Jouni Isoaho

Jouni Isoaho

Auto-generated profile, based on publicly available information

Jouni Isoaho

University of Turku

Research Focus & Metrics

Academic Career Length

31Years (1996-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
25 November 2024 in IEEE Access
Verified
Journal Article
Open access
IHIsmayil Hasanov
Ismayil Hasanov
SVSeppo Virtanen
Seppo Virtanen
AHAntti Hakkala
Antti Hakkala
JIJouni Isoaho
Jouni Isoaho
The emergence of Large Language Models (LLMs) is currently creating a major paradigm shift in societies and businesses in the way digital technologies are used. While the disruptive effect is especially observable in the information and communication technology field, there is a clear lack of systematic studies focusing on the application and impact of LLMs in cybersecurity holistically. This article presents an exhaustive systematic literature review of 177 articles published in 2018-2024 on the application of LLMs and the use of Artificial Intelligence (AI) as a defensive measure in cybersecurity. This article contributes an analytical compendium of the recent research on the application of LLMs in offensive and defensive cybersecurity as well as in research on cyberethics, current legal frameworks, and research regarding the use of LLMs for cybersecurity governance. It also contributes a statistical summary of global research trends in the field. Of the reviewed literature, 68%was published in 2023. Nearly 30% of the articles originate from the USA and 11% from China, with other countries currently having significantly lower contributions to recent research. Most attention in recent research has been given to AI as a defensive measure, accounting for 27% of the reviewed literature. It was observed that LLMs have proven highly effective in phishing attack simulations and in managing cybersecurity administrative aspects, including defending against advanced exploits. Furthermore, LLMs show significant potential in the development of security software, further cementing their role as a powerful tool in cybersecurity innovation.

21 August 2022 in Sensors
Verified
Journal Article
Open access
PDF file
AAAkwasi Adu-Kyere
Akwasi Adu-Kyere
ENEthiopia Nigussie
Ethiopia Nigussie
JIJouni Isoaho
Jouni Isoaho
Autonomous “Things” is becoming the future trend as the role, and responsibility of IoT keep diversifying. Its applicability and deployment need to re-stand technological advancement. The versatile security interaction between IoTs in human-to-machine and machine-to-machine must also endure mathematical and computational cryptographic attack intricacies. Quantum cryptography uses the laws of quantum mechanics to generate a secure key by manipulating light properties for secure end-to-end communication. We present a proof-of-principle via a communication architecture model and implementation to simulate these laws of nature. The model relies on the BB84 quantum key distribution (QKD) protocol with two scenarios, without and with the presence of an eavesdropper via the interception-resend attack model from a theoretical, methodological, and practical perspective. The proposed simulation initiates communication over a quantum channel for polarized photon transmission after a pre-agreed configuration over a Classic Channel with parameters. Simulation implementation results confirm that the presence of an eavesdropper is detectable during key generation due to Heisenberg’s uncertainty and no-cloning principles. An eavesdropper has a 0.5 probability of guessing transmission qubit and 0.25 for the polarization state. During simulation re-iterations, a base-mismatch process discarded about 50 percent of the total initial key bits with an Error threshold of 0.11 percent.

12 June 2023 in IEEE Internet of Things Journal
Verified
Journal Article
Open access
QAQazi Mamoon Ashraf
Qazi Mamoon Ashraf
MTMohammad Tahir
Mohammad Tahir
Mohmed Hadi HabaebiMohmed Hadi Habaebi
Mohmed Hadi Habaebi
JIJouni Isoaho
Jouni Isoaho
With the rise of the Internet of Things (IoT), tiny devices capable of computation and data transmission are being deployed across various technological domains. Due to the wide deployment of these devices, manual setup and management are infeasible and inefficient. To address this inefficiency, intelligent procedures must be established to enable autonomy that allows devices and networks to operate efficiently with minimal human intervention. In the traditional client–server paradigm, autonomic computing has been proven effective in minimizing user intervention in computer systems management and will benefit IoT networks. However, IoT networks tend to be heterogeneous, distributed, and resource constrained, mandating the need for new approaches to implement autonomic principles compared to traditional approaches. We begin by introducing the basic principles of autonomic computing and its significance in IoT. We then discuss the self-* paradigm and monitor, analyze, plan, and execute (MAPE) loop from an IoT perspective, followed by recent works in IoT and key enabling technologies for enabling autonomic properties in IoT. Based on the self-* paradigm and MAPE loop analysis from the existing literature, we propose a set of qualitative characteristics for evaluating the autonomy of the IoT network. Finally, we provide a comprehensive list of challenges associated with achieving autonomic IoT and directions for future research.

22 May 2025 in IEEE Access
Verified
Journal Article
Open access
JTJolly Trivedi
Jolly Trivedi
MTMohammad Tahir
Mohammad Tahir
JIJouni Isoaho
Jouni Isoaho
The adoption of IoT devices for Remote Patient Monitoring (RPM) systems has emerged as a new attack surface and has resulted in an increased risk of cyber-attacks. Besides the traditional security mechanism for RPM systems, Artificial Intelligence (AI), has emerged as a crucial technology to address the security issues in IoT-enabled RPM systems. In this regard, this article provides a survey of the security challenges of RPM systems and discusses the role of AI in strengthening the security of RPM systems through advanced threat intelligence capabilities. The survey analyses 110 research papers from leading databases related to AI, the security of RPM systems, anomaly detection, architectural solutions, and existing RPMs. This review of AI-enhanced Threat Intelligence in RPM systems highlights the research gaps, which include the necessity of comprehensive end-to-end architectures for maintaining security and privacy in the RPM systems. The survey reveals traditional RPM systems are vulnerable to data breaches due to their centralized architecture. AI-powered threat intelligence enhances RPM security significantly by identifying anomalies in patient data, enabling continuous monitoring and early detection of threats. Federated Learning allows for decentralized training of AI models, providing both security and privacy. However, challenges like the explainability of AI models persist, which need continued innovation. This survey paper suggests integrating AI Enhanced Threat Detection as a Service (TDaas) that implements Federated Learning (FL) to transform the existing RPM security system and ultimately contribute to a secure and reliable Threat Detection system in Healthcare. This literature review lays out a roadmap for future research in the area of AI-driven threat intelligence security for RPM systems and offers insights for developing resilient healthcare infrastructures.

30 April 2007 in VLSI Design
Verified
Journal Article
Open access
PDF file
ENEthiopia Nigussie
Ethiopia Nigussie
TLTeijo Lehtonen
Teijo Lehtonen
STSampo Tuuna
Sampo Tuuna
JPJuha Plosila
Juha Plosila
JIJouni Isoaho
Jouni Isoaho
High-performance long-range NoC link enables efficient implementation of network-on-chip topologies which inherently require high-performance long-distance point-to-point communication such as torus and fat-tree structures. In addition, the performance of other topologies, such as mesh, can be improved by using high-performance link between few selected remote nodes. We presented novel implementation of high-performance long-range NoC link based on multilevel current-mode signaling and delay-insensitive two-phase 1-of-4 encoding. Current-mode signaling reduces the communication latency of long wires significantly compared to voltage-mode signaling, making it possible to achieve high throughput without pipelining and/or using repeaters. The performance of the proposed multilevel current-mode interconnect is analyzed and compared with two reference voltage mode interconnects. These two reference interconnects are designed using two-phase 1-of-4 encoded voltage-mode signaling, one with pipeline stages and the other using optimal repeater insertion. The proposed multilevel current-mode interconnect achieves higher throughput and lower latency than the two reference interconnects. Its throughput at 8 mm wire length is 1.222 GWord/s which is 1.58 and 1.89 times higher than the pipelined and optimal repeater insertion interconnects, respectively. Furthermore, its power consumption is less than the optimal repeater insertion voltage-mode interconnect, at 10 mm wire length its power consumption is 0.75 mW while the reference repeater insertion interconnect is 1.066 mW. The effect of crosstalk is analyzed using four-bit parallel data transfer with the best-case and worst-case switching patterns and a transmission line model which has both capacitive coupling and inductive coupling.

26 May 2025
Verified
Conference Paper
Closed access
AAAisvarya Adeseye
Aisvarya Adeseye
JIJouni Isoaho
Jouni Isoaho
MTMohammad Tahir
Mohammad Tahir
Prompt engineering has become an important aspect in optimizing the performance of large language models (LLMs) in diverse applications. This research proposes a systematic framework for system and user prompts by utilizing few-shot learning, chain-of-thought reasoning, role play and iterative refinement. The proposed framework was evaluated on open source LLMs, Llama, Gemma, and Phi, running on local machines to underscore their capability to enhance LLMs' outputs for qualitative data analysis for interview transcripts about security and privacy issues of gamification. Utilizing local LLMs eliminates concerns related to data leakage and privacy, making this approach particularly suitable for organizations that have privacy concerns with publicly available LLM solutions like ChatGPT, Gemini, DeepSeek etc. The LLM output demonstrated improved accuracy, consistency, and scalability in addressing security and privacy concerns with gamification. The validation using manual analysis with NVivo indicates less than 5% error margin for frequency analysis.