Cross Layer Research Articles

Weighted Densely Connected Convolutional Networks for Reinforcement Learning

A weighted densely connected convolution network (W-DenseNet) is proposed for reinforcement learning in this work. The W-DenseNet can maximize the information flow between all layers in the network by cross layer connection, which can reduce the phenomenon of gradient vanishing and degradation, and greatly improves the speed of training convergence. The weight coefficient introduced in W-DenseNet, the current layer received all the previous layers’ feature maps with different initial weights, which can extract feature information of different layers more effectively according to tasks. According to the weight adjusted by learning, the cross-layer connection is pruned to remove the cross-layer connection with smaller weight, so as to reduce the number of cross-layer. In this work, GridWorld and FlappyBird games are used for simulation. The simulation results of deep reinforcement learning based on W-DenseNet are compared with the traditional deep reinforcement learning algorithm and reinforcement learning algorithm based on DenseNet. The simulation results show that the proposed W-DenseNet method can make the results more convergent, reduce the training time, and obtain more stable results.

SWIRL: High-performance many-core CPU code generation for deep neural networks

Deep neural networks (DNNs) have demonstrated effectiveness in many domains including object recognition, speech recognition, natural language processing, and health care. Typically, the computations involved in DNN training and inferencing are time consuming and require efficient implementations. Existing frameworks such as TensorFlow, Theano, Torch, Cognitive Tool Kit (CNTK), and Caffe enable Graphics Processing Unit (GPUs) as the status quo devices for DNN execution, leaving Central Processing Unit (CPUs) behind. Moreover, existing frameworks forgo or limit cross layer optimization opportunities that have the potential to improve performance by significantly reducing data movement through the memory hierarchy. In this article, we describe an alternative approach called SWIRL, a compiler that provides high-performance CPU implementations for DNNs. SWIRL is built on top of the existing domain-specific language (DSL) for DNNs called LATTE. SWIRL separates DNN specification and its schedule using predefined transformation recipes for tensors and layers commonly found in DNN layers. These recipes synergize with DSL constructs to generate high-quality fused, vectorized, and parallelized code for CPUs. On an Intel Xeon Platinum 8180M CPU, SWIRL achieves performance comparable with Tensorflow integrated with MKL-DNN; on average 1.00× of Tensorflow inference and 0.99× of Tensorflow training. It also outperforms the original LATTE compiler on average by 1.22× and 1.30× on inference and training, respectively.

Secure and Proficient Cross Layer (SPCL) QoS Framework for Mobile Ad-hoc

A cross layer QoS framework is a complete system that provides required QoS services to each node present in the network. All components within it cooperate together for providing the required services. In existing QoS frameworks there is no security mechanism provided while Security is a critical aspect for QoS in the MANET environment. Cross layer QoS framework tend to be vulnerable to a number of threats and attacks like, over/under-reporting of available bandwidth, over-reservation, state table starvation, QoS degradation, information disclosure, theft of services timing attack, flooding attack, replay attack, and denial of service (DoS) attack, attacks on information in transit and attacks against routing. So it is necessary when designing protocols for QoS framework, the harmony between security and QoS must be present as one impacts the others. In this work we proposed secure and proficient cross layer (SPCL) QoS frameworks which prevents from various types of threats and attacks. The proposed SPCL QoS framework achieves better performance compared to existing QoS frameworks in metrics of throughput, packet drop ratio, end-to-end delay, and average jitter in both condition when malicious node present in the network and when malicious node not present in the network

Enhanced Power Based Jamming Management Method for Mobile Ad Hoc Networks

The portable mobile network contains adaptable node which are shaped discretionarily. Recalling a definitive goal to keep away from impede, cross layer based blockage manage plot is anticipated for lessening the package occurrences in this system. The planned plot contains three stages. In first stage, the cross layer configuration is proposed to guarantee that the data contribution should be possible between the unquestionable layers in convention stack. In second stage, the blockage unmistakable verification plot is analyzed which accomplishes packet disaster rate and stop upscale factor. In third stage, blockage control is master utilizing cross layer move towards. Here the blockage course is settled in context of the way increase, bolster inhabitancy division. To give smallest centralities utilize and staying away from more blockages, vitality utilizes display and multipath organizing plan is required in structures. . In this study, developed an Efficient Energy based Congestion Control Scheme (EECCS) for jamming prevention and to get better energy efficiency of the mobile nodes[1]. The cross layer propose is deployed to get better the network performance. The multipath direction-finding is concentrating to stay away from jamming and to boost network life span.

On the Realistic Radio and Network Planning of IoT Sensor Networks.

Planning and deploying a functional large scale Wireless Sensor Network (WSN) or a Network of Internet of Things (IoTs) is a challenging task, especially in complex urban environments. A main network design bottleneck is the existence and/or correct usage of appropriate cross layer simulators that can generate realistic results for the scenario of interest. Existing network simulators tend to overlook the complexity of the physical radio propagation layer and consequently do not realistically simulate the main radio propagation conditions that take place in urban or suburban environments, thus passing inaccurate results between Open Systems Interconnection (OSI) layers. This work demonstrates through simulations and measurements that, by correctly passing physical information to higher layers, the overall simulation process produces more accurate results at the network layer. It is demonstrated that the resulting simulation methodology can be utilized to accomplish realistic wireless planning and performance analysis of the deployed nodes, with results that are very close to those of real test-beds, or actual WSN deployments.

Cross Layer Power Control and User Pairing for DL Multi-antenna NOMA

In a previous study, Kim et al. proposed a suboptimal user pairing and optimal power control algorithm for maximizing sum capacity in downlink multi-antenna non-orthogonal multiple access (NOMA) systems. First, Kim et al. calculate the correlation between all users. If the channel correlation is greater than the threshold, the user’s channel gain difference is stored in the channel gain-difference set. Then, the first and second large channel gain-differences of the user pairs are selected from the set. Based on the user pairing of Kim et al. in the physical layer, we propose novel iterative user substitution algorithm where a user is substituted if the peak signal-to-noise ratio (PSNR), the video quality, in the application layer increases after substitution. And we propose an optimal power control that minimizes the sum video distortion to maximize PSNR in the application layer. The numerical results show that the proposed cross layer optimal power control and user substitution outperforms Kim et al. by 1.4 dB in PSNR and only 1 dB away from exhaustive upper bound, when SNR = 15 dB and the number of users is 16.

Wireless sensor network design for smart grids and Internet of things for ambient living using cross-layer techniques

Due to the increased demand of wireless sensor networks for their characteristics like low energy consumption, robustness, and low cost in several demanding and complex applications like smart grid, health and safety, traffic and weather updates, there is need of monitoring the infrastructure in a timely manner with high reliability. In this article, a cross-layer data communication scheme target-aware cross-layer technique is proposed to enhance reliability and to reduce the latency in wireless sensor networks. The proposed scheme uses connected dominating set at network layer where nodes in the connected dominating set are directly connected to all the other nodes in the network. Each sensor node sends the data to the nodes present in the connected dominating set, which forwards it to their respective destinations. The proposed scheme reduces the chances of collision resulting reduced delays. Higher packet delivery ratio is achieved with the proposed scheme results in improved reliability. The proposed scheme is outperforming other state-of-the-art schemes in terms of packet delivery ratio, latency, and data throughput with the default 802.15.4 and delay-responsive cross layer in both static and mobile scenarios using network simulator tool.

CL-mWSNs: Cross Layer Model-Based QoS Centric Routing Protocol for Mission-Critical Cooperative Communication in Mobile WSNs

The paper presents a robust QoS centric routing protocol for mission-critical communication over mobile Wireless Sensor Networks (CL-mWSN) that exploits dynamic network states from the different layers of the IEEE 802.15.4 protocol stack to make the routing decision. The CL-mWSN protocol exploits three key layers: application layer, network layer and MAC layer. It exhibits proactive network and node table management, service differentiation, fair resource scheduling and congestion detection, avoidance at the network layer, as well as dynamic link quality estimation and packet injection rate estimation at the MAC layer to assess its candidature as the best forwarding node for QoS-centric mission-critical communication. Simulation reveals that the proposed routing model exhibits higher throughput, minimum loss and deadline miss ratio that augments QoS provision in mobile WSNs.

Evaluation of the effect of knots on rolling shear strength of cross laminated timber (CLT)

The study experimentally and analytically investigated the effect of knots on the rolling shear (RS) strength of 3-ply southern yellow pine cross-laminated timber (CLT). Center-point bending tests and two-plate shear tests were conducted on six CLT configurations composed of different types of cross layer laminations based on knots conditions (No Knots, Sound Knots, and Decayed Knots) and pith conditions (Pith, No Pith). The shear analogy method was implemented to evaluate the RS strength values from the bending test results, which were compared against the results from the two-plate shear tests. The cross-layer laminations with either sound knots or decayed knots improved the RS strength of CLT. In general, the RS predicted by the shear analogy method was conservative.

Inviwo - A Visualization System with Usage Abstraction Levels.

The complexity of today's visualization applications demands specific visualization systems tailored for the development of these applications. Frequently, such systems utilize levels of abstraction to improve the application development process, for instance by providing a data flow network editor. Unfortunately, these abstractions result in several issues, which need to be circumvented through an abstraction-centered system design. Often, a high level of abstraction hides low level details, which makes it difficult to directly access the underlying computing platform, which would be important to achieve an optimal performance. Therefore, we propose a layer structure developed for modern and sustainable visualization systems allowing developers to interact with all contained abstraction levels. We refer to this interaction capabilities as usage abstraction levels, since we target application developers with various levels of experience. We formulate the requirements for such a system, derive the desired architecture, and present how the concepts have been exemplary realized within the Inviwo visualization system. Furthermore, we address several specific challenges that arise during the realization of such a layered architecture, such as communication between different computing platforms, performance centered encapsulation, as well as layer-independent development by supporting cross layer documentation and debugging capabilities.

Cross-Layer Energy Efficient Resource Allocation in PD-NOMA Based H-CRANs: Implementation via GPU

In this paper, we propose a cross layer energy efficient resource allocation and remote radio head (RRH) selection algorithm for heterogeneous traffic in power domain—non-orthogonal multiple access (PD-NOMA) based heterogeneous cloud radio access networks (H-CRANs). The main aim is to maximize the EE of the elastic users subject to the average delay constraint of the streaming users and the constraints, RRH selection, subcarrier, transmit power, and successive interference cancellation. The considered optimization problem is non-convex, NP-hard, and intractable. To solve this problem, we transform the fractional objective function into a subtractive form. Then, we utilize successive convex approximation approach. Moreover, in order to increase the processing speed, we introduce a framework for accelerating the successive convex approximation for low complexity with the Lagrangian method on graphics processing unit. Furthermore, in order to show the optimality gap of the proposed successive convex approximation approach, we solve the proposed optimization problem by applying an optimal method based on the monotonic optimization. Studying different scenarios show that by using both PD-NOMA technique and H-CRAN, the system energy efficiency is improved.

Leaving a Mobile Footprint: Utilizing Data to Combat the 2017 - 2018 Influenza Season

ObjectiveDuring this session, participants will be able to understand how Harris County Public Health utilized data to make informed decisions on how to combat the influenza season.IntroductionThe 2017 – 2018 influenza season was classified by the Centers for Disease Control and Prevention (CDC) as ‘high severity’ across all age groups. Furthermore, CDC noted that this was the first year to be categorized as such, with the highest peak percentage of influenza-like-illnesses (ILI), since 2009. In Harris County alone, there were 2,665 positive flu tests reported in comparison to the previous season at 1,395 positive tests. In response to the severity of this year’s flu season, Harris County Public Health (HCPH) collaborated across the department to deploy five pop up influenza vaccination events utilizing our Mobile Fleets open to the general public.HCPH epidemiologists are able to collect influenza data from multiple systems and compile it into useful reports/tools. These data include latitudinal and longitudinal data, allowing us to create highly localized maps of where influenza has had impacted communities the hardest. This granular data allowed HCPH to target 5 areas with our Mobile Fleet that had a) high levels of influenza and b) generally limited healthcare/public health infrastructure.Our Mobile Fleet is made up of 8 different Recreational Vehicles that have been retrofitted to offer various public health services including: immunizations, medical visits, dental visits, pet adoptions, mosquito and vector control education, and a fresh food market. The Fleet allows HCPH to offer a full menu of public health services anywhere within the County. While our efforts for this abstract were focused on controlling the influenza outbreak, we leveraged the opportunity to engage with the public on multiple issues such as environmental, veterinary, mosquito control, dental health, and accessible healthy food options.MethodsAs positive flu reports mounted, our epidemiology program provided surveillance data of influenza and ILI in Harris County. Data was obtained through multiple sources including: National Electronic Disease Surveillance System (NEDSS), which includes electronic laboratory reporting; National Respiratory Enteric Virus Surveillance System (NREVSS), which includes all flu tests done in laboratories in Houston; and last, the Flu Portal, which school nurses in Harris County upload school absenteeism rates due to ILI. Once collected and compiled, our Geographic Information System (GIS) team used the data to generate spatial maps of Harris County illustrating the disproportionally high rates. Specifically, our GIS team was able to utilize ArcGIS, and cross layer them with the flu data provided from the epidemiologists. Utilizing these maps, HCPH leadership mobilized the preparedness team to lead a data driven response in five different zip codes throughout the county to hold the influenza vaccination events.ResultsThe Mobile Fleet was operational on five separate dates in five separate zip codes during February and March of 2018. Overall, 477 individuals were provided the influenza vaccine. Of those 477, 304 were 18 years or older, with 173 being under 18 years of age.ConclusionsHaving timely and actionable data is an essential first step to understand and stop an outbreak of any size. However, surveillance data alone won't prevent an outbreak from spreading. That data must be married to effective public health action. Our Mobile Fleet is able to deliver precision public health services by targeting communities most affected and vulnerable to the spread of disease. As surveillance geospatial data becomes more granular so too must our public health service delivery modes become more precise and targeted.

A novel selective cross layer based routing scheme using ACO method for vehicular networks

Internet of Vehicles (IoVs) as one of the developing paradigm in the current scenario intends to present a plenty of inventive applications and services. These developing applications force to attain a specific quality of service (QoS) requirements for Vehicle Ad-Hoc Networks (VANETs). Because of high node density and substantial mobility the complexity of VANET increases tremendously, which results the process of routing, security and QoS challenging. This yields an emerging need to design a more productive and intellectual routing protocol in VANET. In this paper we propose a novel selective cross layer design based Ant Colony Optimized Ad-Hoc On Demand Distance Vector (ACO-AODV) routing protocol. The cross-layer design proposed in this work selectively optimizes the information acquired from Application (APP) layer, Transport layer; physical layer and the Media Access Control (MAC) layer in the networks. We used the advantage of ACO to resolve the route selection issue and Particle Swarm Optimization (PSO)based MAC scheduling mechanism for slot assignment during data transmission. Simulation results & Comparisons are presented to show the significance of the proposed routing protocol with improved QoS.

Design, implementation and analysis of cognitive radio enabled intelligent WBAN gateway for cost-efficient remote health monitoring

In this paper, we design and implement an intelligent gateway for Wireless Body Area Network (WBAN) that seeks opportunistic access of licensed Primary User (PU) channels using Cognitive Radio (CR) to facilitate spectral-efficient, cost-efficient and reliable Non-Real Time (NRT) backhaul transmission of WBAN data for ubiquitous healthcare monitoring. The controller is termed herein as BodyCog-BNC. We develop and implement a cross-layer based modified protocol stack at the BodyCog-BNC, comprising of BodyCog-BNC Management Entity (BME) and BodyCog-BNC CR Medium Access Control (BCR-MAC) units, to exploit cross layer message passing to support the novel idea of spectrum agility. A spectral agile BodyCog-BNC is capable of intelligently switching to conventional licensed Wide-Area Access Network (WAN) technology in case of lack of proper PU channels or occurrence of successive handoffs in multiple PU channels. Furthermore, the protocol stack facilitates session management and energy-efficient backhaul CR transmission over a PU channel employing a proposed Inter-Sensing Time Optimization (ISTO) algorithm. ISTO leverages on convex optimization to select the optimum inter-sensing time for a PU channel to maximize the energy-efficiency within a PU interference constraint under sensing uncertainties. Closed form analytical expressions are also derived to estimate the average consumed energy, switching probability, switching time and the cost-efficiency of BodyCog-BNC. Exhaustive performance analysis shows that the BodyCog-BNC (WBAN gateway), provides a cost-efficient solution for different healthcare applications under varied operating conditions and CR cost regimes. Finally, through a prototype implementation and proof of concept we validate the ability of BodyCog-BNC in enabling NRT remote healthcare monitoring.

Design of Multitier Energy Efficient Secure Sensor Network

Wireless sensor network is one of the fastest growing technology for a variety of applications. The main concern in such WSN applications is how to increase lifetime of such networks as they mostly built on sensor nodes which are typically failure-prone and have limited battery power. One of the major solution identified for this is to reduce the energy consumption, and it mainly depends on two factors as the transmission range and amount of data transferred. In this paper, a cross layer approach is used to propose an energy-efficiency and extend the life time of the sensor network which focuses on inter-working of Network and MAC layer. Along with this approach proposed Multitier Energy Efficient Secure Sensor Network (MEESSN) also introduces low cost finding protocol to save energy consumption at node level. The use of MSP 432 ultra-low power microcontroller with routing protocol in the design also enhances power usage of sensor node. At the same time in order to resolve transmission range, security and energy hole issues; low cost routing protocol and fault tolerance algorithm is used.

An Improved Adaptive Modulation and Coding for Cross Layer Design in Wireless Networks

Recent research in cross layer design focuses on a number of integration techniques with physical layer to maintain a strategic distance from the shortcomings of Open Systems Interconnect (OSI) layer and also to guarantee the Quality of Service (QoS) in heterogeneous systems. Amongst the integration techniques, the Rate-Adaptive Modulation and Coding draws more attention of the researchers. In this paper, an improved Adaptive Modulation and Coding (AMC) scheme for Cross layer approach has been proposed with slight modifications in the existing AMC scheme that considerably improves the spectral efficiency of the transmission framework for time varying channels. The existing AMC conspire uses the well-known modulation techniques like Binary Phase Shift Keying, Quadrature Phase Shift Keying and Quadrature Amplitude Modulation with different rates that yield low spectral efficiency. This paper presents an improved AMC with the Gaussian Minimum Shift Keying (GMSK) modulation that yields a spectral efficiency of 35% more than the traditional AMC. Proposed method also results in a maximum throughput of 20–50% more than the traditional AMC. The evaluation of performance metrics like Signal to Noise ratio and error rate clearly depicts that the proposed approach significantly outperforms the traditional AMC.

Determination of Cross layer interference in Heterogeneous network using stochastic geometry

This document describes the determination of the cross-layer interference for a heterogeneous network.HetNet Stands in the statistical parameters using the mathematical tool StoGeo stochastic geometry that is based on the poisson point process, PPP. A two-tiers network with differing power and distribution patterns is proposed, although defined as independent and identically distributed at each tiers in the low power nodes NBP, NHP respectively, and they are presented under the homogenous PPP. Power is considered at each constant tiers and fading and interference distributions are planned under approximation models for gamma distribution.

Cross Layer NOMA Interference Mitigation for Femtocell Users in 5G Environment

In this paper, we propose a joint channel allocation and power control algorithm by using cognitive radio non-orthogonal multiple access (CR-NOMA) for femtocell users (FUs). The aim is to maximize the sum rate of the FUs for guaranteed quality of service (QoS). With an aim to have guaranteed QoS for FUs, we use CR-NOMA at the femto base station (FBS). Then, an algorithm for pairing among strong and weak users is proposed by using the channel gain difference. Using pairing, the NOMA interference between them reduces which results in better channel utilization. Moreover, we differentiate the even/odd number of FUs in a femtocell to provide the QoS for weak users also. For this purpose, OMA is used to get a predefined data rate using a greedy channel allocation algorithm. The power of each FBS is controlled by using the successive convex approximation for low complexity (SCALE) protocol with Karush-Kuhn-Tucker (KKT) conditions. Numerical results demonstrate that the proposed scheme enhances the sum rate and provides guaranteed QoS for CR-NOMA based femtocell users in comparison to the existing conventional OMA based-femtocell techniques.

A Cross Layer Cluster Based Routing Approach for Efficient Multimedia Data Dissemination with Improved Reliability in VANETs

Vehicular ad-hoc network (VANET) is a self-configuring decentralized dynamic network connects the vehicles through wireless links. Hence during the life time of vehicles, VANET may experience frequent link breaks, demands an efficient routing mechanism for networking of data packets among high speed vehicles with less delay. In this paper based on, cross layer autonomous route recovery (CLARR) mechanism, a new cross layer cluster based routing (CCBR) protocol is proposed. The major objective is to set up stable clusters to improve the routing path lifetime by decreasing the link breaks for better multimedia data dissemination. This is achieved by introducing clustering concept into the existing CLARR using mobility metric, then finding reliable relay vehicles to the destination by measuring relative distances among them. Performance of proposed scheme is evaluated through simulation experiments using network simulator-2.34 for different vehicle velocities and compared with existing cross layer based on demand routing protocol, CLARR. Simulation results indicate that the proposed CCBR routing algorithm outperforms over existing CLARR in terms of throughput, energy consumption per vehicle and reliability. Additionally, due to improved reliability, overall delay is reduced to 60% in proposed CCBR for different vehicle velocity and density.

XMSN: An efficient cross‐layer protocol for mixed wireless sensor networks

SummaryThe application of wireless sensor networks (WSNs) technology in monitoring systems is demanding more efficient services to fulfill the requirements of the monitoring task. For this purpose, the simultaneous presence of features such as different communication mediums (air and water) used by nodes and various sizes of data generated by heterogeneous nodes are the key obstacles to build a communication protocol, which can ensure the reliable data delivery. This work terms such WSNs as mixed wireless sensor networks (MWSNs) which contains the aforementioned features. In this paper, we introduce a new cross‐layer protocol for mixed wireless sensor network (XMSN) which can adapt these features. The proposed cross layer protocol XMSN for such mixed environment is implemented and analyzed extensively in Castalia simulator. The performance of XMSN is compared with composition of well‐known protocols, namely, CTP plus BoX‐MAC‐2. The result shows that XMSN has better efficiency in terms of end‐to‐end delay, energy consumption, and goodput than that of CTP plus BoX‐MAC‐2 protocol.