PaperlayerView exact revision on arXiv

Source-linked AI summary

EdgeChain: An Edge-IoT Framework and Prototype Based on Blockchain and Smart Contracts

Jianli Pan, Jianyu Wang, Austin Hester, Ismail Alqerm, Yuanni Liu, Ying Zhao

arXiv:1806.06185v1cs.DCcs.CRcs.NI

TL;DR

IoT must scale beyond centralized cloud support while protecting weak, vulnerable devices. EdgeChain combines permissioned blockchain, internal coins, credit-based resource management, and smart contracts, and experiments report acceptable integration overhead while providing blockchain-based logging and policy enforcement.

  • Problem

    IoT faces scalability and security challenges because weak devices depend on remote centralized clouds and vulnerable devices can be controlled by malicious hackers.

  • Method

    EdgeChain integrates permissioned blockchain and internal coins with edge resources and IoT accounts, using credit rules, smart contracts, and blockchain records to manage and regulate devices.

  • Results

    Experiments found that integrating blockchain and smart contracts into EdgeChain has overhead within a reasonable and acceptable range while supporting IoT applications.

  • Takeaways & Limitations

    EdgeChain provides a prototype framework for distributed, secure IoT resource management with auditable activities and transactions.

Abstract

from arXiv · show

The emerging Internet of Things (IoT) is facing significant scalability and security challenges. On the one hand, IoT devices are "weak" and need external assistance. Edge computing provides a promising direction addressing the deficiency of centralized cloud computing in scaling massive number of devices. On the other hand, IoT devices are also relatively "vulnerable" facing malicious hackers due to resource constraints. The emerging blockchain and smart contracts technologies bring a series of new security features for IoT and edge computing. In this paper, to address the challenges, we design and prototype an edge-IoT framework named "EdgeChain" based on blockchain and smart contracts. The core idea is to integrate a permissioned blockchain and the internal currency or "coin" system to link the edge cloud resource pool with each IoT device' account and resource usage, and hence behavior of the IoT devices. EdgeChain uses a credit-based resource management system to control how much resource IoT devices can obtain from edge servers, based on pre-defined rules on priority, application types and past behaviors. Smart contracts are used to enforce the rules and policies to regulate the IoT device behavior in a non-deniable and automated manner. All the IoT activities and transactions are recorded into blockchain for secure data logging and auditing. We implement an EdgeChain prototype and conduct extensive experiments to evaluate the ideas. The results show that while gaining the security benefits of blockchain and smart contracts, the cost of integrating them into EdgeChain is within a reasonable and acceptable range.

I. INTRODUCTION

IoT faces scalability and security challenges because constrained devices depend on centralized clouds and are vulnerable to compromise. EdgeChain addresses these challenges with a blockchain-based, resource-oriented framework integrating edge computing, smart contracts, and credit-managed resource use.

  • IoT scalability is challenged by weak devices, remote-cloud dependence, large data volumes, and long device-to-cloud distances.
  • IoT devices are vulnerable to malicious control and botnet formation because many inexpensive devices have limited security capabilities and poor maintenance.
  • Edge computing moves computing, networking, storage, and intelligence closer to IoT devices to address scalability and provide faster response and reduced backbone traffic.
  • EdgeChain operates between edge cloud platforms and IoT applications, providing a framework that can run on platforms such as HomeCloud or Cloudlet.
  • EdgeChain links edge resources to IoT accounts and behavior through permissioned blockchain, internal coins, credit-based management, smart contracts, and blockchain logging.

II. EDGECHAIN KEY APPROACHES AND DESIGNS

EdgeChain uses permissioned blockchain, internal credits, and smart contracts to regulate resource access and IoT behavior without requiring resource-constrained devices to mine. Its design also supports behavior-based security and auditable resource transactions.

  • A. Permissioned Blockchain: Permissioned blockchain provides EdgeChain with a controlled environment and higher throughput without resource-intensive proof-of-work or mining incentives.
  • A. Permissioned Blockchain: Only edge servers perform mining and append blocks, while resource-constrained IoT devices act as blockchain and smart-contract clients.
  • B. Internal Currency and Credit System: Credit coins linked to blockchain accounts determine how much resource each IoT device can obtain from edge servers.
  • C. Resource-oriented, Smart Contracts Based Policy Enforcement and IoT Behavior Regulating: Coin-based management provides auditable, unmodifiable transaction records and enables smart contracts to enforce resource and behavior policies.
  • C. Resource-oriented, Smart Contracts Based Policy Enforcement and IoT Behavior Regulating: EdgeChain regulates devices by behavior and resource use rather than location, addressing limitations of perimeter-based security for compromised internal devices.

D. Evolutionary and Backward Compatible Approach

EdgeChain uses proxies and system-level controls to include legacy IoT devices that cannot install blockchain software. Its modular deployment supports standalone and distributed environments with shared blockchain data and workload handover.

  • D. Evolutionary and Backward Compatible Approach: Some cheap or extremely incapable IoT devices cannot run even lightweight blockchain clients and are classified as legacy devices.
  • D. Evolutionary and Backward Compatible Approach: A proxy lets legacy devices participate without installing new blockchain software or receiving regular updates, while their activities are registered and recorded.
  • E. Standalone Deployment vs. Distributed Deployment: EdgeChain can be deployed as a standalone smart-home box or as multiple distributed boxes in smart-campus and smart-city environments.
  • E. Standalone Deployment vs. Distributed Deployment: Distributed EdgeChain boxes share blockchain and smart-contract data, and edge servers can offload or hand over workloads during busy situations.
  • III. EDGECHAIN FRAMEWORK AND FUNCTIONAL MODULES: The framework sits between IoT devices and edge servers and includes modules for proxying, smart-contract interaction, blockchain services, and application interfaces.

B. Smart Contracts Interface

The Smart Contracts Interface connects IoT applications with deployed smart contracts that automate policy enforcement and link contract outcomes to edge-resource provisioning. Blockchain operations also record activities, while direct device–edge interactions avoid further delay after approval.

  • Pre-programmed smart contracts are triggered by registration, device communication, resource requests, and external data transmission to enforce predefined policies automatically.Web3-based APIs create contract instances that execute encoded functions for specific IoT devices.
  • Smart contracts combine a virtual-coin system representing device trust or resource quotas with account-linked assignments based on behavior and resource type.Each IoT device is bound to a blockchain account and receives coins according to its history and resource requirements.
  • The blockchain server executes smart contracts from device transactions and records device information, requests, and activities in blocks.EdgeChain can use less resource-intensive consensus mechanisms such as PBFT without proof of work.
  • Approved requests grant IoT devices permission to receive extra edge resources, while rejected requests reflect insufficient balance or identified malicious behavior.The Application Interface opens communication channels that trigger resource provision from contract execution results.
  • Smart-contract and blockchain operations add delay, but resource provisioning and subsequent interactions occur directly between devices and edge servers after initial approval.Registration is generally a one-time operation for each device.
  • Edge servers provision computation, memory, storage, networking, and intelligence through individual virtual machines tailored to application requirements.The requirements may include computing capability, bandwidth, latency, and privacy.

IV. EDGECHAIN KEY PROCESSES AND WORKFLOWS

EdgeChain’s key workflows deploy a private, distributed blockchain and smart contracts across edge infrastructure, then expose controlled contract functions for device-driven operations. The design assigns execution authority to edge servers while restricting IoT-device permissions.

  • A. Blockchain Deployment: EdgeChain distributes blockchain software across edge servers, non-legacy devices, and an IoT Proxy, using an Ethereum-based private campus network.Nodes synchronize across the distributed deployment rather than connecting to the live public network.
  • A. Blockchain Deployment: Blockchain deployment begins with a genesis block containing configuration information, including blockhead hash, timestamp, and mining difficulty.Mining difficulty affects mining speed and global system performance.
  • A. Blockchain Deployment: The system restricts mining to the edge server, sets low difficulty, and mines automatically only when unconfirmed transactions exist.These choices balance mining speed, storage use, transaction packing, and edge-server resource consumption.
  • A. Blockchain Deployment: Each node receives the same genesis configuration and a primary blockchain account with public keys and an address for interacting with nodes and smart contracts.The nodes are configured with no discovery to isolate EdgeChain from other blockchains.
  • B. Development and Deployment of Smart Contracts: Smart contracts enforce device registration and edge-resource allocation, and thorough testing is required because redeployment loses data associated with the previous version.The deployment workflow is presented in Fig. 6.
  • B. Development and Deployment of Smart Contracts: After deployment, contracts receive blockchain addresses and are invoked through RPC interfaces, with edge servers executing functions triggered by IoT devices.The edge server has higher authority over contract functions, while IoT devices are limited to basic functions; a proxy supports legacy nodes.

C. Device Registration on Blockchain

Device registration establishes the accounts and attributes EdgeChain uses to manage and monitor IoT devices. The registered data supports admission decisions, behavior auditing, and differentiated handling of legacy and non-legacy devices.

  • C. Device Registration on Blockchain: Registration begins by identifying device type, with a proxy registering legacy devices and non-legacy devices submitting their attributes directly through transactions.The workflow is illustrated in Fig. 7.
  • C. Device Registration on Blockchain: Registered specifications support request admission and include communication details, resource requests, addresses, protocols, data types, and MUD-derived allowed activities.Each device also registers a unique blockchain account address.
  • C. Device Registration on Blockchain: The edge server verifies registration information and controls modification rights while appending priority, coin balance, credit, and request timestamps.Table I summarizes registered attributes, units, examples, and authority differences.
  • D. IoT Behavior Regulation and Activities Management: Behavior regulation and activity management form EdgeChain’s core function for addressing IoT scalability and security.The section covers workflows, resource allocation algorithms, and behavior management.
  • D. IoT Behavior Regulation and Activities Management: Legacy-device traffic is monitored through a gateway sniffer, logged on the blockchain, and matched against registered attributes to identify potentially malicious behavior.The system continues monitoring normal behavior or can trigger blocking actions for malicious activity.
  • D. IoT Behavior Regulation and Activities Management: Non-legacy resource requests are recorded as blockchain transactions, analyzed by allocation contracts, and penalized through reduced coins, lower credit, or service blocking when malicious behavior is identified.The requests support resource-intensive applications such as VR gaming.

2) Resource Allocation Based on Pricing Mechanism:

EdgeChain uses dynamic pricing and credit-based behavior management to allocate scarce edge resources and regulate requests. A heuristic admission algorithm prioritizes requests by calculated price while application priorities reflect QoE requirements.

  • 2) Resource Allocation Based on Pricing Mechanism:: EdgeChain’s resource-allocation objective is to maximize user-request acceptance by linking edge resources, IoT devices, blockchain, and coins through pricing.Pricing determines request ordering and service fees.
  • 2) Resource Allocation Based on Pricing Mechanism:: Request prices dynamically depend on total edge resources, currently available resources, and the requested edge resource.These environmental parameters update the price of each resource request.
  • 2) Resource Allocation Based on Pricing Mechanism:: IoT applications are categorized into four priority levels: urgent monitoring, latency-sensitive tasks, reliable data transmission, and tolerant tasks.Examples include patient monitoring, VR/AR, bank transactions, privacy transferring, light control, and passive monitoring.
  • 2) Resource Allocation Based on Pricing Mechanism:: The pricing mechanism defines unit and total request prices using the symbols and bounds specified in Tables II and the accompanying formulas.The total-price definition constrains c_j to [0, w_j].
  • 2) Resource Allocation Based on Pricing Mechanism:: At each timeslot, the heuristic algorithm rejects requests lacking sufficient resources, then accepts the lowest-price feasible request and updates available capacity iteratively.The algorithm is stated to be solvable in polynomial time.
  • 2) Resource Allocation Based on Pricing Mechanism:: The credit system detects potentially malicious activities or requests and uses incentives or punishments to restrict requests through their effect on edge-server resource allocation.The paper presents this as an ongoing design and primary model rather than a completed scheme.

3) Behavior Management Based on Credit System:

EdgeChain uses credit, pricing, and behavioral rules to regulate IoT resource access. Requests are admitted according to available resources and pricing, while credit changes reward good behavior and restrict persistent misuse.

  • Request Admission: Requests exceeding available resource capacity are denied; otherwise, the system calculates each request’s total price and accepts the request with minimal price.The admission algorithm updates the available edge resources after each accepted request.
  • Behavior Evaluation: A device-specific price threshold classifies requests as potentially bad or good behavior, decreasing or increasing the device’s credit accordingly.The threshold applies when a device runs one application and submits one resource type.
  • Behavior Evaluation: Continuously sending requests at overhigh frequency reduces a device’s credit because it tends to occupy more resources than common use.
  • Behavior Evaluation: Using a network port outside the predefined MUD-file port is treated as abnormal device behavior.
  • Credit-Based Actions: A device with zero credit is blocked from future activities, while other devices receive coin returns based on credit changes.The return calculation uses charged coins, the credit change, and an influence factor.
  • Credit-Based Actions: The credit system controls access to edge services, giving better-behaved devices a higher chance of obtaining more resources.

V. PROTOTYPE AND EVALUATION

The EdgeChain prototype combines an OpenStack edge cloud with heterogeneous IoT clients and blockchain components. The evaluation tests feasibility, application compatibility, and pricing-based resource allocation on this testbed.

  • Evaluation Scope: The evaluation examines prototype feasibility, acceptable performance overhead, compatibility with two IoT applications, and pricing-based resource allocation.
  • Testbed: The testbed contains an OpenStack edge cluster with four Dell PowerEdge R630 servers, one Dell PowerEdge C730x server, and a Cisco 3850 switch.
  • Testbed: Front-end devices include Raspberry Pi 3 computers, Google AIY voice and vision kits, a laptop, and a desktop proxy for legacy devices.
  • Hardware Configuration: Each OpenStack compute node provides 18 CPU cores and 256GB RAM, while mining uses one core and reserves the remaining cores for edge computing.
  • Software Configuration: The software stack uses CentOS 7, Go-ethereum, Solidity, Truffle, Node.js, and OpenStack virtualization for flexible resource scaling.
  • Blockchain Roles: The edge server acts as a full blockchain node and miner, while Raspberry Pis and laptops act as light clients that generate resource-request transactions.

B. Overhead of Blockchain and Smart Contracts Operation

EdgeChain’s blockchain and smart-contract operations add computation, communication, storage, and synchronization costs, but the reported overhead remains compatible with IoT applications. Synchronization and transaction delays remain within the stated real-time requirements.

  • Overhead Evaluation: The evaluation measures computation, communication, and storage overhead from device registration, resource allocation, block mining, and smart-contract interactions.
  • Block Synchronization: 35.9 ms is the average IoT-device time to synchronize a new block, compared with 4.09 ms for the edge server.The paper states that IoT devices still satisfy applications requiring response times below 100ms.
  • Storage Cost: 128.78 KB is the average block size, with up to 208 device registrations per block and approximately 1.8 MB of blockchain data for 1,000 registrations.
  • Smart-Contract Transactions: Less than 50 ms is required to complete one smart-contract transaction, satisfying the stated latency requirement for real-time applications.Regular transactions use a low percentage of CPU resources, while blockchain clients occupy 8% memory even during idle time.
  • Application Overhead: Blockchain has the lowest CPU usage among the evaluated systems and the highest memory usage, though memory remains low when applications run in parallel.
  • Communication Cost: Resource-request transactions impose very low overhead on Raspberry Pi I/O performance and overall network bandwidth.
  • Overall Result: The experiments conclude that blockchain can collaborate with IoT applications in a distributed and secure way with reasonable and acceptable overhead.

D. Resource Allocation Performance of the Pricing Scheme

The pricing scheme is evaluated for request acceptance under different parameter settings and resource supplies. It favors high-priority requests and outperforms FCFS and priority-based multi-level scheduling in the reported comparisons.

  • Evaluation Goal: The resource-allocation evaluation targets request acceptance rates, which mainly depend on the proposed pricing mechanism.
  • Parameter Influence: α does not affect performance because it determines the range in which c_j is located, namely [1, α].
  • Parameter Influence: β values from 1.3 to 1.4 provide the best reported range, while excessively large β values decrease the acceptance rate.β adjusts the impact of application priority, making high-priority requests more likely to be served.
  • Baseline Comparison: With β = 1.35, the proposed pricing mechanism achieves the best acceptance performance compared with FCFS and multi-level priority scheduling.
  • Resource Variation: When total edge resources decrease, the proposed pricing algorithm performs better in the reported acceptance-rate comparison.

VI. RELATED WORK

Related work spans IoT, edge computing, blockchain, and smart contracts, but existing studies often address individual topics or narrowly optimize operational objectives. EdgeChain instead combines these technologies to provision edge resources and regulate IoT-device behavior.

  • Related work covers separate efforts across IoT, edge computing, blockchain, and smart contracts because EdgeChain is interdisciplinary.
  • Closest prior work optimizes edge-server miner profits through game theory and pricing, or device computation offloading through incentives and reputation.
  • Blockchain and smart contracts have been applied to smart homes, 5G fog handovers, virtual-machine orchestration, IoT access control, and data provenance.
  • Edge-computing studies commonly target specific applications or optimize revenue, cost, delay, energy, offloading, migration, chaining, placement, or orchestration.
  • EdgeChain combines a permissioned blockchain, credit-based resource management, and smart contracts to link edge resources with IoT-device accounts, usage, and behavior.
Loading 1806.06185v1…