Catálogo de publicaciones - libros

This paper studies the problem of dimensioning a wireless base station that operates with CAC. It models the base station as a multi-rate, multi-class, multi-resource loss model and proposes two methods for fast estimation of the admission region. The results show that our methods are typically more accurate than other more computationally intensive approximations, such as the reduced-load approximation, and provide an excellent basis for the dimensioning of wireless networks with CAC.

Sensor nodes in large-scale sensor networks autonomously and proactively report diverse information obtained from extensive target area to a base station, which is energy unconstrained node. To efficiently handle the variable traffic, we present Co-MAC, an energy efficient medium access control protocol for large-scale sensor networks. In Co-MAC, an overall network is divided into independent , and each subnet orthogonally operates on time line in a temporal fashion. The novelty in this protocol is to evenly distribute sensor nodes to subnets at the association process with probability . In our simulation, it was observed that energy efficiency of Co-MAC outperforms S-MAC by 6 times and T-MAC by 2 times under identical conditions. The reason for this phenomenon is that the overhearing frequency between sensor nodes is relatively lower in Co-MAC. The results of these analyses demonstrate that Co-MAC obtains significant energy savings.

The paper presents a new analytical saturation throughput model of IEEE 802.11 DCF () with basic access in ad-hoc mode. The model takes into account when a station senses a busy channel. It is shown that taking into account this feature of DCF is important in modeling saturation throughput – yields more accurate results than models known from literature. The proposed analytical model also takes into account the effect of transmission errors.

We analyze the impact of size-based scheduling on the flow level performance of elastic traffic in wireless downlink data channels. The impact is assessed by comparing the flow level delay of the simple RR scheduler to two optimized non-anticipating schedulers (FB and FB°) and SRPT. The optimized distance-aware scheduler FB° is derived by applying the Gittins index approach. Our results show that for Pareto-type file size distributions, the size-based information is more important than the location information. Additionally, FB not only decreases the overall mean delay, but it can also decrease considerably the mean delay of all users independently of their location.

Wireless mobile networks generally provide radio access bearer services over wireless links between a terminal equipment (UE) and a base station (BS). In this paper, a radio resource scheduling scheme is proposed for providing a merged service of both realtime and non-realtime flow connections over the wireless link in the Portable Internet (e.g., IEEE 802.16e) with frequent host mobility. This scheme can efficiently improve the connection blocking probability, connection dropping probability, and bandwidth utilization.

We consider the problem of scheduling of a wireless channel (server) to several queues. Each queue has its own link (transmission) rate. The link rate of a queue can vary randomly from slot to slot. The queue lengths and channel states of all users are known at the beginning of each slot. We show the existence of an optimal policy that minimizes the long term (discounted) average sum of queue lengths. The optimal policy, in general needs to be computed numerically. Then we identify a greedy (one step optimal) policy, MAX-TRANS which is easy to implement and does not require the channel and traffic statistics. The cost of this policy is close to optimal and better than other well-known policies (when stable) although it is not throughput optimal for asymmetric systems. We (approximately) identify its stability region and obtain approximations for its mean queue lengths and mean delays. We also modify this policy to make it throughput optimal while retaining good performance.

This paper presents a unified model of classical teletraffic models covering both circuit-switched and packet-switched networks. The models of Erlang, Engset, Palm, and Delbrouck are generalized to loss and delay systems with multi-rate BPP (Binomial, Poisson, Pascal) traffic streams which all are insensitive to the holding time distributions. The Engset and Pascal traffic streams are also insensitive to the idle time distribution. Engset traffic corresponds to a finite number of users and Pascal traffic results in same state probabilities as Pareto-distributed inter-arrival times. The model is similar to balanced fair scheduling by Bonald and Proutière, but the approach is much simpler and allows for explicit results for finite source traffic and performance measures of each individual service. The model is evaluated by a simple stable algorithm with a minimal memory requirement and a time requirement which is linear in number of streams and number of servers plus queueing positions.

Renewed interest in ALOHA-based Medium Access Control (MAC) protocols stems from their proposed applications to wireless ad hoc and sensor networks that require distributed and low complexity channel access algorithms. In this paper, unlike in the traditional work that focused on mean value (throughput) and stability analysis, we study the distributional properties of packet transmission delays over an ALOHA channel. We discover a new phenomenon showing that a basic finite population ALOHA model with variable size (exponential) packets is characterized by power law transmission delays, possibly even resulting in zero throughput. This power law effect might be diminished, or perhaps eliminated, by reducing the variability of packets. However, we show that even a slotted (synchronized) ALOHA with packets of constant size can exhibit power law delays when the number of active users is random. From an engineering perspective, our results imply that the variability of packet sizes and number of active users need to be taken into consideration when designing robust MAC protocols, especially for ad-hoc/sensor networks where other factors, such as link failures and mobility, might further compound the problem.

In this paper we evaluate the performance of a content distribution service with respect to reliability and efficiency. The considered technology for realizing such a service can either be a traditional client/server (CS) architecture or a peer-to-peer (P2P) network. In CS, the capacity of the server is the bottleneck and has to be dimensioned in such a way that all requests can be accommodated at any time, while a P2P system does not burden a single server since the content is distributed in the network among sharing peers. However, corrupted or fake files may diminish the reliability of the P2P service due to downloading of useless contents. We compare a CS system to P2P and evaluate the downloading time, success ratio, and fairness while considering flash crowd arrivals and corrupted contents.

We study globally optimal user-network association in an integrated WLAN and UMTS . The association problem is formulated as a generic MDP (Markov Decision Process) connection routing decision problem. In the formulation, mobile arrivals are assumed to follow Poisson process and a technique is applied in order to transform the otherwise state-dependent mobile departures into an i.i.d. process. We solve the MDP problem using a particular network model for WLAN and UMTS networks and with rewards comprising financial gain and throughput components. The corresponding Dynamic Programming equation is solved using Value Iteration and a stationary optimal policy with neither convex nor concave type switching curve structure is obtained. Threshold type and symmetric switching curves are observed for the analogous homogenous network cases.