• Edge computing represents a paradigm shift in data processing, where computational resources are moved closer to the data source”typically on the "edge" of the network”such as IoT devices, routers, and gateways. This reduces latency, decreases bandwidth usage, and enables real-time decision-making, making it a significant advancement in distributed computing and an integral part of modern IT architectures.
  
  Edge computing is particularly beneficial in applications where real-time processing is crucial, such as autonomous vehicles, smart grids, industrial IoT (IIoT), and healthcare systems. As organizations collect and analyze massive amounts of data from distributed devices, edge computing alleviates the pressure on centralized cloud servers by offloading some of the computational work to edge devices.
  
  Edge computing is revolutionizing the way data is processed and analyzed in distributed networks, offering faster decision-making, improved efficiency, and enhanced data privacy. By bringing computation closer to where data is generated, edge computing allows for scalable, real-time applications across a range of industries, including healthcare, transportation, and agriculture. In final year projects, edge computing offers students the opportunity to explore cutting-edge technology, enabling them to develop solutions that respond instantly to the challenges of modern IoT environments.

Software Tools and Technologies

• • Operating System: Ubuntu 20.04 LTS 64bit / Windows 10
• • Development Tools: Apache NetBeans IDE 22b / EdgecloudSim 1.0 / CloudSim 4.0.0
• • Language Version: JAVA SDK 21.0.2

List Of Final Year EdgeSim Projects in Edge Computing

• Mobile Cloud Offloading Based Energy-Efficient Decision Making Problem
  Project Description : This project tackles the fundamental challenge for mobile devices: whether to execute a computationally intensive task locally, consuming battery power, or to offload it to the cloud, consuming energy for data transmission. It formulates this as a binary or multi-choice decision-making problem. The proposed framework develops sophisticated algorithms that dynamically analyze factors such as task complexity, network conditions (bandwidth, latency), and the devices current energy state. By solving this optimization problem, the system makes energy-optimal offloading decisions in real-time, significantly extending the battery life of smartphones, IoT sensors, and other mobile devices while maintaining acceptable application performance.
• A Sustainable Platform Based Green Cloudlet Networks in Mobile Cloud Computing
  Project Description : This project designs a sustainable mobile cloud computing architecture centered around a network of cloudlets—small-scale, energy-efficient data centers deployed at the network edge. The platform incorporates "green" IT strategies, such as leveraging renewable energy sources (solar, wind) to power cloudlets and implementing dynamic power management policies that put idle servers into low-power sleep states. The resource management system prioritizes routing user tasks to the "greenest" available cloudlet, minimizing the carbon footprint of the entire network and providing an environmentally sustainable solution for the growing energy demands of mobile computing.
• Enhancing Cloud Service Reliability with Proactive Fault-Tolerance Strategies in Mobile Cloud
  Project Description : Focusing on the reliability of mobile cloud services, this project moves beyond reactive fault tolerance to proactive strategies. It employs predictive analytics and machine learning to forecast potential failures in cloud resources or network connections before they occur. Based on these predictions, the system can proactively migrate virtual machines or tasks from a soon-to-fail node to a healthy one, checkpoint task progress more frequently before instability, or initiate backup processes. This approach minimizes service interruptions, ensures continuous availability for mobile users, and provides a more robust and dependable mobile cloud experience.
• Market Equilibrium-Based Pricing Strategies for Resource Allocation in Edge Computing
  Project Description : This project applies economic theory to edge computing resource management. It models the interaction between edge server providers (sellers) and mobile users or application providers (buyers) as a market. Using concepts from game theory, it designs pricing strategies that reach a market equilibrium—a state where supply (edge resources) meets demand (user tasks) at a fair price. These strategies efficiently allocate scarce edge resources to those who value them most, prevent market manipulation, and incentivize both providers to offer resources and users to use them efficiently, creating a stable and self-regulating ecosystem.
• A Wireless Metropolitan Area Networks(WMAN) Based Optimal Cloudlet Placement and User to Cloudlet Allocation
  Project Description : This project addresses the infrastructure planning problem for deploying cloudlets within a Wireless Metropolitan Area Network (WMAN). It solves two intertwined problems: first, determining the optimal number and physical locations to place cloudlets across the metropolitan area to maximize coverage and minimize access latency for users. Second, it develops efficient algorithms to allocate mobile users to their optimal cloudlet, considering user mobility patterns, wireless link quality, and the current load on each cloudlet. This ensures users within a city-wide network consistently have low-latency access to powerful computing resources.
• Optimizing Joint Scheduling and Cloud Offloading Techniques for Mobile Applications
  Project Description : This research proposes a holistic framework that co-optimizes two critical decisions: the scheduling order of multiple tasks on a mobile device and the offloading decision for each individual task. The joint algorithm considers task dependencies (e.g., Task B requires the output of Task A) and decides which tasks to run locally and in what sequence, and which to offload to the cloud and in what order. This coordinated approach prevents bottlenecks, minimizes the total application completion time, and reduces overall energy consumption more effectively than solving each problem independently.
• Resource Optimization for Efficient Mobile Edge Computing by using Geo-Clustering Strategies
  Project Description : This project leverages geo-clustering, a machine learning technique, to group mobile users based on their geographical locations and similar resource demand patterns. By understanding these spatial clusters, the mobile edge computing (MEC) system can pre-allocate and optimize resources at edge servers that serve each specific cluster. This proactive resource pooling allows for more efficient load balancing, reduces resource fragmentation, and ensures that sufficient computational power is available where and when it is needed most, improving the overall efficiency and responsiveness of the MEC network.
• Optimized Task Scheduling with Deadline-Aware in Mobile Edge Computing
  Project Description : This work focuses on scheduling real-time tasks with strict deadlines in a mobile edge computing environment. The proposed schedulers primary objective is to maximize the number of tasks that complete successfully before their individual deadlines. It employs priority-based queuing models where tasks with tighter deadlines receive higher priority for execution on edge servers. The algorithm efficiently manages the limited MEC resources to meet the temporal constraints of applications like autonomous driving assistance, augmented reality, and industrial automation, where missing a deadline can have critical consequences.
• Energy Efficiency based on the Joint Computation and Communication Co-Operation for Mobile Edge Computing
  Project Description : This project recognizes that energy consumption in MEC involves both computation (on the device and edge server) and communication (data transmission). It proposes a co-operative optimization framework that jointly manages computation offloading policies and communication parameters like transmission power and modulation schemes. By synchronizing these two domains, the system finds a global optimum that minimizes the total energy consumption of the entire mobile-edge system, rather than optimizing each part in isolation, leading to significant gains in overall energy efficiency.
• Efficient Application Offloading based Distributed Multi-dimensional Pricing in Mobile Cloud Computing
  Project Description : This project introduces a distributed market-based model for application offloading where multiple cloud/edge providers compete. It employs multi-dimensional pricing, meaning providers can set different prices for different resource types (CPU, RAM, bandwidth) and service qualities (latency, reliability). Mobile users or brokers then make offloading decisions based on these prices and their applications specific requirements. This distributed mechanism encourages competition among providers, drives down costs for users, and leads to an efficient allocation of resources across the entire mobile cloud ecosystem.
• Optimizing Cloudlet Sharing with Auction-Based Resource Allocation in Mobile Cloud Computing
  Project Description : This project uses auction theory to manage the allocation of shared cloudlet resources among multiple mobile users. In this model, users "bid" for resources based on their urgency and willingness to pay. An auction mechanism (e.g., a Vickrey-Clarke-Groves auction) is then used to determine the winning bids and the corresponding resource allocation. This approach ensures that resources are allocated to the users who value them the most, promotes fair and efficient sharing of public cloudlets, and provides a clear economic model for cloudlet operators to monetize their services.
• Task Allocation with Distributed Truthful Auction Mechanisms in Mobile Cloud Computing
  Project Description : Building on auction-based allocation, this project specifically designs mechanisms that are "truthful" (incentive-compatible). This means that the best strategy for a mobile user is to bid their true valuation for the resources, rather than attempting to manipulate the auction. This property is crucial for preventing fraud and ensuring the auctions stability and fairness. The distributed truthful auction provides a robust and scalable method for task allocation across a wide area mobile cloud network, guaranteeing efficient outcomes even with strategic users.
• Geo-Distributed Mobile Cloud Computing based Decentralized and Optimal Resource Cooperation
  Project Description : This project proposes a fully decentralized framework for resource cooperation among geographically distributed mobile cloud and edge nodes. Instead of a central controller, nodes autonomously negotiate with their neighbors to share computational load, cache data, or forward tasks. Using consensus algorithms and distributed optimization techniques, the system achieves a globally efficient resource pool without a single point of failure. This architecture is highly scalable, fault-tolerant, and well-suited for large-scale, dynamic mobile environments.
• Optimizing Energy Efficient Edge Scheduling with Double Deep Q-Learning in Edge Computing
  Project Description : This project employs advanced reinforcement learning, specifically Double Deep Q-Networks (DDQN), to learn optimal energy-efficient scheduling policies for edge servers. The DDQN agent learns from interactions with the environment by observing system states (server load, task queue, energy levels) and taking actions (scheduling decisions). Over time, it learns a policy that minimizes energy consumption while maintaining performance, effectively adapting to complex and unpredictable workload patterns without requiring pre-defined models or rules.
• Multi-Tiered Services for Resource Provisioning in the Edge for IOT Applications
  Project Description : This work designs a resource provisioning framework that offers multi-tiered service levels for diverse IoT applications. For example, it defines a premium tier with guaranteed low latency and high bandwidth for critical applications (e.g., health monitoring), a standard tier for common applications (e.g., smart home control), and a best-effort tier for non-critical tasks (e.g., data logging). The edge infrastructure dynamically provisions resources to meet the Service Level Agreements (SLAs) of each tier, ensuring efficient resource utilization and cost-effective service differentiation for a wide range of IoT needs.
• Green Mobile Edge Cloud Computing for Multi-User Multi-Task Computation Offloading
  Project Description : This project tackles the complex problem of offloading multiple tasks from multiple users to a green mobile edge cloud. The objective is to minimize the total carbon footprint or energy consumption of the entire system. The proposed solution involves a centralized scheduler that collects tasks from all users and then makes coordinated offloading and scheduling decisions, considering the renewable energy availability at edge locations. This approach ensures that the computational workload is processed in the most energy-efficient way possible across the user population.
• Maximum Processing Capacity with Power Constraints Edge Computing for IoT Networks
  Project Description : This project focuses on maximizing the total processing capacity and throughput of an edge computing system serving a dense IoT network, under a strict total power budget. The optimization involves dynamically adjusting the computational power of edge servers (using DVFS techniques) and the allocation of IoT tasks to balance performance and energy use. The goal is to process the maximum number of IoT data streams or tasks per second without exceeding the available power capacity, which is critical for power-constrained environments like remote sensors or mobile base stations.
• Joint Task Offloading and Resource Allocation Strategies for Multi-Server Mobile-Edge Networks
  Project Description : This research provides a comprehensive strategy for multi-server MEC systems. It jointly solves the problem of which edge server to offload a task to (server selection) and how much computational resource (e.g., CPU cycles) to allocate to that task on the chosen server. This combined optimization ensures that both radio resources (for uplink transmission) and computational resources are used efficiently, preventing scenarios where a task is sent to an overloaded server, thus minimizing overall latency and maximizing system capacity.
• IoT Using Mobile Edge Computing for Privacy Preserving Data Aggregation Scheme
  Project Description : This project leverages Mobile Edge Computing (MEC) as a secure and private intermediary for IoT data aggregation. Instead of sending raw data to the cloud, IoT devices send encrypted or obfuscated data to a nearby MEC server. The MEC server then performs aggregation operations (e.g., calculating averages, sums, max values) on the encrypted data without being able to decipher the individual values. This scheme preserves the privacy of individual IoT devices and users while still providing valuable aggregated insights for data analysts and cloud applications.
• Optimizing the Framework for Edge Node Resource Management in Mobile-Edge Computing
  Project Description : This project designs a comprehensive software framework for managing the lifecycle of resources on individual edge nodes. The framework includes components for resource discovery, monitoring, allocation, and deallocation. It provides APIs for applications to request resources and a policy engine to enforce management rules (e.g., priority, fairness). This optimized framework reduces the overhead of resource management itself, ensuring that the maximum amount of an edge servers capacity is available for executing user tasks efficiently.
• Optimizing Cost Effective Provisioning of Healthcare Data by using Edge-of-Things Computing Framework
  Project Description : This project applies edge computing (the "Edge-of-Things") to healthcare for cost-effective data management. It proposes a framework where sensitive healthcare data from IoT devices (e.g., wearables, sensors) is processed and filtered at the edge before being sent to the cloud. This reduces the volume of data transmitted and stored in expensive cloud storage, lowering costs. It also enables immediate, local analysis for real-time alerts while only sending summarized, anonymized, or long-term trend data to the cloud for deeper analysis, optimizing the entire data pipeline for cost and efficiency.
• Edge Server Placement Solutions for Mobile Edge Computing Environments
  Project Description : This project focuses on the strategic problem of where to physically place edge servers within a network infrastructure. It develops algorithms and models to determine the optimal number and locations of these servers (e.g., at base stations, central offices) to minimize average access latency for mobile users, maximize coverage, and balance load across the servers. The solutions consider user density, mobility patterns, and network topology, providing a blueprint for telecom operators and service providers to deploy a cost-effective and high-performance MEC infrastructure.
• Optimizing Task Prediction and Computation Offloading in Mobile-Edge Cloud Computing
  Project Description : This project introduces a predictive element to offloading. It uses machine learning to predict upcoming computational tasks on a mobile device based on user behavior and application usage patterns. With this forecast, the system can proactively pre-offload necessary code or data to the edge cloud or make preliminary resource reservations. This enables a "just-in-time" offloading process that hides the offloading latency from the user, making the experience seamless and further improving application responsiveness and energy efficiency.
• Improving User Experience Based on Joint Optimization of Data Placement and Scheduling in Edge Computing
  Project Description : This work recognizes that user experience depends on both where data is stored and how tasks are scheduled. It jointly optimizes the placement of popular data (caching) on edge servers to minimize data access time and the scheduling of tasks that need that data. By co-locating tasks with the data they require, the system significantly reduces data retrieval delays. This holistic optimization leads to faster application load times, smoother streaming, and a overall enhanced Quality of Experience (QoE) for the end-user.
• Hybrid Cloud and Edge Environment Based Secure Management
  Project Description : This project creates a unified security management framework for hybrid environments that span public/private clouds and edge nodes. It addresses the unique security challenges of this distributed architecture, such as secure authentication across tiers, consistent policy enforcement, encrypted data-in-motion between edge and cloud, and secure orchestration of workloads. The framework provides a single pane of glass for managing identities, access controls, and threats across the entire hybrid infrastructure, ensuring a consistent and high level of security posture everywhere.
• Optimizing Response Time for Cloudlets in Mobile Edge Computing
  Project Description : This project is dedicated to minimizing the end-to-end response time experienced by mobile users when accessing cloudlets. It involves optimizing every step in the process: the network latency to the cloudlet, the queuing time once a task arrives, the execution time on the cloudlets virtual machine, and the time to send the result back. Techniques include efficient load balancing across cloudlets, implementing low-latency networking protocols, and using lightweight virtualization technologies. The ultimate goal is to make cloudlet access feel instantaneous to the user.
• Mobility-Aware and Caching-Enhanced Task Scheduling Strategies for Mobile Edge Computing
  Project Description : This strategy enhances traditional task scheduling by incorporating two key elements: user mobility prediction and intelligent data caching. It predicts a users movement to anticipate which edge server they will connect to next. Based on this, it can pre-migrate ongoing tasks or pre-fetch relevant data to the next edge server, ensuring service continuity. Furthermore, it caches frequently accessed application data and code on edge servers to reduce redundant downloads, collectively minimizing latency and preventing service disruptions for mobile users.
• Optimizing Task Scheduling in Mobile Edge Computing by using User Mobility Awareness
  Project Description : This project focuses specifically on integrating user mobility awareness into the core task scheduling algorithm. The scheduler does not just see a static user but one that is moving through the network. It uses trajectory prediction to estimate the duration of a users connection to a specific edge server. This allows it to make smarter scheduling decisions, such as prioritizing tasks for users who are about to leave the coverage area or assigning longer tasks to users with stable connections, thereby reducing the number of failed tasks due to user mobility.
• Smart Edge Caching Solutions for Mobile Multimedia Content in Information-Centric Networks
  Project Description : This project designs intelligent caching strategies for edge servers within an Information-Centric Network (ICN) architecture, which focuses on content names rather than locations. For mobile multimedia content (videos, music), the solution uses popularity prediction, user preference analysis, and context-awareness (e.g., location, time of day) to decide what content to cache at which edge node. This maximizes the cache hit ratio, ensuring that popular content is readily available close to users, which drastically reduces video start-up delays, eliminates buffering, and saves significant backhaul bandwidth.
• Resource Augmentation based Economic and Energy Consideration in Mobile Cloud Environments
  Project Description : This project explores the economic and energy implications of resource augmentation—using cloud/edge resources to extend the capabilities of a mobile device. It develops models to analyze the trade-off between the economic cost of renting remote resources and the energy cost of local computation and data transmission. The framework helps users and system designers decide when resource augmentation is economically viable and energy-beneficial, providing a holistic cost-benefit analysis for leveraging mobile cloud computing technologies.