The SIMT execution model is commonly used for general GPU development. CUDA and OpenCL developers write scalar code that is implicitly parallelized by compiler and hardware. On Intel GPUs, however, this abstraction has profound performance implications as the underlying ISA is SIMD and important hardware capabilities cannot be fully utilized. To close this performance gap we introduce C- For- Metal (CM), an explicit SIMD programming framework designed to deliver close-to-the-metal performance on Intel GPUs. The CM programming language and its vector/matrix types provide an intuitive interface to exploit the underlying hardware features, allowing fine-grained register management, SIMD size control and cross-lane data sharing. Experimental results show that CM applications from different domains outperform the best-known SIMT-based OpenCL implementations, achieving up to 2.7x speedup on the latest Intel GPU.

This work studies the joint scheduling - admission control (SAC) problem over a fading channel. In particular, the optimal trade-off between maximizing the throughput and minimizing the queue size (or average congestion) is investigated. The SAC problem is formulated as a constrained Markov decision process (MDP) to maximize a utility defined as a function of the throughput and the queue size. The structural properties of the optimal policies are subsequently derived. When the statistical knowledge of the traffic arrival and channel processes is not available, we propose an online learning algorithm for the optimal policies. The analysis and algorithm development are relied on the reformulation of the Bellman's optimality dynamic programming equation using suitably defined value functions which can be learned using online time-averaging.

We propose a new framework for wireless network virtualization. In this framework, service providers (SPs) and the network operator (NO) are decoupled from each other: The NO is solely responsible for spectrum management, and SPs are responsible for quality-of-service (QoS) management for their own users. SPs compete for the shared wireless resources to satisfy their distinct service objectives and constraints. We model the dynamic interactions among SPs and the NO as a stochastic game. SPs bid for the resources via dynamically announcing their value functions. The game is regulated by the NO through: 1) sum-utility optimization under rate region constraints; 2) enforcement of Vickrey-Clarke-Groves (VCG) mechanism for pricing the instantaneous rate consumption; and 3) declaration of conjectured prices for future resource consumption. We prove that there exists one Nash equilibrium in the conjectural prices that is efficient, i.e., the sum-utility is maximized. Thus, the NO has the incentive to compute the equilibrium point and feedback to SPs. Given the conjectural prices and the VCG mechanism, we also show that SPs must reveal their truthful value functions at each step to maximize their long-term utilities. As another major contribution, we develop an online learning algorithm that allows the SPs to update the value functions and the NO to update the conjectural prices iteratively. Thus, the proposed framework can deal with unknown dynamics in traffic characteristics and channel conditions. We present simulation results to show the convergence to the Nash equilibrium prices under various dynamic traffic and channel conditions.

This work studies the joint scheduling- admission control (SAC) problem for a single user over a fading channel. Specifically, the SAC problem is formulated as a constrained Markov decision process (MDP) to maximize a utility defined as a function of the throughput and queue size. The optimal throughput- queue size trade-off is investigated. Optimal policies and their structural properties (i.e., monotonicity and convexity) are derived for two models: simultaneous and sequential scheduling and admission control actions. Furthermore, we propose online learning algorithms for the optimal policies for the two models when the statistical knowledge of the time-varying traffic arrival and channel processes is unknown. The analysis and algorithm development are relied on the reformulation of the Bellman's optimality equations using suitably defined state-value functions which can be learned online, at transmission time, using time-averaging. The learning algorithms require less complexity and converge faster than the conventional Q-learning algorithms. This work also builds a connection between the MDP based formulation and the Lyapunov optimization based formulation for the SAC problem. Illustrative results demonstrate the performance of the proposed algorithms in various settings.

In this paper, we propose a systematic solution to the problem of scheduling delay-sensitive media data for transmission over time-varying wireless channels. We first formulate the dynamic scheduling problem as a Markov decision process that explicitly considers the users' heterogeneous multimedia data characteristics (e.g., delay deadlines, distortion impacts and dependences, and so on) and time-varying channel conditions, which are not simultaneously considered in state-of-the-art packet scheduling algorithms. This formulation allows us to perform foresighted decisions to schedule multiple data units for transmission at each time in order to optimize the long-term utilities of the multimedia applications. The heterogeneity of the media data enables us to express the transmission priorities between the different data units as a priority graph, which is a directed acyclic graph. This priority graph provides us with an elegant structure to decompose the multidata unit foresighted decision at each time into multiple single-data unit foresighted decisions which can be performed sequentially, from the high priority data units to the low priority data units, thereby significantly reducing the computation complexity. When the statistical knowledge of the multimedia data characteristics and channel conditions is unknown a priori, we develop a low-complexity online learning algorithm to update the value functions, which capture the impact of the current decision on the future utility. The simulation results show that the proposed solution significantly outperforms existing state-of-the-art scheduling solutions.

Existing network pricing solutions mainly focus on congestion-dependent pricing schemes, while ignoring the users' traffic state information, which we shall show in this paper can be exploited to significantly improve the service provider's revenue. In order to derive pricing strategies that explicitly take into account the users' traffic dynamics, we propose a systematic framework of traffic-dependent pricing by focusing on delay-sensitive multimedia networks. First, we introduce a finite-state Markov chain to capture the users' traffic dynamics, and a service demand model that is dependent on the users' traffic state information. Thus, we relate the users' traffic dynamics to the service provider's pricing policy, by means of the traffic-dependent demand model. Then, we formulate the service provider's pricing problem into a Markov decision process, and propose a low-complexity pricing algorithm, i.e., static pricing without considering the resource constraint, which can achieve a close-to-optimal performance. Next, by considering a practical scenario in which the service provider does not know the users' traffic dynamics a priori, we propose a learning-based algorithm that allows the service provider to identify an (locally) optimal pricing policy. Finally, we conduct simulations to quantify the proposed framework of traffic-dependent pricing.

In this paper, we formulate the cross-layer optimization for delaysensitive media transmission over time-varying wireless channels as a finite-horizon Markov decision process (MDP) by explicitly considering the users' heterogeneous multimedia traffic characteristics (e.g. delay deadlines, distortion impacts and dependencies etc.) and time-varying network conditions. Based on the heterogeneous characteristics of the media packets, we are able to express the transmission priorities between packets as a new type of directed acyclic graph (DAG). This DAG provides the necessary structure for determining the optimal cross-layer actions in each time slot. The simulation results demonstrate that the proposed solution significantly outperforms existing state-of-theart cross-layer solutions.

We propose a virtualization framework to separate the network operator (NO) who focuses on wireless resource management and service providers (SP) who target distinct objectives with different constraints. Within the proposed framework, we model the interactions among SPs and NO as a stochastic game, each stage of which is played by SPs (on behalf of the end users) and is regulated by the NO through the Vickrey-Clarke-Groves (VCG) mechanism. Due to the strong coupling between the future decisions of SPs and lack of global information at each SP, the stochastic game is notoriously hard. Instead, we introduce conjectural prices to represent the future congestion levels the end users potentially will experience, via which the future interactions between SPs are decoupled. Then, the policy to play the dynamic rate allocation game becomes selecting the conjectural prices and announcing a strategic value function (i.e., the preference on the rate) at each time. We prove that there exists one Nash equilibrium in the conjectural prices and, given the conjectural prices, the SPs have to truthfully reveal their own value function. We further prove that this Nash equilibrium results in efficient rate allocation in our virtualized wireless network. In other words, there are enough incentives for NO to advertise such a conjectural price and SPs to follow this advice.

In this paper, we systematically formulate the problem of multi-user wireless video transmission as a multi-user Markov decision process (MUMDP) by explicitly considering the users' heterogeneous video traffic characteristics, time-varying network conditions as well as, importantly, the dynamic coupling among the users' resource allocations across time, which are often ignored in existing multi-user video transmission solutions. To comply with the decentralized wireless networks' architecture, we propose to decompose the MUMDP into multiple local MDPs using Lagrangian relaxation. Unlike in conventional multi-user video transmission solutions stemming from the network utility maximization framework, the proposed decomposition enables each wireless user to individually solve its own local MDP (i.e. dynamic single-user cross-layer optimization) and the network coordinator to update the Lagrangian multipliers (i.e. resource prices) based on not only current, but also the future resource needs of all users, such that the long-term video quality of all users is maximized. This MUMDP solution provides us the necessary foundations and structures for solving multiuser video communication problems. However, to implement this framework in practice requires statistical knowledge of the experienced environment dynamics, which is often unavailable before transmission time. To overcome this obstacle, we propose a novel online learning algorithm, which allows the wireless users to simultaneously update their policies at multiple states during each time slot. This is different from conventional learning solutions, which often update the current visited state per time slot. The proposed learning algorithm can significantly improve the learning performance, thereby dramatically improving the video quality experienced by the wireless users over time. Our simulation results demonstrate the efficiency of the proposed MUMDP framework as compared to conventional multi-user video transmission solutions.

Large-scale multimedia semantic concept detection requires realtime identification of a set of concepts in streaming video or large image datasets. The potentially high data volumes of multimedia content, and high complexity associated with individual concept detectors, have hindered the practical deployment of many current solutions. In this paper, we present a summary of our work in building systems and applications for resource adaptive semantic concept detection in multimedia using large-scale distributed stream mining systems. We construct such concept detection applications as a hierarchical topology of individual concept detectors, and deploy them on distributed processing infrastructure. We then focus on dynamically configuring individual concept detectors to meet system imposed resource constraints while minimizing a penalty defined in terms of the misclassification cost. We present multiple centralized and distributed algorithms for this configuration, and describe the implemented application and system. We also verify through simulations that significant improvement in terms of accuracy of classification can be achieved through our approach.

With the current proliferation of high bandwidth and delay-sensitive multimedia applications and services, each wireless user will try to maximize its utility by acquiring as much spectrum resources as possible unless a preemptive mechanism exists in the network. Thus, emerging solutions for dynamic spectrum access in cognitive radio networks will need to adopt market-based approaches in order to effectively regulate the available resources. In this paper, we show how various centralized and decentralized spectrum access markets can be designed based on a stochastic game framework, where wireless users (also referred to as secondary users) can compete over time for the dynamically available transmission opportunities (spectrum “holes”). When operating in such spectrum access “markets,” wireless users become selfish, autonomous agents that strategically interact in order to acquire the necessary spectrum opportunities. We also show how wireless users can successfully compete with each other for the limited and time-varying spectrum opportunities, given the experienced dynamics in the wireless network, by optimizing both their external actions (e.g., the resource bids, power and channel used for transmission, etc.) and their internal actions (e.g., the modulation schemes, etc.). To determine their optimal actions in an informationally decentralized setting, users will need to learn and model directly or indirectly the other users' responses to their external actions. We studied the outcome of various dynamic interactions among self-interested wireless users possessing different knowledge and determine that the proposed framework can lead to multiuser communication systems that achieve new measures of efficiency, rationality and fairness. Lastly, our illustrative results show that the presented game-theoretic solution for wireless resource management enables users deploying enhanced (“smarter”) learning and communication algorithms and being able to make efficient use of the spectrum resources can derive higher utilities. This presents the designers of wireless devices and systems with important incentives to endow their next-generation products and services with enhanced capabilities to gather information, learn, and strategically compete for resources in the emerging spectrum resource markets made possible by the cognitive radio network technologies.

In this paper, we model the various wireless users streaming delay-sensitive data in a wireless network as a collection of selfish, autonomous agents that strategically interact in order to acquire the spectrum opportunities. The spectrum allocation is coordinated by a central spectrum moderator which deploys an incentive compatible mechanism. We further model the repeated spectrum competition as a stochastic game through which we are able to characterize the interaction among wireless users. Based on the observed resource allocation and corresponding rewards from previous allocations, we propose a best response learning algorithm that can be deployed by wireless users to improve the accuracy of their private information at each stage. The simulation results show that by deploying the proposed best response learning algorithm, the wireless users can significantly improve their own performance.

Cross-layer optimization aims at improving the performance of network users operating in a time-varying, error-prone wireless environment. However, current solutions often rely on ad-hoc optimization approaches, which ignore the different environmental dynamics experienced at various layers by a user and violate the layered network architecture of the protocol stack. This paper presents a new theoretic framework in which the cross-layer optimization problem is formulated as a layered Markov decision process (MDP). In this framework, each layer adapts its own protocol parameters and exchanges information (messages) with other layers in order to cooperatively maximize the performance of the wireless user. Hence, this layered cross-layer framework does not change the current layered architecture and is suitable for the delay-sensitive applications over wireless networks.

We design algorithms for optimally configuring, in terms of model complexity as well as operating point, a cascade of exclusive classifiers on distributed systems, given underlying resource constraints. Under rate-dependent constraints we design a Viterbi-like strategy to determine the optimal solution with significantly lower computational complexity than an exhaustive search. We illustrate the performance of this strategy on a cascade of classifiers for a speaker verification task, and highlight performance gains over using a single classifier.

We propose to add a new dimension to existing wireless multimedia systems by enabling autonomous stations to dynamically compete for communication resources through adjustment of their internal strategies and sharing their private information. We focus on emerging spectrum agile wireless networks, where developing an efficient strategy for managing available communication resources is of high importance. The proposed dynamic resource management approach for wireless multimedia changes the passive way stations are currently adapting their joint source-channel coding strategies according to available wireless resources. Each wireless station can play the resource management game by adapting its multimedia transmission strategy depending on the experienced channel conditions and user requirements. The resource allocation game is coordinated by a network moderator, which deploys mechanism-based resource management to determine the amount of transmission time to be allocated to various users on different frequency bands such that certain global system metrics are optimized. Subsequently, the moderator charges the various users based on the amount of resources it has allocated to them, in order to discourage them from being dishonest about their resource requirements. We investigate and quantify both the users' and the system performance when different cross-layer strategies, and hence users' levels of smartness, are deployed by wireless stations. Our simulations show that mechanism-based resource management outperforms conventional techniques such as air-fair time and equal time resource allocation in terms of the obtained system utility. They also provide insights that can guide the design of emerging spectrum agile network protocols and applications

We propose to add a new dimension to existing wireless multimedia communications systems by enabling competing stations to proactively engage in the resource management game by adapting their cross-layer transmission strategies. For this, we model wireless stations (WSTAs) as rational and selfish players competing for available wireless resources in a dynamic game. We focus on polling-based wireless LAN (WLAN) networks, where developing an efficient solution for managing the available transmission opportunities is of paramount importance. The resource allocation game is coordinated by a network moderator, which deploys a novel resource management based on the Vickrey-Clarke-Groves (VCG) mechanism to determine a) the amount of time to be allocated to the various users and b) the transmission cost associated to the allocated resources. The transmission cost is referred to in the VCG mechanism as "transfer" and depends not on the used resources, but rather on the inconvenience (in terms of utility impact) that it causes to other WSTAs. The transfer is introduced in order to discourage WSTAs from lying about their resource requirements. Importantly, this proposed dynamic resource management approach for wireless multimedia applications changes the passive way stations are currently adapting their cross-layer strategies by enabling them to selfishly influence the wireless systems dynamics by proactively adapting their packet scheduling strategies, error protection strategies, etc. Hence, each wireless station can play the resource management game by adapting its multimedia transmission strategy depending on the experienced channel conditions, derived video quality, attitude towards risk, willingness to pay for resources and available information about the wireless network. Our simulations show that using the VCG mechanism the WSTAs do not have any incentives to lie about their resource requirements as otherwise they will be severely penalized by a high transfer. We also show that deploying advanced cross-layer strategies for playing the resource management game significantly benefit the WSTAs' received video quality. The willingness-to-pay for resources is introduced to provide WSTAs a tool to gather additional resources whenever they need to transmit an important (part of a) video sequence by agreeing to pay for resources an increased cost. A novel risk-aware scheduling scheme is also proposed that provides WSTAs the ability to dynamically avoid network congestion and hence, reduce their incurred transfer

We propose a new way of architecting the wireless multimedia communications systems by jointly optimizing the protocol stack at each station and the resource exchanges among stations. We model wireless stations as rational players competing for available wireless resources in a dynamic repeated game. We investigate and quantify the system performance and the impact of different cross-layer strategies deployed by wireless stations onto their own performance as well as the competing station performance. We show through simulations that the proposed game-theoretic resource management outperforms alternative techniques such as air-fair time and equal time resource allocation in terms of the total system utility