Fog Processing Research Articles

Hygroscopic growth and activation changed submicron aerosol composition and properties in the North China Plain

Abstract. Aerosol hygroscopic growth and activation under high-relative-humidity (RH) conditions significantly influence the physicochemical properties of submicron aerosols (PM1). However, this process remains poorly characterized due to limited measurements. To address this gap, we deployed an advanced aerosol–fog sampling system that automatically switched between PM1, PM2.5 and total suspended particulate (TSP) inlets at a rural site in the North China Plain in the cold season. The results revealed that aerosol swelling due to water vapor uptake influenced aerosol sampling under high-RH conditions by shifting the cut-off size of impactors. At subsaturated high RH (> 90 %), over 25 % of aerosol mass with dry diameters below 1 µm resided in supermicron ranges, while in supersaturated foggy conditions, more than 70 % of submicron aerosol migrated to supermicron ranges. Hygroscopic growth and activation particularly affected highly hydrophilic inorganic salts, shifting a significant number of submicron sulfate and nitrate particles to supermicron ranges, with 27 %–33 % at 95 % ≤ RH ≤ 99 % and more than 78 % under supersaturated foggy conditions. Moreover, more than 10 % of submicron biomass burning organic aerosols grew beyond 2.5 µm during fog events, while fossil-fuel-related organic aerosol (FFOA) remained dominantly in submicron ranges, suggesting inefficient aqueous conversion of FFOA. The two secondary organic aerosol (SOA) factors (OOA1 and OOA2) behaved differently under supersaturated conditions, with OOA2 exhibiting a higher activated fraction despite a lower oxygen / carbon ratio. A substantial increase in organic nitrate and organosulfur mass concentrations in activated droplets during fog events suggested aqueous conversions and formations of brown carbon with potential radiative impacts. Overall, our study highlights remarkably different cloud and fog processing behaviors between primary and secondary aerosols, which would benefit a better understanding of aerosol–cloud interactions under distinct atmospheric conditions.

Sulfate Formation Driven by Wintertime Fog Processing and a Hydroxymethanesulfonate Complex With Iron: Observations From Single‐Particle Measurements in Hong Kong

AbstractFog processing has a significant impact on sulfur chemistry in the atmosphere. This study analyzed three winter fog events in Hong Kong using single‐particle aerosol mass spectrometry (SPAMS) and a Monitor for AeRosols and GAses in ambient air (MARGA). Black carbon (BC)‐related carbonaceous particles with substantial sulfate amounts comprised the largest particle number fraction (56.7%). Sulfate mass concentration decreased during fog due to the cloud's effective scavenging, but fog processing notably promoted sulfate formation at the single‐particle level (average peak area increases of 31.2%). Hydroxymethanesulfonate (HMS), an important S(IV) compound and fog tracer, was identified accounting for up to 12% by particle number fraction. Although pH showed a positive correlation (r = 0.53–0.69) with HMS particles in each fog scenario, a negative overall correlation (r = −0.51) was observed. Further analysis revealed that the higher aerosol acidity (pH 0.65–3.11), promoted Fe dissolution, leading to 49% of HMS particles being mixed with Fe, which potentially facilitated sulfate formation via the Fenton reaction. Additionally, around 40% of HMS‐Fe particles are mixed with oxalate, thereby warranting further attention for their potential to cause more intricate sulfur oxidation processes. This study reveals the initial identification of a high mixed‐state of HMS‐Fe, which could potentially serve as a crucial avenue for the formation of sulfate on individual particulate matter. Considering the persistent augmentation of aerosol acidity in the Asian region, this phenomenon necessitates further investigation and attention.

Secure Computing for Fog-Enabled Industrial IoT.

Smart cities are powered by several new technologies to enhance connectivity between devices and develop a network of connected objects which can lead to many smart industrial applications. This network known as the Industrial Internet of Things (IIoT) consists of sensor nodes that have limited computing capacity and are sometimes not able to execute intricate industrial tasks within their stipulated time frame. For faster execution, these tasks are offloaded to nearby fog nodes. Internet access and the diverse nature of network types make IIoT nodes vulnerable and are under serious malicious attacks. Malicious attacks can cause anomalies in the IIoT network by overloading complex tasks, which can compromise the fog processing capabilities. This results in an increased delay of task computation for trustworthy nodes. To improve the task execution capability of the fog computing node, it is important to avoid complex offloaded tasks due to malicious attacks. However, even after avoiding the malicious tasks, if the offloaded tasks are too complex for the fog node to execute, then the fog nodes may struggle to process all legitimate tasks within their stipulated time frame. To address these challenges, the Trust-based Efficient Execution of Offloaded IIoT Trusted tasks (EEOIT) is proposed for fog nodes. EEOIT proposes a mechanism to detect malicious nodes as well as manage the allocation of computing resources so that IIoT tasks can be completed in the specified time frame. Simulation results demonstrate that EEOIT outperforms other techniques in the literature in an IIoT setting with different task densities. Another significant feature of the proposed EEOIT technique is that it enhances the computation of trustable tasks in the network. The results show that EEOIT entertains more legitimate nodes in executing their offloaded tasks with more executed data, with reduced time and with increased mean trust values as compared to other schemes.

Reactive Oxygen Species Formation and Peroxide and Carbonyl Decomposition in Aqueous Photolysis of Secondary Organic Aerosols.

The mechanism and kinetics of reactive oxygen species (ROS) formation when atmospheric secondary organic aerosol (SOA) is exposed to solar radiation are poorly understood. In this study, we combined an in situ UV-vis irradiation system with electron paramagnetic resonance (EPR) spectroscopy to characterize the photolytic formation of ROS in aqueous extracts of SOA formed by the oxidation of isoprene, α-pinene, α-terpineol, and toluene. We observed substantial formation of free radicals, including •OH, superoxide (HO2•), and organic radicals (R•/RO•) upon irradiation. Compared to dark conditions, the radical yield was enhanced by a factor of ∼30 for •OH and by a factor of 2-10 for superoxide radicals, and we observed the emergence of organic radicals. Total peroxide measurements showed substantial decreases of peroxide contents after photoirradiation, indicating that organic peroxides can be an important source of the observed radicals. A liquid chromatography interfaced with high-resolution mass spectrometry was used to detect a number of organic radicals in the form of adducts with a spin trap, BMPO. The types of detected radicals and aqueous photolysis of model compounds indicated that photolysis of carbonyls by Norrish type I mechanisms plays an important role in the organic radical formation. The photolytic ROS formation serves as the driving force for cloud and fog processing of SOA.

Physical and chemical properties of PM1 in Delhi: A comparison between clean and polluted days

Considering the significance of PM1 aerosol in assessing health impacts of air pollution, an extensive analysis of PM1 samples collected at an urban site in Delhi is presented in this study. Overall, PM1 contributed to about 50 % of PM2.5 mass which is alarming especially in Delhi where particle mass loadings are usually higher than prescribed limits. Major portion of PM1 consisted of organic matter (OM) that formed nearly 47 % of PM1 mass. Elemental carbon (EC) contributed to about 13 % of PM1 mass, whereas SO42− (16 %), NH4+ (10 %), NO3− (4 %) and Cl− (3 %) were the major inorganic ions present. Sampling was performed in two distinctive campaign periods (in terms of meteorological conditions and heating (fire) activities), during the year 2019, each spanning two-week time, i.e. (i) September 3rd-16th (clean days), and (ii) November 22nd-December 5th (polluted days). Additionally, PM2.5 and black carbon (BC) were measured simultaneously for subsequent analysis. The 24-h averaged mean concentrations of PM2.5 and BC during clean days (polluted days) were 70.6 ± 26.9 and 3.9 ± 1.0 μg m−3 (196 ± 104 and 7.6 ± 4.1 μg m−3), respectively, which were systematically lower (higher) than that of the annual mean (taken from studies conducted at same site in 2019) of 142 and 5.7 μg m−3, respectively. Changes in characteristic ratios (i.e., organic carbon (OC)/elemental carbon (EC) and K+/EC) of chemical species detected in PM1 show an increase in biomass emissions during polluted days. Increase in biomass emission can be attributed to increase in heating practices (burning of biofuels such as wood logs, straw, and cow-dung cake) in- and around- Delhi because of fall in temperature during second campaign. Furthermore, a significant increase in NO3− fraction of PM1 is observed during second campaign which shows fog processing of NOX due to conducive meteorological conditions in winters. Also, comparatively stronger correlation of NO3− with K+ during second campaign (r = 0.98 as compared to r = 0.5 during first campaign) suggests the increased heating practices to be a contributing factor for increased fraction of NO3− in PM1. We observed that during polluted days, meteorological parameters such as dispersion rate also played a major role in intensifying the impact of increased local emissions due to heating activities. Apart from this, change in the direction of regional emission transport to study site and the topology of Delhi are the possible reasons for the elevated pollution level, especially PM1 during winter in Delhi. This study also suggests that black carbon measurement techniques used in current study (optical absorbance with heated inlet and evolved carbon techniques) can be used as reference techniques to determine the site-specific calibration constant of optical photometers for urban aerosol.

Task Offloading for Cloud-Assisted Fog Computing With Dynamic Service Caching in Enterprise Management Systems

In enterprise management systems (EMS), augmented Intelligence of Things (AIoT) devices generate delay-sensitive and energy-intensive tasks for learning analytics, articulate clarifications, and immersive experiences. To guarantee effective task processing, in this work, we present a cloud-assisted fog computing framework with task offloading and service caching. In the framework, tasks make offloading decisions to determine local processing, fog processing, and cloud processing with the goal of minimal task delay and energy consumption, conditioned on dynamic service caching. To this end, we first propose a distributed task offloading algorithm based on noncooperative game theory. Then, we adopt the 0–1 knapsack method to realize dynamic service caching. At last, we adjust the offloading decisions for the tasks offloaded to the fog server but without caching service support. In addition, we conduct extensive experiments and the results validate the effectiveness of our proposed algorithms.

An Efficient Method for Implementing Applications of Smart Devices Based on Mobile Fog Processing in a Secure Environment

An Efficient Method for Implementing Applications of Smart Devices Based on Mobile Fog Processing in a Secure Environment

The trade-offs between Fog Processing and Communications in latency-sensitive Vehicular Fog Computing

In a vehicular fog computing paradigm, connected autonomous vehicles are envisioned as processing nodes (i.e. fog nodes) so that end-devices may offload processing tasks to them. As such, both local and distributed processing on fog nodes will depend heavily on wireless network conditions and the current traffic demand. In this work, we investigate the trade-offs on the operation of fog nodes under different vehicle densities and network conditions and formalize a Time Constrained One-Shot Open First Price Auction for resource allocation in vehicular fog computing. Through a large-scale simulation study, we assess important aspects of the performance of fog nodes in Vehicular Fog Computing. We show that current wireless network standards may dictate the limits of processing despite the availability of processing power of fog nodes. Our results indicate the existence of trade-offs on the operation of fog nodes regarding message overhead and processing redundancy to achieve high task completion ratio. Finally, we evaluate the social welfare distribution of the task allocation achieved using the auction where higher message rates lead to higher costs.

Design an efficient internet of things data compression for healthcare applications

The internet of things (IoT) is an ecosystem of connected objects that are accessible and available through the internet. This "thing" in the IoT could be a sensor such as a heart monitor, temperature, and oxygen rate in the blood. These sensors produce huge amounts of information that lead to congestion and an effect on bandwidth in the IoT network. In this paper, the proposed system is based on the Zstandard compression algorithm to compress the sensor data to minimize the amount of data transmitted from the IoT level to the fog level and decrease network overloading. The proposed system was evaluated using compression ratio, throughput, and latency time for healthcare applications. The result showed better calculation through decreased response time and increased throughput for transmitted data compared with the case of non-compressed data. It showed the compression data ratio about 70% of orignial data, maximum number of IoT sensor reads as 100, throughput is 85.43 B/ms, and fog processing delay is 6.25 ms.

Primary Emissions and Secondary Aerosol Processing During Wintertime in Rural Area of North China Plain

AbstractThe vast rural areas often experience more severe haze pollution than megacities during wintertime in North China Plain (NCP), yet the sources and evolution processes of aerosol particles, particularly organic aerosol (OA) remain poorly understood. Here we conducted real‐time measurements of submicron aerosol (PM1) species using a high‐resolution aerosol mass spectrometer at a representative rural site in NCP in winter 2019. Our results showed the dominance of organics (36%) and nitrate in PM1 (22%) in 2019. Positive matrix factorization analysis illustrated similarly important primary sources from traffic emissions, coal combustion, and biomass burning, yet secondary OA (SOA) from photochemical and aqueous‐phase related processing exceeded primary OA (53% vs. 47%). Substantial decreases in primary species and considerable increases in nitrate and sulphate were observed since winter 2018, demonstrating enhanced secondary formation in winter 2019. OA composition changed significantly from clean period to fog events with the contribution of aqueous phase‐related oxygenated OA increasing from 6% to 44%, while the photochemical SOA decreased correspondingly from 51% to 19%. The size distributions of aerosol species also changed by shifting toward large sizes during fog events. Elemental analysis of OA and the Van Krevelen diagram (H/C vs. O/C) illustrated the different roles of photochemical and aqueous‐phase processing during daytime and nighttime, respectively, and aqueous‐phase processing is subject to the formation of organic compounds with high H/C and O/C ratios. The large differences in fog processing of submicron aerosol species between 2018 and 2019 due to different temperatures were also elucidated.

Characterization of chemical and physical changes in atmospheric aerosols during fog processing at Baengnyeong Island, South Korea

In situ interactions between aerosols and atmospheric fog droplets were explored using aerosol physical and chemical measurements and fog samples taken at the Baengnyeong Island Intensive Air Quality Monitoring Station, South Korea during June 16th to July 21st, 2014. To investigate wet scavenging by fog, aerosols were characterized by aerosol mass spectrometer before, during, and after five fog events. Total non-refractory PM1 mass decreased by an average of 22.8% with the onset of fog. The scavenging efficiency (η) of organics was 7–34%, where lower η was associated with lower organic oxidation and hygroscopicity. The η was 16–68% for NH4+, -1–35% for SO4, and 17–86% for NO3−, consistent with previous studies and the relatively high NO3− hygroscopicity. The variation in scavenging was size dependent for both organics and NO3−. Particles with lower NO3−η on 7/18 were smaller (mode = 346 nm), while those with higher NO3−η were larger (593 nm). Differences in mass were also explored before and after fog events. Three episodes showed overall decreases in all the components after the fog event, while two episodes showed overall increases, most notably in organics. Both organic mass and SO2+/H2SO4+ ratio were used to indicate the prevalence of hydroxymethanesulfonate (HMS), an aqueous secondary organic aerosol (aqSOA) formation marker, which increased after the 7/18 fog episode. The aerosol carbon oxidation state decreased toward the end of all fog episodes suggesting possible molecular fragmentation and loss of highly oxidized functional groups. On 7/18, the organic content developed lower O:C and higher H:C moving “up” a slope of −1 in a van Krevelen diagram over the course of the fog episode, indicating carboxylic acid or hydroxycarbonyl loss. This study explores relatively rare time-resolved aerosol fog processing measurements. The results confirm previously established relationships between scavenging and particle size, demonstrate overall increases and decreases in particle mass after fog processing based on dominant local chemistry and wet deposition, and show predominant decreases in organic oxidation after fog processing, thereby, contributing to our understanding of particle processing and organic cycling in the atmosphere.

IoT-inspired Framework for Real-time Prediction of Forest Fire

Wildfires are one of the most devastating catastrophes and can inflict tremendous losses to life and nature. Moreover, the loss of civilization is incomprehensible, potentially extending suddenly over vast land sectors. Global warming has contributed to increased forest fires, but it needs immediate attention from the organizations involved. This analysis aims to forecast forest fires to reduce losses and take decisive measures in the direction of protection. Specifically, this study suggests an energy-efficient IoT architecture for the early detection of wildfires backed by fog-cloud computing technologies. To evaluate the repeatable information obtained from IoT sensors in a time-sensitive manner, Jaccard similarity analysis is used. This data is assessed in the fog processing layer and reduces the single value of multidimensional data called the Forest Fire Index. Finally, based on Wildfire Triggering Criteria, the Artificial Neural Network (ANN) is used to simulate the susceptibility of the forest area. ANN are intelligent techniques for inferring future outputs as these can be made hybrid with fuzzy methods for decision-modeling. For productive visualization of the geographical location of wildfire vulnerability, the Self-Organized Mapping Technique is used. Simulation of the implementation is done over multiple datasets. For total efficiency assessment, outcomes are contrasted in comparison to other techniques.

Emerging investigator series: aqueous-phase processing of atmospheric aerosol influences dissolution kinetics of metal ions in an urban background site in the Po Valley.

Metals are an important atmospheric aerosol component; their impacts on health and the environment depend also on their solubility, dissolution kinetics and chemical form in which they are present in the aerosol (e.g., oxidation state, inorganic salt or oxide/hydroxide, organic complex). In this study, we investigated the impact of fog processing on the solubility and dissolution of metals in PM2.5 samples collected in an urban background site in Padova (Italy). For each sample, we determined the solubility and dissolution kinetics of 17 elements in a solution simulating fog water in the winter season in the Po Valley (pH 4.7, T 5 °C, and water content ∼0.5 g m-3). We also determined water-soluble inorganic and organic compounds having ligand properties. We used the model E-AIM IV to calculate the aerosol liquid water (ALW) content and pH, and we used the model Visual MinteQ to determine the speciation picture of the most important elements under conditions of both deliquescent aerosol (ALW and pH calculated using E-AIM IV, ambient temperature) and simulated fog. We found that the dissolution of Al, Cu, and Fe metal ions, predicted to be largely coordinated with organic compounds under fog conditions, was either immediate or considerably faster in samples collected on days with observed fog events compared with those collected on days having drier conditions. For readily soluble elements, such as As, Cd, Cr, Sr, and Zn, such an effect was not observed. Our study highlights the importance of coordination chemistry in atmospheric aerosol and fog in determining the bioavailability of particle-bound metals.

Serverless data pipeline approaches for IoT data in fog and cloud computing

With the increasing number of Internet of Things (IoT) devices, massive amounts of raw data is being generated. The latency, cost, and other challenges in cloud-based IoT data processing have driven the adoption of Edge and Fog computing models, where some data processing tasks are moved closer to data sources. Properly dealing with the flow of such data requires building data pipelines, to control the complete life cycle of data streams from data acquisition at the data source, edge and fog processing, to Cloud side storage and analytics. Data analytics tasks need to be executed dynamically at different distances from the data sources and often on very heterogeneous hardware devices. This can be streamlined by the use of a Serverless (or FaaS) cloud computing model, where tasks are defined as virtual functions, which can be migrated from edge to cloud (and vice versa) and executed in an event-driven manner on data streams. In this work, we investigate the benefits of building Serverless data pipelines (SDP) for IoT data analytics and evaluate three different approaches for designing SDPs: (1) Off-the-shelf data flow tool (DFT) based, (2) Object storage service (OSS) based and (3) MQTT based. Further, we applied these strategies on three fog applications (Aeneas, PocketSphinx, and custom Video processing application) and evaluated the performance by comparing their processing time (computation time, network communication and disk access time), and resource utilization. Results show that DFT is unsuitable for compute-intensive applications such as video or image processing, whereas OSS is best suitable for this task. However, DFT is nicely fit for bandwidth-intensive applications due to the minimum use of network resources. On the other hand, MQTT-based SDP is observed with increase in CPU and Memory usage as the number of users rose, and experienced a drop in data units in the pipeline for PocketSphinx and custom video processing applications, however it performed well for Aeneas which had low size data units.

Deep reinforcement learning for computation offloading in mobile edge computing environment

Recently, in order to distribute computing, networking resources, services, near terminals, mobile fog is gradually becoming the mobile edge computing (MEC) paradigm. In a mobile fog environment, the quality of service affected by offloading speeds and the fog processing, however the traditional fog method to solve the problem of computation resources allocation is difficult because of the complex network states distribution environment (that is, F-AP states, AP states, mobile device states and code block states). In this paper, to improve the fog resource provisioning performance of mobile devices, the learning-based mobile fog scheme with deep deterministic policy gradient (DDPG) algorithm is proposed. An offloading block pulsating discrete event system is modeled as a Markov Decision Processes (MDPs), which can realize the offloading computing without knowing the transition probabilities among different network states. Furthermore, the DDPG algorithm is used to solve the issue of state spaces explosion and learn an optimal offloading policy on distributed mobile fog computing. The simulation results show that our proposed scheme achieves 20%, 37%, 46% improvement on related performance compared with the policy gradient (PG), deterministic policy gradient (DPG) and actor–critic (AC) methods. Besides, compared with the traditional fog provisioning scheme, our scheme shows better cost performance of fog resource provisioning under different locations number and different task arrival rates.

JOI: Joint placement of IoT analytics operators and pub/sub message brokers in fog-centric IoT platforms

Internet of Things (IoT) systems are expected to generate a massive amount of data that needs to be processed. Given the large scale and geo-distributed nature of such systems, fog computing along with publish/subscribe (pub/sub) messaging has been proposed as possible solutions for coping with processing at scale. However, it is still unclear how practitioners can leverage the benefits of fog computing, e.g., how to optimally place data processing operators and pub/sub brokers. Moreover, current IoT systems typically rely on pub/sub brokers at the cloud, which might diminish the benefits offered by edge or fog processing as the communication between IoT operators has to be mediated by the brokers located in the cloud. To address this shortcoming, we propose to place the IoT application operators and the pub/sub brokers jointly on a network of nodes spanning from edge to the cloud considering various factors such as network topology or the locations of the IoT sensors and the consumers of the IoT applications. Different than the prior works, we specifically consider pub/sub brokers and their unique characteristics in the placement decision. First, we formulate the placement of operators and brokers jointly across edge, fog, and the cloud as a cost minimization problem. Next, we design two low-complexity heuristics. Our simulation results corroborate the argument that a placement in the cloud is usually a good option for IoT use cases, but also reveal the gap to the optimal solution in scenarios with heavier clustering of producers and consumers of sensor data. Studying the optimality gap shows that in such a setting heuristic solutions usually stay under a stretch factor of 2, with a worst case factor of 2.5 for a tabu-based solution and 2.85 for a greedy and a fixed placement in the cloud.

High Production of Soluble Iron Promoted by Aerosol Acidification in Fog

AbstractThe current poor understanding of soluble iron (Fe) yield in atmospheric aerosols leaves two observational facts having not yet been correctly simulated in numerical models: the high Fe solubility in aerosols with low Fe content and, hence, the wide range of observed Fe solubility. Our observation at Qingdao, a coastal city of China, revealed that soluble Fe was produced along with aerosol acidification much more efficiently in fog than under other weather conditions. The median Fe solubility in fog aerosols, 5.81%, was 3.3 times of that in haze aerosols, 5.2 times of that in clear days, and 21.5 times of that in dust aerosols. Involving fog processing in models may reduce the discrepancy in the atmospheric flux of soluble Fe to the ocean between numerical simulations and field observations.

Optimized ANFIS Model Using Hybrid Metaheuristic Algorithms for Parkinson’s Disease Prediction in IoT Environment

Throughout recent years, the progress of telemonitoring and telediagnostics devices for evaluating and tracking Parkinson’s (PD) disease has become increasingly important. The early detection of PD increases the consistency of the treatment of patients and ultimately allows it possible to achieve a rapid diagnostic decision from an experienced clinician. In this paper, a proposed fog-based ANFIS+PSOGWO model provided for Parkinson’s disease prediction. The proposed model exploits the advantages of the grey wolf optimization (GWO) and the particle swarm optimization (PSO) for adjusting the adaptive neuro-fuzzy inference system (ANFIS) parameters with the use of chaotic tent map for the initialization. The fog processing utilized for gathering and analyzing the data at the edge of the gateways and notifying the local community instantly. Compared to other optimization methods, many evaluation metrics used like the root mean square error (RMSE), the mean square error (MSE), the standard deviation (SD), and the accuracy and five standard datasets from repository of UCI machine learning that demonstrated the superiority of the model proposed against the grey wolf optimization (GWO), the particle swarm optimization (PSO), the differential evolution (DE), the genetic algorithm (GA), the ant colony optimization (ACO), and the standard ANFIS model. Moreover, the proposed ANFIS+PSOGWO applied for Parkinson’s disease prediction and achieved an accuracy of 87.5%. The proposed ANFIS+PSOGWO compared in producing positive outcomes better than PSO, GWO, GA, ACO, DE, and some recent literature for Parkinson’s disease prediction. The proposed model produced accuracy for the Parkinson’s disease prediction has outperformed its closest competitors in all algorithms by 7.3%.

5GPy: A SimPy-based simulator for performance evaluations in 5G hybrid Cloud-Fog RAN architectures

The joint cooperation of cloud and fog computing emerges as a new architectural pattern for future 5G networks in order to cope with the increasingly number of mobile elements presented in such networks. Through the use of the cloud, power efficiency can be achieved through centralization of processing. On the other hand, the use of fog processing nodes increases power consumption but helps to decrease the latency of delay-sensitive applications and to increase the coverage of the network. As the use of cloud and fog presents conflicting characteristics, it is important to accurately study their behaviour in order to define the best way to use such a hybrid architecture. In this work we present a three-fold contribution to the study of joint cloud and fog computing architectures. First, we present a hybrid architecture called Cloud-Fog RAN (CF-RAN) that focus on dynamic activation and deactivation of both network and processing resources in order to maintain a balanced operation between the cloud and the fog. Second, we present a performance evaluation model used to analyse the performance of different metrics of CF-RAN. Third, as it is very difficult and costly to build cloud and fog real scenarios, we introduce 5GPy, a SimPy event-driven simulator, publicly available, used to perform small and large scale simulations on architectures such as CF-RAN. We present the architectural details of 5GPy and, by using Integer Linear Program (ILP) and graph-based heuristics to allocate resources in 5G networks, we performed simulations of CF-RAN operation in a small network and in a large network based on a Brazilian city. The results show interesting aspects and trade-offs between cloud and fog computing that were possible to be found with the proposed performance evaluation model and with the 5GPy simulator.

Architectural Considerations for Next Generation IoT Processors

Currently, the growth of the first generation Internet of Things (IoT) system for low-end devices having less memory and infrastructure is challenged due to the lack of engineering support in terms of architecture, technology, hardware, privacy, security, and business. This paper discusses system engineering issues relevant in IoT processors and builds a case for exploring rough set as an approximate computing technique in next generation IoT hardware. The proposed approach has a potential of dealing with inconsistencies arising at layers in hardware infrastructure for IoT applications with a dominance at cloud and fog processing. The authors bring out a research direction in IoT as an engineering discipline with an emphasis on the impact of processor hardware extension.