Ransomware attacks cost businesses more than $75 billion/year, and it is predicted to cost $6 trillion/year by 2021. These numbers demonstrate the havoc produced by ransomware on a large number of sectors and urge security researches to tackle it. Several ransomware detection approaches have been proposed in the literature that interchange between static and dynamic analysis. Recently, ransomware attacks were shown to fingerprint the execution environment before they attack the system to counter dynamic analysis. In this paper, we exploit the behavior of contemporary ransomware to prevent its attack on real systems and thus avoid the loss of any data. We explore a set of ransomware-generated artifacts that are launched to sniff the surrounding. Furthermore, we design, develop, and evaluate an approach that monitors the behavior of a program by intercepting the called Windows APIs. Consequently, we determine in real-time if the program is trying to inspect its surrounding before the attack, and abort it immediately prior to the initiation of any malicious encryption or locking. Through empirical evaluations using real and recent ransomware samples, we study how ransomware and benign programs inspect the environment. Additionally, we demonstrate how to prevent ransomware with a low false positive rate. We make the developed approach available to the research community at large through GitHub to strongly promote cyber security defense operations and for wide-scale evaluations and enhancements.

In this paper, we build on top of the MalConv neural networks learning architecture which was initially designed for malware/benign classification. We evaluate the transfer learning of MalConv for malware multi-class classification by extending its contribution in several directions: (1) We assess MalConv performance on a multi-classification problem using a new dataset composed of solely malware samples belonging to different malware families, (2) we evaluate MalConv on the raw bytes data as well as on the opcodes extracted from the reversed assembly samples and compare the results, (3) we validate the MalConv findings about regularization, and (4) we study MalConv performance when using a medium size dataset and limited computational resources and GPU. The obtained results show that MalConv performs equally well for multi-classification and its performance on raw byte sequences is comparable to opcodes sequences. DeCov regularization is shown to improve the accuracy results better than other regularization techniques.

We live in the age of big data and distributed computing. The current large scale computation frameworks are based on a scaling-out approach for distributing tasks over a cluster of commodity machines. Apache Spark is one of these frameworks that has excelled in many computational tasks. Implementation of statistical learning algorithms over Spark is a challenging task. A bad implementation may lead to a significant decrease in performance and a waste of cluster time and money. Poor performance is mostly due to a lack of understanding of the data in hand and Spark's underlying mechanisms more than it is due to a deficit in the framework itself. In this paper, we consider the use case of X2 feature selection which is very popular in supervised learning pipelines. Our implementation follows the algorithm of the Scikit-learn Python machine learning library which is different than the algorithm used by the Spark machine learning library. The Spark ML library implementation of X2 feature selection accepts only categorical features. Our alternative implementation is more suitable for numerical features. We experiment in particular with features of high sparsity such as n-gram counts. We study the best partitioning scheme of the data and the optimal number of partitions. Our experiments are run over the Databricks platform.

The integration of sensors and units in vehicle manufacturing is constantly increasing the data volume generated by vehicles. This requires to support services to handle these data and provide a continuous response to application requests such as collision warnings, lane changing, traffic information, routing information, multimedia streaming, and many others. Add to that the need for achieving the required quality of service is of immense importance. Fog devices deployed along the road are used as a solution to service vehicles. Vehicular ad hoc network (VANET) environment have the property of dynamically changing its cluster connectivity because of the different mobility patterns, which makes it a challenge for vehicular users to access their services. Many preliminary works proposed the use of Software Defined Wireless Network (SDWN) in VANET. In this paper, we study and analyze the use of SDWN in vehicular fog computing environment taking into consideration the suspected high delay between the SDWN controller that is located on the cloud and its switches, the high traffic coming to the controller, the high vehicle speed, and the range of RSU coverage. We have used Mininet-Wifi to implement a vehicular fog computing architecture and simulate different scenarios. Obtained results showed that using SDWN in a VANET environment might not be the best solution in many cases.

Malware detection and classification are attracting more research nowadays due to the increasing number of malware and ransomware instances targeting financial, educational and industrial systems. In artificial intelligence, we witness a resurgence of neural networks against the symbolic school which is manifested by many breakthroughs in gaming, image recognition and natural language processing under the umbrella term of deep learning. Researchers are evaluating deep learning algorithms for static and behavioral malware analysis. In this paper, we benchmark deep learning architectures composed of recurrent and convolutional neural networks. We report results on different techniques such as Long Short Term Memory (LSTM), Gated Recurrent Units (GRU), and one-dimensional Convolutional Neural Networks (1DCNN). We develop an automatic feature extraction component and a hybrid CNN/RNN classification model. We validate our model using the Microsoft Malware Classification Challenge (BIG 2015). Our results show comparable accuracy to approaches requiring manual and thorough feature engineering.

The Long Range Wide Area Network (LoRaWAN) standard was mainly developed to meet the requirements of many emerging Internet of Things (IoT) applications as it provides low power and low cost connectivity over long distances. In LoRaWAN, when an end-device transmits an uplink message, gateways that receive this message transmit it to the network server, which, in turn, selects a single gateway to reply to the end-device. LoRaWAN does not specify how to select the gateway. In this paper, we focus on the gateway selection for downlink communications in LoRaWAN in order to improve the throughput of the network. We aim to present and evaluate several algorithms for selecting the best gateway for downlink while increasing LoRaWAN throughput for different types of gateway deployment. We show that the system throughput depends on this deployment and that balancing the number of end-devices per gateway improves the performance compared to choosing the gateway with the highest signal quality.

Malware detection and malware construction are evolving in parallel. As malware authors incorporate evasive techniques into malware construction, antivirus software developers incorporate new static and dynamic analysis techniques into malware detection and classification with the aim of thwarting such evasive techniques. In this paper, we propose a new approach to static malware analysis, aiming to treat malware analysis as natural language analysis. We propose modeling malware as a language and assess the feasibility of finding semantics in instances of that language. We concretize this abstract problem into a classification task. Given a large dataset of malware instances categorized into 9 classes, we isolate strong semantic similarities between malware instances of the same class and classify unknown instances by strength of similarity to a class. Our approach consists of a proposed method for defining a malware-language, where malware instances are documents written in that language. We use the word2vec model to generate a computational representation of such documents and choose a document-distance as the measure of semantic closeness between them. We classify malware-documents by applying the k nearest neighbors algorithm (kNN). Validating our model using leave-one-out cross validation, we record a classification accuracy of up to 98%. We conclude that we can find, and ultimately manipulate semantics in malware.

When downloading an application by a smartphone user, the application asks the user to accept a set of permissions allowing it to access sensitive information on the phone. It is mostly unclear how, when, and why such data will be used. openPDS is a framework that was proposed to limit such privacy invasion. Although openPDS protects the user's raw data, sensitive personal information, such as location trace, can still be inferred by the service provider by analyzing the accumulated answers. In this paper, we aim to append openPDS with a module that prevents the service provider from reconstructing the trace of users. This module abides with the QoS requirements necessitated by the provider, and defined as the tolerance of the application to inaccurate answers. Our approach was tested on 10 users whose locations traces were recorded for 10 months. Results show that no user trace was successfully reconstructed even when high QoS levels were required.

In the LTE EPC, many network entities and interfaces have to be maintained and updated regularly. Moreover, to accommodate more users, new hardware must be integrated, although rarely used. To address these challenges, the EPC can be moved to the cloud using two modern technologies: SDN and NFV. In this paper, we study the impact of integrating these novel technologies on LTE networks. We propose a hybrid approach for selecting whether to apply NFV or SDN on each gateway at a given time while minimizing the network load taking into consideration key network parameters such as the number of active datacenters, the deployment city population, the intensity at a given time, the QoS class identifier (QCI), and the delay budget. We formulate the SDN decomposition/NFV virtualization selection as an optimization problem where the objective is to minimize the network load subject to a set of constraints. Our results show that our proposed solution is more responsive to the dynamic state of the network such that for a given gateway, at a certain time slot, an SDN decomposition might be the optimal choice; while at another time slot with a different network state, the NFV architecture might be more suitable.

With the rise the of the Internet of Things (IoT) the Routing Protocol for Low Power and Lossy Networks (RPL) gained a lot of interest in the research community mainly for its flexibility to cope with different network topologies and its ability to offer features like Auto-Configuration, Self-Healing, Loop avoidance and detection, etc. Based on certain routing metrics, RPL's Objective Function (OF) assigns ranks to the nodes in the network then selects and optimizes the routes. This paper overviews the most used objective functions then proposes a modification on the Minimum Rank with Hysteresis Objective Function (MRHOF) which takes into consideration two metrics instead of one, to get more reliable and optimized routes.

Long Term Evolution (LTE) was developed by 3GPP to cope with the increasing demand for better Quality of service (QoS) and the emergence of bandwidth-consuming multimedia applications. Today's data transmission networks face extreme challenges in providing high data rate and low latency. Scheduling paradigms such as Round Robin, Best Channel Quality Condition and Proportional Fair are commonly adopted in current LTE downlink scheduling algorithms, but they are far from optimal for satisfying latency requirements. In this paper, we first survey the state of the art downlink scheduling algorithms in LTE and identify their main challenges. We then formulate the LTE downlink scheduling problem as an optimization problem in order to meet the flow deadlines, then incorporate the formulation within the surveyed scheduling algorithms, to produce better performance. We consider strict deadlines for different types of packets with the goal of maximizing resource distribution. Additionally, in our formulation the buffer state for each user is taken into consideration in order to minimize the packet loss. We evaluate the proposed formulation using LTE-Sim and study its positive impact on the existing LTE downlink scheduling algorithms; the performance in terms of QoS, packet loss and fairness is improved throughout all evaluations.

In LTE, tracking Areas (TAs) are used to group cells where each cell is assigned to one TA and each user equipment (UE) registers with one TA. A TA may be connected to more than one mobility management entity (MME), which records the current TAs of its UEs. The home subscriber server (HSS) records the current MME for each UE. When a UE moves to a new TA, a TA update procedure is triggered to update the MME in case of an intra-MME move and the HSS in case of an inter-MME move. When the UE is called, the paging procedure is triggered to determine the UE's current cell, by broadcasting a paging message in all cells of the UE's current TA. The smaller the TAs are, the smaller the number of cells needed to be paged but more often TA updates. On the other hand with a larger TA, the paging cost increases and the TAU signaling decreases. Hence designing TAs is a problem as it affects both paging and TAU signalling cost. An initial optimal TA configuration cannot guarantee a low signaling overhead because the UEs mobility might alter their distribution. Therefore, the TAs need to be reconfigured to account for the new distribution of UEs. In this paper, we use the Integer Programming Model to solve the TAs reconfiguration problem in LTE networks. We then evaluate the performance of the proposed approach and compare it to the tabu search and the genetic algorithm based solutions to study how far from optimality those solutions are.

SIP is amongst the most popular Voice over IP signaling protocols. Its deployment in live scenarios showed its vulnerability to flooding attacks. In this paper, we present a SIP flooding attack detection mechanism that dynamically detects SIP flooding attacks and correlates in real time the temporal characteristics of SIP reliable mechanism and the number of received INVITE requests. Experimental results show that the proposed mechanism is able to detect SIP flooding rapidly and does not suffer from false alarms. When compared to other similar approaches in literature, the proposed approach outperformed the other approaches in terms of detections speed and accuracy.

Huge amounts of data and personal information are being sent to and retrieved from web applications on daily basis. Every application has its own confidentiality and integrity policies. Violating these policies can have broad negative impact on the involved company's financial status, while enforcing them is very hard even for the developers with good security background. In this paper, we propose a framework that enforces security-by-construction in web applications. Minimal developer effort is required, in a sense that the developer only needs to annotate database attributes by a security class. The web application code is then converted into an intermediary representation, called Extended Program Dependence Graph (EPDG). Using the EPDG, the provided annotations are propagated to the application code and run against generic security enforcement rules that were carefully designed to detect insecure information flows as early as they occur. As a result, any violation in the data's confidentiality or integrity policies is reported. As a proof of concept, two PHP web applications, Hotel Reservation and Auction, were used for testing and validation. The proposed system was able to catch all the existing insecure information flows at their source. Moreover and to highlight the simplicity of the suggested approaches vs. Existing approaches, two professional web developers assessed the annotation tasks needed in the presented case studies and provided a very positive feedback on the simplicity of the annotation task.

Accessing remote cloud services usually involves high network latencies. To deal with this issue, several models of local cloudlets have been proposed, but most of which are based on powerful desktop machines running continuously and constantly to provide services to nearby users. In this paper, we propose a more ubiquitous solution based on less powerful but considerably more accessible machines, namely mobile smartphones. We envision a network of cloudlets distributed within a geographic area, connected to a root server that keeps track of their online presence and resource availability. Our framework is especially suitable for environments where cloud clients are interested in network and software services, in information and data generated and updated by applications or cached from accesses to remote cloud centers. We use the CloudSim simulator to evaluate the performance of the proposed solution and compare it with a peer solutions found in the literature.

In LTE networks, tracking areas (TAs) design affects the signaling cost of user equipments (UEs) during mobility. Indeed, when the latter are called, the mobility management entity (MME), which records the TA in which the UE is registered, broadcasts a paging message in the UE's TA to determine the exact cell through which the call should be delivered. When a UE moves to a new TA, a TA update might be performed either to update the MME or the home subscriber server (HSS) depending on whether the move is intra-or inter-MME. TA update and paging result in a significant amount of signaling overhead. The TA design is usually optimized during the network planning phase. However, the mobility of UEs might turn the initial TA configuration inefficient requiring TA reconfiguration. In this paper we use the tabu search (TS) heuristic to solve TAs reconfiguration problem in LTE networks taking into consideration both signaling overhead and reconfiguration cost. We compare our solution with a variation of the genetic algorithm (GA) based solution found in the literature. Preliminary numerical results showed that the TS algorithm outperformed the GA due to the TS different memory mechanisms used to avoid moves that might lead to non-feasible solutions.

The popularity of Mobile Ad Hoc networks (MANETs) continues to increase due to the technological advances in wireless radios and wireless networks. Given the unpredictability of the wireless medium, one of the major challenges in MANETs remains to be the efficiency of the underlying routing protocol. The Optimized Link State Routing Protocol (OLSR) is one of the most popular routing protocols being deployed in MANETs. OLSR uses multipoint relay mechanism (MPR) to maintain the topology information at each node. Instead of using blind broadcasting where each node sends a control message to all its neighbors, and consequently every neighbor forwards the message to its neighbors, MPR selects a small set of nodes to do the broadcasting such that all network nodes are covered. In this paper, we propose improving further the performance of OLSR by introducing the concept of Mobile Agent (MA) when maintaining the MANET topology information, resulting in MA-based OLSR (MOLSR). In MOLSR, MA will substitute MPR and will rely on message unicast instead of broadcast, a major improvement to current OLSR in terms of traffic overhead. In MOLSR, every node creates and launches a MA that intelligently travels the network and returns with the full topology. In case of an MA loss, MOLSR automatically recovers the MA and makes sure it covers the whole network. Extensive simulation results show that MOLSR outperforms OLSR in various areas, namely the total control messages, network utilization, and reliable message transfer.

A peer-to-peer (P2P) system is known for its scalability and dynamic nature, where nodes can join and leave the system easily and at any time. These networks are susceptible to malicious behavior such as nodes dropping messages and misleading other nodes. P2P routing protocols are not immune against such incidents. Additionally, most secure routing protocols in the literature suffer from false negatives. In this paper, we propose to use the K-nearest neighbor (K-nn) algorithm in order to reduce false negatives in P2P secure routing protocols. We incorporate the proposed algorithm in a chord based trust aware P2P routing protocol, and evaluate its performance using the PeerSim simulator. Preliminary simulation results demonstrate that the proposed algorithm reduces the rate of false negatives without impacting the malicious node detection rate.

Abstract This work proposes a replication scheme that is implemented on top of a previously proposed system for MANETs that cache submitted queries in special nodes, called query directories, and uses them to locate the data (responses) that are stored in the nodes that first request them, called caching nodes. The system, which was named distributed cache invalidation method (DCIM), includes client‐based mechanisms for keeping the cached data consistent with the data source. In this work, we extend DCIM to handle cache replicas inside the MANET. For this purpose, we utilize a push‐based approach within the MANET to propagate the server updates to replicas inside the network. The result is a hybrid approach that utilizes the benefits of pull approaches for client server communication and those of push approaches inside the network between the replicas. The approach is analyzed analytically, and the appropriate number of replicas is obtained, where it was concluded that full replication of the indices of data items at the query directory and two‐partial replication of the data items themselves makes most sense. Simulation results based on ns2 demonstrate the ability of the added replication scheme to lower delays and improve hit ration at the cost of mild increases in overhead traffic. Copyright © 2013 John Wiley & Sons, Ltd.

Several scheduling techniques were designed for the base station BS of IEEE 802.16e wireless interoperability of microwave access networks. However, depending on the BS scheduler alone to determine the servicing order of each connection might affect the accuracy of the scheduling process because the BS does not necessarily have enough up-to-date information about the current state of the connections at the subscriber station. In this paper, we propose a preemption-based scheduling algorithm that focuses on improving the quality of service requirements of real-time service flow classes. The proposed algorithm incorporates two schedulers, one at the BS and another one at the subscriber station. We have implemented and integrated the proposed algorithm with the network simulator NS2 using the Network and Distributed Systems Laboratory wireless interoperability of microwave access module. Simulation results have shown that the proposed approach outperforms other scheduling algorithms in terms of enhancing the throughput and the average delay of real-time quality of service classes. Copyright © 2013 John Wiley & Sons, Ltd.

A peer-to-peer (P2P) system is known by its scalability and dynamic nature where nodes can join and leave the system easily and anytime. These networks are susceptible to malicious behaviors such as nodes dropping messages and misleading requesting nodes. P2P routing protocols are not immune against these misbehaviors. Therefore, detecting and dealing with malicious nodes will certainly lead to more reliable and secure system. In this paper, we propose a trust aware system for P2P routing protocols. The proposed system analyzes constantly the behaviors of all nodes to determine their trust-worthiness then classify them accordingly isolating the ones deemed malicious. It tracks the nodes' reputation based on evaluation reports from the nodes themselves. The credibility of nodes that are inaccurately evaluating other nodes is also monitored, thus, malicious evaluations would not affect other nodes' reputation. We have integrated the proposed approach with several P2P routing protocols and evaluated their performance through simulations measuring parameters such as request delivery ratio, malicious detection, and false negatives. Results show that the proposed approach improves significantly the performance of P2P routing protocols.

LTE uplink frequency scheduling algorithms have neglected the user equipment's (UE) QoS requirements, relying only on the time domain to provide such requirements when creating the allocation matrix for the next transmission time interval. Two time domain paradigms exist for creating the resource allocation matrix: channel-dependent and proportional fairness. The channel dependent paradigm considers mainly the channel quality of UEs, allowing for users with high channel quality to get assigned most resources. The proportional fairness paradigm allocates resources to users based on the ratio of their channel condition over their lifelong service rate, allowing for users with low channel conditions to get some resources, but fewer than those with better channel conditions. Even though the proportional fairness paradigm's main focus is to achieve high system throughput without starving any user, it does not account for QoS requirements in many scenarios especially when UEs with high priority data pending for transmission have worst channel conditions than those with lower priority data. In this paper we propose a QoS-aware resource allocation paradigm for LTE uplink scheduling that gives more advantage to UEs having high priority data, while not starving other users. The proposed approach is scalable and mobility aware where the dynamic nature of the network is taken into account while devising the algorithm. When simulated using NS3, the proposed algorithm produced very promising results and outperformed the state-of-the-art approaches presented in literature.

Devices in mobile ad hoc networks (MANETs) have limited power resources and may be equipped with different transmission technologies. These characteristics make MANET traditional routing protocols inconvenient in a heterogeneous environment. In this paper, we propose a power aware routing protocol for a MANET formed of heterogeneous nodes. The proposed approach takes into consideration the battery status of nodes when building the routing table. It aims to avoid exhausting the nodes that are falling on optimal routes across the network, thus providing better connectivity and extending the network lifetime. We implemented the protocol as an extension to JiST/SWANS network simulator, and compared its performance to other heterogeneous and power aware routing protocols found in the literature.

A wireless sensor network (WSN) is a collection of tiny and limited-capability sensor nodes that report their sensed data to a data collector, referred to as a sink node. WSNs are used in many applications, but are challenged by memory and energy constraints. To address these issues, solutions have been proposed on different levels including the topological level where multiple sinks can be used in the network to reduce the number of hops between a sensor and its sink node. Topological level solutions are very crucial in time-sensitive applications where the maximum worst case delay incurred by a message to get from a sensor to the corresponding sink should be minimal or at least less than a certain value. In turn, the maximum worst case delay can be minimized by choosing near optimal locations of the sinks. Consequently the network lifetime will be extended since the energy consumed by the sensor nodes will be reduced. In this paper, we propose an efficient and robust approach based on Particle Swarm Optimization (PSO) heuristic to solve the multiple sink placement problem; more specifically we use Discrete PSO (DPSO) with local search (LS). We start by formulating the problem then discretizing it and finally applying PSO while introducing local search to the inner workings of the algorithm. When compared to Genetic Algorithm-based Sink Placement (GASP), which is considered the state-of-the-art in solving the multiple sink placement problem, our approach improved the results in most scenarios while requiring less runtime.

The IEEE 802.16 standard, the service data units deemed to exceed the granted bandwidth may be delayed or dropped. In addition, nodes would behave maliciously by requesting more resources than their requirements just to deny other nodes from having access to the network. In this paper, we propose a trust-aware adaptive monitoring scheme that can be integrated with the basic architecture of WiMAX network to monitor if a service flow exceeds its maximum sustained traffic rate and whether the resources allocated to subscriber stations are significantly used or not. To detect and react to malicious behaviors, the scheme gradually establishes a trust relationship between the different entities of the network. We have evaluated the performance of the proposed scheme using the network simulator NS-2 while measuring several network parameters such as throughput.

This paper proposes a dynamic energy efficient clustering algorithm for mobile ad hoc networks (MANETs). The proposed algorithm elects first the nodes that have a higher energy and less mobility as cluster-heads, then periodically monitors the cluster-heads' energy and locally alters the network topology or the clusters to increase the network lifetime by reducing the energy consumption of the suffering cluster-heads. The algorithm employs two dynamically-computed energy thresholds: a yellow threshold that is used to achieve some sort of local load balancing by distributing the load equally among the adjacent cluster-heads, and a red threshold that is used to trigger local re-clustering in the network. Simulation results show that the proposed algorithm outperformed the weight clustering approach found in the literature.

This paper introduces a cooperation-based database caching system for Mobile Ad Hoc Networks (MANETs). The heart of the system is the nodes that cache submitted queries. The queries are used as indices to data cached in nodes that previously requested them. We discuss how the system is formed and how requested data is found if cached, or retrieved from the external database and then cached. Analysis is performed and expressions are derived for the different parameters, including upper and lower bounds for the number of query caching nodes as well as the average load they experience, generated network traffic, node bandwidth consumption, and other performance-related measures. Simulations with the ns-2 software were used to study the performance of the system in terms of average delay and hit ratio, and to compare it with the performance of two other caching schemes for MANETs, namely CachePath and CacheData. The results demonstrate the effectiveness of the proposed system in terms of achieved hit ratio and low delay.

Testing web services impose many challenges to existing testing methods, techniques, and tools; especially those available to traditional applications. Composed web services increase these challenges by requiring additional validation and verification efforts. Structural-based testing approaches have been thoroughly researched for traditional applications; however, they have not yet been examined, as a methodology, for testing composed web services. In this work, we introduce a formal model for an abstract-based workflow framework that can be used to capture a composed web service under test. We then define a set of applicable structural-based testing criteria to the framework. Finally we outline a promising line of testing criteria that can be applied to this framework.

In today's MANET, nodes can be equipped with one or more wireless communications interfaces that support different wireless technologies such as WIFI IEEE 802.11 and Bluetooth. Each technology implements distinct hardware and logical specifications. Current AODV routing protocol becomes inefficient when used in a network formed of heterogeneous devices. Indeed AODV in its nature search the paths between homogeneous devices and ignore the ones between heterogeneous devices. In this paper, we propose a heterogeneous AODV protocol that interoperates between WIFI and Bluetooth technologies in a single MANET. The extended algorithm takes into consideration several factors when building the routes such as network stability, traffic, conversion rate; in addition to the shortest path.

IEEE 802.16 standard defines the specifications for emerging WiMAX networks. It did not however define the scheduling algorithms that determine the uplink and downlink bandwidth allocation. This paper proposes a preemptive deficit fair priority queue (PDFPQ) scheduling architecture for QoS management for the IEEE 802.16 standard. The proposed scheduling architecture is an extension of the DFPQ scheduling technique found in the literature. It enhances the QoS requirements of real time polling service (rtPS) flow class, and improves its delay and throughput. Compared to DFPQ, preliminary results show that a significant rtPS delay reduction and throughput increase can be realized with our new scheduling technique.

Handheld devices in mobile ad hoc networks (MANETs) use different communication technologies (e.g., Bluetooth and IEEE 802.11) and are equipped with different software/hardware processing capabilities. This multi-level heterogeneity poses a serious challenge on the performance of MANETs and adds a number of characteristics, complexities, and constraints to their design. This paper proposes a communication abstract model that supports interoperability and coexistence between heterogeneous mobile devices. The model is represented through a three-layered paradigm: a communication service layer that deals with available communication capabilities; a communication capability layer that is responsible for controlling and managing the lower communication service layer and translating the high level functions of the available capabilities; and a communication interface layer that provides the applications with a unified public interface that supports various functionalities. Simulation results show that the proposed model, when used in a heterogeneous MANET environment, increases its performance in term of reliability and power saving, and yields a significant reduction in path length and delay

This paper proposes a model for caching database data in mobile ad-hoc networks through caching the queries and their responses. The approach makes a distinction between the queries and their responses and caches them on separate mobile nodes. This method is driven by the fact that queries do not become invalid unless when the design of the database entities changes while data changes depending on the application that feeds information into the database. The architecture we use is hierarchical and employs three types of designated nodes: a query directory (QD), service manager (SM), and caching node (CN). The one or more QDs are responsible for caching the queries and are assigned and supervised by the SM that also oversees the mobility activities and the availability of nodes in the network and makes "managerial" decisions accordingly, including appointing backup nodes. With this model, any node that joins the ad hoc network will either contribute services to other nodes (willingness to become an SM, a QD, or a CN) or consume services offered by other nodes. This model attempts to coordinate the query executing and query caching mechanisms in a seamless manner while maintaining minimal communication among nodes. We present preliminary results of a model that was simulated using the NS-2 software and show the viability of the proposed approach.

the IEEE 802.11e standard has been introduced recently for providing quality of service (QoS) capabilities in the emerging wireless local area networks. This standard introduces a contention window based enhanced distributed channel access (EDCA) technique that provides a prioritized traffic to guarantee the minimum bandwidth needed for time critical applications. However, the EDCA technique resets statically the contention window of the mobile station after each successful transmission. This static behavior does not adapt to the network state hence reduces the network usage and results in bad performance and poor link utilization whenever the demand for link utilization increases. This paper proposes a new adaptive differentiation technique for IEEE 802.11e wireless local area networks that takes into account the network state before resetting the contention window. The performance of the proposed technique is evaluated compared to the original differentiation techniques of the IEEE 802.11a and IEEE 802.11e standards. Preliminary results show that the proposed adaptive technique enhances the channel utilization and increases throughput.