Image deblurring is important to improve the quality of the image captured from a camera. To well deblur an image, it is an important issue to estimate the distributions of noise and edges, especially for the methods based on maximum likelihood estimation (MLE). In this paper, first, we propose a method for noise estimation based on the smooth prior. That is, for most of natural images, there should be some smooth regions. Then, after predicting the smooth regions and determining their gradient distributions, the noise distribution of the image can be well estimated. Then, based on the estimated noise and the additivity property of the Gaussian distributions, the local edge distributions can be determined. Then, the parameters of the MLE model can be assigned adaptively by the ratio of the standard deviations of edge and noise distributions. Experiments show that the proposed image deblurring algorithm based on the local Gaussian model and smooth prior has better performance than state-of-the-art image deblurring algorithms, including conventional and learning based ones.

Two commonly used methods for traffic information rely on Vehicle Detector (VD) and Global Positioning System-Based Vehicle Probe (GVP); however, they have some confinements, such as high cost for construction and maintenance, and limited coverage. For the sake of overcoming dilemmas happened on VD and GVP, Cellular-Based Vehicle Probe (CVP) comes into being. However, the current applications for CVP mainly focus on arteries or freeways, where traffic information for longer distance is derived from two Inter-Visitor Location Register Location Area Update (Inter-VLR LAU) events with various Location Area Code (LAC) borders, and the one for shorter distance is from two consecutive handover events. The perplexity of available CVP techniques comes about is that there are no two Inter-VLR LAU events with various LAC borders and few handover events in scenic spots. In order to expand the applications for CVP to scenic spots, a cost-effective and flexible method utilizing mobile network signalling called Enhanced CVP (ECVP) is proposed. The key concept for ECVP is that we adopt Inter-VLR LAU events at the origin and all kinds of communication events at the destination to retrieve traffic information. The inaccuracy of ECVP consists in the uncertainty of event occurred time at the destination. Therefore, with a view to acquiring more accurate traffic information, three novel Reinforced CVP (RCVP) algorithms, inclusive of Fixed r percent samples CVP (F-RCVP), Dynamic r percent samples (D-RCVP), and Dynamic r percent samples with Discarding former samples (DD-RCVP), are presented. Numerical results show that F-RCVP is suitable for scenic spots that the LAC border only contain samples resulting from cars. By contrast, if the samples consist of both cars and motorcycles, it is recommended that D-RCVP and DD-RCVP are introduced.

In this paper, we present a quota-based dynamic network selection algorithm for accommodating access requests in a multi-network (HSPA+, HSDPA and UMTS) co-location environment, which is common for today's 3G operators. In most commercial networks, the accesses of individual users are often charged at a flat rate. In such a charging scheme, a few extremely heavy users may degrade the performance and the fairness of the whole network. Therefore, we propose an algorithm called Quota-based Utilization Balance Algorithm (QUBA), designed to settle such a problem and also maximize the overall utility of all users. The design philosophy of QUBA is to let the system utilization in the 2 fastest networks, HSPA+ and HSDPA, be statistically balanced and sufficiently utilized. Performance metrics in terms of the overall system utility (i.e., the satisfaction index) and the blocking probability are investigated. From simulation results, it is shown that the QUBA algorithm has significant improvements in the aspect of the overall system performance than that of existing approaches.

A novel blind statistical algorithm for obtaining the optimal receive beam vector with maximum time-domain effective SNR in OFDMA systems under multipath environment is proposed. The basic procedure of the proposed approach is obtained via the coarse cyclic prefix (CP) detection first and the sample average of the correlation matrices is calculated from selected signal samples for later singular value decomposition (SVD) on the averaged correlation matrix. The proposed approach has the following advantages: First, it is accurate even in the presence of high noise power and when the process of the OFDMA sampled signals are not white (i.e. not all subcarriers are allocated for data transmission). Secondly, it can be operated in time-domain with no training signals and no pre-assumption of the antenna array structure. Hence, it could be applied directly to a variety of existing OFDMA systems and its design can be made relatively simple for implementation. Theoretical proofs of its optimality and the guarantee of convergence are also provided in this paper. Finally, simulation results verifies the theoretical analysis and shows that the proposed approach can accurately achieve a near optimal SNR gain even in low SNR environment and has a convergence speed within two OFDM symbols.

In this paper, we develop a stream-based system, CoreView (CV) having launched at Chunghwa Telecom in Taiwan, that characterizes with T-connection or non-intrusion and is based on the mobile network signalling, to support mobile network diagnosis and comprehensive location services simultaneously in near real-time. With the CV-network diagnosis, can operators have expertized systems and save the expenses for the network maintenance. In addition, they have opportunity to enhance the profitability with the CV-location services.

Carrier Aggregation (CA) is one of the promising features to support the much wider bandwidth for packet transmissions in Long Term Evolution-Advanced (LTE-A) systems. To the best of our knowledge, works regarding the packet scheduling algorithm in LTE-A systems with the CA mechanism have not thoroughly been investigated yet till now. In this paper, we therefore aim to design an efficient packet scheduling algorithm based on the proportional fair criterion in such environment. We focus on downlink transmissions supporting both Real-Time (RT) and non-RT traffics simultaneously. Simulation results show that performance measures in terms of the system throughput, the mean delay, and the fairness are much better than those of a system in which component carriers are not aggregated but used independently.

Carrier aggregation in LTE-Advanced provides the capability to uncoordinatedly deploy home eNodeBs with the opportunity to autonomously mitigate the interferences. Among many improvement approaches, the autonomous component carrier selection (ACCS) scheme to improve the performance of the cell edge users worth our attentions. However, a poorly ordered execution of the carrier selection can lead to waste of radio resources in a typical home Internet access scenario. In order to address this issue, we propose a component carrier reselection scheme which ameliorates the total throughput of the eNodeBs by reselecting the carriers according to an intelligently planned order. Additionally, we also propose to improve the performance of the cell edge users with the low-complexity fractional power suppression technique. System-level simulation of the proposed schemes has shown substantial improvement over the original ACCS scheme.

Traditional spectrum allocation approaches assume that the intended users of BSs are treated equally, without consideration of their distances. Consequently, the efficiency of spectrum usage for downlink transmissions may suffer, because a BS may interfere with other BS transmissions; this would not happen if a proper transmission power is used. In this paper, we propose a user-centric spectrum allocation algorithm based on a physical interference model for downlink transmissions. In our approach, BSs set the power of an allocated spectrum band to ensure an effective transmission range for covering a group of intended users. A spectrum band is allocated to a BS if it can satisfy a group of users with the least amount of power possible (relative to other spectrum bands). Thus, after spectrum bands and transmission ranges are assigned, each BS owns several spectrum bands that can cover a group of users within varying transmission ranges. This approach leads to a higher rate of spectral reuse and improves system performance. In addition, the approach properly considers fairness issues among both users and BSs. Overall, simulation results show that our approach not only outperforms traditional BS-centric approaches by offering a 26% higher throughput in dense networks, but also provides fairer spectrum allocation.

As various wireless technologies and application demands for more and more bandwidth, dynamic channel allocation in multiple bands has become an attractive research direction. In this paper, we propose several channel allocation policies for various types of user profiles under various fluctuation traffic loads in dual-band multi-cell environments. The considered policies include 1) fixed allocation with admission control, 2) reallocation on hourly basis, and 3) reallocation on hourly basis with forced dropping. Performance comparisons of the system utility and the band utilization with various policies are investigated. Also, how the bandwidth concentration ratio is balanced between the operator's revenue and users' satisfaction under a selected policy is illustrated. Simulations show that the proposed dynamic policies in a dual-band environment can achieve much better utility performance than that in a single-band environment when no dynamic policy is employed.

This work addresses the problem of a fast hybrid automatic-repeat-request (HARQ) scheme intended for transporting delay-sensitive flows in a cooperative diversity (CD) environment. This CD with fast HARQ scheme allows the retransmission for only one time via the cooperating user (i.e. partner) or via both the sender and the partner simultaneously. A complementary link packet is used for retransmissions. The retransmission policy can be adjusted based on the qualities of channels among the sender, the partner, and the receiver and the application layer protocol data unit size by using the application throughput as the objective. For this presented scheme, we completely derive the application throughput formulas, which are then validated via simulations. Next, the CD with optimized fast HARQ scheme is shown able to achieve better throughput than effective throughput of other CD with ARQ schemes in various Nakagami-m slow-fading environments. As a result, the fast HARQ scheme should be an excellent ARQ mechanism for the real-time multimedia transport in CD environments.

Both the femtocell technology and the concept of mobile virtual network operators (MVNO) are believed to be key directions for future mobile network evolutions. This paper proposes a detailed model for the femtocell-based MVNO and analyzes the dynamics between the femto base station transmit power and its absorbing effect on the macrocell users via game theoretic techniques. The results show that the power settings given by the Nash equilibrium maintain the required QoS level without causing excessive interferences. Such numerical results should help the feasibility of large-scale deployment of femtocell-based MVNOs.

Dynamic spectrum sharing has become an important approach to improve spectrum efficiency by exploiting traffic loads variations of multiple wireless networks when a radio band is shared. In this study, the authors present an effective centralised decision approach to deal with the problem regarding how a spectrum manager should periodically re-allocate the spectrum between two networks in a single cell, while maintaining both the call blocking and dropping probabilities in their acceptable levels, and improving the spectrum efficiency simultaneously. In the decision algorithm, a two-level model (a spectrum level model and a call level model) is used to characterise the dynamic spectrum sharing mechanism and network resource usage within the two networks. Performance analysis begins with an approximated Markov Chain analysis and an iterative approach. The authors then further validate the approximated numerical results with simulations. These numerical results demonstrate that the call blocking and dropping probabilities of each network employing dynamic spectrum sharing are much better maintained than those using a fixed allocation under heavy traffic load. The authors thus conclude that the dynamic spectrum sharing algorithm can maintain the quality of service (QoS) of each network while the effective spectrum utilisation is improved under a fluctuation traffic environment when the available spectrum is limited.

Capacity has been an important issue for many wireless backhaul networks. Both the multihop nature and the large per packet channel access overhead can lead to its low channel efficiency. The problem may get even worse when there are many applications transmitting packets with small data payloads, e.g., Voice over Internet protocol (VoIP). Previously, the use of multiple parallel channels and employing packet concatenation were treated as separate solutions to these problems. However, there is no available work on the integrated design and performance analysis of a complete scheduler architecture combining these two schemes. In this paper, we propose a scheduler that concatenates small packets into large frames and sends them through multiple parallel channels with an intelligent channel selection algorithm between neighboring nodes. Besides the expected capacity improvements, we also derive delay bounds for this scheduler. Based on the delay bound formula, call admission control (CAC) of a broad range of scheduling algorithms can be obtained. We demonstrate the significant capacity and resequencing delay improvements of this novel design with a voice-data traffic mixing example, via both numerical and simulation results. It is shown that the proposed packet concatenation and channel selection algorithms greatly outperform the round-robin scheduler in a multihop scenario.

There are totally 163 colleges and universities in Taiwan, 90% of them have established their campus WLAN environment and 61% of them have joined the cross-campus WLAN roaming environment under the support of the National Science Council (NSC). This paper describes how we have established such a WLAN roaming environment in Taiwan. Relying on this RADIUS based WLAN roaming environment, a campus WLAN user in Taiwan can enjoy WLAN roaming service via Web portal or 802.11x based AAA. Newly developed applications such as voice/video service, or campus WiMAX networks can also be tested over this environment. This paper will also introduce our research on how the campus wireless network security can be improved.

A packet scheduling scheme in the downlink channel for IEEE 802.16 is proposed.The scheme can be implemented in the base station and is compatible to the MAC protocol of IEEE 802.16. In order incorporate the properties of location-dependent and burst error of wireless channels, we design a novel packet scheduling algorithm named the credit-based algorithm (CBA) to guarantee QoS of both real-time and non-real-time services. Simulation experiments are conducted to evaluate the performance of the proposed scheme. Based on the simulation results, we show the CBA scheme is capable of providing significantly improved system capacity,guaranteeing the packet delay for real-time services, and maintaining fair capacity allocation to non-real-time services.

Among all related issues in shared wireless networks, the bandwidth allocation mechanism should be one of the most important ones that affect system efficiency, QoS, and the revenue of network operators. In this paper we present a model for the bandwidth resource sharing in a wireless access network among multiple mobile virtual network operators (MVNOs). Within this model, we propose an efficient scheme to adaptively allocate the bandwidth in accordance with the traffic fluctuation. Our proposed bandwidth allocation scheme adopts the technique of dynamic programming to maximize the expected total revenue across multiple stages from the perspective of the facility provider. Meanwhile, we also allow various MVNOs to compete for bandwidth resources by setting the prices. The design is then evaluated by simulation results and proved to be efficient and beneficial.

Dynamic spectrum allocation can improve spectrum efficiency by exploiting the traffic loads variations of multiple wireless network systems, when they share neighboring band. In this study, we use a centralized approach in which a spectrum manager periodically re-allocates spectrum among two networks in a single cell. We propose a two levels model (a spectrum level model and a call level model) to characterize the dynamic spectrum allocation. We have further validated the approximated numerical results by simulations. Numerical and simulation results show that the performances of networks with the dynamic spectrum allocation are better than those with fixed allocation and complete its performance analysis via approximation and iterative approach. Hence, we conclude that our model and scheme can improve quality of service (QoS) of wireless networks when spectrum is limited

With advances in variant wireless technologies, many mobile devices equipped with multiple interfaces have emerged. In such a multiple wireless network environment, it is desirable to enjoy streaming video with good quality and low cost. The challenge of providing such services is to take advantage of heterogeneous wireless network characteristics, including their bandwidth, packet lost probability and transmission cost. In this paper, we first present a detailed video distortion model to estimate the distortion experienced by service users. Based on this model, we then propose a 3-stage greedy algorithm, which determines the transmission sequences over two heterogeneous wireless networks, so that distortion controlled streaming video services can be provided with near-optimal cost. Experiment results show that the detailed video distortion model can keep the expected distortion under better control. Besides, the proposed 3-stage greedy algorithm provides a solution which meets the distortion and deadline constraints with the near-optimal cost

Call admission control (CAC) is one important QoS mechanism for maintaining performance of streaming media services. In this paper, we propose a novel CAC algorithm, which handles long-lived and short-lived streams differently to satisfy their different QoS requirements. Under such mechanism, we show that not only one can guarantee the quality of each stream, but also improve the system utilization.

Virtual router redundancy protocol (VRRP) is designed to eliminate the single point of failure in the static default routing environment in LAN. The original VRRP protocol does not support load balancing for both incoming and outgoing traffic. This paper describes EVRRP, i.e. enhanced VRRP. EVRRP supports an efficient multiple-node cluster and symmetric load balancing among routers. Each router periodically exchanges information to determine the status of the master and backups. The master router distributes and redirects the traffic to one of the backup routers by ICMP redirect message. Backup routers accept the traffic from the master and one of the backup routers takes over the master traffic using a gratuitous ARP message when the master fails. The improved election protocol speeds up the original VRRP election protocol and shortens the failover time by adding a new state in the previous VRRP state diagram and a new protocol type. An extensive evaluation of the EVRRP protocol is described in the paper.

We propose an MPLS switch with a GPS-based scheduler to provide guaranteed QoS in the Internet. The considered scenarios include cases with sufficient labels or with a limited number of labels under competing streams. Due to tractability issues in direct analysis, we use a worst case equivalent system model to obtain the theoretical upper bounds of delays and backlogs of the original system model. Simulation results are used to verify the analytical results. We can obtain the worst case delay bounds and the maximum backlogs. Furthermore, the trade-off among the number of labels, the worst delay bounds, and the maximum backlogs can be observed. With our model, one can determine the number of required labels according to the allowable delay and the available buffer size.

The performance of a GMPLS (generalized multiprotocol label switching) architecture with flush capability is studied. For this switching architecture, we propose a queueing model that includes the control plane, the switching buffer mechanism, and the flush mechanism. The flush capability is included to reduce the out-of-sequence problem due to dynamic path changes. The behavior of aggregated streams, the label-setup and release policies, and the mechanisms for efficient resource allocation are all covered. With the proposed model, one can select appropriate parameters for the label-setup policy and the label-release policy to match the traffic load and network environment. Key performance metrics, such as throughput, label-setup rate, and fast path bandwidth utilization, can all be evaluated by this mathematical model. Numerical results and simulations are used to verify the accuracy of our proposed queueing model. Finally, the trade-off among these performance metrics can be observed as well.

We propose a new traffic scheduling scheme, called differentiated multi-layer gated frame queueing (DMGFQ), to provide users with differentiated QoS, better jitter control and fair bandwidth sharing simultaneously. We show that the DMGFQ scheme can achieve excellent performance for well-behaved flows, including delay, jitter and cell discarding ratio, under the presence of misbehaved flows. The residual bandwidths released from other terminated sessions are also fairly distributed among backlogged flows. As a result, the DMGFQ scheme not only supports a wide range of QoS for real-time multimedia, but also achieves the goal of protecting the users conforming to their service level agreements. In addition, DMGFQ also accommodates many flexible realization approaches without sacrificing its performance.

The generalized wireless differentiated fair queueing (GWDFQ) algorithm is proposed to accommodate delay/jitter controls, and fair residual bandwidth sharing for real-time and non-real-time traffic streams simultaneously. The location-dependent channel error property, as appeared in most wireless networks, are considered in the algorithm and the temporary short error burst are compensated by the design of credits. The simulation results show that GWDFQ can achieve excellent performance, including timely delivery of real-time traffic, virtually loss-free transmission of non-real-time traffic, and fair usage of channel bandwidth among remote stations.

In this paper, we propose a framework for real-time multimedia transmission in asynchronous transfer mode (ATM) networks using an efficient traffic scheduling scheme called multilayer gated frame queueing (MGFQ). MGFQ employs only one set of FIFO queues to provide a wide range of QoS for real-time applications. We also propose special cell formats for real-time multimedia transport and a hybrid design to allow MGFQ to combine its scheduling scheme with Age Priority Packet Discarding scheme. For this hybrid design, the cell level performance as well as the packet level QoS can be improved at the same time, Simulation results show that this hybrid design will be useful for packetized voice and progressive layer-compressed video transmission across the backbone networks. With the presented framework and the MGFQ algorithm, real-time multimedia traffic streams can be much better supported in terms of cell/packet delay and jitter.

In this paper, two novel scheduling algorithms with low implementation complexity are investigated. Most scheduling algorithms proposed so far usually involve a sorting operation with the complexity of O(logN) per packet, where N denotes the number of connections sharing the link. To solve this problem, we propose two new scheduling algorithms with the complexity O(1), implemented with only a single FIFO queue in the output scheduler. The proposed scheduling algorithms make use of the concept of gated scheduling, and thus the schedulers are called gated-scheduling servers (GSS). The key contribution of the GSS algorithms is the successful elimination of output sorter in their designs such that the scheduling mechanism can accommodate a large number of flows. Both delay bounds and fairness index for flows scheduled under these two algorithms are validated with simulations.

In the current ATM AAL5 implementation, even a single cell loss event can lead to the corruption of one whole packet. Hence, it has been observed that the throughput of the upper layer protocol may easily collapse on a congested ATM network. We propose a buffer management method called age priority packet discarding (APPD) scheme to be used along with two other schemes: the early packet discarding (EPD) and the partial packet discarding (PPD) schemes. After describing the operations and the pseudo code of the proposed APPD scheme and how it operates with the EPD/PPD schemes, the packet level QoS of APPD and its extended versions are derived analytically under homogeneous on-off source model. Numerical results obtained via, analytical approach suggest that the proposed APPD scheme can more effectively and fairly reduce the packet loss probability than other schemes.

The proposed GCRA traffic shaper consists of a regulator and a scheduler. It can shape multiple incoming variable bit rate (VBR) cell streams simultaneously to be strictly conforming according to the GCRA algorithm when the cells depart for the ATM output link. The impact of cell emission conflicts is considered and resolved by using a feedback signal from the scheduler to the regulator. The call admission control condition and the cell delay bound are derived. Simulation results demonstrate that the output cell streams of the proposed GCRA traffic shaper do not contain any non-conforming cells and the output queue size is significantly reduced.

A circuit-switched network consisting of multiple low Earth orbiting satellites and three geostationary satellites is considered. Crosslinks among low-orbit satellites are used as the major communication channels, while calls are routed via geostationary satellites only when low-orbit routes exceed or equal to a hop-count threshold. Simulation results are used to illustrate the network throughput under various traffic conditions and different link capacities. The impact of satellite failures on the network performance is also investigated.>

An ATM call interworking with the public switched telephone network may experience echo problems due to impedance mismatch. The authors propose two echo control designs to assure the quality of voice services by reducing round-trip echo delay. Performance of these two echo control designs is then analyzed.>

A LAN environment in which two token rings are interconnected via remote bridges is considered. Traffic streams of two priority classes are supported. An approach is presented to characterize the output process of one token ring with a bridge node, using a three-phase Markov modulated Poisson process. Numerical results indicate that this approximate process can be much more accurate than the Poisson process for the performance analysis of the bridge node.

A hierarchical and modular local asynchronous transfer mode (ATM) switch is proposed. In order to improve queuing performance, the output queuing technique is adopted and several outputs are allowed to partially share the same output buffer space. The performance of the proposed switch is evaluated under uniform and nonuniform traffic patterns. Comparisons with the knockout switch, the recursive switch, and the Christmas tree switch show that, in terms of complexity, crosstalk, scalability, and synchronization, the proposed switch is superior.

A symmetric priority-based token network is considered. Messages are divided into two priority classes. High-priority messages are assumed to require tight delay constraints. As a result, each station is allowed to establish, at any time, at most a single real-time high-priority access concentration. High-priority messages are guaranteed access onto the channel within a prescribed limited period. In turn, regular priority messages are only served when the system determines, through the repetitive use of circulating tokens (as used by the IEEE 802.5 token-ring-type protocol), that no high-priority messages are currently waiting in the system. Two token schemes employing different service disciplines are used to provide network access. Exact and approximate mean delay formulas for both message classes are derived. Numerical results are then exhibited to illustrate the network performance under various traffic conditions.>

A symmetric priority-based token network is considered. Messages are divided into two priority classes. High-priority messages are assumed to require tight delay constraints. As a result, each station is allowed to establish, at any time, at most a single real-time high-priority access connection. High-priority messages are guaranteed access onto the channel within a prescribed limited period. In turn, regular priority messages are only served when the system determines, through the repetitive use of circulating tokens, (as used by IEEE 802.5 token ring type protocol) that no high-priority messages are currently waiting in the system. Two token schemes, using different service disciplines, are used to provide network access. Exact mean delay formulas for both message classes are derived. Approximate message delay analysis approaches are also investigated for networks containing large numbers of stations. Numerical results are then exhibited to illustrate the network performance under various traffic conditions.>

A multiplexing communications channel is considered. The network stations share the channel on a time-division multiple-access (TDMA) basis or by using a slotted frequency-division multiple-access (FDMA) scheme. Messages are categorized by their differing priorities. Two service disciplines are considered: nonpreemptive and preemptive-resume. A discrete-time priority queuing system is introduced and investigated. Message-delay distributions are obtained for priority and ordinary messages in a two-class system model, under each service discipline. Next, delay analysis for a p-class model is carried out. The results are then applied to general, slotted, priority FDMA and TDMA schemes.>

We consider a double-tier communication network architecture for metropolitan area and packet radio networks for controlling the access of network terminals into a single shared broadcast channel. Terminals are organized into local groups in accordance with their geographical proximity or other criteria. Each local group can be associated with a primary station. The latter can serve as an unbuffered repeater for terminal transmissions unto the multiple-access channel. A polling policy is used by the primary stations to gain access into the shared communications backbone. Once a primary station is provided access into the channel, it initiates a local access control procedure. Message delay performance results and bounds are obtained for general reservation local-access procedures. In particular, we analyze and present performance results for a polling/tree-random-access procedure which can be effectively used for many packet radio and cellular digital radio networks, as well as for a polling/positionalpriority scheme for CATV and fiber-optic based networks.