different systems, since it can support both multi-user diversity and system fairness guarantee. Brute-force searching is the optimal scheduling, pairing and ordering algorithm. However, the optimal brute-force searching method is too computationally prohibitive, especially in case of the large number of users. This paper proposes a novel 2-dimensional grid based scheduling scheme for uplink MU-MIMO transmission using either the conventional proportional fairness metric or delta scheduling metric. Comparing to the scheduling approach using the conventional priority based pairing metric, the proposed scheduling algorithm based on delta scheduling metric could obtain approximate 7\% gain for both cell and cell-edge throughput gain with neglectful computational complexity increase. With the aid of successive interference cancellation, MU-MIMO with MMSE-SIC receiver could also improve the cell-edge throughput compared to SU-MIMO. The advantage of the proposed scheme is that it can improve the throughput performance without the sacrifice of the user fairness guarantee.

In this paper, we investigate linear precoding and scheduling algorithms for downlink multi-user multiple input multiple output (MU-MIMO) systems. In order to eliminate the multi-user interference and improve the bit error rate (BER) system performance, block diagonalization precoding could be combined with geometric mean decomposition (BD-GMD) scheme. We propose GMD linear precoding approaches utilizing oblique projection (OP-GMD). This paper also derives system capacity in closed-form through applying water-filling power allocation algorithm into the system model. Our simulation results show that the proposed scheduling algorithms perform better with the increase of the number of transmit antennas. In terms of system capacity, the proposed OP-GMD algorithms significantly outperform the precoding scheme based on BD-GMD at low SNRs.

Long period fiber gratings (LPFGs) are fabricated in standard single mode fibers with femtosecond laser. The position of optical fiber refractive index modification is controlled exactly. It finds that the depth of resonant rejection bands rely on the position of optical fiber refractive index modification. Optical fiber refractive index modification is in proper scope, the resonant rejection bands are strong. Changing the duty cycle of LPFGs, the background losses and the harmonic resonance peaks can be sustained. However, when the duty cycle is reduced to a certain extent, although there is a periodic perturbation of the properties of the optical fiber, the attenuation of LPFG is very weak.

Data I/O has become the bottleneck of distributed network servers. This paper proposes a block Correlations Directed Optimizing System (CDOS), which provides high ability of mining block correlations and optimizes the I/O performance based on the mining results, to improve the I/O performance of distributed servers. CDOS proposes a novel block correlations mining method based on noise filter and uses block correlations to optimize the policy of prefetching, cache replacement and data layout on the I/O path of the distributed systems. For most real system workloads, CDOS can reduce the overhead of mining by one or two orders compared with traditional approach and decrease the average of system response time by 50%~79%.

The paper presents a novel network RAID storage system based on the out-of-band virtualization architecture and the backend centralized redundant management. The application servers can inquire the mapping information of virtual disk from the out-of-band virtualization server and directly access the storage nodes. The read request can fetch the data from special storage node, while the write request is not only stored into the storage node, and also mirrored into the redundant server by the storage node. The redundant server can cache updated data in local disks with log-structured mode, and calculate the parity of RAID5 in the background process when the system is idle. It relieves the bottle problem of I/O performance and low reliability danger of single controller in the front-end centralized management system. And the layout of RAID1/RAID5 on the data block has acquired the trade-off among performance, reliability and cost.

Net ecosystem exchange (NEE) is a critical ecological parameter indicating the exchange of carbon dioxide between vegetation and atmosphere, which is widely used in the field of carbon cycle researches. It is common measurement by flux tower based on eddy covariance technique can only represent local status, however regional NEE is much more important. This paper introduces a spatialization method of NEE based on artificial neural network (ANN). 14 input nodes are selected purposefully including meteorological variables, ecological variables, land cover variables and seasonal variables. A feed-forward back propagation neural network is trained by 92 measured samples. Validation results show that ANN is a satisfactory method for the spatialization of NEE-like ecological parameters.

In this paper, a new scheme of deadlock-free routing in wormhole mesh/torus interconnection networks, path-division routing (PDR), is proposed. PDR is designed to provide high routing flexibility and diversity without adopting a deadlock-recovery scheme, thus its implementation is very simple because no deadlock detection mechanism is needed. Two design methods of PDR algorithms, namely the division method and the integration method, are also presented. An implementation example of PDR is also given, and from simulation results, its performance is shown to be satisfying when compared to several well-known routing algorithms adopting traditional deadlock avoidance or deadlock recovery scheme.

IPv6, the successor of IPv4, has gotten a lot of investigation from the academic and research community as well as the manufacturers. Since it is coming near to commercialization, the deployment of IPv6 is becoming more and more urgent. One of the most appealing features in IPv6, stateless auto-configuration mechanism, can make the IPv6 terminal true "plug & play" and also provide convenience for node's mobility. However, this mechanism is originally designed for open friendly computer network environments, it cannot be directly applied in telecommunication access networks. This paper investigates how to deploy of IPv6 stateless auto-configuration in access network. Firstly, the architecture of the future IPv6 access network is described, and the potential problems are identified. Then, stateless auto-configurations in different network models are carefully investigated, including proposing some effective solutions to deploying IPv6 stateless auto-configuration mechanism in them. In addition, the issues related to the stateless auto-configuration mechanism (e.g., network service selection & authentication and multiple prefixes selection in ISP/multi-homing environments) are also discussed. I. INTRODUCTION With the rapid development of Internet, the network architecture is required to evolve to support the fast growing numbers of users, applications, and services. IPv6, which is designed to meet these requirements, provides such a vast number of available addresses that nearly every electrical device/terminal in the world could be connected

Bandwidth demand in the Internet is growing at a very fast rate. While links can provide the requested bandwidth, the situation is much more critical for routers. The design of a high-speed switch fabric is critical in current routers and switches. The paper presents a scalable and high speed switching fabric for Internet routers - the torus network, which before was mainly used in high performance computers. Using wormhole routing and a virtual channel in the torus fabric can achieve a high speed and scalable switching fabric for Internet routers. The performance of the torus network is analyzed as a switching fabric core for Internet routers. Also, an analysis model has been achieved in OPNET simulation tools. Throughput and latency characteristic can be analyzed through this analysis model. It was found that, by using this fabric, a tera-bit throughput switching fabric can be realised.

A k-ary n-cube direct interconnection structure is originally introduced and developed in the context of massively parallel computer (MPC) architectures. In recent years, with the development of VLSI (very large scale integration) technology and requirements of high performance Internet routers/switches, the k-ary n-cube interconnection network is considered to build Terabit routers. Analysis model for three-dimensional torus is presented as switch cores of Internet routes. Delay/throughput characteristics are measured with four different routing algorithms, buffer length and scale of fabric. Also a new algorithm Ladder is presented. All simulation experiments are done on OPNET. Results show that this architecture can build high performance routers. Compared to other routing algorithms, Ladder has better performance.

A k-ary n-cube direct interconnection structure is originally introduced and developed in the context of massively parallel computer (MPC) architectures. In recent years, with the development of VLSI (very large scale integration) technology and requirements of high performance Internet routers/switches, the k-ary n-cube direct interconnection network is considered to build Terabit routers. In this paper, the analysis model for three-dimensional torus as switching fabric of Internet routers is presented. Delay/throughput characteristics are measured in terms of four different routing algorithms, buffer length and the size of fabrics. Also a new algorithm ladder is proposed. Simulations have been done on OPNET. Numeric results show that this architecture can be used to build high performance routers. Comparing with other routing algorithms, the ladder has better performance.