Online services face increasing threats from various anomalous activities. Although regulatory organizations have identified multiple red-flag behaviors, many existing detection methods focus primarily on groups of related accounts. However, these methods are not able to distinguish whether an account that does not relate to others is anomalous or not, because an individual account shows no clustering characteristics on the graph. We name this kind of the lack of detecting vision as “single anomaly problem”. To address this issue, we propose ASIAN—a unified anomaly detection framework. First, drawing from regulatory reports, we define an account or a group sharing an unusually high number of identifiers as abnormal. Unlike traditional approaches that rely on account-to-account relationships, our framework models the connections between accounts and their associated identifiers, formulating an objective function for our nonnegative matrix factorization (NMF)-based method. This low-rank approximation technique filters out low-significance data during loss minimization, enabling the extraction of high-volume patterns of identifier usage that indicate abnormal behavior. Finally, we define the involved identifiers as anomalous identifiers and trace back the accounts associated with them, whether they belong to individuals or groups, marking these accounts as anomalous. Experimental results show that ASIAN can indeed detect individual and group anomalies, while achieving a higher F1-score than alternative methods. This design not only improves the accuracy of the detection, but also clearly interprets how each component contributes to overall performance, thereby providing valuable insights into online service anomaly detection.

5 G networks, essential for applications like smart cities and factories, must ensure high reliability, availability, and security. However, the safety of 5 G remains a matter of concern. Despite its enhanced security mechanisms compared to previous generations, our study identifies two novel denial-of-service (DoS) signaling attack types that can disrupt the 5 G control plane (CP), exposing weaknesses in its current security design. Traditional switches cannot prevent such attacks because they are unable to extract attacking user equipment (UE) information from encapsulated nonaccess stratum (NAS) data. While ensuring the availability of the 5 G CP is essential to protect 5 G services, this issue has emerged as a pressing challenge. To address this challenge, we propose a novel P4-based user equipment fingerprinting (P4USE) method to mitigate 5 G CP DoS signaling attacks. Leveraging P4's programmable packet processing capability, P4USE can parse encapsulated NAS information at line rate, identify UE fingerprints, and drop anomalous NAS messages originating from attackers. Experimental evaluations in Open5GS confirm both the existence of the identified vulnerabilities and the effectiveness of P4USE in mitigating them. Overall, our work reveals two novel vulnerabilities and presents a defense solution to enhance the availability and security of the 5 G CP.

Abstract Static analysis tools are widely used for vulnerability detection as they can analyze programs with complex behavior and millions of lines of code. Despite their popularity, static analysis tools are known to generate an excess of false positives. The recent ability of Machine Learning models to learn from programming language data opens new possibilities of reducing false positives when applied to static analysis. However, existing datasets to train models for vulnerability identification suffer from multiple limitations such as limited bug context, limited size, and synthetic and unrealistic source code. We propose Differential Dataset Analysis or D2A, a differential analysis based approach to label issues reported by static analysis tools. The dataset built with this approach is called the D2A dataset. The D2A dataset is built by analyzing version pairs from multiple open source projects. From each project, we select bug fixing commits and we run static analysis on the versions before and after such commits. If some issues detected in a before-commit version disappear in the corresponding after-commit version, they are very likely to be real bugs that got fixed by the commit. We use D2A to generate a large labeled dataset. We then train both classic machine learning models and deep learning models for vulnerability identification using the D2A dataset. We show that the dataset can be used to build a classifier to identify possible false alarms among the issues reported by static analysis, hence helping developers prioritize and investigate potential true positives first. To facilitate future research and contribute to the community, we make the dataset generation pipeline and the dataset publicly available. We have also created a leaderboard based on the D2A dataset, which has already attracted attention and participation from the community.

Cybercriminals controlling multiple accounts to conduct malicious activities are a threat to the security of online services. These accounts form malicious communities that are difficult to detect using conventional methods with single-factor identity, such as browser fingerprint or internet protocol (IP) address. Single-factor identity is prone to noise and uncertainty and does not capture dynamic relationships between accounts. To solve the problems of insufficient single-factor identity and uncertainty binding between accounts and identity, we propose Ai-Urg, a novel account identity-based uncertain graph with multifactor identity modeling for online service fraud detection. To find account groups, we embed account representation that preserves the uncertain graph’s possible world semantics and use the domain knowledge that accounts of family members also from small communities to filter these benign groups and detect malicious communities. The ablation study demonstrates how each component contributes to the effectiveness of Ai-Urg. The comparison results on two datasets show that Ai-Urg outperforms the alternatives with a higher F1-score (58.0%) and precision (46.0%) on a real-world online bank dataset and with a higher F1-score (36.5%) and precision (96.9%) on a large-scale single-sign-on online service dataset. The experimental results can provide valuable insights into online service fraud detection.

It is expected to be pervasive sensors to integrate physical and digital world. Low-latency and wide-covering networks are hence urged, which lead to costly infrastructure. We are proposing ARCH based on smart contract and Self-sovereign Identity (SSI), to provide telecoms a trusted and transparent multilateral platform sharing their infrastructure to decrease the deployment and agency cost, and carbon emission.

Phishing has persistently posed threats to the World Wide Web as phishing websites evolve over these years. Many previous works were devoted to extracting useful features and focused on the essential components of phishing websites. One of the essential components is sensitive inputs which require sensitive information. Yet, due to a large variety of web designs, detecting the existence of sensitive inputs is not trivial. Some previous works have provided rule-based approaches to detect login forms, which contain sensitive inputs, using HTML codes. However, the novel phishing websites modify HTML codes against the detection rules, which causes less accurate detection.To overcome the limitation of previous works, we proposed SenseInput using hybrid deep learning models to detect the existence of sensitive inputs and sensitive information because phishing websites eventually present sensitive inputs in their visual content. SenseInput achieved 96.94% f1-score for sensitive input detection on our dataset and 96.73% f1-score on a public dataset, Phishpedia Phish30K. Next, we used 22 features involving the proposed seven statistical features and two sensitive input features for phishing detection. The experiment shows that our approach achieves 98.48% and 95.87% f1-score on our validation and Phishpedia datasets, outperforming previous approaches. Finally, we investigated the influence of sensitive input features. The result shows that our sensitive input features are more effective than the rule-based login form. Besides, the experiment also indicates that proposed sensitive input features can reduce the impact of bias between different datasets.

Websites can spread malware and phishing scams, and this represents a significant risk to users. To block such malicious websites, various organizations and individuals, e.g., security vendors, analyze URLs and create blocklists. Some blocklists are paid and some are free public blocklists; however, the effectiveness of public blocklists is generally not guaranteed. Thus, many studies have attempted to verify their effectiveness. To the best of our knowledge, public blocklist providers (PBP) have not been studied as rigorously as blocklists, and it is still unclear how blocklists should be operated and how PBP websites should be designed to maintain the effectiveness. Therefore, to unveil more characteristics of the PBP, we designed a measurement study to analyze PBPs in terms of lifespan, update frequency, entry bias, and user interface metrics. In this paper, we describe the results of measuring seven PBPs according to these four metrics.

With the rising influence of current online services, it is important for service providers to discover related accounts because it helps detect suspicious account groups. Existing research on this topic mostly focuses on a variety of account behaviors. Little attention has been paid to relations among account identity, which unveils the relationship between accounts and real-world people. In this paper, we propose DRAGON for modeling an account identity network and detecting suspicious account groups among this network. To this end, identifiers for tracking physical devices are collected and uncertain graph is used for modeling uncertainty in the network. Within this network, a strategy for detecting suspicious account groups is also investigated in DRAGON. We evaluate DRAGON using a real-world dataset. The results indicate that DRAGON achieves a 280% improvement in precision and 150% improvement in recall compared to a binary classifier.

The Internet of Things (IoT) is a reality that changes several aspects of our daily life, from smart home monitoring to the management of critical infrastructure. The “Routing Protocol for low power and Lossy networks” (RPL) is the only de-facto standardized routing protocol in IoT networks and is thus deployed in environmental monitoring, healthcare, smart building, and many other IoT applications. In literature, we can find several attacks aiming to affect and disrupt RPL-based networks. Therefore, it is fundamental to develop security mechanisms that detect and mitigate any potential attack in RPL-based networks. Current state-of-the-art security solutions deal with very few attacks while introducing heavy mechanisms at the expense of IoT devices and the overall network performance. In this work, we aim to develop an Intrusion Detection System (IDS) capable of dealing with multiple attacks while avoiding any RPL overhead. The proposed system is called DETONAR - DETector of rOutiNg Attacks in Rpl - and it relies on a packet sniffing approach. DETONAR uses a combination of signature and anomaly-based rules to identify any malicious behavior in the traffic (e.g., application and DIO packets). To the best of our knowledge, there are no exhaustive datasets containing RPL traffic for a vast range of attacks. To overcome this issue and evaluate our IDS, we propose RADAR - Routing Attacks DAtaset for Rpl: the dataset contains five simulations for each of the 14 considered attacks in 16 static-nodes networks. DETONAR’s attack detection exceeds 80% for 10 attacks out of 14, while maintaining false positives close to zero.

As network environment becomes more dynamic and complex, deploying distributed firewalls with access control lists in software-defined network can protect internal services and hosts from network attacks and insider attacks. With ever-changing types and sources of attacks, firewall policies need frequent updates. However, most existing integer linear programming-based solutions require recomputing the rule placement to optimize certain performance criterion. This leads to the cost of high computation overhead. To deal with the challenges of dynamic updates of firewall rules and rule placement, we propose a resource constraint splitting algorithm to compute only the rules related to the updated policies and preserve the others. The key idea is to separate the decision variables into disjoint subproblems and to only solve the associated part. Simulation results demonstrate that this approach shows significantly less computation time while maintain the optimized rule placement for network performance.

In order to prevent network services and end hosts from Internet attacks, a firewall is an important protective component to enforce security policy on network packets. A typical firewall sits at the entry point of an Autonomous System (AS). However, it may become the congestion point because of the growing number of security policies and network traffic. Also, a SDN-based firewall can suffer from the TCAM memory limit of SDN devices and thus it can only install a limited number of security policies. This paper presents a robust algorithm to distribute security policies of a firewall into distributed SDN devices in cloud-clustered environment. While this algorithm can obtain a better performance and resolve the TCAM memory limit of SDN devices, it can also guarantee a more complete protection, by stopping insider attacks.

Cyber security has grown up to be a hot issue in recent years. How to identify potential malware becomes a challenging task. To tackle this challenge, we adopt deep learning approaches and perform flow detection on real data. However, real data often encounters an issue of imbalanced data distribution which will lead to a gradient dilution issue. When training a neural network, this problem will not only result in a bias toward the majority class but show the inability to learn from the minority classes. In this paper, we propose a Tree-Shaped Deep Neural Network (TSDNN) which classifies the data in a layer-wise manner. To better learn from the minority classes, we propose a Quantity Dependent Backpropagation (QDBP) algorithm which incorporates the knowledge of the disparity between classes. We evaluate our method on an imbalanced data set. Experimental result demonstrates that our approach outperforms the state-of-the-art methods and justifies that the proposed method is able to overcome the difficulty of imbalanced learning. We also conduct a partial flow experiment which shows the feasibility of realtime detection and a zero-shot learning experiment which justifies the generalization capability of deep learning in cyber security.

IP multicast is a technique to save bandwidth when sending multimedia traffic to multiple users. However, to provide end-to-end quality of service (QoS) for multicast applications is still a challenging problem. The recently emerging technique, software defined network(SDN), provides a flexible network management capability by decoupling the data plane and the control plane. The control plane can be programmable to intelligently allocate network resources based on the information collected from the data plane. In this paper, we propose an end-to-end QoS multicast mechanism in SDN environments. We design an algorithm which adaptively provides the bandwidth provision via a learning mechanism for the multimedia traffic. Thus, QoS of multimedia flows could be guaranteed effectively and efficiently even under heavy-loaded scenarios. In addition, the learning approach is to utilize the link bandwidth efficiently by predicting the required QoS bandwidth. Furthermore, low priority packets are forwarded statistically among N different routes to mitigate the congesting link problem. Such a design could prevent the oscillation problem which could exist in traditional approaches which are to reroute the multimedia traffic to other routes. Our design leverages the flexible functions provided by SDN protocols. The performance is evaluated in a real SDN environment. The experimental results show the effectiveness of the proposed algorithm.

A heterogeneous network (HetNet) contains macrocells and different small cells, such as picocells, femtocells, and relay nodes. In HetNets, small cells are used to increase network capacity. Femtocells are usually used to increase coverage range in indoor environment, and can be deployed by users arbitrary. Besides, carrier aggregation is a major improvement in Long-Term-Evolution-Advanced (LTE-Advanced), and component carriers (CCs) are the basic aggregated units which are shared among cells. Therefore, cells use the same CCs lead inter-cell interference among them. In this work, we propose a resource allocation algorithm in LTE-Advanced HetNet called ”Stochastic Automata component Carrier Selection” (SACS) based on stochastic learning automata. SACS is a fully-distributed and self-optimizing algorithm, and it doesn't any information exchange among cells. SACS has some good properties:(i) energy saving, (ii) no information exchange, and (iii) low complexity. From the simulation results, SACS has fast convergent time and saves more energy than others.

This study proposes a clustering-based Wi-Fi fingerprinting localization algorithm. The proposed algorithm first presents a novel support vector machine based clustering approach, namely SVM-C, which uses the margin between two canonical hyperplanes for classification instead of using the Euclidean distance between two centroids of reference locations. After creating the clusters of fingerprints by SVM-C, our positioning system embeds the classification mechanism into a positioning task and compensates for the large database searching problem. The proposed algorithm assigns the matched cluster surrounding the test sample and locates the user based on the corresponding cluster's fingerprints to reduce the computational complexity and remove estimation outliers. Experimental results from realistic Wi-Fi test-beds demonstrated that our approach apparently improves the positioning accuracy. As compared to three existing clustering-based methods, K-means, affinity propagation, and support vector clustering, the proposed algorithm reduces the mean localization errors by 25.34%, 25.21%, and 26.91%, respectively.

This paper presents a novel approach to building a WLAN-based location fingerprinting system. Our algorithm intelligently transforms received signal strength (RSS) into principal components (PCs) such that the information of all access points (APs) is more efficiently utilized. Instead of selecting APs, the proposed technique replaces the elements with a subset of PCs to simultaneously improve the accuracy and reduce the online computation. Our experiments are conducted in a realistic WLAN environment. The results show that the mean error is reduced by 33.75 percent, and the complexity by 40 percent, as compared to the existing methods. Moreover, several benefits of our algorithm are demonstrated, such as requiring fewer training samples and enhancing the robustness to RSS anomalies.

This study focuses on indoor localization in Wireless Local Area Networks (WLANs). We investigate the unequal contribution of each access point (AP) on location estimation. The main contribution is two parts. First, a novel mechanism is proposed to measure the degrees of the AP importance. The importance of each AP is quantified by the signal discrimination between distinct locations. We utilize such numerical relevancies to select important APs for positioning. Second, the importance is further embedded into our positioning system. We provide a weighted kernel function where the effect of APs is differentiated. That is, the larger weights are assigned to the more important APs. Moreover, we develop a quasi-entropy function to avoid an abrupt change on the weights. Our positioning system is developed in a real-worldWLAN environment, where the realistic measurement of receive signal strength (RSS) is collected. Experimental results show that the positioning accuracy is significantly improved by taking the different importance into consideration.

Wireless spectrum is a limited and valuable resource for communications. However, wireless spectrum is known to be underutilized in spacial, temporal, and spectral domains. The dynamic spectrum access (DSA) of cognitive radio networks provides the capability to improve the spectrum efficiency by allowing secondary users to access the spectrum opportunistically without interfering primary users. The dynamic spectrum access is a joint channel allocation and power control problem with the objective to maximize the aggregated throughput of all secondary users. This problem is especially difficult in multi-channel multi-user cognitive radio networks. In the literature it is often formulated as a mixed integer nonlinear programming (MINLP) problem which is NP-hard. Therefore, some approximation methods are proposed to solve this problem, which lead to suboptimal solutions. In this paper we carefully reexamine the DSA problem, and prove that the original MINLP problem formulation is over-parameterized. We show that the DSA problem could be formulated as a nonlinear programming (NLP) problem without losing globally optimal objective function values. Moreover, the optimal solution to this NLP problem could be obtained by an interior point DSA optimization algorithm in polynomial time. Simulation results show that the proposed method performs better than other approximation methods do.

This paper presents a novel approach to placing access points (APs) in wireless local area networks (WLAN). Unlike the traditional methods focusing on coverage maximization, we investigate how to place APs from the perspective of an indoor location system. We present a framework for linking the placement of APs and the positioning performance. Our algorithm, namely maximizing SNR (signal-to-noise ratio), aims at choosing a proper set of APs' locations so that the signal is maximized and the noise is minimized simultaneously. The proposed algorithm regards the location discriminant information as the signal and the degree of unstable measurements as the noise. Such numerical values provide a good theoretical formulation to measure the influence of the AP deployment. In our location system, we utilize such SNR ratio to deploy APs so as to reduce the positioning errors. Our location system is developed in a real-world WLAN environment, where the realistic measurements were collected. The experimental results show that the positioning accuracy is improved based on our deployment approach in different scenarios.

This study focuses on the localization using Received Signal Strength (RSS) in dense multipath indoor environments. A dynamic system approach is proposed in the fingerprinting module, where the location is estimated from the state instead from RSS directly. The state is reconstructed from a temporal sequence of RSS samples by incorporating a proper memory structure based on Taken's embedded theory. Then, a more accurate state-location correlation is estimated because the impact of the temporal variation due to multipath is considered. An indoor experiment in Wireless Local Area Networks (WLAN) shows the effectiveness of our approach.

Recent advances in mobile devices and ubiquity of wireless infrastructures create the opportunity to utilize heterogeneous wireless networks (HWNs) for localization. To efficiently exploit the spatial correlation embedded in the measurements from HWNs, we proposed two algorithms via a cooperative approach, called Direct Multi-Radio Fusion and Cooperative Eigen- Radio Positioning. The former discovers the spatial correlation after the information of measurements is reorganized to minimize the redundancy. The latter takes a further step to incorporate the spatial discrimination to estimate the location. We have implemented our algorithms for different wireless technologies involving the cellular GSM, DVB, FM and WLAN in realistic outdoor/indoor environments. The results show that the proposed algorithm reduces 44.19-48.88% of the mean error, as compared to the conventional approaches.

Designing efficient search algorithms is a key challenge in unstructured peer-to-peer networks. Flooding and random walk (RW) are two typical search algorithms. Flooding searches aggressively and covers the most nodes. However, it generates a large amount of query messages and, thus, does not scale. On the contrary, RW searches conservatively. It only generates a fixed amount of query messages at each hop but would take longer search time. We propose the dynamic search (DS) algorithm, which is a generalization of flooding and RW. DS takes advantage of various contexts under which each previous search algorithm performs well. It resembles flooding for short-term search and RW for long-term search. Moreover, DS could be further combined with knowledge-based search mechanisms to improve the search performance. We analyze the performance of DS based on some performance metrics including the success rate, search time, query hits, query messages, query efficiency, and search efficiency. Numerical results show that DS provides a good tradeoff between search performance and cost. On average, DS performs about 25 times better than flooding and 58 times better than RW in power-law graphs, and about 186 times better than flooding and 120 times better than RW in bimodal topologies.

IEEE 802.11 wireless local area network (WLAN) physical layer (PHY) offers multiple data rates. In multi-rate WLANs, 802.11 distributed coordination function (DCF) presents the phenomenon so called ldquoperformance anomalyrdquo: when some hosts transmit at lower data rates, throughput of others at high rates will be restricted within the lowest rate, leading to the degradation of overall performance. Therefore, the link adaptation scheme for 802.11 WLAN should consider not only throughput of the observed host but also system throughput in order to optimize the overall performance. In this paper we propose a cross-layer link adaptation algorithm which improves system throughput by regarding the situation of PHY rate degradation more thoughtfully and critically, and meanwhile compensates the throughput of hosts with bad link qualities by applying differentiated channel access parameters. Simulation results demonstrate the effectiveness of the propose algorithm.

For low-delay video transmission, we introduce aq-domain characteristic-based bit-rate model. Specifically, three characteristics are efficiently extracted from the quantized DCT spectra to construct the bit-rate model. Extensive experimental results show that our rate model can provide more accuracy with lower complexity than existing models.

We present a novel approach to the problem of the indoor localization in wireless environments. The main contribution of this paper is fourfold: 1) we show that by projecting the measured signal into a decorrelated signal space, the positioning accuracy is improved, since the cross correlation between each AP is reduced, 2) we demonstrate that this novel approach achieves a more efficient information compaction and provides a better scheme to reduce online computation (the drawback of AP selection techniques is overcome, since we reduce the dimensionality by combing features, and each component in the decorrelated space is the linear combination of all APs; therefore, a more efficient mechanism is provided to utilize information of all APs while reducing the computational complexity), 3) experimental results show that the size of training samples can be greatly reduced in the decorrelated space; that is, fewer human efforts are required for developing the system, and 4) we carry out comparisons between RSS and three classical decorrelated spaces, including Discrete Cosine Transform (DCT), Principal Component Analysis (PCA), and Independent Component Analysis (ICA) in this paper. Two AP selection criteria proposed in the literature, MaxMean and InfoGain are also compared. Testing on a realistic WLAN environment, we find that PCA achieves the best performance on the location fingerprinting task.

This brief paper presents a novel localization algorithm, named discriminant-adaptive neural network (DANN), which takes the received signal strength (RSS) from the access points (APs) as inputs to infer the client position in the wireless local area network (LAN) environment. We extract the useful information into discriminative components (DCs) for network learning. The nonlinear relationship between RSS and the position is then accurately constructed by incrementally inserting the DCs and recursively updating the weightings in the network until no further improvement is required. Our localization system is developed in a real-world wireless LAN WLAN environment, where the realistic RSS measurement is collected. We implement the traditional approaches on the same test bed, including weighted kappa-nearest neighbor (WKNN), maximum likelihood (ML), and multilayer perceptron (MLP), and compare the results. The experimental results indicate that the proposed algorithm is much higher in accuracy compared with other examined techniques. The improvement can be attributed to that only the useful information is efficiently extracted for positioning while the redundant information is regarded as noise and discarded. Finally, the analysis shows that our network intelligently accomplishes learning while the inserted DCs provide sufficient information.

The access of multimedia computing in wireless networks is concerned with the performance of handoff because of the irretrievable property of real-time data delivery. To lessen throughput degradation incurred by unnecessary handoffs or handoff latencies leading to media disruption perceived by users, this paper presents a link quality based handoff algorithm. Neural networks are used to learn the cross-layer correlation between the link quality estimator such as packet success rate and the corresponding context metric indictors, for example, the transmitting packet length, received signal strength, and signal to noise ratio. Based on a pre-processed learning of link quality profile, neural networks make essential handoff decisions efficiently with the evaluations of link quality instead of the comparisons between relative signal strength. The experiment and simulation results show that the proposed algorithm improves the user perceived qualities in a transmission scenario of VoIP applications by minimizing both the number of lost packets and unnecessary handoffs.

Positioning in indoor wireless environments is growing rapidly in importance and gains commercial interests in context-awareness applications. The essential challenge in localization is the severe fluctuation of receive signal strength (RSS) for the mobile client even at a fixed location. This work explores the major noisy source resulted from the multipath in an indoor wireless environment and presents an advanced positioning architecture to reduce the disturbance. Our contribution is to propose a novel approach to extract the robust signal feature from measured RSS which is provided by IEEE 802.11 MAC software so that the multipath effect can be mitigated efficiently. The dynamic multipath behavior, which can be modeled by a convolution operation in the time domain, can be transformed into an additive random variable in the logarithmic spectrum domain. That is, the convolution process becomes a linear and separable operation in the logarithmic spectrum domain and then can be effectively removed. To our best knowledge, this work is the first to enhance the robustness to a multipath fading condition, which is common in the environments of an indoor wireless LAN (WLAN) location fingerprinting system. Our approach is conceptually simple and easy to be implemented for practical applications. Neither a new hardware nor an extra sensor network installation is required. Both analytical simulation and experiments in a real WLAN environment demonstrate the usefulness of our approach to significant performance improvements. The numerical results show that the mean and the standard deviation of estimated error are reduced by 42% and 29%, respectively, as compared to the traditional maximum likelihood based approach. Moreover, the experimental results also show that fewer training samples are required to build the positioning models. This result can be attributed to that the location related information is effectively extracted by our algorithm.

In this paper we propose a cross-layer handoff ordering scheme. The frame success rate (FSR) is adopted as the basis of prioritization. Different quality of service (QoS) requirements of various applications would result in different FSR requirements. In order to indicate how critical a handoff request is, both the FSR requirement from the application layer and the FSR measurement from the medium access control layer are taken into consideration in the proposed scheme. The prioritization of handoff requests follows the most-critical-first policy. Performance analysis shows that the proposed scheme effectively reduces the forced termination probabilities. Under the same forced termination probability requirements, it could provide 1.95% to 11.13% more arrival calls compared to previous works.

In this paper we investigate the throughput performance of IEEE 802.11 distributed coordination function (DCF) in the presence of physical-layer (PHY) inequality, i.e. varied channel conditions and/or unequal data rates determined by the Link Adaptation (LA) scheme. We present a theoretical model for DCF protocol with the LA scheme of auto rate fallback (ARF). The analysis results show that the system throughput of DCF-based WLANs is determined by the lowest data rate used with stations; throughput sharing among stations depends on the variation of link qualities rather than the difference of data rates. The simulation results validate the accuracy of our analytical model.

The authors address the problem of providing fair multimedia quality-of-service (QoS) in IEEE 802.11 distributed co-ordination function-based wireless local area networks in the infrastructure mode where mobile hosts experience heterogeneous channel conditions due to mobility and fading effects. It was observed that unequal link qualities can pose significant unfairness of channel sharing, which may thereby lead to the degradation of multimedia QoS performed in adverse conditions. A cross-layer adaptation scheme that provides fair QoS by online adjusting the multidimensional medium access control layer backoff parameters in accordance with the application-layer QoS requirements as well as the physical-layer channel conditions was proposed. The solution is based on an optimisation approach, which utilises neural networks to learn the cross-layer function. Simulation results demonstrate that the proposed adaptation scheme can tackle heterogeneous channel conditions and random joining (or leaving) of hosts to achieve fair QoS in terms of throughput and packet delay.

Flooding and random walk (RW) are the two typical search algorithms in unstructured peer-to-peer networks. The flooding algorithm searches the network aggressively. It covers the most nodes but generates a large number of query messages. Hence it is considered to be not scalable. This cost issue is especially serious when the queried resource locates far from the query source. On the contrary, RW searches the network conservatively. It only generates a fixed amount of query messages at each hop, but it may take particularly longer search time to find the queries resource. We propose the dynamic search algorithm (DS) which is a generalization of flooding, modified breadth first search (MBFS), and RW. This search algorithm takes advantage of different contexts under which each previous search algorithm performs well. The operation of DS resembles flooding or MBFS for the short-term search, and RW for the long-term search. We analyze the performance of DS based on the power-law random graph model and adopt some performance metrics including the guaranteed search time, query hits, query messages, success rate, and a unified metric, search efficiency. The main objective is to obtain the effects of the parameters of DS. Numerical results show that proper setting of the parameters of DS can obtain the short guaranteed search time and provide a good tradeoff between the search performance and the cost.

We analyze the fairness of IEEE 802.11 DCF in heterogeneous wireless LAN environments where users experience unequal channel conditions due to the mobility and fading effects. Previous works [3] [4] show that the 802.11 CSMA/CA can present fairness characteristics in both long- term and short-term. However, the conclusion is only valid under the condition of homogeneous link qualities, which may be impractical. In this paper, we consider heterogeneous channel conditions based on an analytical approach of extending a verified two dimensional Markov chain model of DCF proposed by Bianchi [10]. From our analytical results, it is shown that 802.11 CSMA/CA can present fairness among hosts with identical link qualities regardless of equal or different data rates applied, which is consistent with the observations of previous works. Our analytical results also demonstrate that the presence of heterogeneous channel conditions can pose significant unfairness of channel sharing even with a link adaptation mechanism since the MCSs (modulation and coding schemes) available are limited.

An accurate rate-distortion model, which characterizes the relationship among bitrate, distortion, and quantization parameter (QP), is very desirable for real time video transmission. It has been reported that the actual coding bitrates can be estimated by a linear combination of two characteristic rate curves in rho-domain where rho is defined as the percentage of zeros among the quantized transform coefficients. The process is referred as the "pseudocoding" process. Unfortunately, since rho values are real numbers, an interpolation process must be required for the domain transformation from rho-domain to q-domain. Thus, the prediction inaccuracy can not be avoided and even be "propagated" due to the interpolation uncertainty error. Hence, three parameters are addressed in this paper to support a more accurate and direct estimate of encoding bit rates based on q-domain. They include the number of nonzero coefficients, the count of zeros before the last nonzero coefficient in the zigzag-scan order, and the sum of absolute quantized nonzero coefficients, respectively. We surprisingly find that the estimation accuracy in q-domain is better than currently well-known rho-domain based R-Q model. In addition, a quantization-free extraction method, which only involves some additions and a few multiplications, is developed. That is, the implementation complexity of the proposed mechanism is very low. Consequently, the proposed R-Q model is very suitable for real time applications

The digital cameras can be easily integrated with other handheld devices in one single chip. For hardware size and cost consideration, the camera sensor array of the PDAs or the cellular phones is therefore limited. However, as the demanding of better image quality is increasing, the producer needs to employ large sensor array for higher image resolutions. Therefore, how to emulate large sensor array to raise resolutions is a key problem to these devices. In this paper, a novel enlargement method on Bayer pattern has been presented. We introduce a unified image processing scheme which effectively interpolates the sensor array directly for a higher resolution. The unified image processing scheme can also lead to efficient hardware implementations. Unlike traditional enlargement methods which operate only within a color plane, our method utilizes the relationship between the distinct channel color differences of neighbors and the non-cross weighting-based approach. Experimental results indicate the proposed method has the ability to emulate efficiently larger sensor array from the relation of color difference over local region and edge-sensing weight coefficients. The proposed algorithm gives sharp edges over the enlarged image without color-bleeding artifacts. Both subjective visual evaluation and objective performance measurement demonstrate the effectiveness of the proposed method.

Packet size is one of the most important factors that would affect the user-perceived multimedia QoS in the wireless LAN environments. The time-varying channel characteristics make it difficult to find the exact relationship between the packet size and the throughput and decide an optimal packet size in advance. Furthermore, every node would suffer different channel conditions. In this paper, we tackle this problem by an optimization approach. A context-aware framework is designed to optimize the packet size adaptively in order to maximize the throughput. In this approach each node abstracts its specific context via the throughput from the time-varying wireless environments. The obtained throughput information is the instantaneous integrated effect of all contexts in wireless LAN environments. This approach adopts neural networks to learn the complex nonlinear function between the packet size and the throughput and adaptively adjusts the packet size. Simulation results show that out method can cope with the time-varying wireless channel conditions and improve the perceived QoS of wireless multimedia services

Captured CFA data by image sensors like CCD/or CMOS are often corrupted by noises. To produce high quality images acquired by CCD/CMOS digital cameras, the problem of noise needs addressing. In this paper, we propose a novel demosaicking algorithm with the ability to handle noisy CFA data directly. By utilizing the proposed spatial filter which can characterize the similarity likelihood in local structure accurately, the noisy pixel is then filtered depending on the degree of similarity between the current pixel and a weighted average of its neighboring pixels. Therefore the edge information can be preserved without the blurring artifacts while the capacity of noise reduction can be adjusted to the maximum degree in the smooth region. Our algorithm is the first one that can accomplish the demosaicking processing and noise removal simultaneously, which contributes to the reduction of hardware cost since one module can achieve two functions efficiently at the same time

This paper investigates multimedia quality fairness in wireless LAN environments where channel are error-prone due to mobility and fading. The experimental results show that using fixed MAC arguments for nodes in heterogeneous channel conditions leads to unequal throughput performance and that may incur the degradation of multimedia QoS. To overcome the unfairness problem for provisioning QoS, we propose a cross-layer adaptation scheme by on-line adapting the multidimensional MAC-layer backoff parameters depending on the application-layer QoS requirements and PHY-layer channel conditions. Our solution is based on an optimization approach which utilizes neural networks to learn the cross-layer function. Simulation results demonstrate that our adaptive scheme can tackle a variety of channel condition to provide fair throughput for nodes in heterogeneous channel conditions

In this letter, we propose a cross-layer adaptation scheme which improves IEEE 802.11e quality of service (QoS) by online adapting multidimensional medium access control (MAC)-layer parameters depending on the application-layer QoS requirements and physical layer (PHY) channel conditions. Our solution is based on an optimization approach which utilizes neural networks (NNs) to learn the cross-layer function. Simulations results demonstrate the effectiveness of our adaptation scheme.

Traditional handoff ordering methods adopt the received signal strength (RSS) as the basis of prioritization. However, the RSS is not the only one metric to represent the user perceived quality of service, since many other factors, like the packet length, interference, and the modulation/codec schemes, would also affect it. In this paper, we propose a handoff ordering method based on packet success rate (PSR) for multimedia communications in wireless networks. The priority of a handoff request is based on its current PSR, the PSR degradation rate, and the minimum PSR requirement of its service class. The major contribution of our method is that we improve the user perceived QoS during the handoff process. Simulation results indicate that our method can effectively improve the handoff call dropping probability with little or no increase of the new call blocking probability.

The access of multimedia computing in wireless networks is concerned with the efficiency of handoff because of the irretrievable property of real-time data delivery. To lessen throughput degradation leading to media computing disruption perceived by users, this paper presents a link quality based handoff algorithm. Neural networks are used to learn the correlation between link quality estimator and the corresponding context metric indictors. Based on a pre-processed learning of link quality profile, neural networks make efficient handoff decisions with an evaluation of link quality instead of a comparison between relative signal strength. The experimental and simulation results show that the number of lost packets is minimized using the proposed algorithm without incurring unnecessary handoffs.

This paper deliberates on various critical aspects in evaluating searching networks. Existing metrics either draw biased conclusions regarding search performance or provide wrong guidelines for algorithm design. We, therefore, define a unified criterion, search efficiency (SE), to objectively address search performance in a comprehensive manner. The goal of SE is to better characterize performance of searching networks than existing metrics do as well as to guide the design of future ones. We first validate the correctness of SE in performance evaluation in an ideal graph, strictly binary tree, by analyzing SE for two typical search methods, breadth first search and random walk. We further show its strength in performance characterization in the real-world topology, power-law random graph, under various network conditions. We finally design an algorithm, dynamic search, based on SE analysis. Its proved outstanding performance demonstrates the strength of SE to provide guidance for the future design of searching networks.

Recently peer-to-peer networks (P2P) have gained great attention and popularity. One key challenging aspect in a P2P resource sharing environment is an efficient searching algorithm. This is especially important for Gnutella-like decentralized and unstructured networks due to the power-law degree distributions. We propose a hybrid search algorithm that decides the number of running walkers dynamically with respect to peers' topological information and search time state. It is able to control the extent of messages generating temporally by the simulated annealing mechanism, thus being a scalable search. Furthermore, we present a unified quantitative search performance metric, search efficiency, to objectively capture dynamic behavior of various search algorithms in terms of scalability, reliability and responsiveness. We quantitatively characterize, through simulations, the performance of various existing search algorithms. The proposed algorithm outperforms others in terms of search efficiency in both the local and global search spaces.

Recently peer-to-peer networks (P2P) have gained great attention and popularity. One key challenging aspect in P2P resource sharing environments is efficient searching algorithm. This is especially important for Gnutella-like decentralized and unstructured networks since they have power-law degree distributions. A robust search algorithm should respond to the query message promptly without generating redundant query messages. We present unified quantitative search performance measurements: query efficiency, search responsiveness, and search efficiency to objectively capture dynamic behaviors of various search algorithms from different perspectives. To gain insight of these search algorithms, we quantitatively characterize, through simulations, their search performance on different network topologies with different query/replication distributions.

It has previously been shown that gradient-descent learning algorithms for recurrent neural networks can perform poorly on tasks that involve long-term dependencies, i.e. those problems for which the desired output depends on inputs presented at times far in the past. We show that the long-term dependencies problem is lessened for a class of architectures called nonlinear autoregressive models with exogenous (NARX) recurrent neural networks, which have powerful representational capabilities. We have previously reported that gradient descent learning can be more effective in NARX networks than in recurrent neural network architectures that have "hidden states" on problems including grammatical inference and nonlinear system identification. Typically, the network converges much faster and generalizes better than other networks. The results in this paper are consistent with this phenomenon. We present some experimental results which show that NARX networks can often retain information for two to three times as long as conventional recurrent neural networks. We show that although NARX networks do not circumvent the problem of long-term dependencies, they can greatly improve performance on long-term dependency problems. We also describe in detail some of the assumptions regarding what it means to latch information robustly and suggest possible ways to loosen these assumptions.