This paper presents current and latest schemes for location management in next generation wireless systems (3G and 4G). First, the hierarchical reference model of 3G wireless systems is introduced. Then a dynamic inter-system location management technique is developed based on new concepts of boundary location area and boundary location register, which are suitable for inter-system roaming scenarios. Moreover, the corresponding mobility application part (MAP) protocol is designed for inter-system location registration and call delivery procedures. The performance is evaluated in terms of signaling costs, latency of location registration and call delivery, as well as call losses due to the inter-system roaming. For the 4G systems, a generic system model is introduced to incorporate the existing wireless local networks and cellular systems and so on. In particular, a new scheme for reducing costs in route optimization is introduced to solve the above problems. In this new scheme, route optimization is performed only when it minimizes the total cost function, which provides the optimal result from the viewpoint of link and signaling costs. The simulation results show that the proposed scheme provides the better performance.

We present a brief account of three problems in wireless networks. First, on the problem of scalability, we show that under a certain model of interference, a network with n nodes can only provide a per-node throughput of order square-root of n bits/sec. However, an experimental investigation with IEEE 802.11 technology yielded the considerably worse power law of n raised to -1.68. We then present a new media access control protocol for ad hoc networks, called SEEDEX. It attempts to make reservations without explicitly doing so for each and every packet. Nodes use random schedule, and can inform each other about their schedules merely by exchanging the seeds of their pseudo random number generators. Last we turn to the problem of power control in ad hoc networks. We present a new protocol called COMPOW, which is designed to find the lowest power level at which the entire network is connected. We provide the theoretical justification for it, and a solution of the architectural issues involved. Using a notion of parallel modularity at the network layer, this protocol provides a joint solution for both the routing and power control problems. It has been implemented in the Linux kernel.

In this work, we consider the problem of utility function-based resource allocation when a mixture of reallocation-tolerant and reallocation-intolerant users are present. Unlike reallocation-intolerant users, reallocation-tolerant users can be reallocated a different amount of resource during the course of their call. We develop a resource allocation mechanism that maximizes the average aggregate utility per unit time. By formulating the resource allocation problem as a Markov decision process (MDP), we determine the optimal quantity of resource to be allocated to newly arriving and the optimal reallocation of resources to reallocation-tolerant calls whenever there is a change in the state of the system. We present numerical results that show that our resource allocation scheme performs better than the greedy resource allocation scheme. To reduce the computational complexity involved in determining the optimal policy, we identify problem-specific model reduction techniques that do not compromise the optimality of the solution.

We consider dynamic resource allocation problems that arise in wireless networking. Specifically transmission scheduling problems are studied in cases where a user can dynamically allocate communication resources such as transmission rate and power based on current channel knowledge as well as traffic variations. We assume that arriving data is stored in a transmission buffer, and investigate the trade-off between average transmission power and average buffer delay. A general characterization of this trade-off is given and the behavior of this trade-off in the regime of asymptotically large buffer delays is explored. An extension to a more general utility based quality of service definition is also discussed.

In a cellular wireless communication system, channel bandwidth is a scarce resource. Even if the network design takes into account the average traffic pattern in a geographical area, time-dependent variation of traffic often creates load imbalance (commonly known as hot spots) in different parts of the network. Some attempts have been made to alleviate the time-varying network congestion in cellular networks. However, because of co-channel interference problem associated with the concept of frequency re-use in a cellular system, maximum possible channel capacity gain cannot be achieved through these conventional load balancing schemes. If somehow the co-channel interference problem could be avoided while balancing the traffic load in different cells, the maximum achievable capacity gain could be attained. In this paper, we provide an analytical framework to show the maximum possible capacity gain in a cellular system. Through an analogy of fluid flow among the connected reservoirs, we study the load balancing dynamics. An example of 3-tier cellular structure is used to demonstrate that the maximum achievable capacity gain could be as much as 69%. In an identical scenario, the best known technique such as load balancing with selective borrowing achieves a maximum capacity gain of only up to 18%.

The efficient management of the radio resource of a 3-G system is important from an operator's perspective. This, however, cannot be the only concern when quality of service (QoS) negotiations have been made for various users and the operator has to uphold these. This leads to a fairness objective that the operator has to keep in mind. In this paper we outline a scheme to perform packet-level scheduling and resource allocation at the wireless node that takes into account the notions of both efficiency and fairness and presents a means to explore the trade-off between these two notions. As a part of this scheme we see the scheduling problem as deciding not just the packet transmission schedule but also the power allocation, the modulation and coding scheme allocation and the spreading code determination since the latter three directly influence the radio resources consumed. Using a utility maximization formulation based on the data-rates that the mobiles can transmit at, we decide on the weights for a weighted proportionally fair allocation based scheduling algorithm. We also show how one can adapt the weights and the algorithm for a time-varying channel. We conclude with a simulation based performance analysis for infinitely-backlogged sources and TCP sources on an EDGE system.

In this paper, we consider a degraded Gaussian broadcast channel, where the transmitter maintains separate queues for each receiver. We present throughput optimal policies that stabilize the queues without knowing the statistics of the arrival processes to these queues.

Traditionally, voice has been transported using circuit switched networks, the Public Switched Telephone Network (PSTN) for example. However, driven by the ubiquity of the Internet and the development of low bit-rate digital voice codecs there has been increasing focus on using packet-switched networks for voice traffic. We focus on one such application. Our model comprises of a relatively slow packet link (between 1.5 and 5.0 Mbps) being utilized for voice traffic. Large number (of the order of 100) voice sources are multiplexed on this link. For such a link we obtain delay distributions seen by a voice source. Specifically, we exploit key characteristics of the model, such as the large number of sources to obtain a heavy traffic approximation for the system. The key result is that the delays can be well approximated by concatenation of two exponential distributions. We also provide valuable insight into how the delay distribution is connected with the statistical properties of the voice sources, in particular their correlation behavior. Our analytical results are validated with simulation results.

Providing packet-level quality of service (QoS) is critical to support both rate-sensitive and delay-sensitive applications in the bandwidth-constrained, shared-channel, multihop wireless networks. This problem is challenging due to the unique issues such as location-dependent contention, inherent conflict between ensuring fairness and maximizing channel utilization, and the distributed nature of packet scheduling in such networks. In order to address these issues, we have taken a new self-organizing approach to QoS solutions for such networks. In this approach, local decision makers self-organize themselves and coordinate among one another, and collectively achieve the desired global property. Some features of our approach include fully localized design, coordination among local decision makers, intentional and optimized information propagation, scaling property and achievable global property. Two key contributions of this work are: (a) a model-referenced self-organizing design methodology for multihop wireless networks; and (b) a table-driven approach and a backoff-based approach to distributed packet scheduling that provides QoS performance bounds in terms of fairness, throughput and delay, maximizes channel spatial reuse, and arbitrates the conflict between fairness and maximal channel utilization. oth proposed designs work within the CSMA/CA MAC framework. We also compare the performance of these two approaches through simulations. Our extensive simulation results have confirmed the effectiveness of the proposed design.

Recent advances in wireless communications and MEMS technology have led to the development of wireless sensor networks with a large number of small sensing devices. Most of these wireless sensor networks are motivated by the increases in sensing coverage and performance that can potentially be achieved through effective sensor data fusion. However, as the density of a wireless sensor network grows, the likelihood of transmission collisions among sensor nodes can increase simultaneously. In this paper, we consider a class of wireless sensor networks and show that there can exist fundamental limits on their performance based on recent results on the capacity of wireless networks. We study a simple distributed detection problem with wireless sensor networks and investigate the scalability of several common data fusion architecture under appropriate assumptions on the sensor models.

This paper discusses the design of soft handoff algorithms for cellular communication systems based on signal-strength measurements. The handoff process is modeled as a hybrid system and handoff design is cast as an optimization problem based on such a model. Performance is evaluated in terms of call quality, average number of active base stations and average number of active set updates. Preliminary designs based on selective diversity combining are investigated. A suboptimal handoff algorithm, which achieves a tradeoff between these three performance criteria, is obtained using principles of dynamic programming.

In this paper, we study transmission techniques for broadcast channels with a single transmitter and multiple receivers. The transmitter is assumed to be equipped with multiple antennas. Further, each receiver is interested only in part of the transmitted information. Using Gaussian code based information theoretic bounds, we analyze two transmit beamforming techniques. The first technique, zero-forcing beamformer, uses the channel information for all users to send spatially orthogonal signals to different users, i.e., no user receives interference from the other user signals. The second method uses the single-user optimal beamformer with no effort to reduce interference at the receivers. It is shown that the above transmit beamforming techniques are similar to multiuser receivers used in non-orthogonal CDMA systems. In particular, the zero-forcing beamformer is similar to a decorrelating detector and the single-user beamformer is identical to a matched filter. The comparison with CDMA multiuser receivers is strengthened by results on spectral efficiency of proposed beamformers, which follow the behavior of decorrelating and matched-filter receiver.

A Monte Carlo simulation was conducted to assess the performance of Multi-Carrier Code Division Multiple Access (MC-CDMA) in a sectored cell, in an indoor environment. In each sector, terminal schedule packet transmission using slotted p-persistent Inhibit Sense Multiple Access, while the packets themselves are transmitted using MC-CDMA. The simulated bit error rate performance of MC-CDMA with combining strategies maximal ratio combining (MRC) and equal gain combining (EGC) used in the frequency domain are presented. A comparison is made between analytically determined bit error probability for a single user in an additive white Gaussian noise channel using binary phase shift keying (BPSK), and simulated bit error rate for a single user using MC-CDMA with MRC and EGC. The determination of the number of sectors in a cell is also given. MC-CDMA with diversity combining outperforms BPSK in a non-fading AWGN channel. Further, as the user-number increases, EGC performs better than MRC if these combining strategies are used with MC-CDMA. Finally, the results show the basis for determining the number of sectors in a cell. When using slotted p-persistent ISMA, it is envisaged that only a single user will transmit at a time in a sector.

In wireless communication systems, power control is a popular technique that can be employed to greatly increase system performance. The general idea is to adjust the transmitted power of each user based on channel characteristics (e.g. path-loss and shadowing) so that each user is received with his/her desired signal-to-interference ratio. Traditional approaches to power control are generally derived assuming that the pertinent channel characteristic on which the power control algorithms are based can be measured accurately. However, due to the effects of the multipath fading and the additive noise inherent to the wireless channel, there can be significant errors in measurements of the average received power. In this paper, the error statistics for average power measurements are considered; in particular, the probability distribution of the value of the average received power at the time of interest conditioned on an outdated measurement is obtained. The resulting expression should have high general utility in the analysis of wireless communication systems; however, in this paper, this expression is employed in the design of optimum power control algorithms that minimize the average transmitted power required to achieve a desired outage probability for a wireless link. It is demonstrated that power control algorithms that accurately take into account uncertainty in the average power measurements can display significant gains over those that employ simple energy margins or approximate models for such purposes.

We consider power control for voice users in a wideband CDMA wireless network. We investigate admission control policies that base a new call admission decision not only upon available capacity but also upon the required downlink transmit power and upon the user's willingness to pay. We assume that each voice user has a utility function that describes the user's willingness to pay as a function of downlink SINR. The network is assumed to desire to either maximize total utility of all users or total revenue generated from all users. We present a numerical study of a single cell. We display the optimal power allocation to each user, as a function of the geographical distribution of users, for a selection of different utility function distributions. We demonstrate how prices per code and per unit transmitted power can be used to achieve the optimal power allocation in a distributed fashion, and the variation of these prices with system load.

Power control has been recognized as an essential requirement in the design of cellular spread spectrum systems, since only by control of transmit powers can users share radio resources equitably and efficiently in a multicell environment. Much of the work on power control for spread spectrum systems found in the literature focuses on static channel models, i.e., models in which the channel gain of every user is assumed constant. In this paper, the design of dynamic power control algorithms for spread spectrum systems is considered. The design problem is posed as a tradeoff between the desire for users to maximize their individual capacity and the need to minimize the transmitted signal energy over the duration of their calls. The dynamic nature of the wireless channel for mobile users is incorporated in the problem definition. Based on a cost minimization framework, an optimal multiuser solution is derived. The multiuser solution is shown to decouple, and effectively converge to a single user solution in the large system asymptote, where the number of users and the spreading factor both go to infinity with their ratio kept constant. The proposed solution for large systems is a simple decision rule that is easily implementable in a practical system.

We consider wideband CDMA networks with many users, and study the system performance when blind linear multiuser receivers are employed for demodulation. We first characterize the signal-to-interference-plus-noise ratio (SINR) for blind multiuser receivers. Our results reveal that the SIR of these blind receivers saturates when the power of the desired user increases, which is in stark contrast to the fact that the SIR achieved by the exact MMSE receiver can get arbitrarily large. This saturation phenomenon of SINR indicates that the capacity of a wireless network with blind multiuser receivers is not only interference-limited, but also estimation-error limited. Furthermore, the effect of estimation error can be quantified. We then show that the residual interference at the output of these blind receivers is asymptotically Gaussian. The Gaussianity enables easy characterization of bit error probability. A parallel result is that the estimation error of the receiver is asymptotically orthogonal to the signal space.

In this paper, we analyze the performance of space-time trellis codes. In particular, we derive the union bound for space-time trellis codes over quasi-static fading channels. We first observe that the standard approach for evaluating the union bound yields very loose, in fact divergent, bounds over the quasi-static fading channel. We then develop a method for obtaining a tight bound on the error probability. We derive the union bound by performing expurgation of the standard union bound. In addition, we limit the conditional union bound before averaging over the fading process. We demonstrate that this approach provides a tight bound on the error probability of space-time codes. The bounds can be used for the case when the fading coefficients among different transmit/receive antenna pairs are correlated as well. We present several examples of the bounds to illustrate their usefulness.

We consider the problem of adaptive modulation and receiver design for DS-CDMA fading channels with imperfect channel state information. The goal is to adapt the geometry of the signal constellation within the subspace spanned by a small number of spreading codes (multicoding) in order to maximize throughput, minimize power consumption, or control bit error rate in the presence of uncertainty regarding instantaneous channel state information. We study two possible cost functions based on J-divergence and investigate their relationship to effective system capacity via bounds on the mutual information between the channel input and output given an estimate of the channel state information. The bounds will be used in future analytical and simulation studies to evaluate the performance improvement associated with the proposed adaptive modulation scheme.

More and more network operators plan to use broadband satellite systems as part of their global broadband multimedia infrastructure because of their large geographic coverage, inherent broadcast capabilities, and fast deployment features. The recent advances on Ka-Band transmission, fast on-board switching and spot beam technologies mean that satellites can be employed efficiently to help provide high-speed, global connectivity to users all around the world. Satellite systems are multiple access systems with limited transmission capacity, very limited on-board buffering facilities, and long propagation delays compared to terrestrial network. Because of the limited on-board buffer space, high packet loss could be expected on board the satellite if no special protections were built-in. End-to-end resource management in on-board processing (OBP) satellite networks is key to provide fair and efficient sharing of traffic resources while delivering acceptable Quality of Service (QoS) to users. In this paper, we propose a Weighted Fair Bandwidth-on-Demand (WFBoD) process, which is a Demand Assignment Multiple Access (DAMA) based resource management protocol. It allows the network operator to implement various levels of service segregation by providing both absolute and differential QoS to the end-users by sharing the transmission resources fairly and efficiently. We formulate the problems associated with the implementation of WFBoD in OBP geostationary satellite networks, and present some simulation results in a Bent-Pipe satellite network scenario.

Among the factors, which can affect the quality of service of the four different levels of the traffic classes differentiated by their sensitivity to the delay tolerance, signaling ranks on the top. There is significant signaling messaging activity in UMTS among core network elements, and between SGSN and UTRAN that needs to be characterized so that its impact on the application architecture could be determined, and appropriate hardware and service software architectures to support signaling could be developed. Signaling affects mobility management, session management, call setup, and teardown times, etc. thus affecting the QoS of applications. In this paper we first examine the resulting signaling load on a core and a radio access networks for a given traffic penetration in the packet domain by considering a signaling procedure `Attach' and then analytically compute the throughput and the latencies for other basic signaling procedures such as Detach, and the PDP context activation in addition to attach by considering a platform which can act as a signaling server for the RNC, the SGSN, and the GGSN.

Multi-beam array antennas are being deployed in wireless cellular networks to implement space division multiple access (SDMA), using various forms of antenna sectorization, as a means to increasing the capacity of these networks. In this paper we propose a dynamic beam-forming strategy which provides superior SDMA performance. Specifically, we propose successive splitting of an original wide angle beam so as to spatially separate a number of initially interfering transmitters. This is shown to result in very drastic reductions in delays and computational loads while incurring no throughput penalties compared to standard random access protocols.

Within the coming decade, there will be a dramatic increase in the availability of inexpensive, computationally powerful mobile devices running applications which use the Internet Protocol (IP) to access multimedia services over broad-band wireless connections. To this end, there has been extensive research and standardization in the areas of Mobile IP and IPv6. The purpose of this paper is to apply this work to the issues involved in designing a mobility model able to adapt to different wireless mobile IP scenarios. We describe the usefulness of this model in the 4th generation mobile multimedia systems to come. This new model has been synthesized through a comparative analysis of current mobile IP models where particular attention has been given to the problems of mobile IP handoff and mobility management and their impact on QoS. By applying a unique perspective to these problems, our model is used to set a roadmap for future mobile IPv6 testbed construction.

Mobile IP is used to keep track of location information and make data available to the Mobile Host (MH) anytime, anywhere. In order to maintain uniform connectivity during mobility of the MH from one cell to the next, handoff latency must be minimized. In Mobile IP, when the MH is in a foreign domain far away from its home domain and Home Agent (HA), latency for registrations is high. This latency implies that many data packets could be lost during handoff from one domain to another. Hierarchical mobility management and forwarding pointers are two methods of reducing handoff latency. We propose a mobility management scheme called Hierarchical mobility management with Forwarding Pointers (HFP), that combines the two methods. In HFP, when the MH moves within a foreign domain, registration requests are handled locally. When the MH moves from one foreign domain to another, forwarding pointers between the two domains are used to track the location of the MH and perform smooth handoff. We show that HFP performs better than hierarchical mobility management during inter-domain mobility and better than forwarding pointers during intra-domain mobility. Simulation results that validate the performance improvement are presented and discussed.

Current handoff techniques in integrated services wireless networks focus on the mobile-to-fixed node handoff and have not addressed issues regarding mobile-to-mobile handoff. Virtual connection extension based mobile-to-mobile handoff is a fast solution, but results in inefficient routes. Crossover switch based techniques provide better routes, but at the cost of call dropping during simultaneous handoff. We propose three crossover switch based techniques aimed at preventing call dropping due to disjointed connection and also improving the latency during handoff. Partial- rerouting/path-extension, which uses the features of mutual exclusion principle effectively, prevents call dropping and also invokes a route optimization phase. The route information based partial-rerouting/path-extension technique is aimed at preventing call dropping and also reducing handoff latency due to rerouting. Finally we propose a dynamic route information based technique that further reduces the handoff latency. A comparative evaluation of handoff latency and complexity of the proposed techniques with crossover switch based mobile-to-mobile handoff scheme is provided.

In this research, we first address the QoS issue in different levels of wireless multimedia networks, and present a generic QoS framework to meet the requirements of different applications and services adaptively. Then, we focus on the connection-level QoS, measured by the connection blocking and dropping probabilities. A service model consisting of three service classes designed for connection-level QoS provisioning is proposed. The underlying network of consideration employs different call admission control and resource reservation schemes to allocate resources adaptively to each service class according to their haracteristics and requirements. The system is analyzed by a multi-dimensional model. Simulations are conducted based on the model analysis to evaluate the system performance.

Mobile Ad hoc Networks (MANET) have been the focus of many recent research and development efforts because of the necessity of decentralized network configuration. Ad hoc networks are characterized by multi hop wireless connectivity and frequently changing network topology, which supports both mobile users and services. To facilitate communication in this dynamically changing network topology several routing protocols such as Dynamic Source Routing protocol, Ad hoc On-demand Distance Vector protocol, and Destination Sequenced Distance Vector protocol, etc are proposed in the literature. Most of the research works are carried out by assuming random mobility models for speed and direction of the mobile. Such a model can generate an unrealistic behavior such as sudden stopping and turning. Later, researchers used uniform mobility models for both speed and position of the node. In this work, we have studied various mobility models, which are more realistic than the earlier assumptions. We have also investigated the effect of exponential and normal distribution models for mobiles speed and position. Based on our study, the throughput, end-to-end delay and the control overhead performance, it is found that the exponential model outperforms the uniform and normal models for a wide range of speeds.

The shadow cluster concept is used to estimate the future resource requirements, to reserve resources accordingly, and to admit only those calls which can be supported adequately. This paper describes the use of shadow cluster concept to allocate resources in Ad hoc networks using Cluster Head Gateway Switch Routing (CGSR) . With shadow cluster, the QoS of mobile calls can be improved by reducing the number of dropped calls during handoff, and by disallowing the establishment of new call that are highly likely to result in dropped calls later. In our work, we have used the shadow cluster concept to predict the demand by calculating the active mobile probability (AMP) of a call. Using this information, the resources are reserved in advance in the shadow area of the call. The performance results show that the shadow cluster scheme has less call drop, less call acceptance and fair bandwidth utilization.