With the rapid growth in demand for mobile data fueled by the emergence of new consumer applications that require high quality of service, existing static mobile data pricing plans are no longer suitable. Although several dynamic pricing mechanisms have been proposed in the literature, they have not been widely adopted due to their associated implementation complexity, the lack of trust in the associated platforms, and the absence of automatic enforcement. To address these challenges, we propose a credible trust framework (CREDIT) that leverages well-established Ethereum smart contracts for service-level agreement (SLA) enforcement. CREDIT introduces a novel SLA verification mechanism through fair auditor selection and auditor payoff to ensure truthfulness, and hence reinforces trust between the various parties involved in CREDIT. A detailed game-theoretic analysis is provided to prove the credibility of CREDIT by using the principle of a strong Nash equilibrium. In addition, CREDIT is prototyped by leveraging the smart contracts of Ethereum Blockchain. The results achieved provide a valuable validation of the feasibility of CREDIT.

Network Slicing has emerged as a keystone for supporting multiple service verticals over shared 5G network infrastructure. It leverages virtualization techniques to allocate programmable network instances matching service requirements. This requires novel resource allocation mechanisms, as well as new admission control and load balancing policies, to satisfy the diverse quality of service (QoS) requirements of different services, while optimizing the resource utilization across slices. This in turn requires accurate determination of loading on a slice, which not only depends on the type of application, but also on the stringency of QoS requirements, and hence existing metrics such as resource utilization may be inadequate. We introduce a novel load metric that can be used across slices, and illustrate how it can be used for load balancing, slice capacity resizing, and admission control in the proposed cross-slice resource orchestration framework.

Online social networks have been charged with enhancing and augmenting polarization in society. This polarization can have negative repercussions on the health of both individuals and society on the whole. Hence, it is vital that our online social networks are powered by algorithms that avoid polarisation and seek to curtail it. One target would be the curated news feed supplied to users in online social networks.

Many Telecommunications providers now support multiple generations of wireless standards in their network. For example, some providers simultaneously support 2G, 3G, 4G and even 5G standards. Upper management generally need a high level view of the performance being provided to consumers on a regular basis. In addition, engineers need to monitor the health of each technology in order to determine problems and be able to react quickly. We provide a framework for determination of a single Figure of Merit (FoM) that can be used for high level monitoring while at the same time providing sufficiently valuable low level indicators to assist with the isolation and detection of problems. We illustrate this framework using data from a real cellular network.

One new trending feature of Smartphones is the support for E-SIM (Embedded Subscriber Identification Module) cards. These allow the user to simultaneously subscribe to multiple cellular providers while also supporting at most one physical SIM (Subscriber Identification Module) card. This feature allows customers to easily switch between providers and is especially useful for those who use prepaid plans which are popular in developing countries. A customer may have multiple providers and, at any point in time, can choose the provider with the most cost effective data plan. This means that cellular providers must now take into account “soft-churn”, where the consumer dynamically switches between multiple plans from multiple providers, in addition to the more traditional churn where a consumer switches providers. This means that data pricing for such consumers must now be more personalized in order to be competitive and maximize profits. We determine the optimal personalized prepaid plan for such users while providing a competitive advantage to the provider. Examples are provided to demonstrate the benefit and numerical results corroborate our premise that these personalized pricing plans can, in fact, increase provider revenue.

Data and Artificial Intelligence are changing the business models of many financial institutions. The availability and granularity of customer data allows for the development of a personalized banking experience which has been shown to improve customer relationships and increase retention. We present a Machine Learning approach to providing personalized overdraft protection. The approach simultaneously provides benefits to both customer and bank and hence increases customer retention while improving the bank's revenue. We illustrate the approach with examples.

The Network Neutrality principle states that, In order to preserve the openness of the Internet, Internet Service Providers (ISP) should be prohibited from blocking or differentiating between content from different application providers, and the end users should have equal access to all legal content on the Internet. The Network Neutrality issue is critical as it can have a considerable impact on the future Internet design principles and associated laws and policies. In this paper, we discuss some of the challenges in monitoring Network Neutrality conformance, focusing on the content provider discrimination. In addition we discuss the emerging network technologies such as Giga-LTE, LTE in Unlicensed spectrum (LTE-U) or Licensed Assisted Access (LAA), LTE Broadcast, LTE Wi-Fi Aggregation (LWA), MulteFire, etc., and discuss how they may affect Network Neutrality.

Recent increases in the adoption of Online Social Networks (OSNs) for advertising has resulted in the research and development of algorithms that can maximize the resulting revenue. OSN users are likely to be influenced by their friends; therefore, one can leverage friendship relationships to determine how advertisements should be distributed among users. If a user is given an indication that their friend clicked on an advertisement link (called an impression), then they are more likely to also click on the impression if it were to be provided to them. The problem of assigning impressions can be modeled as an optimization problem in which the goal is to maximize the expected number of clicks achieved given a fixed number of impressions. Hosein and Lawrence [1] formulated this as a Stochastic Dynamic Programming problem in which impressions are provided in stages and the outcomes of previous stages are used in making impression allocations for the present stage. However, the determination of the optimal solution is computationally intractable for large problems; hence we require heuristics that are efficient while providing near-optimal solutions. In this paper we provide and compare various heuristics for this problem.

Network Neutrality states that users should have equal access to all Internet content and that Internet Service Providers (ISP) should not unfairly block or degrade access to any site. The Network Neutrality controversy has been in the spotlight for several years due to political, economic, social, and technical implications. While proponents argue that enforcing Network Neutrality by law is important to preserve the openness of the Internet and boost innovation, the opponents counter this argument by stating that such rules provide disincentives for the ISPs to invest in their network. Among the various issues related to Network Neutrality, we focus on the issue of detecting whether or not an ISP is Network Neutral (or Content Provider neutral). We propose a methodology for monitoring ISPs to ensure compliance. We monitor the packet loss statistics and determine if the rate adaptation in case of streaming media is being influenced by throttling at the ISP, and support our proposal through simulations.

The rapid increase in the global availability and use of the Internet has made it the most effective platform for the distribution of product information and advertising when compared to the traditional media such as print, television and radio. In addition, in the case of Online Social Networks (OSN), one can take advantage of the influences that one's actions has on other users and, in particular, one's friends. Even if a user does not purchase a product, their online mention of the product can have positive or negative influences. This social behavior can be used to improve advertising strategies. For example, if a user is given an advertisement impression (a linked icon) and clicks on it then it is more likely that their friends (if told of the action) will also click if they are subsequently given the impression. In this paper we provide a Stochastic Dynamic Programming formulation of this problem together with its solution. Because of the computational complexity of the solution we provide a simple heuristic that we show to be computationally much faster.

We consider the downlink of an Orthogonal Frequency Division Multiple Access (OFDMA) network. OFDMA transmission technology will be used in the next generation of wireless networks such as the LTE (Long Term Evolution) and WiMAX wireless standards. In such a channel, multiple User Equipments (UE) may each be allocated multiple Resource Blocks (RBs) in each frame. A RB is a collection of subcarriers spanned over all symbols that are available for data transmission. We are concerned with the optimal allocation of power and bandwidth (number of subcarriers) with the objective of maximizing the sector-wide throughput. Note that since we are considering multi-sector performance then the influence of intersector interference is also taken into account. In general, this problem is intractable and so we focus on the two sector case to gain intuition that can then be applied to more general algorithms. We determine the regions within which (a) a frequency reuse factor of unity is optimal, (b) orthogonal frequency allocations is optimal and (c) joint processing of signals from both sectors is optimal. We also show that it is never optimal to combine two or more of these solutions for a given scenario.

The next generation of cellular networks (Long Term Evolution) uses SC-FDMA (Single Carrier Frequency Division Multiple Access) on the uplink. In SC-FDMA, the uplink resources assigned to a user are called Resource Blocks (RBs) each of which consists of a set of consecutive subcarriers spanning all symbols available for data. Since at most one User Equipment (UE) is assigned to each RB then uplink transmissions within a sector are orthogonal. However, the transmission is affected by intersector interference since users in adjacent sectors may also have been assigned to the same RB. We assume that, within a site (i.e., basestation) interference cancellation is used to address this interference problem. However, for sectors across sites, interference management must be used to ensure efficient utilization of network resources. In this paper we investigate the issues of power allocation and interference management for the LTE uplink. In particular, we address the case in which QoS guarantees are made for uplink connections.

The latest generation of wireless networks use OFDMA (Orthogonal Frequency Division Multiple Access) transmission technology in the downlink (e.g, WiMAX and LTE). The downlink is divided into bandwidth units and UEs (User Equipments) are each allocated a subset of these. Since at most one UE is assigned to each of these units, downlink transmissions within a sector are orthogonal. However, the transmission is affected by intersector interference since UEs in adjacent sectors may also have been assigned to the resource. If the transmissions in neighbouring sectors are strong then the intersector interference may severely limit the SINR achieved by the UE. In this paper we consider a single site consisting of multiple sectors and consider the problem of maximizing the downlink capacity. We show that the sum rate, of a single resource unit, is maximized using a binary power allocation strategy. We then provide numerical results for a simple example to illustrate the gains that can be potentially achieved.

The next generation of wireless networks (e.g., Long Term Evolution (LTE)and WiMAX) use multiple techniques to improve channel spectral efficiencies. In this paper we focus on one such technique, namely Coordinated Beam Switching (CBS). With CBS, each sector determines a sequence of beams (possibly with repetitions) over which it continuously cycles. Each sector independently determines its beam pattern with the only constraint being that all sectors use a common cycle period. A beam can be used for the entire frame or one may use different beam patterns for each subband of the frame. Therefore the interference pattern of each subband of a sector repeats with the same period and hence the channel quality can be predicted if we assume slow moving UEs (User Equipment). Such a method may not be suitable for delay sensitive applications but, in the case of VoIP, the periodic arrivals of VoIP packets can be synchronized with the periodic service provided to the UE. Furthermore, VoIP service is limited by UEs at the edge but these are precisely the UEs that benefit from beamforming. In this paper we provide algorithms for accomplishing this synchronization and illustrate the corresponding VoIP capacity gains through simulations.

We present a novel solution to the problem of handover between two different wireless technologies. We illustrate this solution with the example of LTE (Long Term Evolution) to CDMA (Code Division Multiple Access) handover. Current inter-technology handover can take approximately 6 to 7 seconds and hence can heavily impact the network performance. We assume that two receivers (primary and secondary) are used in LTE, and that the UE performs an inter-technology handover between LTE and CDMA. The basic concept is that the UE uses two receivers (which are needed in MIMO mode) to enhance the network capacity when at the cell center but switches the secondary receiver, to the other network, when at the network border to obtain system information, signal strength, and perform CDMA synchronization. This switching function can be controlled by either the network or the UE. In this paper, the switching function is evaluated for the network controlled scenarios and is compared with non-switched scenarios. This study focuses on the effects on diversity gain and network coverage. Simulation results show that the loss in diversity gain has minimal impact on the network performance and the inter-technology handover delay can be reduced to less than 1 second.

With the increasing throughput demands of wireless users, operators will need to build wireless networks with significantly larger capacities than those presently available. Therefore for the next generation of wireless networks (4G) a significant effort is being placed on increased spectral efficiencies. One means of achieving higher spectral efficiencies is through the use of Distributed Antenna Systems (DAS) whereby a single basestation controls transmissions from multiple geographically distributed antennas. This allows for the placement of the antennas closer to the UEs (User Equipments) so as to use reduced transmission power while at the same time increasing the spatial diversity of the transmissions from such antennas. In this paper we investigate various strategies for serving UEs using DAS. In particular, we investigate the traditional frequency reuse one and orthogonal frequency allocation schemes as well multipoint transmission schemes in which multiple transmitters can be used to serve a single user. In order to gain some intuition on the effectiveness of each approach we analyze a simple network model and show that, in practice, it may be possible to reduce the set of strategies one uses when serving users by removing those strategies that are optimal in only small regions of the cell. We use simulations to further support our conclusions. We believe that the effectiveness of DAS depends on the radio resource management applied and that investigations such as those in this paper can increase our intuition in this area.

We consider the downlink of an Orthogonal Frequency Division Multiple Access (OFDMA) network. OFDMA transmission technology will be used in the next generation of wireless networks such as the LTE (Long Term Evolution) and WiMAX wireless standards. In such a channel, multiple User Equipments (UE) may each be allocated multiple Resource Blocks (RBs) in each frame. A RB is a collection of subcarriers spanned over all symbols that are available for data transmission. For each UE we assign a utility function that depends on the UE's average throughput. Such a function can be used to achieve specified Quality of Service (QoS) guarantees of the UE. We are concerned with the optimal allocation of power and bandwidth in each sector with the objective of maximizing the sum utility over all UEs in all sectors. Note that since we are considering sector-wide performance then the influence of intersector interference is also taken into account. In general, this problem is intractable and so we focus on the two-sector case. We determine an upper bound on performance for any set of utility functions and for any strategy for scheduling, power control and interference management. This bound can therefore be used to evaluate the optimality of any particular strategy.

In the present and future generation of wireless networks, link adaptation is used for better channel efficiency. With link adaptation the modulation and coding scheme used for transmissions to a particular user is adjusted based on the user's channel conditions. However, because of delays and measurement errors in the reported channel quality information, the actual transmission may experience a significantly different channel than anticipated. This can either lead to transmission errors (if the channel has deteriorated) or wasted transmission resources (if the channel is actually better than predicted). To avoid the former scenario, a conservative mapping from channel quality information to modulation and coding scheme is used. However, the degree of this conservatism depends on the Doppler Frequency of the channel with higher values requiring a more conservative mapping. In this paper we propose an algorithm for dynamically adjusting the mapping from channel quality reports to the modulation and coding scheme. Hybrid ARQ is assumed and we use the outcome of the first transmission to adjust the mapping. Simulation results are used to illustrate the effectiveness of the approach.

The next generation of wireless networks (e.g.,) will use multiple techniques to improve channel spectral efficiencies. In this paper we focus on one such technique, beamforming, in the downlink channel. In each sector, a number of fixed beams are used to serve UEs (User Equipment). For each frame, the scheduler chooses a beam and then schedules one or more UEs that lie within that beam. Since the power is focused on a narrow beam then the served UEs can achieve higher rates and the sector coverage is also increased. However, this beam causes more interference in adjacent sectors. If a neighboring sector schedules a UE that lies within the coverage of this beam then the rate for that UE will be low. The neighboring sectors must therefore schedule UEs outside of the beam or at least know of the interference that will be experienced. The first approach (interference avoidance) requires coordination among the sectors with the associated communication overheads and delays. In this paper we follow the second approach and assume no coordination among sectors except for the fact that they all use the same beam cycle period. Although all NBs use the same beam cycle period, they may occasionally change the specific beam schedule based on local factors such as the geographical distribution of UEs within its beams. Such changes should be made infrequently since the approach is based on the assumption that the interference experienced by a UE repeats with the beam cycle period.

The next generation of wireless networks (e.g.,) will use multiple techniques to improve channel spectral efficiencies. In this paper we focus on one such technique, beamforming, in the downlink channel. In each sector, a number of fixed beams are used to serve UEs (User Equipment). For each subband in each frame, the scheduler chooses a beam and then schedules one or more UEs within the corresponding beam. Since the power is focused on a narrow beam then the served UEs can achieve higher rates and the sector coverage is also increased. However, this beam causes more interference in adjacent sectors (flashlight effect). The neighboring sectors must therefore schedule UEs outside of the beam or at least know of the interference that will be experienced. In this paper we follow the second approach and assume no coordination among sectors except for the fact that they all use the same beam cycle period but this can be changed occasionally.

In this paper we consider the problem of allocating resources to users in the downlink of a 802.16 network. This network uses orthogonal frequency-division multiple access (OFDMA) as the downlink transmission technology. The transmission resources available in the downlink include bandwidth (subcarriers), time (OFDMA symbols) and transmission power all of which are limited. The frequency-time resource allocations for each user must satisfy certain constraints (e.g., must form logical rectangular bursts). The focus of this paper is on the determination of the size and locations of these bursts within the frame for the given scheduler decisions. Prior work has addressed this problem independently of other protocol layers. In our approach we simplify the burst construction problem, without significant performance degradation, by taking into account higher layer information such as the tolerable delay of the underlying application.

Multiple Input Multiple Output (MIMO)antennasystemsare being used to increase the spectral efficiency for the next generation of wireless networks (LTE, WiMAX). In this paper we focus on beamforming techniques for the downlink channel. In particular, we consider the problem of downlink user selection to maximize the downlink resources while satisfying the QoS requirements of all users. We do this by assigning suitable utility functions to user flows that are designed to maintain the specified QoS constraints. Since multiple users can be simultaneously scheduled in a time-bandwidth resource block, the scheduling problem becomes multi-dimensional and, in general, its computational complexity grows non-linearly with the number of users. We propose a simple but sub-optimal coordinate ascent algorithm for scheduling users. Prior work on this problem included simple greedy algorithms but with inferior performance or near-optimal algorithms that are computationally intensive. We will demonstrate that our proposed solution, although computationally simple, provides near-optimal performance.

Orthogonal frequency division multiple access (OFDMA) transmission has been chosen for the next generation of wireless networks. Multiple input multiple output (MIMO) techniques are well suited for OFDMA and will be used for increased transmission efficiency for both downlink and uplink. In this paper we focus on space division multiple access (SDMA) techniques for the uplink of such a network. In particular, we consider the problem of scheduling uplink users so as to maximize the uplink resources while satisfying the QoS requirements of each userpsilas flows. We do this by assigning suitable utility functions to user flows that are designed to maintain the specified QoS constraints. Since multiple users can be simultaneously scheduled, the scheduling problem becomes multi-dimensional and, in general, its computational complexity grows non-linearly with the number of users. In addition to choosing the users to be scheduled one must also determine the optimal transmission power for each scheduled user. For the case of two users per frame we show that the optimal power allocation scheme is such that either both users transmit with full power or only one transmits with full power. We then propose a coordinate ascent algorithm for solving the utility-based scheduling problem. Simulations are then used to show that the solution produced by this approach is close to optimal and is quite robust.

The next generation of wireless networks (LTE, WiMAX and UMB) will have sufficient capacity to support video broadcasts in addition to unicast services such as voice and data. Since broadcast services are resource intensive, radio resources must be carefully managed to ensure proper operation of the unicast services. One approach is to allocate a fixed set of resources (a zone) specifically for broadcast services. Multiple multimedia streams are then multiplexed within this zone for broadcast over the air. In this paper we focus on the case of variable bit rate (VBR) streams with long range dependency. In this case buffering of streams to smooth out bursts is not sufficient and efficient resource multiplexing must instead be used for adequate delay performance. We present a novel scheduling algorithm that achieves significant multiplexing gains. This scheduler is designed for low terminal power utilization as well as high broadcast channel utilization.

The next generation of wireless networks (4G) will use OFDMA (orthogonal frequency division multiple access) on the reverse link. In OFDMA, the reverse link resources assigned to a user are called tiles each of which consists of a subset of consecutive subcarriers. Since at most one user is assigned to each of these tiles then reverse link transmissions within a sector are orthogonal. However, the transmission is affected by intercell interference since users in adjacent sectors may also have been assigned to the same tile. If those users in the adjacent sectors transmitted with high power then the intercell interference may severely limit the SINR achieved by the user. Therefore user transmission powers must be carefully managed to avoid excessive intercell interference. In this paper we formulate and investigate this optimization problem. Numerical examples are used to illustrate our conclusions.

In this paper we consider the reverse traffic channel of the 1xEV-DO Revision A wireless network standard. We develop an analytic model for determining the VoIP capacity of this channel. This model is then compared with published simulation results to demonstrate its accuracy. Results for the model can be determined in real-time and the intent of this work is to be able to use such models to dynamically adjust network parameters to more efficiently use radio resources.

In traditional wireless networks, voice service is supported with synchronous power controlled channels and hence latency requirements were easily met. Fourth generation networks (3GPP2, 3GPP, WiMAX), are packet based and use orthogonal frequency division multiple access (OFDMA) on the forward link. In this case voice will typically be transported using the voice over IP (VoIP) protocol. With this approach, there is a trade-off between delay performance and user capacity since by using queuing one can more efficiently utilize radio resources but at the expense of additional queuing delays. In this paper we provide models for determining various performance metrics of such a system and compare the accuracy of the models with published simulation results. These models can be used to develop an intuitive understanding of the underlying performance as well as for performing sensitivity analysis.

The next generation of wireless networks will use OFDMA (orthogonal frequency division multiple access) technology in the uplink. In OFDMA, the uplink resources assigned to a user are called tiles each of which consists of a subset of consecutive subcarriers. Since at most one user is assigned to each of these tiles then reverse link transmissions within a sector are orthogonal. However, the transmission is affected by intersector interference since users in adjacent sectors may also have been assigned to the same tile. If those users in the adjacent sectors transmitted with high power then the intersector interference may severely limit the SINR achieved by the user. Therefore user transmission powers must be carefully managed to avoid excessive intersector interference. In this paper we investigate a framework in which each sector periodically broadcasts information on its intersector interference. Each access terminal (AT) then uses this information to adjust its transmission power in order to maintain intersector interferences below some specified threshold. We provide an algorithm for doing this and we show that the algorithm is robust, efficient and fair.

Orthogonal frequency division multiple access (OFDMA) has been chosen as the radio transmission technology for future 4G wireless networks. One important reason for this choice is that OFDMA offers various forms of diversity (time, frequency, space, etc). These diversity dimensions can be used to provide channels with reduced variance and hence more predictable performance. In this paper we focus on the time and frequency dimensions and propose a metric that can be used to quantify diversity. We use the proposed diversity metric to design highly diverse data subchannels and demonstrate the effectiveness of the approach through simulations. With the use of this diversity metric one can more easily design OFDMA channels because of the reduced computational time required to compare layouts.

The third generation of wireless networks such as 1xEV-DO, 1xEV-DV and HSPA were designed to support high speed data services. The CDMA versions (1xEV-DO and 1xEV-DV) uses a 1.25 MHz bandwidth while the WCDMA version uses a 5 MHz bandwidth. Recent standardization work on 1xEV-DO revision B has been focused on the efficient use of multiple carriers to increase the peak and average rates available to users. In the reverse direction this allows the access terminal (AT) to simultaneously use multiple reverse links for its applications. Each of these reverse links has an associated cost for its maintenance (pilot, signaling etc.). Two extreme strategies for allocating reverse link resources are: (a) have each AT use the minimum number of reverse links necessary to support its desired rate and (b) have each AT use all available reverse links and maintain the same rate on each of them with the resultant total rate equal to the desired rate. In this paper we investigate the trade-offs between these two approaches and show that, in general, minimizing the number of reverse links that an AT uses is the more efficient strategy

The present suite of 3G wireless network standards make use of forward link channel quality reports to more efficiently schedule users on a forward time-shared packet data channel. As the feedback rate and precision of these reports is increased, the performance of the scheduling decision is also increased since it is based on more recent and accurate data. However, this also increases the reverse link resources needed for these reports. In order to reduce the mobile power required to make these reports, some standards (e.g., IxEV-DO (N. Bhushan et al., 2006) (2004)) take advantage of time diversity gains by repeating the report for a specified number of times but with an overall reduction in the power required. Naturally this reduces the maximum rate at which these reports can be made and increases the staleness of the information provided to the scheduler. The users with the poorest reverse link conditions dictate the number of repetitions. This means that reports made by users in good radio conditions will likely be received after a smaller number of repetitions. Presently the scheduler does not use these early terminated reports in making scheduling decisions. In this paper we show that if the scheduler uses this information then the forward link performance can be increased

In traditional CDMA wireless networks, real-time services, such as circuit-switched voice, are transported over the air via synchronous channels because of their stringent delay requirements. In the third generation networks (3G), IxEV-DV [A.Soong et al., 2003] and HSDPA [E. Dahlman et al., 1998], an additional time shared channel was introduced on the forward link to support data services for which delay requirements are less stringent. The IxEV-DO [P. Bender et al., 2000] standard provides a single time-shared forward link channel and no synchronous channels since it was designed primarily for data services. It has recently been suggested that certain voice services (e.g.. Voice over IP (VoIP) and push-to-talk (PTT)), can be efficiently transported over such time-shared channels since they have less stringent delay requirements compared to circuit-switched voice. In this paper we investigate the capacity of VoIP users over these time-shared channels and investigate the sensitivity to various base station (BS) and mobile station (MS) design parameters, algorithms and features. Note that detailed simulations of each standard is not provided but rather a comparative approach is used whereby we investigate specific features of each standard. Our focus is on the forward channel since code division multiplexing is performed in the reverse link and hence comparable user capacities are achieved.

Third generation (3G) networks will provide support for high speed applications to its users in both the forward and reverse directions. In the forward direction, this is achieved by time sharing users over one or more high speed packet data channels. Users are scheduled based on their application needs as well as the mobile operator's objectives. On the reverse link, two options are possible for serving users, rate control and scheduling. Scheduling provides fine control of reverse link resources but at the expense of complexity and signaling overhead. Two forms of rate control are possible, common rate control and dedicated rate control. Dedicated rate control is similar in concept to scheduling, but is limited to one rate level change per user. In common rate control, each mobile independently determines its reverse link rate using load information that is broadcast by the base station on a periodic basis. The algorithm used by each mobile to determine its transmission rate should again be determined based on the needs of the application as well as the operator's overall objectives. One option is to be completely fair and provide all mobiles (as much as is physically possible) with identical transmission rates (i.e., the performance objective used for voice users). Another option is to allow mobiles to use rates based on their reverse link channel conditions (the proportionally fair performance objective used in the forward packet data channel). We compare the throughput and fairness of these two approaches.

Third generation (3G) wireless networks have been designed to provide high speed data services on both forward and reverse links. In addition to increased data capacity, they also support QoS (quality of service) for mobile users. Users may run multiple applications (service instances) simultaneously and QoS guarantees must be assured for one or more of them. We address the problem of providing QoS guarantees on the reverse link of a 3G (CDMA) network. We present a flexible, low overhead framework for efficient management of the reverse link resources to satisfy the QoS guarantees of each user's applications. We show how the QoS demands of all service instances of a particular user can be converted and combined into a single resource demand, namely the data rate of the corresponding reverse link channel.

With the proliferation of data services over wireless networks, it has become increasingly important to understand the user perceived application performance aspects of such networks. In this paper we provide simple, yet accurate, models for the physical and MAC layers of a CDMA2000 supplemental channel. We illustrate how these models can be used to determine TCP performance over such channels.

CDMA/HDR (High Data Rate) is a high-speed wireless data service that is optimized for best-effort data services (for a description of HDR see [1]). As with all best-effort services, as the number of users increases, the quality of service experienced by each user decreases. In order to provide an acceptable level of service to users, the system must be engineered for a limited number of connected users and ensure (by admission controls) that this number is rarely exceeded. In this paper we provide a capacity model that can be used to determine user perceived delays and throughput as a function of the total number of users. This can then be used to determine the maximum number of simultaneous users that the system can support for specified limits on user delay and throughput.