Unmanned Aerial Vehicles (UAVs), have been widely studied as aerial Mobile Edge Computing (MEC) systems for resource-constrained Internet of Things (IoT) devices. Yet, battery limitations on both sides call for integrated energy harvesting alongside computation offloading. This paper studies a wireless-powered aerial MEC system, where a UAV harvests wireless energy from a terrestrial laser source, transfers wireless energy to IoT devices and serves as a computing unit to process their offloaded tasks, all at the same time. We formulate a long-term trade-off between the processed tasks and UAV energy consumption in this system by jointly optimizing system decision variables, including offloading decision, transmit power, computing resources and UAV motion trajectory. This trade-off adheres to the system constraints such as stability of data and energy queues, as well as limitations of transmit power budget, computing capacity and flight trajectory. Due to non-convex and long-term nature of this formulation, we initially leverage Lyapunov drift-plus-penalty theory, initially, to convert it into an instantaneous equivalent form. This step is subsequently followed by reformulating the problem as a Markov Decision Process (MDP), which facilitates the training of a Soft Actor-Critic (SAC) agent. To improve the adaptability of the SAC algorithm to system heterogeneity, such as differences in resource capacity, we integrate meta-learning and propose Meta Soft Actor-Critic (MSAC). As a result, the proposed agent is capable of efficiently optimizing system decision variables in real time. Numerical results demonstrate that our proposed resource allocation approach achieves its strongest gains under cloud-capacity scaling, delivering up to 26% improvement over SAC and up to 43% over TD3, representing the highest performance margins observed across all experiments.

Integrated into existing Mobile Edge Computing (MEC) systems, Unmanned Aerial Vehicles (UAVs) serve as a cornerstone in meeting the stringent requirements of future Internet of Things (IoT) networks. The current endeavor studies an MEC system, in which a computationally-empowered UAV, wirelessly linked to a cloud server, is destined for task offloading in uplink transmission of IoT devices. The performance of this system is studied by formulating a resource allocation problem, which aims to maximize the long-term computed task efficiency, while ensuring the stability of task buffers at the IoT devices, UAV and cloud. The problem jointly optimizes the uplink transmit power of IoT devices and their offloading decisions, the trajectory of the UAV and computing power at all transceivers. Regarding the non-convex and stochastic nature of the problem, we devise a multi-step solution approach. Initially, by invoking the fractional programming and Lyapunov theory, we transform the long-term optimization problem into an equivalent per-time-slot form. Subsequently, we recast the reformulated problem as a Markov Decision Process (MDP), which reflects the network dynamics. The MDP model, eventually, serves for training a Meta Twin Delayed Deep Deterministic Policy Gradient (MTD3) agent, in charge of adaptive resource allocation with respect to the MEC system variations derived from the mobility of the UAV and IoT devices. Simulations reveal the dominance of our proposed resource allocation approach over its Deep Reinforcement Learning (DRL)powered counterparts, increasing computed task efficiency and reducing task buffer lengths.

Unmanned aerial vehicles (UAVs) have emerged as a promising platform for mobile edge computing (MEC) due to their ability to provide flexible and efficient on-demand computing and communication services. However, deploying UAVs in MEC systems raises concerns about energy efficiency since both the UAVs and users' devices are typically energylimited. In this paper, we consider a multi-UAV assisted MEC heterogeneous network with two types of UAVs: edge UAVs equipped with computing resources and relay UAVs that assist users in transmitting their tasks to Ground Base Stations. We aim to minimize the total energy consumption of users and UAVs by optimizing task offloading decisions. We use a metaheuristic genetic algorithm to solve the optimization problem, and propose a novel multi-task convolutional neural network (MT-CNN) to efficiently predict near-optimal solutions. Simulation results demonstrate the effectiveness of our proposed model.

With the development of next-generation wireless networks, the Internet of Things (IoT) is evolving towards the intelligent IoT (iIoT), where intelligent applications usually have stringent delay and jitter requirements. In order to provide low-latency services to heterogeneous users in the emerging iIoT, multi-tier computing was proposed by effectively combining edge computing and fog computing. More specifically, multi-tier computing systems compensate for cloud computing through task offloading and dispersing computing tasks to multi-tier nodes along the continuum from the cloud to things. In this paper, we investigate key techniques and directions for wireless communications and resource allocation approaches to enable task offloading in multi-tier computing systems. A multi-tier computing model, with its main functionality and optimization methods, is presented in detail. We hope that this paper will serve as a valuable reference and guide to the theoretical, algorithmic, and systematic opportunities of multi-tier computing towards next-generation wireless networks.

Multi-tier computing effectively enables flexible computation and communication resource sharing by offloading computation-intensive tasks to nearby servers along the cloud-to-thing continuum. In essence, multi-tier computing networks can distribute computing, storage, and communication functions anywhere between the cloud and the endpoint to take full advantage of the resources available along this continuum, thus extending the traditional cloud computing architecture to the edge of the network. With multi-tier computing, some application component processing, such as delay-sensitive components, can take place at the edge of the network, while other components, such as time-tolerant and computation-intensive components, can be performed in the cloud. To best meet user requirements, centralized cloud computing with extensive resources, secure environments, and powerful algorithms is still needed, but also must be complemented by distributed fog and edge computing with shared resources, accessible environments, and simple algorithms for real-time decision-making. Given heterogeneous computing resources and collaborative service architectures, future multi-tier computing networks will be capable of supporting a full range of computing and networking services for different environments and applications. This Special Issue aims to provide a forum for the latest advances in multi-tier computing for next-generation wireless network research, innovations, and applications. Multi-tier computing enables low-latency processing by allowing data to be processed at the network edge close to end devices. It also facilitates the distribution of fog/edge nodes to collect data from end devices. Therefore, multi-tier computing effectively complements the cloud computing architecture.

Multi-tier computing effectively enables flexible computation and communication resource sharing by offloading computation-intensive tasks to nearby servers along the cloud-to-thing continuum. In essence, multi-tier computing networks can distribute computing, storage, and communication functions anywhere between the cloud and the endpoint to take full advantage of the resources available along this continuum, thus extending the traditional cloud computing architecture to the edge of the network. With multi-tier computing, some application component processing, such as delay-sensitive components, can take place at the edge of the network, while other components, such as time-tolerant and computation-intensive components, can be performed in the cloud. To best meet user requirements, centralized cloud computing with extensive resources, secure environments, and powerful algorithms is still needed, but also must be complemented by distributed fog and edge computing with shared resources, accessible environments, and simple algorithms for real-time decision-making. Given heterogeneous computing resources and collaborative service architectures, future multi-tier computing networks will be capable of supporting a full range of computing and networking services for different environments and applications. Multi-tier computing enables low-latency processing by allowing data to be processed at the network edge close to end devices. It also facilitates the distribution of fog/edge nodes to collect data from end devices. Therefore, multi-tier computing effectively complements the cloud computing architecture.

Most IoT devices have limited or no computing capability while many emerging IoT applications require both computing and communications services. Moreover, low latency requirements of numerous applications such as autonomous driving and augmented reality are becoming critical. In this paper, we propose a novel framework that can utilize the edge-computing facilities and the full-duplex technique at the edge nodes to address computing and communication services with low latency to IoT terminals for different applications. We then formulate an application-aware edge-computing problem for IoTs with the target to minimize the average latency. We propose a machine learning algorithm to solve this problem by achieving the best user-edge-node assignment and developing an optimal assignment and scheduling algorithm for the communication and computing resources. We evaluate the performance of the proposed machine learning algorithm (via Python and Tensor ow) and present results and comparison with other methods.

The high agility and maneuverability of the unmanned aerial vehicles (UAVs) provide a unique opportunity to carry communications and edge-computing facilities on board to serve mobile users in the cellular networks. An important problem would be to maximize the average aggregate quality-of-experience of all users over time slots. However, this is a non-convex, nonlinear and mixed discrete optimization problem, which is difficult to solve and obtain the optimal solution. We thus propose a deep reinforcement learning algorithm to solve this problem by considering UAV path planning, user assignment, bandwidth and computing resource assignment. The UAVs and base stations are to serve mobile users in multiple continuous time slots, and machine learning is leveraged to facilitate joint resource allocation and path planning in provisioning UAV-assisted edge computing. We compare the performance of our proposal with two baseline cases through simulations 1) with fixed UAV locations and 2) without UAVs. We demonstrate that the deep reinforcement learning algorithm performs better than these two baseline cases.

Unmanned aerial vehicles (UAVs) are playing a critical role in provisioning instant connectivity and computational needs of Internet of Things Devices (IoTDs), especially in crisis and disaster management. In this work, we focus on optimizing trajectories of UAVs along which IoTDs are served with communication and computing resources in multiple time slots. The Quality of Experience (QoE) of an IoTD depends on its latency performance; we thus aim to maximize the average aggregate QoE of all IoTDs overall time slots. However, this is a nonconvex, nonlinear, and mixed discrete optimization problem, which is difficult to solve and obtain the optimal solution. We thus propose two deep reinforcement learning algorithms to solve this problem by considering UAV path planning, user assignment, bandwidth, and computing resource assignment. We compare the performance of our proposed algorithms through simulations with three baseline cases: 1) with fixed UAV locations; 2) without UAVs; and 3) the fixed UAV trajectories. We demonstrate that the deep reinforcement learning algorithms perform better than all baseline cases.

We develop an analytical framework to model the temporal characteristics of Wi-Fi channel statistics, particularly, the duration of white spaces resulting from idle periods in Wi-Fi traffic. We employ a bivariate Markov process that captures the underlying dynamics of Wi-Fi network in formulating group Poisson arrivals corresponding to aggregated Wi-Fi frames and determine the parameters of this Markov modulated batch Poisson process by matching the associated first-order traffic arrival statistics. While this model is drastically simpler than others such as BMAP, nevertheless, it provides remarkably similar accuracy in modeling the white spaces. The results from analytical models are corroborated with those of the simulation platform we developed in NS3. Next, we use this model to study the feasibility of exploiting Wi-Fi white spaces for unlicensed LTE (U-LTE) transmissions. We further propose an opportunistic coexistence algorithm that enables the U-LTE base station to dynamically estimate the duration of upcoming Wi-Fi white spaces and determine the U-LTE ON/OFF epochs. The proposed scheme demonstrates performance limits when the latency imposed on Wi-Fi activity is minimized, while U-LTE maximally utilizes the available spectral resources. A useful application of this model is when integrated with stochastic geometry to form a basis for optimal resource allocation in 5G small cells to guarantee network quality of service and low latency.

Multi-tier structure is recognized as an assuring architecture to enhance the capacity and reliability of wireless networks, especially when it comes to the new generation of networks such as 5G. However, having multiple cell selection options in heterogeneous multitier networks introduces more uncertainty in end users QoS and network reliability. To address this issue, this paper proposes a real-time Mobile Terminal (MT) and Base Station (BS) association scheme for the multi-tier heterogeneous networks based on the cooperative game. The proposed algorithm models the base station association utility function based on Nash Bargaining, considering the interrelationship of base stations of different tiers in the solution. Unlike the static association algorithms which lead to either a large number of re-associations or suboptimal solutions, the proposed algorithm selects the best (MT, BS) pair upon new mobile terminal arrival and maximizes the product of objective functions to capture the tiers correlations and guarantee the improved performance. Simulation results show that the proposed cooperative cell selection scheme properly distributes the traffic load between multiple tiers of base stations and enhances users QoS.

In the coexistence problem of LTE-U and WiFi in unlicensed spectrum, interference can result in serious degradation in the performance of the two networks, especially WiFi. Conventionally, LTE is designed as the exclusive operator of the spectrum, whereas WiFi nodes contend in a distributed fashion based on CSMA-CA to access the spectrum. Consequently, WiFi network performance will be pushed to the limit as a result of backing-off for continuous LTE-U transmissions. To mitigate WiFi bandwidth starvation, in this paper we introduce opportunistic utilization of white spaces in the WiFi channel for LTE-U transmissions. Recognizing the dynamic and bursty nature of WiFi traffic, we develop an analytical framework based on Markov modulated batch Poisson processes (MMBPP) to stochastically model the temporal behavior and the idle gaps, or white spaces, in the unlicensed bands. Closed-form expressions are derived and approximations are provided to characterize the duration and the frequency of occurrence of WiFi white spaces. Noticing the considerable availability of idle resources, we discuss the possibilities of exploiting these white spaces for LTE-U transmissions without disturbing and delaying WiFi traffic. Next, under our MMBPP framework, we propose an opportunistic coexistence mechanism which enables the LTE-U base station to dynamically schedule transmissions by estimating the duration of WiFi white spaces, so as to maximize its utilization while limiting the disturbance introduced to the WiFi system. The latency experienced by the coexisting WiFi network is reduced to 90% of the case with ABS, while LTE-U utilization remains comparable.

This paper analyzes the downlink outage probability and spectral efficiency in the multi-tier heterogeneous networks according to the punctured Poisson Point Process (Punctured-PPP) base stations (BSs) distribution. To achieve a successful communication, we consider an exclusion region, centered at macro BSs locations, in order to restrict the small BSs transmission and avoid their high interference which causes macro users outage. The resulting interference field of the lower tier BSs is modeled by the thined-PPP. Using stochastic geometry, we derive an accurate interference model to explore the spectral efficiency of the multi-tier networks. We discuss the effect of the radius of the exclusion region as a design parameter in evaluating the network performance. Our results reveal that extending an exclusion area, by either increasing its radius or raising the density of macro BSs, significantly reduces the network outage probability. Analytic results are in a good agreement with the numerical simulations and demonstrate the accuracy of our proposed model.

LTE in unlicensed spectrum, also referred to as LTE-U, is a promising solution introduced towards capacity enhancement and efficient spectral utilization through data offloading in the 5GHz unlicensed spectrum. This is especially applicable to LTE operated small cells, which are supposed to boost capacity and expand coverage as a second tier of cellular networks, by allowing seamless flow of data across both licensed and unlicensed bands via a unified core network. To ensure this, fair and stable coexistence with incumbent technologies, most important of which is WiFi, is crucial. In this paper, considering the dynamic and bursty nature of WiFi traffic, we propose an analytical framework based on Bivariate Markov Processes to stochastically characterize the temporal activity of WiFi network in the unlicensed band. In particular, we use Markov Modulated Batch Poisson Processes (MMBPP) to formulate the idle gaps, aka WiFi white spaces, by capturing the underlying states of the WiFi network, as well as the group packet arrivals. Relying on this model, we investigate the possibilities of LTE-U opportunistically accessing the WiFi channel, so as to fully utilize white spaces, while ensuring a fair coexistence. Simulation results based on developed models using NS3 validate the accuracy of our analytical model and demonstrate the potential opportunities available for LTE-U exploitation.

In spectrum overlay cognitive wireless networks, when the secondary users are blind, i.e. have no prior knowledge of primary users' activities, a major question remains as to what strategy to adopt to learn the primary state information online while sensing the spectrum for exploiting idle bands. To maximize secondary network spectral utilization while minimizing interference and collision, requires searching for a balance between choosing empirically best channel while investigating other channels for potential opportunities. Moreover, competition should also be resolved to prevent congestion and collision among secondary users who tend to access the same idle channel. In this paper, by compressing the inessential exploration phase, we first propose an asymptotically optimal, fast-converging channel selection algorithm referred to as modified-myopic strategy for the single-user scenario based on the result of multi-armed bandit. Through analysis and simulation modeling, we evaluate the performance of our approach and compare it to that of other strategies in the literature. Next we adopt a game-theoretic model and extend our algorithm to design a fair and low-complexity access strategy for multi-user case based on a generalized CSMA/CA scheme. Finally, we show the improved efficiency of the proposed approach, in particular significantly enhanced for dense networks.

A MAC-layer channel selection scheme is proposed in this paper for efficient discovery of spectrum opportunities in the cognitive radio networks. In a practical uncertain cognitive environment, secondary users are unaware of the primary traffic statistics. Thus, in order to maximize the utilization, they need to obtain fairly accurate estimations of spectrum availability along with the actual sensing procedure, so as to learn when to sense which frequency channel. However, it is important not to waste the time, energy and the potential opportunities to explore primary's behavior in all frequency bands. In this work we investigate a low-complexity method referred to as modified myopic scheme, inspired by myopic strategy, whose gain converges to the highest possible throughput shortly after learning primary's activity and finding the most likely available channels. We also show that the secondary performance is inversely proportional to the lowest activity level of primary among all channels, rather than its average level. Therefore, highly-congested channels do not reduce the cognitive user's gain. The effect of sensing errors is also studied. Simulation results indicate the accuracy of our statements as well as the effectiveness of our scheme compared to the other strategies.

The femtocell — Wi-Fi integration is a promising approach to solve the problem of inter-tier or intra-tier interference in heterogeneous networks. In this paper, multimode femtocell base stations are deployed which form a Wi-Fi mesh network to communicate between themselves while deploying cellular technology (e.g., LTE) for communications with mobile users, thus ensuring ultimately connectivity between mobile users and their macro base stations to improve the network coverage. We propose a tractable model for coverage/outage to evaluate the benefits of such integration in terms of SINR and received signal strength. Our work is based on point processes in two-dimensional plane that models locations of femtocell — Wi-Fi (also referred to as multimode) nodes. The proposed model is more realistic than the classical Poisson point process, as the distribution of points is more homogeneous and it ensures that the nodes are not too close to each other. The derivation of coverage/outage formula allows us to determine operational parameter ranges for the Wi-Fi network to form a mesh network. In addition, it helps in the design of the femtocell network to ensure a suitable coverage for users in terms of SINR and received signal strength.

In this paper, we present a generalized multipath load sharing framework for Internet. Under the hypothesis that edge networks are fully independent in establishing Internet routing policies, we use game theory and a vectorized utility function to solve the Internet multipath routing problem. We first briefly review the existing techniques that apply game theory to improve the Internet routing resiliency. We propose a vectorized routing cost model to consider multiple routing metrics in the network setting, along with a universal refinement method to quantify the profile performance and predict the behavior of the vectorized routing game. Based on the universal refinement method as well as a linear traffic distribution algorithm, we define a generalized multipath load sharing framework to improve the routing resiliency in traffic exchange between two distant edge networks. Running simulations with real measured Internet metrics, we find that the proposed generalized edge-to-edge multipath load sharing framework is able to offer far more resilient solutions than legacy and alternative protocols such as standard BGP and BGP with LISP-based traffic engineering.

This paper proposes a dynamic resource allocation scheme based on Nash Bargaining Game (NBG) for the uplink of single-cell multiuser OFDM cellular networks. Although Nash Bargaining Solution (NBS) provides a fair and optimum approach in order to maximize the total rate of the network, there is no simple solution for the case of #36;K#36; users. NBS relates to maximizing the user utilities multiplication rather than the summation, hence, it makes the optimization problem harder to solve. Here, by decomposing the NBS problem into two sub- problems, the power allocation reduces to the well- known water-filling algorithm and the subchannel assignment leads to a simple algorithm which takes the total channel gain of each user as the fairness factor. Simulation results show that the proposed algorithm keeps a balance between the Max-Min approach and the Max-Sum where Max-Min aims at maximizing the worst user rate and Max-Sum maximizes the sum of the rates by blocking the users in the poor channel conditions.

This paper considers a half-duplex relay channel with a single source, relay, and destination, where each node has multiple antennas and the relay operates in decode-and-forward (DF) mode. The additional degrees of freedoms introduced by the MIMO channels entail increased complexity in comparison with the single antenna case. We propose a new transmission strategy for the relay channel that is able to take advantage of the MIMO gains while employing practical techniques that help reduce complexity. In the proposed scheme, the source splits its message in two parts and sends them to the destination, one part directly and the other with help from the relay. For such a transmission strategy, we formulate the problem of obtaining the rate-maximizing linear precoding matrices with time, power, and spatial stream allocation and transform the problem into a convex form. We also propose a suboptimal algorithm that provides simplifying expressions to solve the resulting problem with far less computational complexity. The numerical results show that the achievable rate of our scheme is greater than the DF rate in certain scenarios, as well as that of other practical existing strategies. In addition, the rate obtained by the suboptimal method approximates the optimal rate extremely well in all considered scenarios.

Considering both spectrum underlay and opportunistic spectrum access scenarios in cognitive wireless networks, we seek to solve the following problem in this paper: Given that the secondary nodes have imposed upper limits on their transmission power levels to coexist with the primary network, what is the achievable throughput by the power-constrained secondary nodes? We first obtain statistics for some spatial properties in the secondary network, e.g., relative distances and angles between nodes. Assuming that the secondary transmitters choose the maximum permissible power level to coexist with a primary link and using the obtained spatial statistics, we find a lower bound for the throughput of secondary nodes by considering different forwarding models in the secondary network. We show that the approach is applicable for finding the achievable throughput in general power-constrained random wireless networks.

In this paper, we propose a collaborative spectrum sensing method based on Stackelberg game in order to improve the sensing performance of cognitive radio users, especially when they are operating under severe channel fading. In this scheme the users with acceptable received SNR are allowed to lead the network sensing process and share their observations with the ones experiencing weak channel conditions. By defining a new and more appropriate metric to evaluate the performance of the collaborative detection, the proposed method addresses the drawbacks of the conventional centralized or distributed spectrum sensing. Moreover, it does not require exchange of channel information among nodes and only minimum reporting of local observations is needed. The simulation results indicate that the proposed scheme improves the network sensing performance and reduces the overhead as compared to the noncooperative case and the conventional collaborative schemes, respectively.

The significant growth in the global Internet traffic and routing table size requires solutions to address Internet scalability and resiliency. A number of proposals have considered moving away from the flat legacy Internet routing to a two-level hierarchical routing, separating edge networks from transit carrier networks. In this paper, we study the extended inter-domain traffic engineering capabilities arising in a transit-edge separated Internet routing, focusing on those multi-homed edge networks (e.g., small ISPs, content providers, large corporations) that aim at increasing their Internet resiliency experience. We model using game theory the interaction between distant independent edge networks exchanging large traffic volumes, with the goal of seeking efficient edge-to-edge load-balancing routing solutions. The proposed traffic engineering framework relies on a non-cooperative potential game, built upon path prepending- and path diversity- dependent costs, that indicates efficient equilibrium solution for the edge-to-edge load-balancing coordination problem. Simulations on real instances show that, in comparison with the alternative multipath BGP and normal LISP, we can achieve significantly higher resiliency and stability. In particular, our simulation for an illustrating case shows four-times more stable multipath routing solutions with a five-times larger path diversity.

In this paper, we propose a cooperative framework for spectrum sharing between a secondary random ad hoc network and a primary network with capacity constraint. While the interference from secondary network must satisfy a power constraint, there is no stipulation on its higher order statistics. Considering a generalized Gaussian model for interference, we show that the capacity of a primary link can be improved by increasing the kurtosis of interference. We show that by using power control, the secondary network can increase the kurtosis of its interference and thereby can help the primary links to maintain their minimum capacity constraint. Secondary network is allowed to share the spectrum for a fraction of time proportional to the capacity gain obtained by the primary link. Simulation results show that by cooperation from the secondary network, both secondary and primary networks can achieve their goals in accessing the spectrum and satisfying the capacity constraint.

We consider coexistence of overlapping but distinct primary and secondary random ad hoc networks. A secondary node has node-level and network-level obligations to avoid interference on primary receiving nodes. At the node level, it is allowed to transmit only if it does not detect data reception from its primary neighbors. At the network level, it needs to make sure that the aggregate secondary interference on an arbitrary primary receiving node satisfies an interference constraint. In this paper, we show that both of these obligations are non-trivially related to transmission attempt probabilities in primary and secondary networks and our aim is the optimization of throughput in these networks. We show that the interference constraint limits the feasible space for transmission probabilities of both systems. Looking at the secondary network, as a network plugged into the primary, we propose a progressive transmission probability optimization algorithm and show through simulation that it converges to the optimum throughput in a few time slots. Results show that at the optimum transmission probabilities, while the primary network deviates slightly from its optimal throughput (obtained when secondary network is not present), a considerable throughput can be gained by the secondary network. Results also show that the sum throughput also increases as a result of this optimization.

The problem of location estimation and its accuracy in wireless networks with fixed beacons has been studied extensively. However, the problem becomes more challenging when the beacons are non-stationary, or when the location of the beacons is not precisely known. In this paper, we analyze the effect of the uncertain beacon positions on the localization error. First we examine the performance of the conventional localization techniques. Subsequently, we study the effectiveness of a novel technique, which utilizes the topological diversity provided by the various combinations of beacon nodes. Results from our modeling effort show that this approach helps reduce the errors caused by imperfect knowledge of beacon positions. The results also indicate that the use of this approach improves the robustness of location estimation as compared to existing methods.

In spectrum underlay cognitive wireless networks, secondary nodes need to limit their aggregate interference on the primary receiving nodes. The trends for interference modeling has been either indiscriminate use of Central Limit Theorem to model the aggregate interference as a Gaussian random variable or by application of the Campbell's Theorem and approximating the probability density function of interference from its cumulants (e.g., by using Edgeworth or Gram-Charlier series). In the latter case, the theorem can be applied only when the interfering nodes have the same power level. In this paper, we deviate from the previous trends of interference modeling in following ways: (1) We allow the secondary neighbors of a primary node to have arbitrary power levels. (2) We split the set of interfering neighbors of a primary node into non-Gaussian (close neighbors) and Gaussian (far neighbors) interferers. For the case of Log-normal fading, we show that an accurate model for interference is sum of a Normal and a Log-normal random variables. We proceed to obtain an upper bound for the complementary cumulative distribution function of interference and show its tightness through simulation. Simulations results confirm the accuracy of the proposed model. Finally, we propose adjustable interference avoidance strategies and show that interference constraint is satisfied using these strategies.

Non-line-of-sight (NLOS) signal propagation is a major issue in location estimation. Several application oriented approaches have been proposed in recent years to mitigate the errors induced by the biased distance estimates from NLOS propagation. In this paper, we present a new algorithm which reduces the adverse effects of bias by utilizing the topological diversity provided by L unique beacon combinations from the power set of N(>3) beacons. The algorithm consists of two complementary weighted average techniques combined to provide high level of robustness. Our simulation and empirical work shows that the proposed algorithm is effective independent of the bias distributions. Thus, it can be implemented over a variety of existing localization methods in wireless networks to mitigate the adverse effects of NLOS propagation and achieve high accuracy in practical scenarios.

In this paper, we consider N ad hoc networks, with nodes randomly and uniformly distributed, coexisting and sharing an unlicensed band. The objective of this paper is the optimization of the transmission probabilities in these networks to maximize the throughput in each network. Throughput in each system is limited by the self interference (the interference inflicted by a network on its own nodes) as well as the interference from the other networks. We consider the coexisting wireless networks being independent and random. Each network uses a random access method with its own transmission attempt probability. We show that the throughput in each network depends on the choice of transmission probabilities of all networks. We use the game theory approach to solve the transmission probability control problem. Simulation result indicate that at the Nash equilibrium a closely optimal solution can be obtained. We also consider the fairness and show that the transmission chances of systems are fairly equalized at the NE solutin. This contrasts with the centralized solution in which there is high discrepancy among the transmission probabilities.

Game-theoretic analytical techniques have been applied, of late, to uplink power control in DS-CDMA networks. We extend our previous analysis [1] of game-theoretic power control under successive interference cancellation performed under a dynamic cancellation ordering. Both continuous and discrete power control are investigated. Under continuous power control, we prove the nonexistence of a symmetric power control equilibrium, illustrating that no easily applicable equilibrium existence result applies to this power control problem. Under discrete power control, we demonstrate empirically a high probability of equilibrium if the action set is not too precise, an encouraging result in the absence of a pure-strategy equilibrium guarantee.

In cognitive radio networks, we assume that primary and secondary nodes form a bivariate spatial Poisson point process. In this paper, we derive the probability distribution and some statistics of the internodal distances between nodes of different types. When the number of primary and secondary nodes are weakly correlated, we devise two power control strategies with (i) constant power levels and (ii) power levels decreasing/ increasing with distance, which assure satisfaction of interference constraint at the primary nodes with some given probability. For the second strategy, it is sufficient for the nth secondary neighbor of a primary node to know n in order to satisfy the interference constraint. Furthermore, n can be found if secondary nodes exchange their relative positional information and can be assumed fixed if the rate of change in relative nodal positions, due to mobility of nodes, is slower than the minimum time required for power control.

This paper addresses the problem of end-to-end path computation in a transport network with multiple switching technologies, for which the label switched path (LSP) traffic engineering (TE) in the multi-layer networks is an important application. By transforming a network graph to a channel graph, we provide a novel and general solution to find optimal paths in multi-layer networks through vertically searching across layers and horizontally searching on the same layer. The channel graph yields an explicit view of the constraints associated with the nodes and links that otherwise are hidden in the network graph. The approach can be applied to constraints such as the wavelength continuity, encoding type, and switching bandwidth granularity. The proposed solution has been implemented in software and deployed in an experimental optical network.

In hybrid optical/packet networks, wavelength and VLAN tag continuity along a label switched path are two common constraints. These two types of constraints, referred to as generalized label continuity constraints, have global significance in a network; namely, the label should appear only in other link- or node-disjoint paths. In considering these constraints, a solution is the common vector approach, which seeks a common available label along the path determined by CSPF without label constraints. In this paper, we provide an estimation method to compute the upper bound of the blocking probability for the above approach. We also address the scalability problem of the existing traffic engineering algorithms. We then present the simulation results corroborating with our analytical model.

A new breed of network intensive "e-science" applications are emerging that require a new perspective on how networks provide resources to the user. These e-science applications often have large raw capacity requirements (tens of gigabits/second) and require global reach. The discipline scientists expect this type of performance to be deterministic and repeatable, and available in conjunction with other resources such as sensors, instruments, computational clusters, storage arrays, and the like. The DRAGON project, an NSF funded testbed in the Washington DC metropolitan area, has been developing open source generalized multi-protocol label swapping (GMPLS) based service concepts that provide dedicated and deterministic network resources to the e-science application. Using the GMPLS control plane and DRAGON extensions, an entire "application specific topology" can be established in an automated fashion across administrative boundaries, across heterogeneous network technologies, and can be instantiated in seconds rather than weeks or months. This paper describes the technologies being developed by the DRAGON project and some of the ways application specific topologies are being applied to advanced networked applications.

We consider a ring topology with limited or full switching capability as deployed in high bandwidth metro optical networks and develop a model to estimate the probability of blocking for interconnecting links. The model is based on the lower and upper bounds of link blocking probability. We analyze the performance for a homogeneous traffic case and present simulation results for representative ring networks. We demonstrate that the bounds are very tight with an error of less than 2% when the traffic load is modest. The approach is based on partitioning the state space into subspaces and weighting the upper bound of blocking probability in each subspace with the occurrence of the states. The computational complexity of the approach is comparable to solving a degree of N polynomial equation.

Emerging network applications tend to be built over heterogeneous network resources spanning multiple management domains. Many such applications have dynamic demands for dedicated, deterministic, high-bandwidth connections. The generalized multi-protocol label switching (GMPLS) networks under development can address these kinds of demands by using policy-based resource management and service provisioning technologies. In this paper, we present the architecture and implementation for policy-based resource management and service provisioning as part of the work on the NSF funded dynamic resource allocation via GMPLS optical networks (DRAGON) Project. This work captures several critical features of service-oriented GMPLS networks, including a) collaborative interdomain resource management; b) interdomain end-to-end path computation; c) advance scheduled provisioning; and d) authentication, authorization and accounting (AAA). These features rely on the capability of exchanging and coordinating resource and policy information among multiple participating domains. In pursuit of this capability, we propose a three-dimensional (3D) resource computation model (RCM). The three broad dimensions of this model are traffic engineering (TE) constraints, time schedule constraints, and AAA policy constraints. The 3D RCM facilitates policy based resource allocation based on many specific constraints within these three broad categories. Based on this model we describe our approach to GMPLS interdomain end-to-end path computation, advance scheduled provisioning, and AAA policy based provisioning, respectively. Our current DRAGON implementation status is also reported in this paper.

Inter-domain path computation, or the ability to compute end-to-end paths across multiple domains, is the next step toward wide deployment of a distributed control plane with support for traffic engineering. A key enabler to achieve this goal is the introduction of a Path Computation Element (PCE) in each domain. There are various ways these elements can collaborate to compute an end-to-end path; of particular interest to us in this paper is cooperative path computation, a scheme where PCEs exchange path information in the context of a specific end-to-end path computation instance, often prior to signaling the path. We show that depending on the information available to each PCE, cooperation can take one of two forms, which we call model-based and ad hoc. We demonstrate that model-based cooperation is essentially a multistage decision problem, and offer a probabilistic analysis which we believe is the key to understanding the problem and developing efficient inter-domain path computation heuristics. In particular, we argue that having an estimate of the blocking probability in each domain can be helpful in determining the path computation effort needed to find an end-to-end path.

A power control framework for selfish users in a wireless DS-CDMA uplink with imperfect successive interference cancellation is offered. After presenting a feasibility constraint and a centralized power control solution, the system is analyzed and simulated as a noncooperative game for both continuous and discrete power control cases. Assuming a fixed a priori cancellation order, a unique Nash equilibrium corresponding to a centralized solution is shown to exist in the continuous case. It is shown that the realizable gains of the SIC receiver in a game-theoretic context, although strongly dependent on spreading gain and residual cancellation error, nevertheless are significant when compared to traditional matched filter detection, as expected. If the action space is discrete and sufficiently coarse, empirical results indicate a high probability of Nash equilibrium existence.

A stochastic game-theoretic framework for calculating transmit power strategies for nodes in a competitive CDMA wireless network is presented. Unlike traditional game-theoretic power control, delay sensitivity is explicitly included. Stationary Nash equilibria that are a function of congestion and that allow for multirate transmission are determined. In both static path loss and dynamic shadowing path loss environments, the stochastic game equilibria calculated compare very favorably to traditional single-agent and often approach the performance of centralized optimization

We consider the problem of throughput modeling of wireless multihop packet CDMA networks with cellular overlay using simple forwarding strategies in the upstream. Considering the effect of shadowing and distance-dependent path loss, we approximate the probability density of interference at each base station (BS) and compare numerical and simulation results for different path-loss parameters. We derive the probability density of the received power at each BS due to transmission of one packet from a random node, as well as the probability distribution of the number of packets received at each node per time slot. Subsequently, we use the above results to approximate the probability density of the total received power at each BS based on calculations of moments. We observe that the probability density of intercell interference due to transmissions from terminals and routers may be approximated by normal and log-normal densities, respectively. We quantify the network performance based on throughput, total consumed power, and outage probability for different system parameters. For homogeneous link efficiencies, introducing routers into the network while reducing the transmission power increases the mean and variance of interference to the desired signal, hence higher outage probability. However, there are ample opportunities inherent to multihop structure, applicable to any of the physical, data link, and network layers, which help increase the overall achievable network throughput.

The demand assigned capacity management (DACM) problem in IP over optical (IPO) network aims at devising efficient bandwidth replenishment schedules from the optical domain conditioned upon traffic evolution processes in the IP domain. A replenishment schedule specifies the location, sizing, and sequencing of link capacity expansions to support the growth of Internet traffic demand in the IP network subject to economic considerations. We develop mathematical models to address the DACM problem in IPO networks within a deterministic context based on a class of inventory management replenishment methods. We apply the technique to IPO networks that implement noncapacity adaptive routing in the IP domain and analyze the performance characteristics in terms of minimizing cumulative replenishment cost over an interval of time. This study represents an application of integrated traffic engineering which concerns collaborative decision making targeted towards network performance improvement that takes into consideration traffic demands, control capabilities, and network assets at different levels in the network hierarchy.

We present the development of a framework for a high capacity wireless network for Internet applications. By replacing some of the network nodes with wireless routers, we form a wireless multihop network overlaid an a cellular structure. This paper addresses the problem of jointly optimizing routing, channel scheduling, and power control to maximin the total system throughput under a transmit power constraint. We demonstrate that by considering the interrelationship among the network, data link, and physical layers, we increase the network throughput. The work is based on minimizing a defined objective function which includes the cost related to the transmit power for emptying the buffer with a certain amount of information and the selected route to the final destination.

Provides an overview of the technical articles and features presented in this issue.

We consider the problem of throughput modeling of multihop wireless networks with a cellular overlay using simple forwarding strategies in the upstream. Considering the effect of shadowing and distance loss, we derive a closed-form formula for the probability density function of power and interference at each receiver and compare the numerical and simulation results for different path loss parameters. When link efficiencies of all connections are the same, relaying packets increases the mean and variance of interference and hence results in higher outage probability. However, using capacity enhancing techniques at the physical, link and network layers, we are to benefit from the reduced transmit power due to multihop operation and further trade off power for additional capacity.

Internet traffic engineering is concerned with the performance optimization of IP networks. Multiprotocol Label Switching (MPLS) is an important technology that supports the Internet traffic engineering function. This paper considers a network performance optimization problem related to traffic engineering over MPLS. We investigate the use of a dynamic traffic partitioning and assignment methodology to adaptively map ingress traffic into several parallel Label Switched Paths (LSPs) in MPLS based IP networks. We present a stochastic framework for the traffic partitioning problem. Within this framework, a set of parallel edge disjoint LSPs is modeled by parallel queues and a partitioning algorithm is devised for different service classes that is adaptive to the prevailing state of the network. The performance of this approach is illustrated with numerical examples.

We investigate and evaluate the performance of two scheduling algorithms for forward link packet access: in CDMA based cellular networks. Both algorithms use round-robin scheduling, ensuring fair access to users in overlapping and non-overlapping areas. One method provides time diversity whereas the other provides time as well as space diversity to users located in the handoff regions. The performance is also compared with that of a conventional hard handoff.

Packet transfer using a connectionless mode provides considerable simplification in mobility and handoff functionalities in wireless networks. We develop an architecture using label switched packet forwarding to enhance traffic control and accommodate service differentiation in the network. We present the protocol structure, mobility support, and routing, which are essential in this architecture. The use of the label merging technique is discussed to support multiple radio links between a mobile and a multiplicity of base stations. Preliminary performance results are presented for the effect of diversity areas on the admission control policy.

A performance model of a reservation medium access protocol is developed for wideband CDMA cellular networks. The protocol is implemented to perform statistical multiplexing considering different bit rates and QoS requirements. An analytical method has been presented to calculate users' power and maximum bit rate in a heterogeneous traffic environment. A Markovian approach is used to model the system behavior. System performance is evaluated considering packet dropping by excessive delay and packet corruption due to interference for voice only traffic in a multi-cell environment.

Third generation wireless networks are expected to carry traffic from heterogeneous sources of voice, data, video or multimedia with a specified quality of service. We investigate how packet transfer mode over the radio interface can increase the efficiency of radio resources when there is multiple coverage of overlapping cells. Our discussion is focused primarily on multiple link packet protocol operation and its performance. We present an analysis of the system performance using a Markov modulated Poisson process (MMPP) for the aggregate of ON-OFF state traffic and for the overflow processes and quantify the performance improvement through a numerical example.

Hierarchical structuring of cellular systems is an effective way to address the problem of continuing growth in traffic volume from users with varying degrees of mobility in wireless networks. Overflow and take-back strategies integrate the two tiers (microcell and macrocell) smoothly, within which both microcells and macrocells serve as a secondary resource to each other, taking into account the different mobility classes the subscribers belong to. We develop an analytical model based on MMPP (Markov modulated Poisson processes) to evaluate the performance of such hierarchical systems. We obtain the probability of call loss and forced termination for calls in progress and we contrast them to those from systems with incremental traffic handling capabilities in overflow and take-back.

Hierarchical cellular networks with subscribers of varying mobility are considered. Microcells are used to address the high-intensity traffic of mainly slow mobility areas, and macrocells are overlaid over the microcells to cater mainly to high-mobility lower density traffic. The two tiers of microcells and macrocells provide a secondary resource for new traffic as well as handoffs for mobile subscribers of different mobility classes. Furthermore, resources in alternate layers are monitored to assign the appropriate resource types when they become available. We develop an analytical model to evaluate the performance of such systems, and quantify the gain obtained by providing overflow to alternate resources as well as the advantages in resource reassignment according to the speed classification.

Guglielmo Marconi was born on April 25, 1874, in Bologna, Italy. He contributed significantly to the development of wireless communications technology, demonstrated for the first time a successful setup for radio telegraphy (in 1895), and was a pioneer in the application of electromagnetic waves. Radio communications is one of the most fascinating and perhaps spectacular successes of mankind in advancing science and technology. Few technological developments have had such great and far-reaching consequences as wireless communication. Founded upon the great scientific discoveries of the nineteenth century, it created a spectrum of technologies, each requiring scientific insight of its own, and fostered innovations in many related disciplines. Wireless communication involves technologies with huge potential for aiding world progress, improving the quality of life for every human being on this earth and leading to a better future for mankind. Many persons have contributed to the development of radio communications [1], [2]. The theoretical foundation of electromagnetic waves was laid out brilliantly in 1864 by J. C. Maxwell, the British physicist who introduced the basic equations to relate the electric and magnetic fields and developed the foundation for electromagnetic waves. Maxwell’s ideas, however, had to be confirmed. In 1887, H. Hertz, a physics professor in Karlsruhe, Germany, carried out laboratory experiments and demonstrated the existence of the electromagnetic waves as predicted by Maxwell. Telegraphy, i.e., the transmission of Morse codes representing different characters via wire line, had already been in operation, and the idea of deploying wireless radio for distant sources was clearly present. Furthermore, the importance of using the propagation of electromagnetic waves for radio communication was known and its development had been predicted. For example, W. Crooke in 1892 had written about radio telegraphy at a wavelength of a yard or more [3].

The performance of an asynchronous transfer mode (ATM) multiplexer is evaluated, with the aggregate arrivals modeled as a Markov-modulated Poisson process (MMPP). The analysis is based on two simplifying assumptions: the probability that the MMPP goes through multiple state transitions between two successive departures is negligible, and state transitions occur at departure points. The transition probability matrix that describes the number of cells in the buffer after a departure can then be partitioned into submatrices, each of which is analogous to that of an M/D/1/K queue. These assumptions are reasonable for ATM traffic models in which the arrival rates are large and cell size is small. The accuracy of the analysis is evaluated, using a four-state MMPP model to represent the aggregate arrival process. The departure point and arrival point queue-length distributions, cell loss probabilities and average queuing delays are obtained analytically and compared to simulation results.>

A multiscale video representation using wavelet decomposition and variable-block-size multiresolution motion estimation (MRME) is presented. The multiresolution/multifrequency nature of the discrete wavelet transform makes it an ideal tool for representing video sources with different resolutions and scan formats. The proposed variable-block-size MRME scheme utilizes motion correlation among different scaled subbands and adapts to their importance at different layers. The algorithm is well suited for interframe HDTV coding applications and facilitates conversions and interactions between different video coding standards. Four scenarios for the proposed motion-compensated coding schemes are compared. A pel-recursive motion estimation scheme is implemented in a multiresolution form. The proposed approach appears suitable for the broadcast environment where various standards may coexist simultaneously.>

The statistical characteristics of full-motion video sources using motion-adaptive variable-bit-rate coding techniques is studied. Analytical models are developed to describe the behavior of the coded video signals based on the encoder structure. The video-compression algorithm used is in compliance with the general MPEG syntax and bit-stream definition. Statistical characteristics associated with each block type and their aggregate are presented. A composite model to represent the number of bits per field for the encoded video traffic that comprises multiple autoregressive models for the number of blocks per field and the number of bits in each coded block is derived. The statistics measured from a sample video sequence are compared to those obtained by the model and it is observed that the model captures the coded video behavior for each block type and their combination reasonably well. This model can be used to further study the cell-generation process of full-motion video codecs and the aggregation of such video sources at the statistical multiplexers.>

A motion-adaptive variable-bit-rate (VBR) video codec is considered, and a motion-classified model is developed to represent the characteristics of various classes of motion activities, including scene changes. The codec switches between interframe, motion-compensated, and intraframe coding corresponding to low, medium, and high amounts of motion and scene changes, respectively. The model captures the motion of various video scenes by providing the statistics of VBR-coded video traffic through a first-order autoregressive process with time-varying parameters. The parameters of this model are obtained from a VBR-coded sample video sequence with the objective of matching the bit-rate distribution and the autocorrelation among the bit rates. The validity and accuracy of the model are evaluated, and the characteristics of aggregated traffic sources obtained with the model are discussed.>

The authors use a motion adaptive variable-bit-rate (VBR) video codec and propose a motion classified model to represent the characteristics of various classes of motion activities. The codec switches between interframe, motion compensated, and intraframe coding corresponding to low, medium, and high motions and scene changes, respectively. The model captures the motion of various video scenes and the codec structure by providing the statistics of VBR-coded video' traffic through a first-order composite autoregressive process with three motion classes. The parameters of this model are derived from a VBR-coded sample video sequence such that the bit rate distribution and the autocorrelation in bit rates of two successive frames are matched. The validity and accuracy of the model are verified. Using this model, the characteristics of aggregated traffic sources are discussed.>

The authors develop an analytical model of satellite communication networks using time-division multiple-access (TDMA) and multiple-carrier TDMA (MC-TDMA) systems to support circuit-switched traffic. The model defines the functions required to implement fixed assignments (FAs), variable destinations (VDs), and demand assignments (DAs). The authors describe a general system model for the various allocation schemes and traffic activity. They define analytical expressions for the blocking and freeze-out probabilities. This is followed by the derivation of the satellite capacity requirements at a specified performance level for FA, VD, and DA systems with and without digital speech interpolation. The analysis of disjoint pools and combined pools in DA systems is presented and attention is given to MC-TDMA with limited connectivity demand assignment. Expressions for the required satellite capacity for specified traffic and performance are derived along with numerical results. The degree of complexity and implementation alternatives for the various allocation schemes are considered.>

The authors present modeling of an integrated terrestrial-satellite network using very small aperture terminals (VSATs), for transaction-oriented computer applications. They address the analytical and simulation modeling techniques used to study the network performance. More specifically, these models are used to obtain network response times for an integrated terrestrial-VSAT network for inquiry-response applications under various load conditions at the cluster controller units. The major result obtained through this analytical and simulation modeling suggests that communications satellite channels can be used efficiently for the interactive inquiry-response traffic at an acceptable response time, provided that the system parameters including the link transmission rates are chosen appropriately.>

A method of improving the quality of service in mobile cellular systems based on prioritization of handover requests is presented. The objective is to improve perceived quality of cellular service by minimizing both the probability of forced termination of ongoing calls due to handover failures and the degradation in spectrum utilization. A model based on a multiple-priority nonpreemptive queuing discipline is developed. New calls are blocked if all channels are occupied. Handover requests are queued such that as soon as a channel is available, it is offered to the mobile subscriber with the measurement results closest to the minimum acceptable power level for communication. Service rate is given by channel occupancy time distribution and is assumed to be exponential. The performance of a cellular system employing the proposed handover policy is evaluated analytically and by simulation, and results are compared to those obtained when the cellular system employs nonprioritized call handling and first-in/first-out queuing discipline. This provides lower probability of forced termination and less call blocking, less reduction in traffic, and less delay.>

The routing and congestion control function of Signaling System No.7 (SS7) are described. The elements of the SS7 protocol functional division include message transfer part (MTP), signaling correction control part (SCCP), ISDN user part (ISUP), and transaction capabilities (TC). The routing function, which takes place at the MTP and SCCP, and the congestion control function, which is present in multiple layers, are discussed. This includes MTP level 2, MTP level 3 by signaling traffic flow control procedures, and flow control for connection-oriented services of SCCP. To illustrate the unique capabilities present in SS7, the routing and congestion control functions in SS7 are compared to other common connectionless network layers. Performance considerations in routing and congestion control are discussed.>

The authors discuss the architecture and protocol for broadband ISDN (B-ISDN) based on the CCITT standards. The discussion attempts to address the general concept of B-ISDN architecture and protocol and, whenever possible, present alternatives and the rationale for decisions in the selection of the protocol. B-ISDN is presented as a network evolution, and the impact of the asynchronous transfer mode (ATM) on the network is described. The role of virtual channel and virtual path in B-ISDN is discussed. The B-ISDN protocol structure and lower layer functions comprising the physical, ATM, and adaptation layers are presented, and the tentative trends of signaling and traffic control for the B-ISDN are delineated.>

Various functional parts of signalling system number 7 and the underlying concepts are described. The unique attributes of the lower and higher functional layers are discussed. Applications of signaling system number 7 for call control and for transaction services are presented. The signaling transfer point makes signalling networks possible. Using the lower layers of the protocol, it provides routing capability for signaling messages between the exchange offices and access of these exchange offices to the network databases. The author discusses the performance parameters associated with the signalling transfer point and examines its implementations.>

This paper focuses on the performance of the time-division multiple-access technique with fixed allocation scheme when the memory buffer size at each ground station is finite. Specifically, a queueing model is formulated in order to derive the expected waiting time of the messages in the buffer and the probability of message blocking. Two cases are discussed. single-packet messages and multipacket messages, both with multiple time slots available to the ground station. Numerical results also are provided to facilitate determining the best tradeoff between memory buffer size and the performance parameters. These results suggest that assigning a single slot per ground station is more efficient than multiple slots.