Catálogo de publicaciones - libros

Regions Of Interest (ROI) image coding is one of the most significant features in JPEG2000. It allows ROIs of the image are of higher importance than background (BG). In this paper, a new and efficient scaling-based method so-called Partial Bitplane Alternating Shift (PBAShift) is described. The new algorithm firstly shifts up partial most significant bitplanes of ROI. Then, the most significant bitplanes of BG coefficients and general significant bitplanes of ROI coefficients are shifted up by bitplanes alternating scaling method. The least significant bitplanes of ROI and BG coefficients are obtained in the original position. The PBAShift method, in addition to supporting efficient single ROI coding, can flexibly code multiple ROIs with degrees of interest in an image. The experiments on remote sensing images show the presented method supports ROI coding of both arbitrary shape and arbitrary scaling without shape coding. Additionally, it can handle complexity multiple ROIs of arbitrary shapes efficiently.

Obtaining traffic load on internal links of the network is crucial for network management and control. Though collecting can be available on each link, such as applying traditional SNMP scheme, the approach would be expensive because it may cause heavy processing load and sharply degrade the throughput of the core routers in high-speed IP backbone. Then monitoring merely at the edge and estimating traffic in the core provides a good alternative way for overcoming such functionality limitations. In this paper, we explore a scheme on deriving internal link load of network with edge-based measurements. Contrast to collecting routing data from core routers that costs much, we propose a route probing method based on hash sampling techniques and IP Measurement protocol between node-pairs. Based on statistical theory, we prove that our approach is effective and present the algorithm. Performance simulation results show the potential of our approach.

Although several promising multicast solutions have been proposed till now; however, the support of multicasting still remains notoriously difficult for switches or routers in networks because of the traffic expansion due to multicast replication. In this paper, we propose to use a Combined Input Separate Output Queued Switch (CISOQ for short) to achieve high performance when loaded with multicast traffic. By giving novel definitions for the waiting time and the queue occupancy of multicast cells, we extend the use of oldest cell first (OCF) and longest queue first (LQF) algorithms from the unicast-only traffic load to the multicast traffic load. Furthermore, we show that 100% throughput can be obtained by a CISOQ switch when it is scheduled by OCF and LQF without speedup or by any maximal matching algorithms, just used in the unicast-only traffic load before, with a speedup of 2. The only assumptions on the multicast traffic pattern are that it is and and that it does not any inputs or outputs. As far as we know, this result is the first theoretical analysis of multicast traffic arrival process till now.

This paper presents a control mechanism in SON(Service Overlay Network) architecture to provide de lay guarantees of different aggregate traffics over IP network. In addition to the bandwidth provision problem of aggregate-traffic in service overlay network, queuing delay in the service gateway is another critical QoS parameter, and diverse multimedia flows have different delay requirements. To assure the delay requirements of aggregate flows in service overlay network, we present a simple but effective adaptable control mechanism, which consists of two-queues with exhausitive service in cyclic order and characterizes the abilities to ensure delay requirements of various aggregate flows according to their burstiness. Moreover, Associated performance is analyzed by the Markov chain and probability generation function, and simulations validate this analysis.

Thanks to the highly popular dual channel capabilities (e.g., GSM plus Bluetooth) in modern handheld personal communication devices, an integrated cellular and ad hoc peer-to-peer network (i.e., a wireless network) has already been widely reckoned as a readily practicable and attractive mobile computing environment. In this paper, we propose a co-operative relaying scheme, called , for such a quasi-ad hoc network, to extend the life-time of low energy level users significantly. More importantly, the energy efficiency of the whole network is also remarkably increased.

The performance of wireless sensor networks (WSNs) is greatly influenced by their topology. WSNs with patterned topologies can efficiently save energy and achieve long networking lifetime. In this paper, we discuss different patterned topologies for constructing WSNs which can provide the required coverage area and the connectivity of all sensor nodes. We compare different performance measures among all patterned topologies, and find that WSNs in strip-based topology can provide the greatest coverage area with the same number of sensor nodes as used for WSNs in other patterned topologies. Strip-based topology also consumes least energy in the routing protocol of flooding. We show that triangle-based topology can provide the highest reliability among all patterned topologies. We also propose several routing protocols for WSNs in patterned topologies, which are based on different parameters when selecting next-hop neighbor. Our protocols require only local information and work in a simple and effective way to achieve energy efficiency.

An efficient interleaving algorithm is applied to reduce the loss of ciphered information when a cipher system transmits over a wireless fading channel environment. As such, a new scheme for deciding the interleaving depth over a wireless environment is described. Simulations confirm that the proposed effective algorithm with a variable interleaving depth produces a better performance over a fading channel than a static depth algorithm with a fixed interleaving depth. Experimental results showed that the BER performance of the proposed efficient interleaving scheme was higher than that of the fixed interleaving depth scheme. Of particular note is that the dynamic allocation algorithm (DAA) reduces degraded error bits by up to 51.5%, compared with static allocation algorithm (SAA) of depth 48 in 224MHz.

Density control is a promising approach to conserving system energy and extending lifetime of wireless sensor networks. Most of previous work in this field has focused on selecting a minimal subset of active sensor nodes for high efficiency while guaranteeing only 1-coverage (or plus 1-connectivity of the network). In this paper, we address the issue of constructing a -connected -cover set of a wireless sensor network for fault tolerance and balance efficiency. We propose a distributed, localized algorithm based on self-pruning for selecting active sensor nodes to form a -connected -cover set for the target region. The performance of the proposed algorithm is evaluated through numerical experiments.

Connecting Ad Hoc Network (MANET) to the Internet will extend Internet into a new territory by making web service available “anytime, anywhere”. However, many simulations and practical experiments have shown that when communication is between mobile nodes in MANET and fixed nodes on Internet, the bandwidth is asymmetrical and the available throughput is greatly limited. In this paper, we present a gateway-centric multi-path routing protocol (GCMPR), which ensures every mobile node has a set of link-disjoint and loop-free paths to gateway. The novel routing algorithm over multiple paths can provide higher aggregate bandwidth and improve the end-to-end delay. Simulations for comparing performance between GCMPR and other connectivity methods show that GCMPR achieves 40% throughput enhancement and remarkable improvement for the end-to-end delay. Furthermore, GCMPR can be easily applied to other scenarios in MANET when some hosts provide special service and are visited frequently.

Ubiquitous computing pursues naturalness and harmony. We think that semantic and adaptation are two important aspects. This paper presents a semantic and adaptive context model for ubiquitous computing. We emphasize the fusion of semantic information and context model in smart vehicle space to support ubiquitous computing. In our work, we use a web ontology language to model ontologies of context including the common essential ontology and the application domain-specific ontology. In addition, we present an application scenario in smart vehicle space. Also, we propose a prototype system of our model.