Search

This paper reports defect detections of polymers based on bi-material microcantilever focal plane array (FPA) at 3.1 terahertz (THz). The FPA was realized by integrating the metamaterial absorbers (MAs) into a bi-material microcantilever, which thermomechanical sensitivity was measured to be 3.86 µm/K. An imaging system was then established by a parabolic mirror group for THz beam collimation and expansion, which can decrease the THz energy loss by about 44% compared with traditional lenses. The proposed system together with the fabricated FPA realizes defect detections of different polymers at room temperature with a resolution of 1mm.

With the rapid development of online courses, the requirements of personalized course recommendation have been increasing. The traditional collaborative filtering algorithm confronts with the challenge of cold start, which is difficult to settle on online course recommendation effectively. In this paper, we propose a novel click through rate (CTR) model for personalized online course recommendation, with discriminative user features, item features and cross features. The feature representation ability of the CTR model is improved and the serious challenge of cold start is alleviated. Furthermore, transfer learning is introduced to deal with the problem of insufficient data in models training. More specially, eye tracking technology is applied to capture the users’ cognitive styles, which are visualized with the heat map and fixation point trajectory. Finally, the recommendation interface sent to the learners, according to the user’s cognitive style. The experiments show that the novel CTR model improves the performance of the personalized online course recommendation.

During the period of COVID-19 epidemic situation, in order to ensure that students can study normally, the Ministry of Education has issued measures to “suspend classes without stopping learning”. Therefore, the traditional forms can’t meet the needs of learners, more and more people like online education. The existing online learning platforms cannot perfectly meet the learning needs of students, so the students need an online learning system that can meet their individual needs and improve learning efficiency. This article introduces an intelligent learning companion system, the system based on the advantages of WeChat Mini Programs can help learners achieve online learning and meet personalized needs.

In recent years, vocational education has attracted more and more attention. Teaching evaluation, as a test of teaching effect, is one of the vital factors to promote the development of vocational education. Therefore, it is very necessary to sort out the current teaching evaluation in the field of vocational education in China. In this paper, we will use literature metrology method and CiteSpace to analyze the data from China National Knowledge Infrastructure (CNKI) from 2010 to 2019. The development and hot spots of teaching evaluation in vocational education field are excavated by the visualizations on knowledge maps, which include the key words co-occurrence map, the author cooperation map and the organization cooperation map. Then, based on the current situation of online teaching, this paper analyzes the teaching evaluation from the micro aspect. Finally, the development of teaching evaluation in the future is predicted.

Vocational education is an important part in modern education and its teaching methods appear distinguishing characters. In recent years, the research on teaching models in vocational education has attracted great attention and obtained significant achievements. In this paper, the knowledge mapping software of CiteSpace is introduced to bibliometric analyze and study the hotspots in vocational education based on the data from the China National Knowledge Infrastructure (CNKI). With the visualizations on knowledge maps including institutional cooperation, author cooperation, keyword co-occurrence and time zone map, the development veins of vocational education is explored, the frontiers and research focus on vocational education are achieved. Finally, the research results are applied to practical teaching mode in the course of Python Programming. The results of the teaching practice show that studying passion is promoted and the effect of teaching are improved in the class with novel teaching mode.

In this work, we propose an ultrasensitive detection of aflatoxin B1 using a piezoresistive microcantilever biosensor, which is composed of doped piezoresistive layer, top polyimide (PI) passive layer and bottom silicon oxide (SiO 2 ) passive layer. Four microcantilevers with the same dimensions consist of a wheatstone bridge to achieve a higher sensitivity and better noise performance. The fabricated PI/Si/SiO 2 microcantilever shows a steady output with less than 1 μV@3V fluctuation in PBS buffer, while the spring constant is 21.307 nN/μm and the deflection sensitivity is 3.54×10-7 nm-1. Detections on aflatoxin B1 with a minimum concentration of 1 ng/mL were achieved.

The rapid emergence of intelligent interaction technology has sparked a great deal of research interest in the field of contemporary education. Teaching students in accordance with their aptitude and satisfying the individualized development of students have become the core of individualistic education. There are a large number of educational information resources which increase the difficulty of selection for learners and further bring the problem of "information trek". As an effective tool to address those challenges, the research of personalized recommendation system has become a hotspot in the field of educational services. In this paper, we first classify and summarize the recommendation algorithms which are related to the personalized educational recommendation system. Then we conduct a detailed analysis on the advantages and disadvantages of existing algorithms. Finally, some evaluation indicators are outlined to analyze the performance of current personalized recommendation systems thoroughly.

Although the traditional silicon-based microcantilever working with intrinsic piezoresistive effect has high sensitivity, it is easier to be broken in practical application because of its high elastic modulus. In this paper, we develop a piezoresistive microcantilever with a flexible polymide (PI) layer, whose Young’s modulus is only about one-tenth of silicon oxide, as the top insulator of the single-crystalline silicon piezoresistor. The polymide-based piezoresistive microcantilever was fabricated on a SOI wafer and the experimental output voltage fluctuation of the fabricated microcantilevers in the phosphate buffered saline (PBS) solution is within 2μV at 3V bias voltage supply. Finally, a sensitive detection of abrin toxin was realized with a limit of detection of 30ng/mL (0.01M) by modifying the anti-abrin PcAb on the top gold surface of the microcantilever. Due to the fact that the fabrication process is compatible with CMOS, it provides a promising possibility to monolithically integrate the piezoresistive microcantilever biosensor with SOI CMOS circuits.

We characterized the terahertz emission from a thin metal film with quasi-randomly distributed porous nanostructures. Unlike many demonstrated nanostructured metal terahertz sources, the emission is in the forward direction when the laser excitation illuminates the metal film at normal incidence, making its integration in a time-domain terahertz system straightforward. We attribute the generation mechanism to optical rectification.

This work introduces a flexible piezoresistive microcantilever by using polyimide (PI) and silicon oxide (SiO 2 ) as the insulation layers of titanium (Ti) strain resistors. The microcantilever is elaborately devised including the strain and piezoresistive characteristics of Ti and the film formation and patterned process of the PI. The fabricated PI/Ti/SiO 2 microcantilever is steady and flexible with less than 3 μV@3V output voltage fluctuation in PBS buffer environment and a spring constant of 21.28 nN/μm. An aptasensor is developed through functionalizing the microcantilevers, which is able to detect ricin of 10 ng/mL concentration with good specificity. All the experiment results show that the PI/Ti/SiO 2 microcantilever-based aptasensor enables label free and high sensitivity chemical/biological molecules detection and analysis.

In this paper, a three-layer cascaded complementary metasurfaces for complete terahertz (THz) phase control was designed at 3.11 THz. On this basis, a transmissive lens was fabricated and experimentally demonstrated with a focal length of 23 mm and a full width at half maximum (FWHM) of 1.1 mm at the focal plane. Simulations and calculations reveal that the lens is insensitive to incident angle and polarization direction, which is favorable for the development of compact THz imaging and spectroscopy systems.

This paper reports an ultrathin flat parabolic reflector to focus a laser beam of 1.47μm wavelength based on gradient metasurface. The metasurface consists of an array of sub-wavelength cross resonators fabricated on a continuous gold film spaced by a layer of SiO2. The cross resonators are arranged in a radially periodic manner following a hyperboloidal phase distribution, which is controlled by varying their arm lengths and widths. Transverse profiles of the laser spot were captured by an IR detector along the propagating direction, showing the minimum spot size to be 167μm at focal point with the focal length of 5.8 cm and the overall reflection efficiency of 44%. The symmetric design in the resonators also renders its performance of polarization independence to the incident light, which was also experimentally verified.

As an important branch in the field of frequent pattern mining, approximate frequent pattern(AFP) mining attracts much attention recently. Various algorithms have been proposed to discover long true AFPs in presence of random noise. This paper considers the key issues of AFP mining in noisy databases, and categorizes the previous approaches according to the ways they cope with missing items in the transactions. And then a study of different data models on AFP is presented, in which the merits and defects are analyzed. Finally, we draw a conclusion and propose some solutions to deal with the problems in the field of AFP mining.

Nowadays, mining approximate frequent itemsets from noisy data has attracted much attention in real applications. However, there is not widely accepted algorithm at present to solve the problem under noisy databases, which dues to two key issues. Firstly, the anti-monotonicity property does not hold which is used to prune candidate itemsets efficiently. And secondly, the computation of support counting turns out to be NP-hard. In this paper, we propose a novel model which is based on rough set theory and capable to recover the noise-tolerant frequent itemsets from "reduced itemsets". The novel model applies depth-first growing method to generate candidate itemsets and exerts effective pruning strategies, which narrows the searching space and mines indeed meaningful noise-tolerant frequent itemsets efficiently.

This paper reports ultrasensitive detections of DNA and protein by multi-silicon-nanowire field-effect-transistor (SiNW-FET) biosensors. Single-SiNW-FET, double-SiNW-FET and quad-SiNW-FET have been designed on a single chip and fabricated by a simple top-down fabrication process. Compared with single-SiNW-FET, the multi-SiNW-FET biosensor shows a higher ability of decreasing the noise by averaging the source-drain currents. Both Norovirus DNA and IgG have been detected with the sensors at concentrations as low as 1 fM and 10 fM respectively. The sensor also shown high specificity in 1bp mismatch DNA disturbance.

Nowadays, frequent item set mining is the major task in association rule mining. With the observation that the support plays an important role in mining frequent item sets, in this paper, we review the previous efficient algorithms and study the effect of different order of support in the performance of frequent item sets mining algorithms and propose our improved schedule. In our new algorithm, items are sorted in descending order according to the frequencies in transaction cache while item sets use ascending order of support during support count. Compared with other algorithms, the results of experiments show that the new algorithm gains better performance on on well-known benchmark data sets.

Personalized recommendation is one of the most effective methods to solve the problem of information overloading. As many real existing systems in nature, a recommendation system can also be considered as a complex network system, so we can do personalized recommendation by using the link prediction method which is a new one in complex networks research area. In this paper, we present personalized recommendation method based on the link prediction in Super-networks. Firstly, we give several definitions such as a Super-network, a dynamic Super-network and a utility matrix etc. Secondly, we construct a personalized recommendation model based on these definitions. Thirdly, we define a similarity metric for users and some similarity criteria, put forward five link prediction related algorithms in dynamic Supernetworks and present our recommendation algorithms based on these link prediction algorithms. Finally, we apply our methods to classic datasets in order to evaluate the performance of our algorithms.

This paper presents the design, fabrication and performance of a 256×256 bimaterial cantilever focal plane array (FPA) which is able to work in the three infrared (IR) atmospheric windows of 1~2.5μm, 3~5μm and 8~14μm simultaneously. The FPA employs a silicon-framed structure by selectively etching away the substrate with Deep Reactive Ion Etching technique, and a stacked layer of chromium and SiNx serves as the multi-band absorber. The images of short wavelength, middle wavelength and long wavelength infrared were captured successfully with the same FPA by combining the Chromium nano-film with silicon nitride as the multi-band IR absorber. The measured sensitivity of the FPA is 0.18μm/K.

We design, fabricate and characterize dual-band terahertz (THz) metamaterial absorbers with high absorption based on a structures consisting of a cobalt silicide (Co-Si) ground plane, a parylene-C dielectric spacer and a metal top layer. By combining two periodic metal resonators that couple separately within a single unit cell, a polarization-independent absorber with two distinct absorption peaks was obtained. By varying the thickness of the dielectric layer, we obtain absorptivity of 0.76 at 0.76THz and 0.97 at 2.30THz, which indicates the Co-Si ground plane absorbers present good performance.

Online traffic classification plays an important role in network management such as network security, accounting and provisioning. Traffic classification based on the first few packets (CFFP) reflects great promising prospective. However, not all flows can be observed the needed packets, especially in high-speed network. In this paper, classification based on arbitrary conjoint few packets (CACFP) and classification based on arbitrary disjunctive few packets of a flow (CADFP) are proposed. Meanwhile, classification based on entire packets of a flow (CEP) and classification based on the first few packets (CFFP) is used as baseline. First, mutual information of features with applications are analyzed under above four methods. Then, a real-time trace is used to validate our assumption. Experimental results show that CACFP and CADFP can obtain similar mutual information and classification accuracy with CFFP, which are also suitable for online traffic classification and make it possible for deploying practical online traffic classification system.

A microcantilever-based biosensor with self-sensing piezoresistors integrated is developed for glucose detection. The microcantilevers were fabricated using a series of bulk silicon process. The glucose detection was achieved by immobilizing a self-assembled monolayer of 4-mercaptophenylboronic acid (4-MPBA) on the surface of a gold-coated microcantilever (sensing cantilever) and then detecting the mechanical bending induced by the glucose capture using a Wheatstone bridge. The relationship between cantilever bending and glucose concentration was obtained at glucose concentrations of 5mM and 25mM. Experimental results demonstrated that the magnitudes of bending are proportional to the glucose concentrations and the device is sensitive enough for the glucose trace detection.

Accurate online network traffic classification plays an important role in many areas such as traffic engineering, QoS and intrusion detection. In this paper, four traffic classification methods are compared, which are classification based on entire packets of a flow (CEP), classification based on the first few packets of a flow (CFFP), classification based on arbitrary conjoint few packets (CACFP) and classification based on arbitrary disjunctive few packets of a flow (CADFP). Experiment results demonstrate that, without contribution of port feature, CACFP and CADFP achieved almost the same high classification accuracies in comparing with CEP and CFFP. The most important contribution of our method is that it significantly enhanced the robustness of online traffic classification. It can efficiently adapt to half-baked flows, which take over a certain proportion in real network condition. The method makes it possible for deploying practical online traffic classification system.

With the development of the modern educational technology, more and more teachers are looking for efficient methods and creative ideas, both to get their ideas across and to hold the interest of their students. In this paper, we present our approach in incorporating LEGO Mindstorms robots as simulating instruments in an undergraduate course of Information Technology, which are used in the context of a series of Lab exercises and assignments, helping students construct models of real world situations - load their algorithms to the robots and test their behaviors. As we have seen, teaching IT with LEGO mindstorms is helpful for the students to understand concepts and learn programming in the curriculum.

In this paper, a monolithic integration structure with TSV interconnections is introduced for un-cooled infrared FPA to do easy wafer-level-package. Firstly, the challenging process for making the structure will be reviewed and identified. And then process sequence for making the TSV interconnections and RDLs and CMOS compatible surface process for IR FPA will be developed. In the end, a WLP scheme will be developed to show the benefit of easy WLP.

In this paper, we developed a hybrid process of fabricating single crystal silicon nanostructures by HNA selective etching followed by an oxidization with a common photolithography. HNA etch heavy and light doped silicon with the etching selectivity about 160∶1. On the other hand, silicon will be consumed by an oxidization process. Based on these characteristics, we first fabricate some bigger structures with an optical lithography, and then processed the structures with heavy doping, diffusion, HNA etching, oxidization and BHF etching. At last, silicon nano-structures with the line width of ∼50nm were successfully fabricated. The results are almost catches up with the electron beam lithography, but with much low cost, time saving and reliability. The line width fabricated with this method can be further decreased by a precise controlling of the process parameters, include doping energy and concentration, annealing and oxidization time.

In the design of infrared detectors, nano films are more and more frequently used to improve infrared absorption and mechanical sensitivity. In this paper, a Titanium nitride (TiN x ) film was studied theoretically and experimentally as an infrared enhanced absorption layer. The TiN x films at two different thicknesses and in three different content ratios were deposited and infrared absorption experiments in the range of 8∼14µm were done. Experiment results show that the infrared absorbance is enhanced through depositing 5nm thick TiN x nano films on SiN x layer and the infrared absorbance of the films can be increased as the flow ratio of N 2 /Ar get increased.

Two simple lithography-independent methods by underetching the metal and using SiO 2 sidewall as sacrificial spacer to fabricate nanogap electrodes are demonstrated in this work. In the first method, by wet etching the metal layer after lithography and using the undeveloped photoresist (PR) as the sacrificial layer of the subsequent second metal lift-off process, a metal electrode with nanogap was fabricated. The width of the metal gap is mainly defined by the overetch under the PR. For the sidewall method, first depositing a conformal SiO 2 layer on a polysilicon pattern and then etching back the SiO 2 layer on the polysilicon, a sacrificial spacer is obtained after selectively remove the polysilicon. At last, a metal nanogap was fabricated after the metal sputtering and lift-off process. The width of the metal gap is mainly determined by the thickness of conformal SiO 2 layer. With the conventional lithography methods, metal electrode with the gap width smaller than 100nm can be obtained.

This paper investigates anisotropic wet etching characteristics of (110)-oriented silicon wafers in KOH etchant for obtaining vertical trench. To obtain a near rectangled trench instead of parallelogrammic trench with concave corners of 70° and 110°, a new compensation method was proposed. With this design, a trench with all sidewalls vertical was obtained by an over etching of silicon. Due to different widths of the trench, the over etching rate in releasing inclined crystal plane {111} varies from 0.32µm / min to 0.43µm / min .

Polyimide (PI) was a good candidate as the sacrificial layer for its compatibility with CMOS technology. This paper first presented a new patterning method of PI film and then investigated the relationships among undercut rate, the undercut limit length and the releasing hole size in the releasing step, which was helpful and important for its popularity and its application in MEMS capacitive FPA (Focal plane array) fabrication. A new patterning approach of PI film was successfully developed in ICP chamber. The patterning approach selected a PECVD SiO 2 layer as patterning mask. The optimized ICP PI recipe was composed of an O 2 flow of 180sccm, an electrode power of 400W and bias plate power 200W. With the optimized ICP PI recipe, a vertical etching rate about 0.5–0.6μm/min with a lateral etching rate 0.13μm/min was realized. With 1μm PI sacrificial layer, a 0.22μm/min undercut rate was achieved in a normal barrel etcher. Based on our experimental facts, the optimized releasing hole size was 5μm×5μm and the distance between lateral releasing holes should be fewer than 17 μm for effectively and completely releasing.

Wired and wireless networks are significantly different in terms of bandwidth, speed, propagation delay, and channel reliability. In particular, pocket loss due to bit corruption over wireless links may be misinterpreted by TCP as the indication of congestion that forces TCP to reduce its congestion window and enter the congestion avoidance phase, which reduces TCP's throughput dramatically. In this paper, we propose three techniques that operate at different layers of the traditional protocol stack in order to gain better performance in heterogeneous networks. We have combined these techniques together and established a detailed experimental evaluation both individually and in concert. The simulation results show that it has improved network performance and achieved much more goodputs than TCP-SACK and Snoop.

This paper presents an uncooled microcantilever infrared (IR) detector, of which the focal plane array (FPA) is fabricated based on bulk silicon technique. Compared with conventional cantilever IR detector fabricated with surface micromachining method, the IR absorption efficiency can be improved by 48% due to the selective removal of the substrate silicon with deep reactive ion etching (DRIE) technique at the area where each FPA pixel is located, and hence the noise equivalent temperature difference (NETD) can be decreased by 32%. Human body's images were captured by the fabricated FPA at room temperature with an optical readout system. The thermo-mechanical sensitivity of the FPA pixel was measured to be 0.12µ/K and the NETD of the IR detector was 310 mK. The image quality was improved by post image processing technique.

MEMS resonators are promising as micromechanical oscillators and filters in wireless communication systems. This paper reports on the fabrication and characterization of single crystal silicon (SCS) disk resonators with sub-micro capacitive gaps. Vertical capacitive gaps as small as 100nm have been demonstrated in this work. High frequency disk resonators have been implemented and tested.

With the development of telecommunication engineering and mobile computing, some related curricula are provided in lots of universities and colleges in our country. However, many students complain that they have to spend a lot of time in learning materials and are frustrated with them. This paper describes WNTLS, a wireless network teaching and learning system based on the network simulator (NS-2) that facilitates students to model, tune and analyze the performance of various types of wireless networks in the telecommunication engineering curricula.

This paper presents the design, fabrication and performance of a bimaterial microcantilever focal plane array (FPA) of 256×256 pixels for uncooled infrared (IR) imaging. The microcantilevers made of silicon nitride (SiNx) and gold (Au) convert incident IR energy to a distribution of deflection angles among the FPA pixels according to the IR source, which is read out by an optical method. The thermo-mechanical sensitivity of the cantilever pixel was measured to be 147 nm/K. The noise equivalent temperature difference (NETD) of the FPA was theoretically estimated to be 130 mK and the response time was calculated to be 12 ms. Thermal image of the heated brand iron were captured by the fabricated FPA. The most valuable feature of the implemented FPA is their scalability to high resolution formats without progressively growing fabrication complexity and cost.

Network protocols and architectures are a crucial component in the computer network curricula. Demonstration based on the simulation tools can be of great help for a better understanding of this curricula. This paper describes a computer network simulating and teaching system (CNSTNS), which is based on the NS-2, and able to fully simulate the topologies and traffic patterns of large scale networks. It also demonstrates the capabilities and usefulness of the simulator with case studies. Finally, based on our experiences using the system, ire summarize some guidelines for network simulation, and present further work.

In order to increase the sensitivity of a cantilever-based biosensor, piezoresistive cantilevers with stress concentration holes were designed and fabricated. Firstly the stress distributions and the vertical displacements of the designed cantilevers were simulated through ANSYS analyzing system. Then the relative resistance changes of the piezoresistors as a function of the cantilever vertical displacements were measured. As expected, the stresses are highly concentrated for the designed cantilevers, and this design can obviously result in an improvement on the sensitivity of the surface-stress mode cantilevers

Gold-coated microcantilevers were used to develop sensitive micromechanical biosensor for detecting mismatch mutations in DNA hybrids. The selective 5'-end immobilization of 25-mer single-stranded (ss) DNA probes to a gold-coated cantilever surface and the specific hybridization of ssDNA targets to the immobilized probe were performed by the direct electric field control. The controlled electric fields were used to regulate the transport of charged oligo-nucleotides. More than 70% of the immobilization and hybridization can be carried out in 2-3 seconds, and the detection rates are at least 102 times faster than in the passive control reactions performed without electric field.

A novel micro-jet pump based on the principle of macroscopic jet pump are presented. The micro-jet pump can be used very easily for sampling in micro-fluidic systems under gas actuation. Vacuum suction mode provided by the pump can effectively decreases the formation of air bubbles and overcomes the interface tension between the liquid and solid wall in the micro-fluidic systems. Structurally the pump consists of a nozzle, a throat pipe and a pump chamber, which can be fabricated by means of silicon micro-machining technology. To obtain optimized parameters for the future designs, various pump configurations with different geometries were simulated through FLUENT, and the characteristics of the gas-actuated micro-jet pump was studied experimentally. The pump began to work at a threshold gas actuation pressure of 2 kPa, which can be realized easily by a medical syringe.

This paper presents a monolithic integration of piezoresistive micro-cantilevers with CMOS (complementary metal oxide semiconductor) analog amplifier and digital multiplexer based on the techniques of SOI (silicon on insulator) CMOS and MEMS (micro electronic mechanical system). The three modules, signal detection, signal selection and signal amplification, of the design are proposed step by step. Then, SOI-CMOS-PD (partially depleted) is compared with SOI-CMOS-FD (fully depleted) to ensure its advantages in analog IC. And the post-CMOS process is designed as manufacture solution. Finally, simulations and test results are presented to prove the availability of the whole project.

A good cantilever should possess a small minimum detectable deflection (MDD), which depends on the measurement sensitivity and noise. Although increasing the measurement sensitivity can decrease the MDD of a cantilever, the noise will be a mainly limited factor when the measurement sensitivity is further increased. Therefore, the noise analyses of a piezoresistive cantilever are very important in improving cantilever resolutions. In this paper we analyze the noise characteristics of three silicon materials based on a large number of noise measurement results, which show that the 1/f noise is the dominant noise source at low frequencies and noise at high frequencies is Johnson noise. With the linear relation between 1/f noise and 1/L (piezoresistor length), Hooge factor /spl alpha/ of three materials was calculated. Optimized piezoresistor dimensions for a piezoresistive cantilever are given at last.

Protein-protein interaction (PPI) networks, which are caused by extracellular stimuli in the signal transduction pathways, are not fully understood for all cells. Mass spectrometric analysis is one of the powerful tools available for revealing these interactions. However, this analytical method is sometimes inefficient because of the many false positive candidates resulting from extracellular matrix or exogenous protein contaminants. To efficiently identify PPIs in signal transduction pathways, we developed a software program ‘PROMISS’ for LC/MS/MS. PROMISS contains all known PPIs and domain information, which are very useful for deducting the PPIs in different signal transduction pathways. By referring to the domain and interaction information in PROMISS, users can effectively select from enormous numbers of candidates the proteins that are related to specific PPIs. We also show that gel filtration chromatography is useful for pretreating and fractioning the interacting protein complex in the cell. In this study, we report a method of identifying epidermal growth factor receptor (EGFR) binding protein in EGF-treated A431 cells using PROMISS and gel filtration. Based on mass spectrometry analysis and bioinformatics approaches, this strategy is effective for identification of PPIs in signal transduction pathways.

In this paper, we proposed a novel weight-based bandwidth allocation algorithm (WBA) for transporting compressed video traffic using feedback. Extensive simulation using a modified NIST simulator is conducted to evaluate its performance under a general weight-based share policy. Our results demonstrate that the WBA algorithm is capable of allocating bandwidth fairly for the minimum cell rate (MCR) plus weighted allocation fairness criteria among competitive compressed video sources, and is capable of reallocating smoothly when there are renegotiations of the minimum guaranteed cell rate and weight by some connections. We also show that WBA algorithm is flexible for the reallocation when a new connection is admitted. Furthermore, when unutilized network bandwidth becomes available, our algorithm is capable of exploiting it to dynamically allocate unused bandwidth.