Conference PaperPDF Available

Edge AI and Blockchain for Privacy-Critical and Data-Sensitive Applications



The edge and fog computing paradigms enable more responsive and smarter systems without relying on cloud servers for data processing and storage. This reduces network load as well as latency. Nonetheless, the addition of new layers in the network architecture increases the number of security vulnerabilities. In privacy-critical systems, the appearance of new vulnerabilities is more significant. To cope with this issue, we propose and implement an Ethereum Blockchain based architecture with edge artificial intelligence to analyze data at the edge of the network and keep track of the parties that access the results of the analysis, which are stored in distributed databases. A use case of edge AI for ECG feature extraction and real-time support of multiple sensor nodes is analyzed in the experiments.
Edge AI and Blockchain for
Privacy-Critical and Data-Sensitive Applications
A. Nawaz1,2, T. N. Gia2, J. Pe˜
na Queralta2and T. Westerlund2
1Shanghai Key Laboratory of Intelligent Information Processing, Fudan University, China
2Turku Intelligent and Embedded Robotic Systems (TIERS), University of Turku, Finland
Emails: 1{nanum18, hbkan}, 2{jopequ, tunggi, tovewe}
Abstract—The edge and fog computing paradigms enable
more responsive and smarter systems without relying on cloud
servers for data processing and storage. This reduces network
load as well as latency. Nonetheless, the addition of new layers
in the network architecture increases the number of security
vulnerabilities. In privacy-critical systems, the appearance of
new vulnerabilities is more significant. To cope with this issue,
we propose and implement an Ethereum Blockchain based
architecture with edge artificial intelligence to analyze data at the
edge of the network and keep track of the parties that access the
results of the analysis, which are stored in distributed databases.
Index Terms—Blockchain; Edge Computing; AI; E-Health;
U-Health; IoT; Internet of Things; Fall Detection; Ubiquitous
Health; Ethereum;
Users and organizations are becoming increasingly aware
of the importance and significance of protecting personal data
and online privacy. This is a particularly critical issue in the
IoT, where numerous security challenges have been identified
by the research community. In recent years, a wide variety
of IoT platforms and applications have adopted the use of
blockchain technology to mitigate multiple privacy risks and
allow secure transactions without the need for a trusted party.
Nevertheless, current integrations of blockchain within the
IoT have been focusing on securing communication without
changing the interaction topology. Exploiting the fog and
edge computing paradigms, we propose an extension of the
Ethereum blockchain to resource-constrained devices. With
our proposed platform, end-devices can negotiate directly with
third parties regarding the use of their data. This ensures
data owners are always aware of transactions involving their
data. In addition, because of the immutable nature of the
blockchain, all transactions are recorded and auditable, which
further reduces the possibilities of misuse of private data. We
have implemented and validated the proposed platform in a
real application, demonstrating its potential for integration of
IoT devices with scarce computational capabilities. To enhance
privacy-critical systems, edge based AI techniques has been
implemented to restrict raw data to its producers only. But,
this domain still lacks the owner control over their sensitive
health data, where owner can process sensitive information
by using neural networks and sell statistics to the interested
clients. Furthermore, this reduces the network load and the
latency of Blockchain transactions [1], [2].
To make data access policies accessible at each level re-
searchers proposed blockchain based systems integrated with
edge computing. By implementing AI at edge nodes further
decreased privacy vulnerabilities. Mamoshina. P et. al pro-
posed access policies to accelerate the private patients data
and implement deep learning algorithms to turns raw data
into strong useful information which can be used in bigger
perspectives [1]. In [3], Mackey et. al proposed blockchain
based data privacy control opportunities and challenges which
are significant enough in healthcare applications. A similar
approach was presented by Peterson et. al [4].
To exclude intermediaries involve in data transactions in
edge devices, we define a platform in which the Blockchain
paradigm is extended into scarce computing devices. Ethereum
blockchain is used as a service platform to run smart contracts
to make the system autonomous in therms of its’ commu-
nication, processing and data dealing. A private ethereum
blockchain network is created by creating a genesis file. To
add every device, a pair of private and public key is generated
which will be used as a identifier of a device.
In our proposed system, all resource constrained sensor
nodes are directly connected to and rely on the edge gate-
ways which are often implemented by powerful single board
computers able to work as miner nodes to store, analyze and
aggregate raw data. Miner nodes can run neural networks
to process the raw data received from sensor nodes. With
a predefined time interval, edge nodes process the raw data,
and save this processed information into a new data block
by creating a unique hash. This data block consists of two
parts, header and body. The body apart contains processed
information and header part consists of general characteristics
about the processed information. This includes hash of previ-
ous block, time stamp, raw data definition and the type of data,
which can be further use for combining heterogeneous data at
bigger level for the sake of intelligent systems. To protect
the hash of data block, symmetric cryptography is used. After
encryption, the data blocks are saved on a blockchain cloud
and key is only hold by the end-device. Which will be later
used by a client to decrypt the desired data. Moreover, every
access to the data will be recorded. The proposed architecture
is illustrated in Figure 1, which is composed of four layers.
Global Storage for Encrypted Data
Cloud Services
Web/Mobile Application Servers
Bio-signals analysis
ECG Feature Extraction and Storage in the Blockchain
Fig. 1. Proposed System Architecture
0 0.511.5 2
(a) Fragment of raw data over 2.3 seconds
(b) Extracted cycle template
2 4 6 8 10
(c) Heart rate over 10 seconds
Fig. 2. Results of the data analysis at the edge gateway.
36 35
Fig. 3. Execution time of the different processes (ms).
The raw ECG data collected from a healthy 30 year-old
male person is shown in Figure 2. The data is sent to a smart
Edge-assisted gateway which extracts different ECG features
such as heart rate [5]. We have utilized a Raspberry Pi model
3 as the edge gateway, which in turn runs a node of the
Ethereum Blockchain. In order to test the feasibility of the
proposed model, we accumulate data for 10 seconds and then
analyze it. The data analysis requires around 150ms for the
feature extraction. Then, the results are encrypted and stored
in a distributed storage solution. The metadata is stored in
the blockchain. Figure 3 shows the execution time of the
analysis process (ECG), a data retrieval transaction (TRD),
a transaction confirmation (TCT) and the response time (RT).
In total, the system needs around 300ms to process one batch
of data, which runs every 10 seconds. Therefore, one gateway
could support up to 20 or 30 end-devices with the proposed
Integrating Blockchain with Edge computing opens new
paradigms in privacy-critical and data-sensitive applications.
Our proposed architecture, combining a distributed ledger
with AI at the edge, creates secure database of processed
information which can only be used with the permission of its
owner. By Edge AI we refer to local decision making and data
processing at the edge computing layer. End devices can di-
rectly control all the processing, analyzing and sharing of their
data by updating their policies via ethereum smart contracts.
Implementing AI at edge nodes reduces resource consumption
like bandwidth required to upload data to blockchain cloud as
well as local storage. This data analysis step also increases
privacy by storing only processed information rather than raw
In future work, we will analyze the scalability of the
proposed approach and alternative applications and experiment
with the integration of more complex deep learning algorithms.
We will study the utilization of Ethereum 2 for more scalable
[1] P. Mamoshina et al. Converging blockchain and next-generation artifi-
cial intelligence technologies to decentralize and accelerate biomedical
research and healthcare. Oncotarget, 9(5):5665, 2018.
[2] J. Pe˜
na Queralta et al. Edge-AI in LoRa-based healthcare monitoring:
A case study on fall detection system with LSTM Recurrent Neural
Networks. In 42nd TSP, 2019.
[3] T. K. Mackey et al. ‘fit-for-purpose?’–challenges and opportunities for
applications of blockchain technology in the future of healthcare. BMC
medicine, 17(1):68, 2019.
[4] K. Peterson et al. A blockchain-based approach to health information
exchange networks. In NIST W. Blockchain Healthcare, 2016.
[5] C. Carreiras et al. BioSPPy: Biosignal processing in Python, 2015–.
[Online; accessed Aug. 2019].
... It is made up of a two-tier privacy arrangement with a blockchain section that sends information privately, as well as a principal component analysis module to transform raw data and an anomaly detector using gradient boosting [321]. Additionally, anonymity-critical apps at the network's edge may guarantee that privacy is preserved by utilizing edge machine learning to evaluate information at the edge network and the Ethereum blockchain for maintaining details of users who access result analysis [322]. Furthermore, a framework for Privacy-Preserving Big Data (PPBD) transfer uses graph modeling and extracts subsets of nodes using artificial intelligence, while blockchain-based resources may be communicated with privately and secretly utilizing symmetric-based digital tokens [323]. ...
... Packet forwarding Secure routing [268], GAR [269], trusted routing [270], ENIR [271], CDRL [272], FLEA-RPL [273] Load optimization LB-DRL [274], ECRL [275], BCLB [276], Fuzzy [277], IVEC [278] QoS offering MLBQR [280], MLSMBQS [281], side chaining [282], ATQMB [283], QoS-ledger [284] Network administration User administration CGT [285] Mobility (Generic) MADRL [286], DRL [287], QRM [288], DMM [289] Mobility (Authentication handover) IoTAH [290], deep learning [291], RLAC-FNN [292], AEFO [293], ATLB [294] Mobility (Channel scheduling) BBAIoT [295] Mobility (Offloading) ACDRL [296], SOM [297], DRL-CO [298], SCRDO [299], Edge-cloud CO [300], DCRM [301] Spectrum administration SMS [302], 6GSH [303], CR-IOT [304], DITEN [305], spectrum trading [306], SSA [307] Fault administration FIRP [308], PRLB [309] NAT administration QRM [288], IoMT [310] Energy administration UAGV [311], RM [312], EE [313], pre-caching [314], Block5GIntell [315], VEN [316], DETF [317], DET [318], SDEM [319] Security and privacy Privacy PPSF [321], EAI [322], PPBD [323], FFIDS [324], SWS [325], FDEMATEL [326] Authentication, access control, and encryption ...
Full-text available
Knowledge-Defined Networking (KDN) necessarily consists of a knowledge plane for the generation of knowledge, typically using machine learning techniques, and the dissemination of knowledge, in order to make knowledge-driven intelligent network decisions. In one way, KDN can be recognized as knowledge-driven Software-Defined Networking (SDN), having additional management and knowledge planes. On the other hand, KDN encapsulates all knowledge-/intelligence-/cognition-/machine learning-driven networks, emphasizing knowledge generation (KG) and dissemination for making intelligent network decisions, unlike SDN, which emphasizes logical decoupling of the control plane. Blockchain is a technology created for secure and trustworthy decentralized transaction storage and management using a sequence of immutable and linked transactions. The decision-making trustworthiness of a KDN system is reliant on the trustworthiness of the data, knowledge , and AI model sharing. To this point, a KDN may make use of the capabilities of the blockchain system for trustworthy data, knowledge, and machine learning model sharing, as blockchain transactions prevent repudiation and are immutable, pseudo-anonymous, optionally encrypted, reliable, access-controlled, and untampered, to protect the sensitivity, integrity, and legitimacy of sharing entities. Furthermore, blockchain has been integrated with knowledge-based networks for traffic optimization, resource sharing, network administration, access control, protecting privacy, traffic filtering, anomaly or intrusion detection, network virtualization, massive data analysis, edge and cloud computing, and data center networking. Despite the fact that many academics have employed the concept of blockchain in cognitive networks to achieve various objectives, we can also identify challenges such as high energy consumption, scalability issues, difficulty processing big data, etc. that act as barriers for integrating the two concepts together. Academicians have not yet reviewed blockchain-based network solutions in diverse application categories for diverse knowledge-defined networks in general, which consider knowledge generation and dissemination using various techniques such as machine learning, fuzzy logic, and meta-heuristics. Therefore, this article fills a void in the content of the literature by first reviewing the diverse existing blockchain-based applications in diverse knowledge-based networks, analyzing and comparing the existing works, describing the advantages and difficulties of using blockchain systems in KDN, and, finally, providing propositions based on identified challenges and then presenting prospects for the future.
... Users are becoming more aware of the importance of protecting personal data. Edge AI also finds its use in data encryption by combining it with blockchain technology [38]. ...
... IoT system includes heterogeneous and diverse devices, this will result in a serious issue that needs to be handled. [15] Inflexibility A significant workload is generated by the centralised server's control over activities of communication and processing between all IoT network nodes. The centralized server schedules the workload to manage this workload and avoid peak-load problems. ...
... The other important design issue is the cost of on-chain testing by the smart contract. Testing each model is time-consuming and computationally intensive, so this is done locally by one of the nodes in a random fashion, similar to how decentralized learning solutions achieve model merging [23]. Then the smart contract selects the top few with the highest accuracy rates for further processing, which greatly reduces the gas fees. ...
... Presenting a data trading platform using Ethereum with consensus of proof-of-concept (PoC). No [159] Data Management Security and privacy-assured data management framework for privacy-critical systems. ...
Full-text available
As an important technology to ensure data security, consistency, traceability, etc., blockchain has been increasingly used in Internet of Things (IoT) applications. The integration of blockchain and edge computing (IBEC) can further improve the resource utilization in terms of network, computing, storage, and security. This paper aims to present a survey on the IBEC. In particular, we first give an overview of blockchain and edge computing respectively. We then present a general architecture of an IBEC system. We next study the various applications of the IBEC in IoT. We also discuss the optimizations of the IBEC system and solutions from perspectives of resource management and performance improvement. Finally, we analyze and summarize the existing challenges posed by the IBEC system and the potential solutions in the future.
Digital healthcare systems and the Internet of Medical Things (IoMT) aim to make it simple for consumers to receive high-quality treatment in the convenience of their own homes. So, the universal deployment of home healthcare systems is the goal of IoMT. Millions of lives can be saved by making such systems intelligent and effective at foretelling major diseases early on, while also lightening the load on current healthcare systems like hospitals. Patients and doctors may now obtain real-time data thanks to the development of Internet of Things. Digital healthcare systems now use less money and energy because they can communicate and use efficient sensors. This article presents a number of investigations done to advance and enhance IoMT. The IoMT combines the IoT with medical equipment to increase patient comfort, provide cost-effective healthcare, speed up hospitalizations, and provide even more individualized healthcare.KeywordsIoMTDigital healthIoTHealthcareNetwork architecture
Smart devices contributes unique and improved functionality to the smart healthcare system, a smart device is in possession of a substantial quantity of sensitive patient data. In addition, the many aspects of healthcare systems are now interconnected through the use of the internet, which has resulted in substantial shifts in the manner in which individuals receive healthcare services. Having simple access to healthcare services and applications, on the other hand, has resulted in the emergence of significant risks and vulnerabilities that hinder the performance of a smart healthcare system. In addition, a huge number of heterogeneous devices generate data with varying sizes and formats. This makes it difficult to manage the data in the healthcare repository and protect it from attackers that want to make money off of the data. Therefore, intelligent healthcare systems are vulnerable to a wide variety of security risks and threats, such as assaults based on hardware and software, attacks directed at the system level, and attacks directed at the network, all of which have the potential to put patients’ lives in jeopardy. In this paper, AI, big data analytics and cognitive computing role in healthcare sector is explained. At the end variety of unresolved concerns and research obstacles that are specific to smart healthcare systems.KeywordsSmart healthcare systemsArtificial intelligenceMedical devicesBig data analyticsCognitive computing
With the rapid advances in information and communication technologies, the Internet of Things (IoT) has become large and complex, bearing tremendous amounts of data and running devices in various scenarios. Leveraging Artificial Intelligence (AI) technologies, IoT can achieve superior information extraction, data analytics, and decision-making, which has resulted in the revolutionized Artificial Intelligence of Things (AIoT). AIoT can alleviate the pressure of storage, computation, and communication. Despite the promising features brought by combining AI technologies into IoT infrastructure, AIoT systems still face some serious challenges including inadequate efficiency, violation of security and privacy, lack of trust, and insufficient incentive. Blockchain featured by its distributed consensus and incentive mechanisms can be a promising technology for addressing the challenges in AIoT. AIoT employing blockchain is evolving with expectations of achieving efficient, secure, and trusted network activities. In this paper, we first introduce the background of AIoT and blockchain. Then, we discuss the motivations for employing blockchain with its characteristics in AIoT. Furthermore, we comprehensively review existing solutions on blockchain for AIoT systems from aspects of efficiency, security, privacy, trust, and incentive. Finally, we discuss the challenges and future research directions on blockchain for AIoT.
Artificial Intelligence (AI) at the edge is the utilization of AI in real-world devices. Edge AI refers to the practice of doing AI computations near the users at the network's edge, instead of centralised location like a cloud service provider's data centre. With the latest innovations in AI efficiency, the proliferation of Internet of Things (IoT) devices, and the rise of edge computing, the potential of edge AI has now been unlocked. This study provides a thorough analysis of AI approaches and capabilities as they pertain to edge computing, or Edge AI. Further, a detailed survey of edge computing and its paradigms including transition to Edge AI is presented to explore the background of each variant proposed for implementing Edge Computing. Furthermore, we discussed the Edge AI approach to deploying AI algorithms and models on edge devices, which are typically resource-constrained devices located at the edge of the network. We also presented the technology used in various modern IoT applications, including autonomous vehicles, smart homes, industrial automation, healthcare, and surveillance. Moreover, the discussion of leveraging machine learning algorithms optimized for resource-constrained environments is presented. Finally, important open challenges and potential research directions in the field of edge computing and edge AI have been identified and investigated. We hope that this article will serve as a common goal for a future blueprint that will unite important stakeholders and facilitates to accelerate development in the field of Edge AI.
Conference Paper
Full-text available
Remote healthcare monitoring has exponentially grown over the past decade together with the increasing penetration of Internet of Things (IoT) platforms. IoT-based health systems help to improve the quality of healthcare services through real-time data acquisition and processing. However, traditional IoT architectures have some limitations. For instance, they cannot properly function in areas with poor or unstable Internet. Low power wide area network (LPWAN) technologies, including long-range communication protocols such as LoRa, are a potential candidate to overcome the lacking network infrastructure. Nevertheless, LPWANs have limited transmission bandwidth not suitable for high data rate applications such as fall detection systems or electrocardiography monitoring. Therefore, data processing and compression are required at the edge of the network. We propose a system architecture with integrated artificial intelligence that combines Edge and Fog computing, LPWAN technology, IoT and deep learning algorithms to perform health monitoring tasks. In particular, we demonstrate the feasibility and effectiveness of this architecture via a use case of fall detection using recurrent neural networks. We have implemented a fall detection system from the sensor node and Edge gateway to cloud services and end-user applications. The system uses inertial data as input and achieves an average precision of over 90\% and an average recall over 95\% in fall detection.
Full-text available
Blockchain is a shared distributed digital ledger technology that can better facilitate data management, provenance and security, and has the potential to transform healthcare. Importantly, blockchain represents a data architecture, whose application goes far beyond Bitcoin – the cryptocurrency that relies on blockchain and has popularized the technology. In the health sector, blockchain is being aggressively explored by various stakeholders to optimize business processes, lower costs, improve patient outcomes, enhance compliance, and enable better use of healthcare-related data. However, critical in assessing whether blockchain can fulfill the hype of a technology characterized as ‘revolutionary’ and ‘disruptive’, is the need to ensure that blockchain design elements consider actual healthcare needs from the diverse perspectives of consumers, patients, providers, and regulators. In addition, answering the real needs of healthcare stakeholders, blockchain approaches must also be responsive to the unique challenges faced in healthcare compared to other sectors of the economy. In this sense, ensuring that a health blockchain is ‘fit-for-purpose’ is pivotal. This concept forms the basis for this article, where we share views from a multidisciplinary group of practitioners at the forefront of blockchain conceptualization, development, and deployment.
Full-text available
The increased availability of data and recent advancements in artificial intelligence present the unprecedented opportunities in healthcare and major challenges for the patients, developers, providers and regulators. The novel deep learning and transfer learning techniques are turning any data about the person into medical data transforming simple facial pictures and videos into powerful sources of data for predictive analytics. Presently, the patients do not have control over the access privileges to their medical records and remain unaware of the true value of the data they have. In this paper, we provide an overview of the next-generation artificial intelligence and blockchain technologies and present innovative solutions that may be used to accelerate the biomedical research and enable patients with new tools to control and profit from their personal data as well with the incentives to undergo constant health monitoring. We introduce new concepts to appraise and evaluate personal records, including the combination-, time- and relationship-value of the data. We also present a roadmap for a blockchain-enabled decentralized personal health data ecosystem to enable novel approaches for drug discovery, biomarker development, and preventative healthcare. A secure and transparent distributed personal data marketplace utilizing blockchain and deep learning technologies may be able to resolve the challenges faced by the regulators and return the control over personal data including medical records back to the individuals.
BioSPPy: Biosignal processing in Python
  • C Carreiras
A blockchain-based approach to health information exchange networks
  • peterson
K. Peterson et al. A blockchain-based approach to health information exchange networks. In NIST W. Blockchain Healthcare, 2016.